This page intentionally left blank
Copyright © 2006, New Age International (P) Ltd., Publishers Published by New Age International (P) Ltd., Publishers All rights reserved. No part of this ebook may be reproduced in any form, by photostat, microfilm, xerography, or any other means, or incorporated into any information retrieval system, electronic or mechanical, without the written permission of the publisher. All inquiries should be emailed to
[email protected]
ISBN (13) : 978-81-224-2486-7
PUBLISHING FOR ONE WORLD
NEW AGE INTERNATIONAL (P) LIMITED, PUBLISHERS 4835/24, Ansari Road, Daryaganj, New Delhi - 110002 Visit us at www.newagepublishers.com
To My Parents Nabanita Das & Barun Das
This page intentionally left blank
Preface I have written the textbook keeping in mind the problems which I have faced in my earlier days of studies as a student and for the same reason I have tried to place this book for the students to have better understanding of the subject and I believe this book will serve them as a book ‘‘made easy’’. This book is intended to be used as an introductory text for the study of line communication system. In our present age of advanced telecommunication, the terms switching, sampling, BPS, broadband, are not foreign words. The present book is written for understanding the concept of computer communication, simplex/duplex communication, detailed knowledge of telephony upto the present age key switching i.e., ISDN. This book can be served as the textbook for undergraduate courses (BTech/B.E./B.Sc.) of information technology, electronics and communication engineering. An enormous research and developments are undertaken under various industries in the fast growing field of telecommunication switching. The present book provides best knowledge in depth on line communication system. Though the book can be considered as a textbook for any university, the content is designed specially for the subject ‘‘Line Communication System’’ (ECE dept., 5th semester) introduced by West Bengal University of Technology. Moreover, the approach of presentation is such that, students can easily understand the concept and they can memorize the same without much effort. I am really greatful to some of the persons for their kind co-operation without whom I believe I’ll not be able to complete this book. Firstly I want to mention my family. My brother (Anibarn), Sister (Adwitia), parents (Barun & Nabanita) inspired me a lot to go ahead with my goal. Two of my beloved students namely Pabitra Mukhopadhaya and Koyel Kujan Kundu helped me a lot in collecting data and other information while I am fortunate enough to be gifted with the blessings of the two great personalities Dr. Arun Kr. Ghosh and Dr. Shankar Sen. I am really inspired by them to go ahead with my mission. Apurba Das e-mail :
[email protected]
This page intentionally left blank
Contents Preface 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
... Introduction ... Advancement of Telecommunication Switching - a brief history ... Computer Communication System ... Automatic Switching System ... Crossbar Switching System ... Electronic Switching System (Stored Program Control) ... Traffic Engineering ... Time Division Switching ... Frequency Division Switching ... Transmission Line ... Telephone Network ... Routing in Telephony ... Wave Propagation ... ISDN ... Appendix I : Questions for Individual Chapters ... Appendix II : Sampling Theorem ... Appendix III : Previous Year Question Paper of WBU of Tech with Hint of Answers according to the instruction of head Examiner. ... Bibliography ...
(vii) 1 8 11 17 29 32 47 53 82 85 99 112 124 136 154 159
162 168
This page intentionally left blank
1 Introduction 1. Concept 2. Point-to-Point Communication 3. Basic Telephone Communication 4. Switching System ➢ Concept Telecommunication networks carry information from one place to another situated at a certain distance apart. The word ‘tele’ means distant and ‘communication’ is the process of exchanging feelings and ideas. In telephone communication, the person who originates a call is referred to as the calling subscriber and the person for whom the call is originated is known as the called subscriber or called-for subscriber. In some cases like computer communication and to some extent in telephone communication, the communicating bodies or entities are also known as source (one who transmits a signal) and destination (one who receives a signal). The main idea behind modern telecommunication is to allow anybody in any part of the world to be able to communicate with anybody in any part of the world. The first technological development in the field of telecommunication was the transmission of telegraphic signals over wire. In March 1876, Alexander Graham Bell invented telephone and demonstrated it. It was basically a point-to-point connection. His discovery laid the foundation for telephone communication. A communication system can be classified into three categories as: 1. Simplex 2. Half Duplex 3. Full Duplex Simplex communication is a unidirectional communication system i.e., communication is possible in one direction only. Typically, the transmitter (the one talking) sends a signal and it's up to the other receiving device (the listener) to figure out what was sent. This type of communication is most efficient when there is a lot of data flow in one direction, and no traffic is needed in the other direction. Broadcast systems like the T.V and radio signals, fire alarm systems etc. are good examples of Simplex communication system. 1
2
LINE COMMUNICATION SYSTEM
In Half Duplex communication system, bidirectional communication is possible, but only in one direction at a time. That means one can either transmit or receive a signal at a particular instant of time in this system of communication. One cannot transmit and receive a signal simultaneously. The walkytalky used in defense and by police is a good example of Half Duplex communication system. In Full Duplex communication system simultaneous two-way communication is achieved. Unlike half duplex communication system, one can both transmit and receive a signal simultaneously. Telephone conversation is an appropriate example of Full Duplex communication system.
Fig. 1.1
➢ POINT-TO-POINT COMMUNICATION
Fig. 1.2
The above figure shows the point-to-point network connection for six subscribers. Here each subscriber is connected directly with the other five subscribers. For six subscribers, the number of links required is equal to 15. Let us assume that there are n number of subscribers. In order to connect the first subscriber to all other subscribers, we need n-1 links. With this, the second subscriber is already connected to the first one. So, we need another n-2 links to connect the second subscriber to other subscribers. Similarly for the third subscriber we need n-3 links and so on. If the total number of links required in the entire network be L, then we can find it as under.
3
INTRODUCTION
L = (n – 1) + (n – 2) + (n – 3) +………. + 2 + 1 + 0 = n(n – 1)/2 Networks with point-to-point connections among all the subscribers or nodes are known as fully connected networks. For a fully connected network, the total number of links required increases with increase in number of subscribers. Even for a moderate value of n, we find that L is quite high and is not possible to implement on a large scale practically. For example for n =10 we have L = 45; for n = 50 we have L =1225; for n =100 we have L = 4950 and for n = 1000 we have L = 499,500 !!! As the number of subscribers increased beyond a certain limit, the need for a so-called switching system or switching office or switching exchange was realized. In this new system, the subscribers are directly connected to the exchange and not to one another as in the case of point-to-point communication. All the calls are now completed and put through by the switching system which establishes a connection between the communicating parties. Now since each subscriber is directly connected to the switching system, the total number of links required for n subscribers is n.
Fig. 1.3
This switching system now needs a proper signaling to establish or disconnect a call. Also it should be able to detect whether the called subscriber is busy and convey the same message to the calling subscriber. All the operations which are performed by the switching exchange while establishing or disconnecting a call are done by control functions. Earlier the switching systems were manual i.e. operator oriented. It was a slow process. Moreover as the number of subscribers increased, more man-power was required and the system became more complex. As a result of the limitations of manual switching, automatic switching systems came into the picture. Fig. 1.4 An operator based manual telephone switching system.
4
LINE COMMUNICATION SYSTEM
Switching systems in general can be classified as follows: Switching System
Automatic
Manual
Electromechanical
Electronic (Stored Program Control)
Strowger or step-by-step switching
Crossbar
Space division switching
Time division switching
Digital
Space switch
Time switch
Analog
Combination switch
➢ BASIC TELEPHONE COMMUNICATION (A) Simplex
Fig. 1.5
The microphone normally used in telephone is made of carbon. It has an acceptable fidelity with large electrical output. There are carbon granules in a metal box. One of the faces of the metal box is used as diaphragm. As the voice signal impinges the diaphragm, the concentration of the carbon granules in the box changes. This leads to change in resistance and accordingly the current through the MIC also changes and we get an audio or speech signal equivalent to the voice signal. If a voltage is applied, current through the circuit varies according to the vibration, i.e. the function of MIC regarding the audio signal can be designated as amplitude modulation (AM).
5
INTRODUCTION
Instantaneous resistance of the circuit = ri = r0 – r sin ωt Where, r0 = quiescent resistance of MIC when there is no speech signal r = maximum variation in resistance offered by the carbon granules, r < r0 here the negative sign indicates, resistance decreases when the carbon granules compressed. Ignoring the external impedances, Current I = V/ (r0 – r sin ωt) = Io (1 – m sin ωt)– 1 Therefore modulation index m = r/r0 < 1 and Io = V/r0 = quiescent current Therefore, I ≈ I0 (1 + m sin ωt) ⇒ AM
Fig. 1.6 Sound to electrical energy and vice versa.
In the loudspeaker there is an electromagnet and a diaphragm is placed in front of the electromagnet. As the current varies, the flux linked with the electromagnet changes and the diaphragm vibrates accordingly and we get the required sound output. In simplex communication, the inductor behaves as a short circuit for dc supply and allows the current to flow in both the MIC and Loudspeaker circuits. The voice signal being an ac signal is blocked by the inductor and hence it goes to the loudspeaker circuit. As that off MIC side, in EAR side, AM also is being done. Here the instantaneous flux liking the poles of magnet and diaphragm, Φi = Φ0 + Φ sin ωt using alike notations. (B) Half Duplex
Fig. 1.7
6
LINE COMMUNICATION SYSTEM
This system works in the same way as the simplex does. The difference here is that both subscribers 1 & 2 can communicate to each other, which was not possible in simplex system. But this system has a drawback. This circuit suffers from the problem of side tone i.e., speech can be heard on both sides simultaneously which is undesirable. So, the side tone needs to be coupled. The modified circuit diagram of the half duplex communication is given below.
Fig. 1.8
In this figure only one section of the entire circuit is shown. The other portion is the exact replica of this one. There are two inductors P & Q of same value. Let the impedance viewed from this section into the other be Zin. An impedance of Zb=Zin is connected in the circuit as shown in the figure. When a subscriber speaks into M1, equal currents flow through inductors P and Q (since P = Q & Zin = Zb) but in opposite direction. So, no voltage is induced in the secondary coil and hence E2 has no output. When a subscriber speaks into M2 equal currents flow through inductors P and Q, but now in same direction. So, a net voltage is induced in the secondary of the coil and the speech is heard at E2. The same is valid for the other section as well. ➢ SWITCHING SYSTEM The function of a switching system is to establish a electrical connection between the inlets (inputs) and outlets (outputs) of an exchange. This is done with the help of switching matrix or the switching network. A switching network is a part of a switching system. The figure below shows a switching network with N input lines and M output lines. When N = M, the network is called a symmetric network. The inlets and the outlets can be connected to local subscribers or to trunks. Trunks are the links that run between two or more switching systems. If all the inlets and outlets of the network are connected to the subscriber lines, the network is then called a folded network. In a folded network with N subscribers, there can be at most N/2 number of simultaneous calls. If the system design permits N/2 simultaneous calls, the network is said to be nonblocking network. But the probability that all the subscribers will make a call or speak to each other simultaneous is very low, unless everyone goes crazy! So, it is economical to design a network which will meet the estimated maximum average simultaneous traffic, known as busy hour traffic. In this case,
7
INTRODUCTION
Fig. 1.9 Symmetric Network.
it may occasionally happen (e.g., in New Year’s Eve or Christmas Eve) that a subscriber requesting a connection is denied a connection due to availability of free path in the network. In such an event, the subscriber is said to be blocked and the network is termed as a blocking network.
Fig. 1.10 Folded Network
If in a switching network, all the inlets and outlets are used for inter-system or inter-exchange transmission, then the switching exchange is termed as a non-folded network or a transit network. A transit network doesn't support local subscribers. A switching system provides three different forms of signaling: 1. Subscriber Loop Signaling 2. Inter-exchange Signaling 3. Intra-exchange Signaling
Fig. 1.11 Non-folded Network.
2 Advancement of Telecommunication Switching—A Brief History 1. Manual systems 2. Electromechanical systems 3. Digital, computer-controlled systems
➢ MANUAL SYSTEMS In the infancy of telephony, telephone exchanges were built up with manually operated switching equipment. The first manual exchange was installed in New Haven, USA, in 1878. The operators received calls and switched them manually to the called subscriber (the operator set up a "circuit" between two subscribers, hence the term circuit switching). When the call was complete, the operator released the connection. We might say that the operator was the "control system" of that era. ➢ ELECTROMECHANICAL SYSTEMS In the years that followed, the manual exchanges were replaced by automatic electromechanical switching systems. Although these systems required more maintenance, their overall impact was positive since the number of operators could be reduced significantly (resulting in cut labour costs). The systems provided increased traffic capacity at a lower cost, preparing the way for a continued rapid expansion of the telecommunications network. The new systems also made it possible to route traffic more efficiently through the transmission network, reducing the need for cable capacity. Almon B. Strowger, Kansas City, USA, is regarded as the father of automatic switching. In 1889, he applied for a patent for an automatic telephone exchange. Since then, Strowger's name has been associated with the step-by-step selector (controlled directly from the dial of the telephone set) which was part of his idea. The concept of Strowger telephony will be described in the corresponding chapter. Afterwards, developments followed in the direction of register-controlled systems, in which the number information does not directly control selector set-up (as in the step-by-step selector), but is first received and analyzed in a register. One of the functions of the register is to select alternative switching paths, meaning that the transmission network can be used more efficiently. 8
ADVANCEMENT OF TELECOMMUNICATION SWITCHING—A BRIEF HISTORY
9
Examples of register-controlled systems are: • the 500-line selector (1923); and • Crossbar systems (1937).
➢ DIGITAL, COMPUTER-CONTROLLED SYSTEMS Telephone Exchanges As time went by, transmission as well as switching systems continued to develop, contributing to the total economy. A technique for saving expensive connections was introduced into the long-distance networks: frequency division multiplexing (FDM). This technique was developed around 1910, but was not implemented until 1950 when about 1,000 channels were transferred on the same cable (the coaxial cable). Digital multiplexing (based on PCM), which was introduced around 1970, also made transmission networks less expensive while at the same time improving transmission quality. Costs were further reduced when digital group switches (the actual switching equipment in the telephone exchange) were combined with digital transmission systems, eliminating the need for many relatively expensive analogdigital (A/D) converters. It now became necessary to computerize the control of the exchanges, and before long, not only the group switches but the entire exchanges were digital. The first computercontrolled exchange was put into service in 1960 in the US; the first digital exchange in Europe was opened for traffic in 1968 (Tumba, Sweden). Today’s telephone exchanges use circuit switching technology, just like their predecessors at the end of the 19th century.
Special Nodes for Data Communication The strong growth in data traffic and in the number of users of data communication has resulted in the development of separate data networks and data switches. In many cases, these can meet users' increasingly stringent quality requirements and the need for higher transmission rates in a better and less expensive way. Packet mode and frame relay, for example, provide efficient network utilization, enable packets to be retransmitted when errors occur on a link (applies to packet mode only), and allow for sorting, routing, and buffering.
Nodes for N-ISDN Developments for providing service-integrated networks (for voice, video, and data services) require both public and private N-ISDN nodes (N stands for narrowband). In principle, a complete ISDN node can be seen as a combination of today's telephone exchanges and packet data switches (circuit switching and packet mode), with an important sorting function for subscriber traffic.
Nodes for B-ISDN The technology called ATM, which applies the cell-switching technique and which forms the basis of B-ISDN (B stands for broadband), is not yet completely standardized. See also ISDN chapter, where ATM and B-ISDN are described.
10
LINE COMMUNICATION SYSTEM
Optical Switches It is primarily the switching equipment that limits the bandwidth of a connection. Today, we can make use of very high bit rates, up to tens of billions of bits per second (tens of Gbit/s) in optical transmission systems. However, in switching equipment, we must change over to electrical signals and considerably lower bit rates. The next step is to use optical switching with electronic switch control. And in time, we will most assuredly have fully optical switching systems. Indeed, in view of the intensive research and development that is being carried out in this area, it should not be long before the first optical space switches are commercially available. Figure below describes technical developments in the field of switching (public switching only).
Fig. 2.1
3 Computer Communication System 1. Concept 2. Circuit Switching 3. Store and Forward Switching (a) Message Switching
(b) Packet Switching
➢ Concept The five desired features for a reliable computer communication are as follows: 1. No excessive delay during communication. 2. It should be efficient enough i.e., no substantial part of the network should remain idle for a long time. 3. Data arriving at the host computer (receiving end) must be capable of rapid access. 4. The cost of establishing, maintaining and operating the network should be minimal. 5. If any transmission calls for a response, the response time should be very small.
Local Computer
Remote Terminal
Fig. 3.1
11
Switching Computer
12
LINE COMMUNICATION SYSTEM
A computer network has two parts: 1. Communication sub-network 2. User sub-network A small portion of a computer network is shown in the above figure. The part of the network which comprises of the switching computers and the high speed links is termed as the communication sub-network and the remaining is called the user sub-network. Switching computers are generally connected to local computers and the local computers are connected with the remote terminals. However, in some cases remote terminals are also connected directly with the switching computers. These remote terminals which are directly connected with the switching computers are known as Orphans. The links that connect the switching computers with the local computers and remote terminals are generally low speed links, but can be upgraded to high speed links if there is a requirement. The switching computers are involved in bulk data transfer and the data are of various forms like text, multimedia etc. Hence high speed and high capacity mediums or links are required for transmission of data. The switching computers are neither the source nor destination. They just make or break connections and transfer data from one place to another. The local computers are the sources and destinations. The remote terminals can not access the switching computers directly, but via the local computers. Here, the local computers can be considered to be acting as servers. However, the Orphans can access the switching computers directly as they are directly connected. Computer communication system can be classified as shown below. Computer Communication System
Forward Switching (Circuit Switching)
Store and Forward Switching
Message switching
➢ CIRCUIT SWITCHING
Fig. 3.2
Packet Switching
COMPUTER COMMUNICATION SYSTEM
13
The following are the three basic steps of circuit switching: 1. Circuit establishment 2. Communication 3. Circuit break The choice of the communication link or path depends on the following: 1. Shortest path 2. Availability of free path or a path with low traffic at that instant 3. The selected path should not be faulty In circuit switching, there exists a dedicated path between the source and the destination and no third party can share it with them. Once the communication path is established, data transfer can take place till the connection is broken. The path is chosen as per the criterion mentioned above. It is the property of circuit switching that it needs to set up an end-to-end path before any data can be sent. For this reason a long set-up time is required because a call request from the source must propagate all the way to the destination and be acknowledged. Sometimes this long set-up time becomes undesirable. On the other hand, once the path is reserved, the communication is secured, reliable and free from congestion. After setting up of the path only delay associated with data transfer in circuit switching is the propagation time of data from source to destination. In the above figure, if S (source) wants to send data to D (destination), then a path is reserved between S and D depending on the status of various paths at that instant. Telephone communication is an example of circuit switching. But incase of a node failure or link failure the communication comes to a complete halt. If a node failure or link failure occurs after a connection has been established between the source and destination, then the connection is lost and data transfer stops. Also, the resource utilization is poor in circuit switching. If a link capacity is under used by the source and destination, one cannot feed the rest of the capacity to other users of the network. To overcome these deficiencies of circuit switching, a concept of store and forward switching is introduced.
➢ STORE AND FORWARD SWITCHING In store and forward switching, no physical link is established or reserved between the transmitter and the receiver in advance. The main components of a store and forward switching are the buffers that store data and the relays that send data from one computer to another.
Message Switching The Fig. 3.3 shows a part of a computer network working on store and forward switching. S is the source and D is the destination. There are two paths for data to travel from source to destination. They are S → 1 → 2 → D and S → 1 → 3 → 2 → D. Here data are not continuously sent over the links as in the case of circuit switching. Rather data are sent in the form of blocks called messages. The block of data to be sent is first stored in router or switching computers and forwarded one hop at a time. For each hop, the free path at that instant is chosen for the data transfer. Each block is received in its entirety, checked for errors and then retransmitted. The message blocks are stored in the buffers provided at the intermediate nodes while they are transmitted from source to destination. So, in this way we can avoid the problems of node failure and link failure. If a node or link fails during transmission, then data will be routed via other routes. Also if a data is lost or has errors, it can be re-sent from the immediate or nearest intermediate node and not from the source. To make the system reliable, after receiving a message, the receiver transmits an acknowledgement signal back to the transmitter. The transmitting router then erases the content from its buffer if the received message is okay, otherwise it re-transmits the message.
14
LINE COMMUNICATION SYSTEM
Fig. 3.3
But the size of the message block is a matter of concern. As there is no upper limit of the size of message, and the messages are of different sizes, there is no justification for using a huge buffer. Also, if the number of links is T and the delay time for each node (neglecting other sync. timings) is Rm, then the total delay is T × Rm. There is also no provision for interruption during a long message relay. Another delay associated with store and forward switching is to send the transmitter, a receipt signal or acknowledgement signal. This also takes some time. The acknowledge signaling system is not a mandatory, but it is an optimization of system performance and to make it more reliable. Due to the various delays associated with message switching, the grade of service is poorer than circuit switching.
Packet Switching Packet switching is actually a modified version of message switching. The demerits of message switching have been eliminated to some extent, if not completely, in packet switching. In this switching technique, data is first fragmented into various small blocks of a fixed size, known as packets. Packetizing of data is done at the source terminal. The principle of transmission of data is same as that in message switching. The difference here is that, the whole message is not transmitted at a time, but in small bursts which are in the form of packets. Generally all the packets are of same size. The size of the packets is often determined by the type of network and its design. The data or the information is sandwiched in between a header and a trailer during formation of packets. Packet headers and trailers are of great importance in packet switching. As the packets are transmitted at different times, they may follow different paths depending on the path status at that instant. It may so happen that the second packet has already been received at the receiver while the first packet is in the transit. So, the packets arriving at the final destination are random or out of order though they were transmitted in order from the source. Hence these packets are to be arranged in proper order at the destination. This is ensured by the header and the trailer.
Fig. 3.4 Schematic diagram of a frame
COMPUTER COMMUNICATION SYSTEM
15
Actually, a packet is converted into a frame with the help of headers and trailers and the data transmission takes place in the form of frames. During the formation of frames, they are numbered so that they can be arranged in proper order at the receiver. The header also contains the address and the control information for the frame. Each frame contains the destination address as they are independent of each other. At the destination the packets are extracted from the frames and are checked for errors. This task is accomplished by the trailer which contains the error detection information (checksum). After error correction the packets are assembled as per their numbers in order to get the full data. There are some other control and synchronizing bits that are stuffed in the header and the trailer of the packets. These are called overhead bits. Some of the overhead bits and their purpose are given below: 1. Destination address: to tell each of the switching routers that how far the frame is from the destination and to decide in which direction the frame should be transmitted. 2. Source address: this intended for the destination. It helps the destination to send back an acknowledgement signal to the source or the previous transmitter. As it has been said earlier that sending of acknowledgement signal in the data link layer is just an optimization and never a requirement. 3. User identification: this field is optional. It is generally used for secured or confidential communication. 4. Synchronizing bits: this is required for proper synchronizing of the data, header and the trailer. In this way some of the demerits of message switching are avoided here but it also has some inherent problems. By setting an upper limit on the size of packets, the problem for variable size buffers has been avoided. Also, the network is free from congestion created by long messages and as a result the delay time is also less. So, its grade of service is in between that of message switching and circuit switching. But in order to avoid the problems of message switching, one has to put in extra hardware and software at the source and destination. The source needs packetizing and the destination needs depacketizing hardware and software.
➢ COMPARISON OF DELAY TIME BETWEEN MESSAGE SWITCHING AND PACKET SWITCHING
Message Switching Total delay = Tm = k.(B + b) . Tb = k . (B + b) . (1/fb) = k.B/fb ; (b < < B) where k = number of links B = number of information bits required b = number of overhead bits required Tb = bit duration Fb = bit frequency = 1/Tb
16
LINE COMMUNICATION SYSTEM
Packet Switching Total delay = Tp = k.{(B/P) + b}.(1/fb) + (p – 1).{(B/P) + b}.(1/fb) where P = number of packets. Taking derivative w.r.t. number of packets (P) and setting the result to zero, we get the condition for minimum delay. It is found that the delay is minimum when P = (B/b).(k – 1) Therefore, Tp = (b/fb) B/b + k + 1 If B/b >> k then, Tp : (b/fb). ( B/b )2 = B/fb Therefore, Tp min/Tm = (B/fb)/(k.B/fb) = 1/k If k > 1, then Tp min < Tm
4 Automatic Switching System 1. Advantages of Automatic Switching over Manual Switching 2. Concept of Dialing 3. Strowger Swtiching Components 4. Selector Hunter 5. Line Finder 6. Flow-Chart for Strowger System 7. Differences Between Line Finder and Selector Hunter 8. Design Parameters
➢ ADVANTAGES OF AUTOMATIC SWITCHING OVER MANUAL SWITCHING 1. 2. 3. 4. 5. 6.
With the use of automatic switching a greater privacy is acquired. Automatic switching systems are more reliable than manual switching systems. Time required for making and breaking of connections is less in automatic switching. Time required for making and breaking of connections is independent of load. No operators are involved and hence reduction in man-power. In manual switching due to involvement of operators, errors in connection were unavoidable. In automatic switching there is no scope for such errors.
➢ CONCEPT OF DIALING Generally, dialing of telephone numbers is done in two ways: 1. Pulse or decadal dialing 2. Tone or DTMF dialing The following is the dialing waveform for the telephone number 24 in pulse mode:
17
18
LINE COMMUNICATION SYSTEM
Fig. 4.1
The duty cycle of a pulse is Ton/(Ton + Toff) = 33% The bit duration is set such that 10 pulses can occur in 1 second duration. Therefore, the time period of each pulse is 0.1 second. In pulse mode dialing, a digit represents as many pulses, e.g., one pulse is generated for 1, two for 2 and so on. For 0, ten pulses are generated. In this way the switch recognizes the dialed number. The inter-digit gap is kept for distinguishing between two dialed digits. Its interval is generally 400-500 msec. The following are the few signaling tones for automatic exchange:
Dial-tone
400 Hz continuous
Fig. 4.2
Fig. 4.3
Fig. 4.4
19
AUTOMATIC SWITCHING SYSTEM
Fig. 4.5
➢ STROWGER SWITCHING COMPONENTS
Fig. 4.6
When a call is made, the dialed numbers are used by various sections or parts of the strowger switch for establishment of desired connection. The three main stages of strowger switch are shown in the above figure. Depending on the capacity of a switch and other factors, these stages are designed to suit the requirement. The line finder or selector hunter finds a free line in the switch for connection. Group selector(s) and final selector are used for selecting the dialed number. The selectors are of two types: 1. Uniselector: In uniselector there is a stack of ten contacts and a wiper is used to make a connection out of these contacts. Here the wiper is capable of moving in one direction only i.e., either upwards or downwards. 2. Two motion selector: Here the contacts are arranged in an arch fashion in the form of rows and columns. Generally, there are ten rows and ten columns. Here the wiper is capable of two dimensional movements. First, the desired row is selected vertically and then the wiper rotates in the horizontal direction in that row to select the required contact.
20
LINE COMMUNICATION SYSTEM
Fig. 4.7
Fig. 4.8 Sketch of a uniselector
The uniselector has a rotating centre wheel with arms extending from it. An electro-magnet pulls the armature (top left) downwards. The pawl (centre), which is held against the ratchet wheel operates the ratchet when the armature restores. As the wipers are fixed to the ratchet wheel they therefore rotate one step. When the call clears - the wipers are driven to the home position.
21
AUTOMATIC SWITCHING SYSTEM
Fig. 4.9 Two motion selector.
The following are the steps by which the selectors move to the required contact: Step 1: A digit is dialed. Step 2: The pulse generated is attached to the power of the electro-magnet. Step 3: When the electro-magnet energizes, it attracts the pawl, resulting the pawl to connect to the next tooth of the ratchet wheel. Step 4: As we are using the detent for supporting purpose, the ratchet wheel will not be able to move any further. Step 5: When the electromagnet de-energizes, the pawl will move to previous position forcing the ratchet wheel upward. Step 6: It will result in movement of wiper by only one contact. Step 7: If the electromagnet gets a constant high voltage for a long time (400 ms, which is the inter-digit gap), then the control action will transfer the control to the next stage and it will be reset.
22
LINE COMMUNICATION SYSTEM
➢ SELECTOR HUNTER
Fig. 4.10
The steps followed for connecting a subscriber to a selector hunter are as follows: The calling subscriber lifts his handset
Interrupt mechanism in his selector hunter is activated
Wiper starts scanning the first selector stages until it gets a free group selector
The free group selector is sensed
The interrupter is disabled
The corresponding first selector is marked 'Busy'
23
AUTOMATIC SWITCHING SYSTEM
The first selector sends a dial-tone to the corresponding subscriber via a selector hunter stage. Actually the selector hunter stage provides only the electrical path
The first selector is now ready to receive the dialing pulses
➢ LINE FINDER
Fig. 4.11
The steps involved in selecting a line finder are as follows: Normally the wiper of the allotter switch is connected to one of the free line finders or free first selector The calling subscriber lifts his receiver The start signal is passed through the relays to line receiver Line finder commences to find free line As soon as the calling line is found, the allotter switch will jump to the next free line finder or first selector
The line is now ready to receive the next calling subscriber
24
LINE COMMUNICATION SYSTEM
➢ FLOW-CHART FOR STROWGER SYSTEM
➢ DIFFERENCES BETWEEN LINE FINDER AND SELECTOR HUNTER 1. Line finder finds a free line before lifting of handset and selector hunter finds a free line after lifting of handset. 2. Line finder arrangement is more complex. 3. Delay is less in case of line finder as compared to that in selector hunter. 4. Line relays, Start circuit and Allotter switch concepts are introduced in line finder arrangement.
➢ DESIGN PARAMETERS A switching network may be realized using several stages or switching elements. As the number of stage increases, the delay also increases. Every switching system is designed to support a certain maximum number of calls at a time. This particular number is known as switching capacity.
25
AUTOMATIC SWITCHING SYSTEM
There are ten switching parameters: 1. Number of subscriber lines, N 2. Total number of switching elements, S. 3. Cost of switching system ,C C = Cch + Cc + S × Cs where Cc = cost of common control system (for Strowger switching system Cc is zero) Cs = cost per switching element Cch = cost of common hardware (this is a small fraction of total hardware cost for Strowger Switch) Therefore for Strowger Switching system C = S × Cs 4. Switching Capacity (SC) 5. Traffic Handling Capability (THC or TC) TC =
(Switching capacity) (Theoretical value of total load)
6. Equipment Utilization Factor (EUF) EUF =
No. of switching elements in operation when total switching capacity is utilized Total no. of switching elements in the system
7. Number of switching stages (K) 8. Average switching time per stage, (Tst) 9. Call set-up time, Ts = T0 + K . Tst Where T0 is the quotient time required to set-up a call 10. Cost Capacity Index (CCI) =
(Switching capacity) (Cost per subscriber line)
● 10,000 LINE BLOCKING EXCHANGE For 10,000 line blocking the caller id number or telephone number must ranges from 0000 to 9999. Therefore, the digits to be dialed must be 4. The steps involved in this context are shown below with an example of called for party of telephone number 2369. Getting Dial Tone The caller picks up the telephone handset and this puts a loop condition on the line, which energises a couple of relays on the extension line circuit. These relays are part of a Linefinder and this allocates the Subscribers Uniselector, which operates and finds a free 1st Group Selector. A Uniselector is a mechanical device which rotates a set of springs around a bank of 25 or 50 contacts. Once a free 1st Selector is seized then dial tone is returned to the caller.
26
LINE COMMUNICATION SYSTEM
First Digit is Dialed On receipt of Dial Tone the caller dials the first digit (2 in this case). The 1st Selector steps up 2 levels vertically and then rotates freely horizontally until a free 2nd Group Selector is found. Selectors have a set of springs connected to a shaft which can be stepped vertically and then horizontally. Second Digit is Dialed By now the caller is dialing the second digit (3) and the 2nd selector steps vertically 3 levels and then rotates freely horizontally to find a free Final Selector. Last two Digits are Dialed The caller dials the third digit (6) and the Final Selector steps vertically 4 levels and then steps horizontally according to the forth digit dialed (9). Once the Final Selector has stopped on the required outlet, the outlet is tested for three conditions, these being: busy, free or unobtainable. Ringing the Called Extension If free, the Final Selector connects ringing current to the called extensions line and gives the caller ringing tone. Should the extension line be busy then busy tone is returned to the caller and if unobtainable or in line lockout condition (the extension has left their handset off the hook and the linefinder has parked the extension line) then NU tone. The tones are fed from a ringing machine known as a Dynomotor, or in the case of very small exchanges, a vibrating relay. The Called Extension Answers When the called extension answers the Final Selector removes ringing current and ring tone and connects the speak path via a transmission bridge. The two extensions are now free to talk. The Uniselector and all the selectors are held whilst the call is in progress. Clear Down When the calling party hangs up all the selectors restore to normal. They are automatically driven clockwise to the end of the bank of contacts, the shaft then drops down and finally rotates anticlockwise, under the bank, to the parked position. The selectors in the picture are shown in the parked position.
27
AUTOMATIC SWITCHING SYSTEM
Fig. 4.12
Fig. 4.13
Problem. In a pulse dialing arrangement, pulse in-time is 20 µs. Find the total time required to dial the number 4021 if the duty cycle (off) is 64% and inter-digit gap time is 200 µs. Solution. If the time period of one pulse is T and duty cycle (off) = 64% Then, off time = 0.64 T or, on time = T – 0.64 T = 0.36 T = 20 µs or, T = 20/0.36 µs = 55.56 µs Number of pulses to be counted for digit 4 ⇒4
28
LINE COMMUNICATION SYSTEM
Number of pulses to be counted for digit 0 ⇒ 10 Number of pulses to be counted for digit 2 ⇒2 Number of pulses to be counted for digit 1 ⇒1 Therefore, total number of pulses = (4 + 10 + 2 + 1) = 17 And, total time required for entire time period of pulses = 17 × T = 17×55.56 µs = 944.44 µs As this is a 4 digit number, the inter-digit gap will be counted for (4 – 1) =3 times Time required for 1 inter-digit gap = 200 µs Time required for 3 inter-digit gap = 3×200 µs = 600 µs Therefore the overall time required for dialing = (600 + 944.44) µs = 1544.44 µs.
5 Crossbar Switching 1. Concept 2. Principle of Operation ➢
Concept
Telephone communication follows the circuit switching arrangement, i.e. it follows the three steps of communication circuit making → communication → circuit breakin. We know that telephone exchanges use a wide variety of electromechanical automatic switching gears as Strowger Switching arrangement. At the present time, of the electromechanical systems are in use, Crossbar Switching is the most common one. Actually this can be treated as electromagnetic switching system. As the name implies, it uses vertical and horizontal arrangement of electromagnets. ➢ PRINCIPLE OF OPERATION
As shown in the Fig. 5.1, in a crossbar switch an array of vertical and horizontal wires are connected (solid lines) to the separated contact points of the switches. Horizontal and vertical bars (dotted lines) are connected to these contact points. The bars are connected to the electromagnets.
Fig. 5.1 Crossbar switching principle.
29
30
LINE COMMUNICATION SYSTEM
When a vertical bar (i.e. calling subscriber) is energized to face its corresponding bar (i.e. called for subscriber), an excitation of the magnet causes a rotation of the bar. This much motion of the contact unit is not sufficient to cause a complete contact. If now an electromagnet connected to the horizontal bar along with the electromagnet connected to a vertical bar energizes simultaneously, then only the contact is completed. There is an arrangement of mechanical latching which latches (holds) the contact even after one bar (say horizontal) de-energizes. If and only if the other bar de-energizes, the contact is broken up. In the Fig. 5.1 a, b, c and d are calling subscribers and A, B, C and D are treated as called for subscriber. But actually ‘a’ is not different from ‘A’. The notations are taken for the ease of understanding. Let's assume that subscriber ‘a’ is calling for subscriber ‘D’. The steps of making connection are as follows.
Fig. 5.2 Example of Crossbar switching in step-by-step basis.
31
CROSSBAR SWITCHING
Cross Connection Problem From the basic cross-matrix, shown in Fig.5.3 we can look through the problem occurred in crossbar switching.
Fig. 5.3 Crossbar matrix.
Say at time instant t = t1, a is calling for D and c is calling for B. When (for the 1st case) magnets corresponding to a and D energize, aD connection made. But when (for the 2nd case) c energizes, cD connection will be made as D was being energized before. But this is not a desired connection. And D subscriber will be connected with a and c. This is called as cross-connection problem. We can overcome the problem very easily by following a particular sequence for a successful call make up. That is Horizontal magnet energize
Vertical magnet energize
Horizontal magnet de-energize
Or we can follow another sequence as below : Vertical magnet energize
Horizontal magnet energize
Vertical magnet de-energize
Fig. 5.4
Say the 1st sequence is being followed. Taking into account the same example, if the horizontal magnet is de-energized, no chance of cross-connection is there as c will not find D corresponding magnet energized to make a contact.
6 Electronic Switching System (Stored Program Control) 1. Concept 2. Real-Time Control 3. Processor Structure 4. Centralized SPC 5. Distributed SPC 6. Software Architecture 7. Application Software ➢
Concept
The concept of stored program was first introduced in modern digital computers. Here, programs (i.e., set of instructions) are stored in the memory of the computer. These instructions are executed automatically one after another by the processor of the computer. The exchange control functions are processed and carried out through the execution of the programs stored in the computer memory and hence named as STORED PROGRAM CONTROL. Our digital telephone exchanges are called SPC (stored program control) exchanges and, consequently, are controlled by software stored in a computer. The programs contain the actual intelligence, and the computer (processor) sees to it that the control functions are performed. From the start, great expectations were attached to the success of the SPC nodes. Some expectations were met, while others were not. For instance, it was soon evident that the early systems did not provide the flexibility that was wanted. The software was too complex, and the smallest intervention could lead to quite unexpected side-effects. Today, all functions are divided into well-defined blocks. Modularity, as this is called, also makes the systems less complicated to maintain or extend as required. To some extent, the systems carry out troubleshooting on their own functions, indicating what measures need to be taken–measures that in many cases can be handled from the maintenance staff's terminal. To summarize, today’s SPC systems are characterized by: • simple handling of the equipment; • flexible structure; • low overall costs (investment, operation, maintenance); 32
ELECTRONIC SWITCHING SYSTEM (STORED PROGRAM CONTROL)
33
• extended functions/services; and • high degree of reliability. ➢ REAL-TIME CONTROL
When we are driving a car, we exercise real-time control: we must be watchful of what happens in the traffic around us. Each impression is processed in our brains and immediately results in “control measures” being taken: steering the car in the right direction and regulating its speed by means of the accelerator and brakes. As drivers we are more or less programmed to detect potential hazards, access them, decide what to do, and do it instantly. In telecommunication nodes, the same requirements for real-time control apply, but the pace is even higher: 100,000 or more decisions are made and executed per second. A large part of the tasks are simple and routine; for example, scanning all subscriber lines to look for subscribers that have lifted their handsets. Other tasks, such as selecting the path through the node for setting up a call, are more complicated. ➢ PROCESSOR STRUCTURE
Although processor capacity can be implemented in several ways, two main divisions have been made: Centralized control, where all work to set up a connection is controlled from a central processor system; and distributed control, where the control functions are shared by a number of processors that are more or less independent of one another. ➢ CENTRALIZED SPC
Fig. 6.1
34
LINE COMMUNICATION SYSTEM
The above figure shows the general organization of a centralized SPC. Today, digital technology is used both in the control part (processors) and in the switching part (in the form of specialized hardware). There is a distinct division between the processor part and the switching part - an arrangement called centralized control. However, a modern alternative is distributed control: by using microprocessors, processor capacity can be located close to the object to be controlled. Processor capacity can also be located far away from the equipment, provided the control logic is centralized. In the most refined form, only one processor is used (a single-processor system) to perform both routine work and advanced operations. The processor must be dimensioned according to the most difficult tasks. At the same time, however, because the routine tasks are the most time-consuming, the processor may have difficulty getting all things done. One solution to this kind of problem is to let several processors share the work load (multiprocessor system). Another model for sharing work is based on a hierarchically designed processor structure, in which the routine-like work is handled by several regional processors (RPs) and their coordination and more complex tasks are handled by a central Fig. 6.2 processor. Here, too, we distinguish between single- and multiprocessor systems. In the hierarchically designed multiprocessor system, the central processor consists of several processor units working in parallel. A hierarchical single-processor system has only a central processor. Centralized SPC using two processor configuration is generally classified into three categories on the basis of their mode of operation. These are as follows: 1. Standby Mode 2. Synchronous Duplex Mode 3. Load Sharing Mode
1. Standby Mode The entire exchange environment is scanned through the scanner by the processor P1 and the signals are distributed accordingly through the distributor. The scan result is stored in the secondary memory. For large exchanges it is difficult to store the scan results at every instant as the data is very bulky. So, the results are stored after a fixed amount of time. The processor P2 is the stand-by processor and remains idle as long as P1 function properly. If somehow processor P1 fails, the processor P2 becomes active. The data of the scanned results are retrieved by P2 from the secondary memory. The calls which were being processed in between the last saved result and the failing of P1, are lost.
ELECTRONIC SWITCHING SYSTEM (STORED PROGRAM CONTROL)
Fig. 6.3
2. Synchronous Duplex Mode
Fig. 6.4
35
36
LINE COMMUNICATION SYSTEM
Here the entire exchange environment is scanned synchronously through the individual scanners of the processors P1 and P2. Distribution of signals is done only by P1. Both P1 and P2 have independent dedicated memories. C is a comparator which checks the state of health of both the processors. Here the comparator C actually functions according to XOR decision making. We all know that from the XOR truth table that if two different states occur, output is '1'. In errorless operation, P1 and P2 both will give the same output. When error occurs, C comparator indicates the error and asks for a check out program. If it finds P1 as faulty, it gives the entire control of the exchange to P2. M1 and M2 have check-out programs in them which are run separately to detect the faulty processor. However, if both P1 & P2 fail, the comparator can't detect the fault. Also, transient errors can't be detected by check-out programs. In that case one may ignore the transient error and continue to work with the same processor. It may also be possible to run the exchange by the other processor leaving the faulty one.
3. Load Sharing Mode
Fig. 6.5
In this configuration both the processors scan and distribute different signals together. They share the entire load between them. It is the first step towards distributed processing. The allocation of environment can be done randomly or to the one which has just finished processing an ongoing call. E.D. is an exclusion device that excludes out the faulty processor from the system. In that case the entire exchange is run by the healthy processor. For Single Processor System, A = availability =
MTTR (MTTR + MTBF)
37
ELECTRONIC SWITCHING SYSTEM (STORED PROGRAM CONTROL)
MTTR (MTTR + MTBF) = MTTR/MTBF (MTBF >> MTTR) where MTTR = mean time to repair MTBF = mean time between failure For Dual Processor System,
U = unavailability = 1 – A =
(MTBF)D = AD =
UD =
(MTBF)2 (2 MTTR)
from Baye's theorem
MTBF2 MTBF 2 + 2 MTTR 2 2 (MTTR) 2 (MTBFD) 2
Mainly there are four major functions of a control sub-system. They are Event Monitoring, Call Processing, Charging and Operation & Maintenance (O & M). These functions can be grouped into three categories depending on the real time constraints. Event Monitoring and Distribution
Level 3
Call Processing
Level 2
O & M and Charging
Level 1
Real-time constraint increases
Fig. 6.6
Event Monitoring has the highest and O&M and charging has the lowest real time constraint. This necessitates a priority interrupt facility for processing in centralized control. When a process with higher priority than the ongoing current process occurs, the current process has to be interrupted. The process with the higher priority is taken up and completed (unless another process with higher priority than this one occurs) and then the previous process which was interrupted earlier is completed wherefrom it was left. This requires that the nesting of interrupts. It allows one to suspend any low priority process and take up the processing of higher priority process. When an interrupt occurs, the execution of programs is transferred to appropriate Interrupt Service Routine (ISR). Before transferring the program to the ISR, the relevant data and information of the currently assigned process is in some memory location known as stack. Stack is a data structure and serves as a temporary depository of data and address in some predefined memory location during execution of a program. The interrupts can be either vector interrupts or non-vector interrupts. For vector interrupts, the branching address is fixed. Each vector interrupt is mapped on to a predefined address wherefrom the corresponding ISR starts. This is the branching address for that interrupt. Hence in vector interrupts there is one-to-one correspondence between an interrupting signal and its branching address. But for the non-vector interrupts, the branching address is not known to the processor. The branching address for that interrupt is expected to come to the processor from the interrupting device. These are generally interfaced via interrupt controllers. Vector interrupts are therefore faster than non-vector interrupts because their branching addresses are hardware fixed. On the other hand in case of non-vector interrupts some time is taken by the interrupt controller to send the branching address to the processor.
38
LINE COMMUNICATION SYSTEM
Processing of a process with priority x ← Process with priority y occurs (y > x) Suspend the process with priority x Modify the stack pointer (SP) ← (SP) + 1 Save the PSW to stack (PSW) → ((SP)) Modify the stack pointer (SP) ← (SP)+1 Save the PC to the stack (PC) → ((SP)) Load PC with the starting address of ISR Execute the process with priority y
Suspend the process with priority y Modify the stack pointer (SP) ← (SP) + 1 Save the PSW to stack (PSW) → ((SP)) Modify the stack pointer (SP) ← (SP) + 1 Save the PC to the stack (PC) → ((SP)) Execute the process with priority z
← Process with priority z occurs (z > y)
Process with priority z compete Resume the process with priority y Load the address to PC from stack (PC) ← ((SP)) Modify the stack pointer (SP) ← (SP) – 1 Load the PSW from stack (PSW) ← ((SP)) Modify the stack pointer (SP) ← (SP) – 1
Process with priority y compete Resume the process with priority x Load the address to PC from stack (PC) ← ((SP)) Modify the stack pointer (SP) ← (SP) – 1 Load the PSW from stack (PSW) ← ((SP)) Modify the stack pointer (SP) ← (SP) – 1
Fig. 6.7 An example of 3 level interrupt processing
ELECTRONIC SWITCHING SYSTEM (STORED PROGRAM CONTROL)
39
➢ DISTRIBUTED SPC In distributed control systems there is no central processor for the overall functions. Instead, the switching equipment is divided into a number of switching parts, each of which has its own processor. When the processors have complete control over all the work in the respective switching parts, we have a genuine distributed system. Systems that have centralized control of certain functions, and in which the processors of the switching parts control work to only a lesser degree, are called false distributed systems. Systems that have centralized control of certain functions, and in which the processors of the switching parts control work to only a lesser degree, are called false distributed systems. DSPC provide better reliability and availability than CSPC. In DSPC, the entire control functions can be decomposed in two ways: 1. Horizontally 2. Vertically
Fig. 6.8
40
LINE COMMUNICATION SYSTEM
Level 3 Processing The Level 3 processor is mainly concerned with scanning, distribution and marking functions. The processing involved are very simple, well defined and specialized in nature. The processing generally results in setting, resetting or sensing and modifying flip-flops or registers. These operations in a predefined order constitute a control function. These functions can be implemented either by microprogrammed devices or wired logic. When a control unit is designed as a collection of conventional logic circuits, electronic devices or otherwise it is known as a hardwired control unit. A hardwired control unit lacks flexibility but can be easily implemented for a given task. A microprogrammed control unit is more universal and can be put to many different uses by simply modifying the micro program. Microprogramming is the art of developing micro programs. A micro program is an ordered sequence of micro-instructions. A microinstruction is a list of micro operations that can be performed simultaneously. A micro operation is an elementary operation that can be directly performed on micro-architecture which is basically an interconnection of registers, ALU, and memory through one or more buses. Microprograms for a particular processor are stored in memory (ROM) called Control Memory. Generally microinstructions are fixed for a particular processor. Each word in the control memory contains all control variables which are either 1 or 0 depending on which control signals are to be activated corresponding to a particular microinstruction. A more advanced development known as dynamic microprogramming permits a microprogram to be loaded on to the control memory to suit the processor for a particular use. Here the control memory is obviously erasable. This memory is generally a part of the processor. A major design effort has its emphasis on minimizing the length of the micro instruction. The length of the micro instruction decides the size of the control memory as well as the cost involved with the approach. A control information can be organized in various ways. A trivial way to organize the control field would be to have one bit for each control line that controls the data processor allowing full parallelism and there is no need for decoding the control field. This method however leads to inefficient use of control memory space when it isn't required to invoke all control operations simultaneously. This process is known as horizontal format. Another approach called Vertical format is also used. Here the microinstructions are encoded and then stored in the memory. These are short in size but the processor can execute only one microinstruction at a time needing an extra external decoder to decode the microinstruction. This makes the processor slower and it supports no parallelism. The memory becomes cheaper but we need external decoders. A via media solution is deviced which is known as mixed format. Here micro-operations are subgrouped on the basis of need at several levels. As the levels are increased, the parallelism increases and the cost become higher because this needs decoders at each level. This in turn makes the processor slow. Microprogrammed control
Hardwired control
Flexible
Not flexible
Slower
Faster
More expensive for moderate processing functions
Less expensive
Easier to implement
Difficult to implement
Introduction of new services is easy
Not easy
Easier to maintain
Difficult to maintain
ELECTRONIC SWITCHING SYSTEM (STORED PROGRAM CONTROL)
41
Level 2 Processing This is the call processing stage. This is accomplished by a call processor also known as a switching processor. The architecture, these processors are designed to ensure over 99.9% availability, fault tolerance and security of operation. The transfer of data is done with the help of program controlled I/O and DMA. Other I/O controllers are also employed as and when required. The called for number is the input to the processor from the subscriber. The processor then processes the call and distributes the signal. The traffic handling capacity of the control equipment is usually limited by the capacity of the switching processor. The load on the switching processor is measured by its occupancy t as given below t = a + bN where
a = fixed overhead depending upon the exchange capacity and configuration b = average time to process a call N = number of calls per unit time Usually, the switching processor is designed to handle a traffic load which is 40% higher than the normal load. When this overload occurs, the processor may be loaded to 95% of its capacity so that traffic fluctuations can be absorbed.
Level 1 Processing Level 1 processing is basically a man and machine interface. It has the least real time constraint. Following are some of the important functions performed in this level of processing : 1. Administer the exchange hardware and software 2. Add, modify and delete information in translation tables 3. Change the class of service of a subscriber 4. Put a new line or trunk in operation 5. Supervise operation of exchange 6. Traffic monitoring 7. Fault detection 8. Run test programs and check-out programs Since this level of processing has the least real time constraint, many exchanges are remotely connected to a common O&M. This enables a maintenance personnel to attend several exchanges from a central location.
Fig. 6.9
42
LINE COMMUNICATION SYSTEM
➢ SOFTWARE ARCHITECTURE There are two types of softwares used in SPC. They are system software and application software. Software architecture deals with the system software environment and language processors. The O&M operations are performed by a general purpose computer and the software architecture required for such a computer is similar to a general data processing system. Call processing is specific to the switching systems and demand real time responses. This requires a special software system that will help the associated hardware to function properly. This software is known as System software. The processes that are processed in a processor are not done in one go. Rather there are several elementary steps involved in completion of a process. Here a process means a program in execution and a program by itself is not a process. One important criteria of the system software is that it should support multi-processing because there are many calls at a given time for the processor to attend. A processor in multiprogramming environment can be in any one of the three states: running, ready or blocked. A processes is said to be running, if it currently has the processor allocated to it. A process is said to be ready, if it could use a processor if one were available. A process is said to be blocked if it is waiting for some event to occur before it can proceed.
Fig. 6.10
While only one process may be running at any time, several processes may be ready and several blocked. The ready processes are ordered according to their priority so that the next process to receive the processor is the first ready process in the ordered list. There may be several ready lists for each level of priority. The blocked processes are unblocked in the order in which the events they are awaiting occur. Timers are set to prevent any one process from monopolizing the processor, either accidentally or maliciously. Each process is represented by a program control block (PCB) in the Operating system. PCB is a data structure containing all key information about the current process such as memory allocation, current state, interrupt status, register contents, I/O resources etc., in a multiprogramming environment when a processor switches from one process to another, it save the PCB of the stalled process. When the stalled process needs to be restarted, the PCB of that process is loaded. Several processes share common resources and data. So, the processes should be mutually excluded from accessing shared data simultaneously otherwise it will result in unnecessary data collision.
43
ELECTRONIC SWITCHING SYSTEM (STORED PROGRAM CONTROL) Process A
Operating System
Process B
Idle
Running
Blocked
Save registers (A) Save PSW (A) Load registers (B) Load PSW (B) Running
Unblocked ready state
Save registers (B)
Blocked or
Save PSW (B)
interrupted
Load registers (A) Load PSW (A)
Running
Fig. 6.11
The above figure shows the switching of processes by an operating system. When process A is allocated the CPU, another process B with higher priority interrupts and it needs the same data which are being accessed by process A. Since B has a higher priority it will acquire the CPU and when it is finished the data will be used by process A. But in the mean time if the data is changed by process B, it will give wrong result from process A. This problem can be solved by giving each process exclusive access to a shared table or data. When a process accesses a shared resource, all other processes are kept waiting. Thus, there is mutual exclusion of processes in accessing shared data.
★ Semaphore Semaphore is a software solution devised to ensure mutual exclusion of processes while accessing shared data i.e., in its critical region or critical section. Here we treat semaphore as a process variable. It generally holds integer type values. We all know that, System = Input + Process + Output Therefore system variable is actually the summation of input-output variables and process variable. Semaphore is a protected variable that can assume non-negative integer values. There is one semaphore for each shared resource in the system. Two standard individual operations test (P) and increment
44
LINE COMMUNICATION SYSTEM
(V) are possible on a semaphore (S). Each P operation tests S to see if it is non-zero, and if so, decrements the value of S by 1, indicating that a resource has been removed from the common pool. If S is zero, the process is blocked to be released by a V operation which increments the value of S by one signifying the return of a resource to the pool. The implementation of semaphore with a blocked queue may result in a situation where two or more processes are waiting indefinitely for a V operation that can be caused only by one of the blocked processes. If this happens, the processes are said to be deadlocked. The following is an example to illustrate a deadlock condition. P0 P(S0) ; S0 = S0 – 1 P(S1) ; blocked (S1)
P1 P(S1) ; S1 = S1 – 1 P(S0) ; blocked (S0)
The process P0 having seized the semaphore S0 gets blocked on S1 and the process P1 having seized the semaphore S1 gets blocked on S0. Both the processes are blocked for a V operation to be performed by the other, and there is a deadlock.
➢ APPLICATION SOFTWARE The application software of a switching system can be divided into three categories: 1. Call processing software 2. Administrative software 3. Maintenance software Application software packages of a switching system use a modular organization. Each module deals with a specific task and the size of a module depends on the task. The modules aren't self contained and can exchange data from other modules and data tables. A module may be a part of more than one functional unit. A functional unit generally runs separately in a system. The modules are strung together through special programs or chaining tables. Associated with every module is a pointer to a set of entries in the chaining table pertaining to that module. Each entry in the chaining table consists of a key and a module number. Whenever a module completes execution, its corresponding entries in the chaining table are scanned and the keys are compared to a function key. If a match occurs, the corresponding module in the chaining entry is executed next. This approach allows flexibility for adding new features and functions or deleting old modules by simply modifying the chaining table.
ELECTRONIC SWITCHING SYSTEM (STORED PROGRAM CONTROL)
45
Fig. 6.12
Application software accounts for about 80% of the total volume of software in a switching system. Administration and maintenance software accounts for about 65% of the total volume. This entire software need not be core resident. Considering real time constraints, the call processing software are usually core resident. The administration and maintenance software modules reside in a back up storage and are brought into the main memory as and when required. Depending on the architectural support available from the switching processor, the OS may use overlay or Virtual Memory technique in this purpose. Switching software almost always use a parameterized design. This enables the same package to be used over a wide range of exchanges by adapting the package to specific exchange characteristics. The parameters may be divided into office and system parameters. The system parameters offer flexibility at the overall level while office parameters define program execution limits at specific exchanges. The nature of the parametric data may be classified as either semi-permanent or temporary. Semipermanent data consist of parameters that describe the hardware characteristics of the exchange and its environment. These data are updated rather infrequently, say when there is a need for expansion of exchange capacity. Temporary data have a lifetime equal to that of the process they pertain to.
46
LINE COMMUNICATION SYSTEM
Fig. 6.13
In the figure, the En when multiplied by 2 provides the offset value. There are two words per entry in first level and hence double equipment number is needed. The first part is the directory number of the calling subscriber which is needed for transmitting calling party identification and billing purpose. The second part contains a pointer to the optional service table. The next block determines the class of service whether the calling subscriber is a coin box or business or domestic telephone. In the second level, each entry takes as many words as the number of optional services offered by the system if we assume that each word in this entry stores information regarding one optional service. The value at first level is suitably adjusted taking this into account. When the PTR is added to BA2 the first address of the entry obtained the word corresponding to that service. In this figure, the calling subscriber requests for abbreviated dialing facility which is given the service request number 01. The word 1 corresponding to the abbr. dialing service, contains a pointer ADD-PTR to the third level dialing table which is the abbr. dialing directory for the subscribers. The main purpose of this system is to identify the equipment number corresponding to the directory line number and class of service information.
7 Traffic Engineering 1. Concept 2. Full-Availability Models 3. Difference Between GOS & Blocking Probability 4. Proof of Blocking Probability = Grade of Service 5. Erlang Calculations 6. Erlang-B Congestion Formula 7. Erlang-B Example ➢
Concept
The estimating of the switching quantities for an automatic telephone exchange is based primarily on the amount of traffic which it will be required to handle. Calculations of the probable average and maximum amount of traffic are necessary to determine the optimum number of switches required in the various switching ranks. The more important quantities used in traffic calculations are: ❖ Traffic Unit (T.U). The traffic unit is used to measure the traffic flow. The number of traffic units carried by a group of switches is numerically equal to the product of the number of calls per hour and the average duration of the calls in hours. The traffic flow in traffic units is equal to Erlang. Erlang is a unit of traffic measurement. Traffic in erlangs (E) can be defined as: E=
usage time for all resources total time
The usage time for all resources is divided by the total time interval. An example should clarify the definition. Consider a digital trunk with 31 voice circuits. If each circuit has been busy for half an hour during a one hour measurement interval, the erlang calculation would be as follows: 1. Usage time for all resources is 30 min * 31 = 930 minutes 2. Total time is 60 minutes Thus the total traffic in erlangs is 930/60 = 15.5 erlangs to the average number of simultaneous calls. 47
48
LINE COMMUNICATION SYSTEM
❖ Busy Hour. This is defined as the hour of the day in each exchange when the originating traffic carried is the greatest. The load handled by a system varies a lot based on the time of day and day of the week. Most systems are heavily loaded for a few hours in a day. The main objective of resource dimensioning is to make sure that the system performs well during these busy hours. This will make sure that the system has adequate resources to handle peak as well as off-peak traffic. ❖ Call completion rate. Ratio of number of successful calls to the number of call attempts is known as CCR . CCR value of 0.75 is considered excellent but in INDIA CCR is almost 0.70. ❖ Busy hour call attempts. The number of call attempts in busy hour is called BHCA.
Fig. 7.1 Working Day Telephone Traffic Pattern
❖ Busy Hour Calling Rate. This is the average number of calls initiated per subscriber during the busy hour. ❖ Grade of Service. The flow of traffic initiated by subscribers varies widely throughout the day, and from one day to the next. If during any traffic peak the number of outlets from a switching stage are all occupied, additional calls at that instant cannot be extended and completed and will thus be lost. The proportion of calls that are lost to the total number of calls in the busy hour determines the quantity known as the grade of service for that particular switching stage. Grade of service is directly related to the blocking probability. A higher grade of service guarantee to the customer means ensuring a low blocking probability during the busy hours. Providing a higher grade of service requires increasing the number of resources in the system. Conversely, you can reduce the number of resources to lower the cost, but at the expense of grade of service. GOS = ( A – A0 ) / A ; A – offered traffic; A0 – carried traffic; A – A0 – lost traffic
Service Time Service time is the total time a resource is needed to handle one customer's request.
49
TRAFFIC ENGINEERING
Wait Time The total time customers will have to wait in the queue before they get any service. Availability: The number of trunks to which a switch has access is known as the availability. If a subscriber's uniselector has, say 25 outlets, each of which is coupled to the first selector, then the availability of the uniselector is 25.
➢ FULL-AVAILABILITY MODELS A full-availability system is one in which there are S sources of traffic and N outlets and any free source can seize any free outlet, regardless of the state of the system. The simplest example is a matrix with SXN cross-points as shown below.
Fig. 7.2 SXN Matrix
Probability of Blocking In order to find out if the system will perform as it should, some measure of performance is required. The most often used measure is the probability of blocking, which is the proportion of calls, in the long term that are rejected. If we consider the general case of a two stage link system in which the corresponding outlets on the second-stage switches from a route as indicated in the diagram below. The blocking probability defines the chance that a customer will be denied service due to lack of resources. For example, a blocking probability of 0.01 means that 1% of the customers will be denied service. Most of the time blocking probability calculations refers to the busy hour only. Blocking probability during the busy hour can be decreased by: • Increasing the resources in the system • Offering incentives and discounts to encourage usage during off-peak hours
50
LINE COMMUNICATION SYSTEM
Fig. 7.3 Stage Link System
There are two ways in which a call may be blocked. 1. There may be no free route circuit 2. There may be no path available between the inlet carrying the incoming call and the free route circuits. The diagram shows both of these conditions; route 3 has no free circuits available, whereas if inlet A on first-stage switch 1 is wanting to be connected to route 2, on which there are free route circuits, it cannot be, because there is no available link.
➢ DIFFERENCE BETWEEN GOS & BLOCKING PROBABILITY Grade of service [GOS]
Blocking probability
(i) Calculate from subscriber point of view.
(i) Calculate from switching center (Exch) viewpoint (ii) blocking probability is time congestion (iii) Determined by observing busy servers in switching system.
(ii) GOS is call congestion. (iii) Determined by counting lost calls out of total offered calls.
➢ PROOF OF BLOCKING PROBABILITY = GRADE OF SERVICE Ci = t ; 0 ⇐ t < R, t = average poission call arrival rate R = number of servers in the system
51
TRAFFIC ENGINEERING R –1
C0 =
∑ t Pi
i=0
Pi = probability that the system is in state i P0 + P1 + P2 + ...... + Pr = 1 C0 = t(P0 + P1 + ...... + Pr – 1) = t(1 – Pr) mean traffic carried by network is : A0 = C0 * k A0 = t * (1 – Pr)* k A0 = A * (1 – Pr) as A = t * k A = offered for poission arrival process Pr = (A – A0)/A ; so from the above relation it is clear that blocking probability becomes equal to Grade of service.
➢ ERLANG CALCULATIONS There is a trade off between resource dimensioning and grade of service. In this section we will examine these trade offs using the Erlang-B formula. The choice of the formula to use is dependent upon the handling of customers when all resources are busy. • Erlang-B should be used when failure to get a free resource results in the customer being denied service. The customers request is rejected as no free resources are available.
➢ ERLANG-B CONGESTION FORMULA Erlang-B formula allows you to calculate the probability that a resource request from the customer will be denied due to lack of resources. The formula is:
PB =
EN N! X=N
∑
X =0
EX X!
Where: • N is the total number of resources in the system • E is the total traffic in erlangs Pb is the probability that a customer request will be rejected due to lack of resources. The formula works under the following conditions: • The number of customers is much larger than the number of resources available to service them. In general, the formula gives acceptable results if the number of customers is at least 10 times the total number of resources (N). • Requests from customers are independent of each other. This formula does not work if customer requests have been triggered by some common event like calling a talk show, natural calamity etc. • Customer requests are blocked only when no resources are available to service them.
52
LINE COMMUNICATION SYSTEM
• When a customer cannot be serviced, the resource request is simply rejected. No attempt is made to queue the customer request. • The customer does not retry the request after being denied service. • The resource is allocated exclusively to one customer for the specified period. The resource cannot be shared with other customers.
➢ ERLANG-B EXAMPLE Let’s consider an example where Erlang-B formula is used in analyzing performance requirements. Consider the resource dimensioning of DTMF (Dual Tone Multi Frequency) receivers in the Xenon switching system. DTMF receivers are used to receive tones from the phone keypad and recognize the dialed digits. Thus a DTMF receiver should be allocated before dial-tone is fed to the subscriber. The DTMF receiver can be freed after digit dialing has been completed. The resource dimensioning analysis for DTMF receivers follows: • Performance Requirements • Requirement Analysis
Performance Requirements • Every XEN processor in the switch shall be equipped with DTMF receivers • DTMF receivers shall be allocated before dial-tone is fed and can be freed after digit dialing has been completed. • Average duration of the digit dialing phase is 30 seconds. • The probability of a call getting dropped due to lack of dial-tone shall not exceed 0.1%. • A XEN processor shall handle at least 20,000 originations in the busy hour.
Requirement Analysis We need to calculate the total number of DTMF receivers needed in XEN processor. We decide to use the Erlang-B formula as the system rejects a call if a DTMF receiver cannot be found. Most conditions for Erlang-B use are met. (The only condition that is not met is the requirement that failed customers do not retry.) The total erlang traffic for the DTMF receivers is: B.H.T. = (20,000 orig/sec * 30 sec) / (3600 sec) = 167 Erlangs Blocking probability = 0.1 % = 0.001. With these numbers, the calculator returns a total number of lines as 202. Thus we need a total of 202 DTMF receivers.
8 Time Division Switching 1. Concept 2. Time Division Space Switching 3. Time Division Time Switching 4. Memory Management 5. Time Multiplexed Space Switching 6. Time Mulitplexed Time Switching 7. Combination Switching
➢ Concept We know from the previous chapter that in single stage Space Division switching one cross-point between the input stages (inlets) and output stages (outlets) can establish any connection only between one pair of subscribers. In multistage Space Division switching a cross-point is usable for establishing, more than one connection, i.e. sharable among many subscribers. This sharing leads to a reduction in the number of switching hard-wares. But the connection once made have to be a dedicated one during the speech conversation, because continuous analog speech waveform is passed through the switch. Sharing of cross-point occurs from one connection to the next, therefore. It means that, the switching array or switching network arrangements are shared/ scaled only in space domain. But after introducing the new type of switching concept named time division switching, the time scale can be divided according to the requirement of subscriber. In a digital transmission, sample values of speech are sent as PAM (Pulse Amplitude Modulated) or PCM (Pulse Code Modulated) signal. If sampling frequency = Fs = 8 kHz The sampling rate = Ts = 1/8 kHz = 125 µs In digital domain a sample value can be passed through a switching n/w from inlet to outlet in a few µs. Therefore during 125 µs interval most of the time is left unused, say more than 120 µs. If we establish a dynamic control function mechanism whereby a switching cross-point is assigned a number of inlet-outlet pairs each of few µs, that can be used to transmit speech samples from inlets to their corresponding outlets. It is now very clear that number of switching hard-wares needed is much less 53
54
LINE COMMUNICATION SYSTEM
than that of the multistage space division switching. Time division switching hierarchy is given below :
Fig. 8.1 Classification
➢ TIME DIVISION SPACE SWITCHING The basic N × N Time Division Space Switching can be represented as a two stage network. First stage is N × 1 and second stage is 1 × N. Here each inlet/outlet is a single speech circuit which corresponds to a subscriber communicating line. When PAM samples are switched in time division manner it is analog time division space switching and for PCM binary data the switching is called digital time division space switching. The connecting bus structure between the inlets-outlets also change its nature according to the choice of speech modulation technique, i.e. these two different types of buses are used, digital and analog. We can formulate Simultaneous Conversation (SC) =125/Tst ...(1) where, Tst = time to set up a connection and transfer the sample value.
55
TIME DIVISION SWITCHING
Fig. 8.2(a) Basic time division space switching.
Fig. 8.2(b) Stage Equivalent.
From the Fig. 8.2(a) the cyclic control can be described similar to that in case of space division switching. If N = number of inlets/outlets, then a = log2 N and N = 2ª Now by using a to N decoder synchronously with a clock, we can cyclically choice the inlets as 1, 2, 3, ………, N – 1, N, 1, 2,………and so on.
56
LINE COMMUNICATION SYSTEM
Fig. 8.3 Cyclic Control.
But it can only make up one-to-one connections (ith inlet to ith outlet). Therefore it lacks full availability. We cannot connect any inlet to any outlet. The method by which the sample values of subscriber m are transferred to subscriber n, and which also allows the reverse transmission from subscriber n to m at the same time, is known as resonant transfer. The method is also characterized as being lossless since it allows the transmission without attenuation; consequently no repeaters or amplifiers are needed. It is an effective system when the transmission distance is short as in the case of private branch exchange (PBX). PBX facility spatially are used in close proximity such as the phones of a company, all are in the same building.
Input Controlled Time Division Space Switching Here the new improvement over the previous one is introduced as control memory (CM). The control memory has N words corresponding N inlets and has a width of a (= logN) bits which are used to address N outlets.
TIME DIVISION SWITCHING
57
Fig. 8.4 Input controlled time division space switching (Block).
In the arrangement shown above, the modulo-N counter of cyclic control also acts as the memory address register (MAR) of the control memory. And the address decider which decodes the outlet address from control memory to o/p stage, also acts as the memory data register (MDR) of the control memory. The principle of operation is described schematically in Fig. 8.5.
58
LINE COMMUNICATION SYSTEM
Fig. 8.5 Input controlled time division space switching (step-by-step).
If one inlet i is not active, the corresponding location of the control memory is filled up by a null (Φ) value. It is then very clear that full availability of switching is obtained in this type of arrangement. For example an address sequence 7, 4, N, 3, …….., Φ stored in locations 1, 2, 3, 4,………, N of the control memory (see Fig. 8.4), implies that the connections below are made.
TIME DIVISION SWITCHING
59
This switch is said to be input controlled or input driven as the outputs are chosen depending on the inlet that has been scanned by any instant. Since one switching element, the bus is being shared by N connections, all of which can be active simultaneously, and a physical connection is established between the inlet and outlet for the duration of the same transfer, this switching technique is known as time division space switching.
Output Controlled Time Division Space Switching It is possible to organize a fully available time division space switch with cyclic control for the outlets. In such a case the switch is said as output controlled or output driven time division space switch. The block diagram is shown in Fig. 8.6.
Fig. 8.6 Output controlled time division space switching (Block).
60
LINE COMMUNICATION SYSTEM
The principle of operation is described schematically in Fig. 8.7
Fig. 8.7. Output controlled time division space switching (step-by-step).
For both input and output controlled configurations, the number of inlets or outlets N, which is equal to the switching capacity (SC) , is given by N = SC =
125 Ti + Tm + Td + Tt
...(2)
where Ti = time to increment the modulo-N counter Tm = time to read the control memory Td = time to decode the address Tt = time to transfer the sample value. It must be noted that the clock input of the modulo-N counter must be such that N sample transfers are organized in 125 µs, i.e. clock rate = 8 N kHz. Broadcast takes place only for output controlled switching. Actually broadcast can be defined as same data transfer from one source to several destinations. It can be done by filling all the control memory fields by same inlet address. By this way the particular data from specified inlet is transferred to all the outlets. In practical telephone conversations, speech samples have to be exchanged both ways. We can use two separate buses. Alternatively single bus can be used to organize the two way data transfer—1st in forward direction next in the reverse. PCM bus only supports parallel data transfer. So for serial data,
TIME DIVISION SWITCHING
61
serial to parallel and parallel to serial converters are needed for inlets and outlets respectively. Thus data is transferred both ways in 125 µs cycle. Number of switching elements: i/p side = N o/p side = N Total = 2N SC = N Therefore, traffic handling capability TC = SC / Theoritical maximum load = SC / (Total number of switching elements/2) = SC / (2N/2) = N/N =1 ...(3) Cost of switching network = cost of ( switching elements + control memory) = 2N+N units = 3N units Therefore, cost capacity index CCI = SC/ Cost per subscriber line = N/(3N/N) = N/ 3 ...(4)
Memory Controlled Time Division Space Switching By cyclic control in case of input controlled and output controlled configurations it scans all the input/ output lines where only 20% are active subscribers, generally. Thus, an unnecessary delay is introduced. The versatile configuration, memory controlled time division space switching was introduced then. Here the addresses of both inlet and outlet are stored into the control memory. That's why the word width of the control memory is 2a = log2N bits. If a connection is to be set up between inlet k and outlet j, two addresses are entered into a free location of the control memory by data write facility and the location is marked busy. After the conversation terminated, the location is filled up by Φ and made free. Here for marking purpose the concept of bit vector in h/w comes. Bit vector < = 1, for busy control memory location, and bit vector < = 0, for free control memory location. The memory controlled time division space switching is described with a block diagram in the Fig. 8.8.
62
LINE COMMUNICATION SYSTEM
Fig. 8.8. Memory controlled time division space switching (Block).
In this arrangement, SC = where,
125 , Clock frequency = 8SC kHz Ts
Ts = Ti + Tm + Tt + Td
...(5)
TIME DIVISION SWITCHING
63
The principle of operation is described schematically in Fig. 8.9
Fig. 8.9 Memory controlled time division space switching (Step-by-step).
➢ TIME DIVISION TIME SWITCHING If we compare the time division time switching with the time division space switching, the first improvement is to be noticed must be the introduction of memory block in place of bus. The second one is the sample speech must be PCM, not PAM.
64
LINE COMMUNICATION SYSTEM
Fig. 8.10(a) Basic time division time switching (block).
Fig. 8.10(b) Two stage equivalent.
Physically, the 2 MDRs (data in and data out blocks) are not separate. Only one MDR is there in reality.
TIME DIVISION SWITCHING
65
The principle of operation is described schematically in Fig. 8.11
Fig. 8.11 Time division time switching (Step-by-step).
As in case of time division space switching, time division time switching also is divided in three ways (as i/p controlled, o/p controlled and memory controlled). (i) Sequential read/ Random write (ii) Sequential write/ Random read (iii) Random write/ Random read (The data memory is accessed sequentially) These three ways of switching is exactly similar to that of three ways of time division space switching, except the basic difference in between time switching and space switching. There are two methods of operations. (a) Phased operation (b) Slotted operation
Phased Operation The entire operation is divided into two phases. In the 1st phase the memory write operation is involved, i.e. this is the inlet phase operation. Whereas in the 2nd phase, the two memory read operations are involved, 1st control memory and the 2nd data memory.
66
LINE COMMUNICATION SYSTEM
The total time required for 2-phased operation is Ts = NTd + N(Td + Tc)
...(6)
where,
Td = read/write time for data memory Tc = read/write time for control memory If, Td = Tc = Tm; Ts = 3NTm ...(7) Since the entire operation is to be completed in 125 µs, therefore, the maximum number of subscribers can be defined as, N=
125 Tm 3
...(8)
Fig. 8.12 Phased operation.
It is very clear from the above Fig. 8.12 that, during the last cycle of phase one (Ph. 1), 1st location of the control memory (containing inlet address, say 1 of a1) is read, 2nd location of the control memory (containing inlet address, say 2 of a2) is read, and so on. As a result, extra time required for control memory read is not needed separately for CM read is completed before the completion of data memory (DM) read. Therefore here Ts = 2NTm, and the allowable maximum number of subscriber is N=
125 Tm 2
...(9)
67
TIME DIVISION SWITCHING
Slotted Operation In slotted operation the total period is divided into N sub-periods of duration 125/N µs. In such one of the sub-periods, say sub-period i, the following operations are performed :
k
Fig. 8.13 Slotted operation of time division time switching (Step-by-step).
Here we just divide the 125 µs duration in N number of slots. Each slot is again divided by two separations as shown in the Fig. 8.14
Fig. 8.14 Slotted operation.
In the random write/sequential read operation, the control memory contains the address of outlets corresponding to the inlets. Similarly, in random read/sequential write operation, the control memory contains the address of inlets corresponding to the outlets. In random write/random read operation, the form of control permits a large number of subscribers than switching capacity of the system. There are two control memories (CM). CM1 and CM2 hold the address of active inlets and outlets respectively as shown in Fig. 8.15.
68
LINE COMMUNICATION SYSTEM
Fig. 8.15 Random read/random write time switch.
It is very clear that, there is a one-to-one correspondence between the locations of control memory. If the address of an active inlet is in location x of CM1, the address of outlet to which this inlet would be connected, is placed in c location of CM2. In the 1st phase the addresses of active inlets are read out from CM1, data is transferred to data memory starting from the 1st location. In 2nd phase, the addresses of active outlets are read out from CM2, data is transferred to outlet from the data memory. Therefore,
SC =
125 4 Tm
...(10a)
Overlapping CM and DM, we can make SC =
125 2 Tm
...(10b)
By advent of semiconductor dual port memory chip, the two different control memory access (CM1 & CM2) can be made simultaneously. As a result only one control memory can be used as this purpose of time division time switching. i.e.
SC =
125 Tm
...(10c)
Memory management procedure is shown in the next topic.
➢ MEMORY MANAGEMENT Linked list is maintained for memory management. The list is maintained such that, there are two different fields. One for data and other for the pointer, i.e. address of next location as shown in the Fig. 8.16. Φ pointer denotes no next location is there. There are two fundamental list management operations. One is adding an item and another removing an item.
69
TIME DIVISION SWITCHING
The actions involved in the application are as follows—
Fig. 8.16 Memory management.
70
LINE COMMUNICATION SYSTEM
(1) Allocate a location from the free list to the occupied list Location to be allotted, F < = Content of FLP Content pf FLP < = Pointer of location F Pointer in location F < = Content of OLP Content of OLP < = Location F According to the example discussed in Fig.8.16, location to be allotted must be location number 3. Before allocation, free list is 3—6—7—9—4—1; and occupied list is 2—11—12—8—5—10. After allocation operation, free list is 6—7—9—4—1; and occupied list is 3—2—11—12—8— 5—10. (2) Free a location x from the occupied list and add to the free list Pointer in predecessor in x < = Pointer in x Pointer in x < = Contents of FLP Contents of FLP < = x According to the example, location to be freed must me location number 11. Before deletion, free list is 6—7—9—4—1; and occupied list is 3—2—11—12—8—5—10. After deletion operation, free list is 11—6—7—9—4—1; and occupied list is 3—2—12—8— 5—10.
➢ TIME MULTIPLEXED SPACE SWITCHING In time division space & time switching, there is only one-to-one inlet/outlet correspondence to a single subscriber line each. They are used in local exchanges. Now we'll consider which are used in transit exchanges, as in case of subscriber trunk dialing (STD). Here inlets and outlets are trunks which carry time division multiplexed data stream. N inlets and N outlets each carrying a time division multiplexed data stream of M samples per frame (M samples /frame) are there. Frame duration is made 125 µs. It means in one frame duration MN speech samples have to be switched. Here one sample duration = 125 / M µs; which is usually referred to one time slot. According to the arrangement made, in one time slot N samples can be switched. The number of channel multiplexed in each input trunk is M. The arrangement is done as the Fig. 8.17.
Fig. 8.17 Source multiplexing and destination demultiplexing.
TIME DIVISION SWITCHING
71
In Fig. 8.18 output controlled time multiplexed space switching is shown.
Fig. 8.18 Output controlled time multiplexed space switching (Block).
There is a 1-to-M relationship between output and control memory location. The control memory has MN words, M blocks of N words each. If the location address is (i, j) signifies— i = block address , where 1 ≤ i ≤ M; and j = word within the block , where 1 ≤ j ≤ N; Here i corresponds to time slot and j corresponds to outlet. There are MN words in the control memory. The division of MN words is done as in N word blocks each as follows. 1st time slot = 1st N locations of CM, 1 to N i.e. (1, 1) to (1, N) 2nd time slot = 2nd N locations of CM, N + 1 to 2N i.e. (2,1) to (2, N) As the total time allotted = 125 µs Therefore MNTs = 125 µs where, Ts = (switching time + memory access time of inlet/outlet pair.) or, N = 125/ MTs ...(11) Total cost, C = (Number switches + Number of memory words) units = 2N + MN units ...(12) Problem. Calculate the number of trunks used that can be supported for a TMSS, for given • 64 number of channels are multiplexed • The control memory access time = Bus switching and transfer time = 100 ms
72
LINE COMMUNICATION SYSTEM
Solution. Given that number of channels multiplexed = M = 64 CM access time = 100 ms Bus switching and transfer time = 100 ms Therefore total time Ts = (100+100) ms = 200 ms = 200×0.001 µs Now, we know that, in TMSS MNTs = 125 µs, where N = number of trunks Therefore, N = 125/ (MTs) = 125/ (64×200×0.001) = 9.765625 ≈ 10 The number of trunks supported can be increased now by splitting the control memory into N modules each of M words. If the module serves the output line by switching appropriate input, then it is output controlled. And if the module serve the input line by switching appropriate output, then it is input controlled. The arrangement is shown in the Fig. 8.19.
Fig. 8.19 TMSS with M time slots and N control memory (Block).
73
TIME DIVISION SWITCHING
According to the arrangement shown above, each location corresponds to a particular time slot. They are read out in parallel. Therefore,
M=
125 Tm
...(13)
Where, Tm = control memory access time. And obviously Tm << NTs. The switch has N switching matrices of size N × 1 each at the input. There are MN memory words totally. Therefore the cost = N (N × 1) + MN units = N ² + MN units. ...(14) A time multiplexed time division space switch can be configured around a space array as shown in Fig. 8.20. The array consists of M input horizontals and N output verticals. If M = N, switch is non-blocking M < N, switch is expanding M > N, switch is concentrating. In any time slot, one gate for vertical and upto N gates for horizontal can be enabled for these connections.
Fig. 8.20 Space array of TMSS.
In every time slot, upto N or M samples are switched at a time. The number of words in CM is N with log2N width each. As parallel transfer is possible here, a large number of channels can be multiplexed per incoming trunk.
74
LINE COMMUNICATION SYSTEM
➢ TIME MULTIPLEXED TIME SWITCHING In time multiplexed time switching, the concept of time slot interchange (TSI) comes. It permits the interchange of one input time slot to another output time slot of sample values. Generally it introduces delay between transmission and reception. One trunk is shown in the next Fig. 8.21.
Fig 8.21 Time Multiplexed Time Switching (TSI).
According to the arrangement shown above, 1st inlet to 1st outlet, 5th inlet to 2nd outlet, 4th inlet to third inlet and so on. If the minimum time required to transfer the sample value = Tts, the 1st sample experiences a delay = Tts 2nd sample experiences a delay = [(M +1) – (5 – 2)] Tts = [M – 2]Tts 3rd sample experiences a delay = [(M +1) – (4 – 3)] Tts = [M]Tts mth sample experiences a delay = [(M +1) – (X – m)] Tts There are two sequential memory accesses per time slot interchange. Therefore Tts = 2Tm and 2MTm = 125 µs ...(15)
Fig. 8.22(a) Memory division.
75
TIME DIVISION SWITCHING
Fig. 8.22(b) Time slot arrangements.
Total cost of switching elements = Number of Memory locations = 2M units Problem. Calculate the maximum access time that can be permitted for the data and control memory and number of channels that can be multiplexed in a TSI switch with a single input single output trunk. The cost of switch is 5000 units. Compare the cost with that of a single stage space division switching. Solution. Given that the cost = 2M = 5000 units Therefore number of channels multiplexed = M =
Memory access time = Tm =
5000 = 2500 2
125 125 Tts = µs = µs = 0.025 µs 2M (2 × 2500) 2
= 25 ns Cost for TSI = 5000 units Cost for single stage space division switching (SDS) use array of 2500 ×2500 = 6250000 units Therefore, cost advantage = C(SDS) / C(TSI) = 1250 The switching is called expanding if DM write < DM read And called concentrating if DM read < DM write Handling the MN subscribers in 125 µs is done by four ways. They are :
1. Serial In/Serial Out This configuration is exactly same as the arrangement shown in the Fig. 8.21. The difference is that, N inputs and N outputs are chosen instead of 1 trunk each (as shown). Multiplexed data arrival and departure is done serially. The storage capacity of each of the memories is NM words each. The width of data memory word is 1 byte and that of control memory is
The width of time slot counter is also
log NM . log 2 log NM .The sample of 1st time slot is stored in the 1st N log 2
locations of DM, 2nd time slot corresponds to N +1 to 2N, and so on. The time factors can be analyzed as Tts = 2NTm, 125 = 2NMTm
...(16)
76
LINE COMMUNICATION SYSTEM
2. Parallel In/Serial Out In this type of arrangement, the DM is organized as N modules of M words as in Fig. 8.23. Each module is associated with a i/p line. The width of CM word module is
log M log N + log 2 log 2
For non-overlapped operation, Tts = (2N + 1)Tm, or 125 = M(2N + 1)Tm ...(17a) For overlapped operation, Tts = M(N + 1)Tm ...(17b) Tts is divided into (N + 1) sub-slots (T0 to TN). During T0, all data are written into respective DM modules simultaneously. The two dimensional address structures supports the module address and word address, which are used to access the DM words. Word address corresponds to the time slot and module address to inlet number.
Fig. 8.23(a) Parallel in/Serial out configuration.
77
TIME DIVISION SWITCHING
Fig. 8.23(b) Timing Diagram.
3. Serial In/Parallel Out The arrangement in serial in/parallel out is dual to that of parallel in/serial out switching. Here also, the DM is organized as N modules of M words. Each module is associated with an o/p line. The width of CM word module is | log2 N + log2 M | Output operation is done during sub-slots T0 to TN.
4. Parallel In/Parallel Out This is the most complex of all TMTS (time multiplexed time switch) configuration. Here both DM and CM are organized as N independent module of M words each. Therefore it shows the one to one correspondence between CM and DM modules. The operation is shown in the next figure step-by-step.
78
LINE COMMUNICATION SYSTEM
Fig. 8.24 Parallel in/parallel out TMTS (Step-by-step).
Fig. 8.25(a) Configuration of parallel in/parallel out switching.
79
TIME DIVISION SWITCHING
x = log2 N, y = log2 M Fig. 8.25(b) Contents of control memory.
As this is a three phased operation, Tts = 3Tm ...(18) Problem. Calculate the number of channels can be multiplexed per stream in parallel in/parallel out non-overlapped operation, if the access time of the memory modules is 60 ms. Here, 32 input trunks and 32 output trunks are being used. Solution. Using the equation (number 17b) we get Tts = M(N + 1)Tm or, 125 = M(33)0.06 or,
M=
125 = 63.13 (33 × 0.06)
i.e., 63 channels can be multiplexed
➢ COMBINATION SWITCHING Reviewing the concepts of time switching and space switching under analog time division switching it is clear that in TMSS line communication can be done between any two trunks; on the other hand in TMTS, i.e. employing the time slot interchange concept, line communication can be done between two different time slots-channels. In both the cases we have some lack of availability, therefore. Combination switching is the new concept, introducing which one can get full availability from the switching system, i.e any time slot channel (x, where 1 ≤ x ≤ M) from any trunk (i, where 1 ≤ i ≤ N) can communicate with other time slot channel (y, where 1 ≤ y ≤ M) of any trunk (j, where 1 ≤ j ≤ N). It can be done in two ways. According to the sequence applied in combination switching it is divided in following two parts :
1. Time Space Switching (TSCS) It is the type of combination switching where time switching (TSI) is done prior to the space switching. That is, we can say TSCS = Time switching + Space switching
80
LINE COMMUNICATION SYSTEM
Fig. 8.26 Time space combination switching.
In the Fig. 8.26, the time space combination switching is described. It is clear that the number of trunks is N and number of channel multiplexed in each trunk is M. Therefore total number of channels is MN. Let we call the ith input slot channel into the xth trunk is Ixi and jth output slot channel into the yth trunk is Oyj. Say 3rd time slot channel of 7th input trunk requests a communication between itself and 1st time slot channel of 9th output trunk. That means communication has to be made between I73 and O91. In the 1st step, time switching (TSI) is done. I73 channel changes only its time slot number from 3 to 1, i.e. I73 becomes I71. The next stage is space switching (N × N). In this stage two different things done. Firstly, input trunk becomes output trunk; and secondly, the trunk number can be changed here, i.e. I71 becomes O91. Finally the operation is completed. The two step operation is shown here. I73 ——→ I71 ——→ O91 (TSI)
(N × N)
2. Space Time Switching (STCS) It is the type of combination switching where time switching (TSI) is done after the space switching. That is, we can say STCS = Space switching + Time switching
81
TIME DIVISION SWITCHING
Fig. 8.27 Space Time Combination Switching.
In the Fig. 8.27, the space time combination switching is described. It is clear that the number of trunks is N and number of channel multiplexed in each trunk is M. Therefore total number of channels is MN. Let we call the ith input slot channel into the xth trunk is Ixi and jth output slot channel into the yth trunk is Oyj. Say 5th time slot channel of 8th input trunk requests a communication between itself and 2nd time slot channel of 1st output trunk. That means communication has to be made between I85 and O12. In the 1st step, is space switching (N × N). In this stage two different things done. Firstly, input trunk becomes output trunk; and secondly, the trunk number can be changed here, i.e. I85 becomes O15. in the next stage time switching (TSI) is done. O15 channel changes only its time slot number from 5 to 2, i.e. O15 becomes O12. Finally the operation is completed. The two step operation is shown here. I85 ——→ O15 ——→ O12 (N × N)
(TSI)
9 Frequency Division Switching 1. Concept 2. Frequency Division Multiplexing 3. Frequency Division Switching and Grouping ➢
Concept
As in case of frequency division switching is also dependent upon the frequency division multiplexing. In case of time division multiplexing, we have divided the signals according to the division of time scale. This is very easy to understand. But a signal may or may not contain a large number of frequencies. Then there must have some designator. That's why we need modulation by which we can change different waveforms from different channels to different predetermined frequencies. Actually the voice telephone in AT &T type L4 carrier system, FDS (frequency division switching) is of use. ➢ FREQUENCY DIVISION MULTIPLEXING
Fig. 9.1(a) Block diagram and i/p spectra of FDM.
82
FREQUENCY DIVISION SWITCHING
83
Fig. 9.1(b) Multiplexed spectrum.
As illustrated in the figure above, in FDM scheme, we are using different carrier frequency for modulation of different information signals. At first by using a low pass filter, we have truncated the upper sideband part and we are employing here SSB am modulation. As we are using three different carrier frequencies for three different information signals, the spectrum will be positioned according to the central frequency, i.e. the carrier frequency as shown in Fig. 9.1(a). Next the modulated signals are added to form xm(t) which is again to be modulated using carrier of frequency fc to form the transmittable signal x123(t) . As the Fig. 9.1(a) shows, the output is obtained in a same channel multiplexed in according to the division of time. Here to avoid the interference or unwanted coupling between two signals, named as cross talk, we may use guard band as shown in Fig. 9.1(a). ➢ FREQUENCY DIVISION SWITCHING AND GROUPING
Here we will describe frequency division switching and grouping taking the help of one practical example. In case of AT and T type L4 carrier system, there are 3600 channels each of bandwidth 4 kHz to be multiplexed to be transmitted together via a single cable. Modulation is done using the scheme of SSB, both USB and LSB (but not both at a time), and thus the final base band spectrum ranges from 0.5 to 17.5 MHz, including pilot carrier and guard band. As we use 4 kHz signal band width and a 512 kHz pilot carrier, excessive guard band width, which is actually a frequency occurrence of ‘loss region’, can be reduced by using group of channels. The following figure shows the first two stages of telephone multiplexing. The grouping multiplexer takes 12 i/p channels and deviate their to carrier frequencies 64, 68, 72, 76, ………, 108 kHz using LSB modulation scheme. Five such groups are again multiplexed to sub-carriers 420, 468, ……………, 612 kHz using LSB to form a super-group spectrum from 312 to 552 kHz. It is very important to note that suppressed sub-carrier for group 1 would fall into the group 3 spectrum. Similarly voice channel 1 will appear on the USB signal of 354 kHz, after grouping is done. Here the following figure will describe the concept of grouping more clearly. And the table of AT and T type FDM hierarchy shows the frequency range allocation for each stage of grouping in L4 system. We can also replace video or any other type of data of same frequency distribution instead of voice signal/ speech signal.
84
LINE COMMUNICATION SYSTEM
Fig. 9.2 Group and super-group formation of FDS.
Table 9.1 : Grouping hierarchy Designation
Frequency Range
Bandwidth
Voice Channel Number
Group
60 – 108 kHz
48 kHz
12
Super-group
312 – 552 kHz
240 kHz
5 × 12 = 60
Master group
564 – 3084 kHz
2520 kHz
60 × 10 = 600
Jumbo group
500 – 17500 kHz
17 MHz
600 × 6 = 3600
10 Transmission Line 1. Concept 2. Different Types of Transmission Lines 3. Equivalent Circuit of Transmission Line 4. Transmission Line Equations 5. Standing Wave 6. Voltage Standing Wave Ratio and Reflection Coefficient 7. Group Velocity and Phase Velocity ➢
Concept
When our objective is to communicate between two points through electrical energy, it can be done in one of the two ways namely by radiation of electromagnetic waves through free space or by the use of electrical conductors in any compatible form. A transmission line may be considered as a guide that guides the electrical energy to the desired destination. Transmission line may be classified as lumped lines and distributed line. In distributed lines, the electrical parameters like inductors, resistors, capacitors are distributed uniformly across the entire line length; whereas lumped lines have the parameters lumped at intervals along the line. Most of the transmission lines are of distributed type as they are easy to manufacture and have better characteristics than the lumped one. The commonly used transmission lines are open wire line, parallel or strap line, micro strip line and wave guide. An introduction of optical fiber in transmission line family is used for transmission of light signal. ➢ DIFFERENT TYPES OF TRANSMISSION LINES
Transmission lines are considered to be impedance matching circuits designed to deliver RF power from the transmitter to antenna and maximum signal from antenna to receiver. The parallel wire line is employed where balanced property of transmission line is required: in connecting a folded dipole antenna to a TV receiver or rhombic antenna to an HF antenna. The coaxial wire line is employed where balanced property of transmission line is not required; generally it makes the interconnection of broadcast transmitter to its grounded antenna to avoid radiation: in UHF and microwave frequencies it is employed. 85
86
LINE COMMUNICATION SYSTEM
(a) Co-axial cable
(b) Parallel wire line Fig. 10.1
➢ EQUIVALENT CIRCUIT OF TRANSMISSION LINE
Any transmission line can be treated as the circuit as in following figure. There is a conductance and a capacitance between the two wires separated by a dielectric. As it has a length, the communicating line must have some resistances and inductances. All the quantities (L, R, C and G) shown are proportional to the length of the line. Unless measured and quoted per unit length, they are meaningless.
Fig. 10.2
At RF, L > R and G > C, therefore the circuit can be drawn as Fig. 10.2. ➢ TRANSMISSION LINE EQUATIONS
The simplest transmission line configuration is open wire line consisting of a pair of conductors placed in parallel and separated by a dielectric. A small section of such a transmission line is used to deduce the transmission line equations.
Fig. 10.3 A small section of transmission line.
87
TRANSMISSION LINE
Assume a small section of transmission line which has a small length of dx. The value of measured voltage and current at point P are V and I respectively. After length dx at point Q the value of measured voltage and current are (V + dV) and (I + dI) respectively. If the primary constants of the transmission line are R, L, G and C, then the series resistance is (R + jωL) and shunt admittance is (G + jωC) measured in per unit. Therefore for dx length series resistance is (R + jωL)dx and shunt admittance is (G + jωC)dx. The potential drop across the transmission line is given by V – (V + dV) = (R + jωL) I × dx or, – dV = (R + jωL)I × dx or,
–
dV = I(R+ jωL) dx
...(1)
Similarly the change in current due to the shunt admittance may be determined by the following analysis : I – (I + dI) = (G + jωC)V × dx or, – dI = (G + jωC)V × dx or,
dI = V(G + jωC) dx
–
...(2)
Now, differentiating both sides of equation (2) w.r.t. x –
d 2V dx
2
d 2V
or,
dx 2
= (R + jωL)
dI dx
= (R + jωL)(G + jωC)V (putting the value of dI/dx from eq. 2)
d 2V
or,
dx 2
= γ2V
...(3) (where, γ =
( R + jωL )(G + jω C ) )
Now, differentiating both sides of equation (2) w.r.t. –
or,
or,
d 2I dx
2
d 2I dx 2
d 2I dx 2
=
(G + j ωC )dV dx
= (R + jωL)(G + jωC)I
= γ2I
(putting the value of dI/dx from eq. 1)
...(4)
88
LINE COMMUNICATION SYSTEM
The term (R + jωL)(G + jωC) is a complex quantity having real and imaginary parts. If the term is denoted by γ2 , then
or,
γ=
( R + jωL )(G + jωC )
γ=
( RC − ω2 LC ) + j ω(CR + LG )
The real and imaginary parts of γ are usually designated as attenuation constant (α) and phase constant (β) respectively. Therefore, γ = α + jβ =
( RC − ω2 LC ) + j ω(CR + LG )
...(5)
The two equations (3) and (4) when solved, gives the following results ...(6) V = A e– γx + B eγx – γx γx +De ...(7) I=Ce Where, A, B, C, D are constants and have values that depend upon the conditions existing in the line. From the equations (6) and (7) it is very clear that the 1st term has the maximum value at source, as x increases, i.e. as wave propagates from source towards load, the 1st term value decreases. That component is called as incident component. Similarly, the 2nd term has the maximum value at load, as x increases in reverse direction, i.e. as wave propagates from load towards source, the 2nd term value decreases. That component is called as reflected component.
CASE I: Transmission Line of Infinite Length Assume a transmission line hypothetically of length infinite. It is clear that the voltage and current value will diminish to zero at the distant end (∞), if a signal source is connected. Such a line will therefore have no reflected component of signal (voltage/current). If IS is the source current, the at distance x = 0, i.e. at source end I = IS Substituting the boundary condition x = 0, I = IS at equation (7) we get IS = C Substituting the boundary condition x = ∞, I = 0 at equation (7) we get 0=D Putting the values in equation (7) ...(8) I = IS e– γx – αx – jβx = IS e e Differentiation equation (8) w.r.t. x and euating with equation (2) dI = – IS γ e –γx dx
= – V (G + jωC)
or,
Now, if x = 0, V = VS VS (G + jωC) = IS γ (e –γx)x = 0 VS (G + jωC) = IS γ
89
TRANSMISSION LINE
VS γ = IS (G + j ω C )
or,
=
(R + jωL ) (G + j ω C )
[putting the value of γ]
...(9) = Z0 The term Z0 now can be treated as the input impedance of the infinite length transmission line and is commonly referred to as Characteristic Impedance. It is defined as the input impedance of a transmission line of length infinite.
CASE II: Transmission Line Terminated in a Load Impedance ZR Let’s consider the case of a transmission line terminated in a load impedance ZR. Let VS and IS are the sending voltage and current respectively. If the characteristics impedance is Z0, the voltage (V) and current (I) × distance apart from source end are to be determined. Using equations (6) V = A e–γx + B eγx = A (cosh γx – sinh γx) + B(cosh γx + sinh γx) = (A + B) cosh γx – (A – B) sinh γx = M cosh γx – N sinh γx ...(10) [Taking M = (A + B), N = (A + B)] or,
dV = γ(M sinh γx – N cosh γx) dx Now, from eq. (1) dV = – I(R + jωL) dx
i.e.
γ(N cosh γx – M sinh γx) = I(R + jωL) γ I= (N cosh γx – M sinh γx) ( R + jωL)
or,
=
(N cosh γx − M sinh γx ) Z0
...(11)
At x = 0, I = IS, V = VS, cosh γx = 1 and sinh γx = 0 Therefore VS = M IS =
N Z0
Substituting the values into equation number (10) and (11) we get V = VS cosh γx – ISZ0 sinh γx I = IS cosh γx – VS/Z0 sinh γx
...(12) ...(13)
90
LINE COMMUNICATION SYSTEM
The above equations (i.e. equation (12) and (13)) are general in nature and can be applied for any values of ZR and any distance x.
CASE III: Input impedance of a transmission line terminated in a load impedance ZR Let’s consider the case of a transmission line terminated in a load impedance ZR. Let VS and IS are the sending end voltage and current respectively. If the length of the line is assumed to be l, then at x = l, I = IR and V = VR such that ZR =
VR IR
Therefore, from equation (12) and (13) VR = VS cosh γ1 – ISZ0 sinh γl IR = IS cosh γl –
IR =
But
VS sinh γl Z0
VR ZR
Therefore IS cosh γl –
(VS cosh ã1 I S Z0 sinh ã1) VS sinh γl = ZR Z0
or
IS(cosh γl +
cosh γ1 + 1 Z0 sinh γl) = VS ZR Z 0 sinh γ1
or
input impedance = ZS cosh γ1 + Z 0 VS Z R sinh γ1 = = IS 1 1 + Z R cosh γ1 Z 0 sinh γ1 ( Z cosh γ1 + Z 0 sinh γ1) = Z0 R ( Z 0 cosh γ1 + Z R sinh γ1)
...(14)
CASE IV: Input Impedance of a Transmission Line of Length l Terminated in a Load Impedance Z0 In this case ZR = Z0 Therefore using equation (14) we get the input impedance of the line is ( Z cosh γ1 + Z 0 sinh γ1) Zin = Z0 0 = Z0 ( Z 0 cosh γ1 + Z R sinh γ1)
...(15)
91
TRANSMISSION LINE
From equation (15) it is very clear that input impedance of a finite length transmission line can be defined as the characteristics impedance if the line is terminated at Z0. It is another definition of characteristics impedance. It proves again the maximum power transfer theorem (individually voltage and current transfer) i.e. maximum power can be transferred if the load impedance is equal to the input impedance of the network. Here if the load impedance is equal to the characteristics impedance of the transmission line, the total current and voltage signals are absorbed, no reflected signal is obtained.
CASE V: Input Impedance of a Transmission Line Terminated in a Short-Circuit If the load is short-circuited, then ZR = 0, Then from equation (14) we get Input impedance = Zin (Z cosh γ1 + Z0 sinh γ1) = Z0 0 (Z0 cosh γ1 + ZR sinh γ1) (0 + Z0 sinh γ1) = Z0 (Z0 cosh γ1 + 0) = Z0 tanh γ1 = Zs.c.
...(16)
CASE VI: Input Impedance of a Transmission Line Terminated in a Open-Circuit If the load is short-circuited, then ZR = ∞, IR = 0 Then from equation (13) putting x = l, I = IR= 0 we get 0 = Iscosh γl –
ZS =
VS sinh γl Z0
VS = Z0 coth γl IS
Input impedance = Zin = ZS = Z0 coth γl = Zo.c
...(17)
CASE VII: Determination of Z0 and γ using Zs.c. and Zo.c. The above two equations (16) and (17) represents the open circuit and short circuit impedances. Using those two equations we get Zo.c × Zs.c = Z0 coth γl × Z0 tanh γl = (Z0)2 × 1 Therefore, And
or,
Z0 =
(Zo.c. × Zs.c. )
...(18)
Z 0 coth γ1 Z o.c. = Z 0 tanh γ1 Z s.c.
tanh2 γl =
Z s.c. Z o.c.
...(19)
92
LINE COMMUNICATION SYSTEM
The equations (18) and (19) respectively represents the deterministic equations for characteristics impedance (Z0) and propagation constant (γ). ➢ STANDING WAVE
Say the incident wave is given by (– x direction) Ey0 = E0 sin (ωt-βx)
...(20)
Fig. 10.4 A small section of transmission line.
And the reflected wave is given by (+ x direction) ...(21) Ey1 = E1 sin (ωt – βx + δ) δ = Time lead/phase shift of Ey1 with respect to Ey0 at the point of reflection (x = 0) E0 and E1 are the amplitudes of incident and reflected waves respectively. Now, adding equation (20) and (21), for δ = 0° or 180° we get Ey = Ey0 + Ey1 = E0 sin(ωt – βx) + E1 sin(ωt – βx) ...(22) = (E0 + E1)sin ωt cos βx + (E0 – E1)cos ωt sin βx CASE I: If the load end is open or short circuited, the amplitudes of reflected and incidence wave would be equal. If x = 0 is taken as boundary, the boundary relation for the tangential component of E requires that Ey = 0 0 = (E0 + E1) | sin ωt |max = (E0+ E1) (at boundary δ = 180°) Therefore E0 = – E1 Then the equation (22) becomes Ey = 2E0cos ωt sin βx ...(23) Now we can again write the above equation for time = constant and for x scale as Ey = K sin βx, when K = 2E0 cos ωt = constant. That is the equation represents a wave which is stationary in space. A stationary wave of this type is a pure standing wave.
93
TRANSMISSION LINE
Fig. 10.5 Standing wave pattern for VSWR = ∞.
As you know from the maximum power transfer theorem that maximum power can be transferred from source to destination if the load impedance is equal to the total equivalent internal impedance of the source. Similarly in transmission line if the load end is terminated by impedance equal to the characteristics impedance (Z0), then only total current or voltage signal is absorbed, no parts are reflected. As the mismatch between load impedance (Zl) and Z0 increases, the amount of reflected wave amplitude increases and amount of absorbed signal amplitude decreases. Maximum mismatch is Zl = 0 (i.e. short circuit) or Zl = ∞ (i.e. open circuit). Voltage standing wave ratio (VSWR) is defined as the following ways VSWR(ρ) =
Z1 Vmax = Z0 Vmin
Where, Vmax and Vmin are the maximum and minimum values of voltage signal at any distance x from source. And reflection coefficient (r.c.) =
|Vref | |Vinc |
Where, Vref and Vinc are the reflected and incidence values of voltage signal at any distance x from source. Therefore ρ =
|Vinc +Vref | (1 + | r.c. |) Vmax = = (1 − | r.c. |) |Vinc – Vref | Vmin
From equation 23 putting different values of t we get, t = 0, Ey = 2E0 cos ω(0) sinβx = 2E0 sin βx t=
T T Π , E = 2E0 cos ω sin βx = 2E0 cos sin βx = √2E0 sin βx 8 y 8 4
t=
T T Π , Ey = 2E0 cos ω sin βx = 2E0 cos sin βx = 0 4 2 4
94
LINE COMMUNICATION SYSTEM
t=
3T 3T 3Π , Ey = 2E0 cos ω sin βx = 2E0 cos sin βx = – √2E0 sin βx 8 4 8
t=
T T , E = 2E0 cos ω sin βx = 2E0 cos Π sin βx = – 2E0 sin βx 2 y 2
It is to be noted from the above equations and the figure obtained from them (Fig. 10.3) that a constant phase point (P) doesn’t move in x direction but remains at fixed position as time passes. CASE II: Let’s now examine the condition that E1 = – 0.5E0 Then the equation (22) becomes Ey = (0.5E0)sin ωt cos βx + (1.5E0)cos ωt sin βx Ey = 0.5E0 (3cos ωt sin βx + sin ωt cos βx) ...(24) Now we can again write the above equation for time = constant and for x scale as Ey = K1 cos ωt sin βx + K2 sin ωt cos βx ,when K1 and K2 are constants. From equation 24 putting different values of t we get, t = 0, Ey = 0.5E0 (3 cos 0 sin βx + 0 cos βx) = 1.5E0 sin βx Π Π 3 cos sin βx + sin cos βx = – 1.06 sin βx +0.35 cos βx 4 4
t=
T , E = 0.5E0 8 y
t=
T , E = 0.5E0 (cos βx) 4 y
t=
3T , Ey = 0.5E0 8
t=
T , E = – 0.5E0 (3 cos 0 sin βx) = – 1.5E0 sin βx 2 y
Π Π – 3 cos sin β x + sin cos β x = – 1.06 sin βx + 0.35cosβx 4 4
Fig. 10.6 Standing wave pattern for VSWR = 3.
95
TRANSMISSION LINE
It is to be noted from the above figure that a constant phase point (P) moves in -x direction. Here the standing wave pattern is not sharp instead the pattern envelop is treated as the pattern boundary i.e. the envelop is stationary. That’s why the minimum value of minimum amplitude of voltage is not equal to zero.
➢ VOLTAGE STANDING WAVE RATIO AND REFLECTION COEFFICIENT As you know from the maximum power transfer theorem that maximum power can be transferred from source to destination if the load impedance is equal to the total equivalent internal impedance of the source. Similarly in transmission line if the load end is terminated by impedance equal to the characteristics impedance (Z0), then only total current or voltage signal is absorbed, no parts are reflected. As the mismatch between load impedance (Zl) and Z0 increases, the amount of reflected wave amplitude increases and amount of absorbed signal amplitude decreases. Maximum mismatch is Zl = 0 (i.e., short circuit) or Zl = ∞ (i.e., open circuit). Voltage standing wave ratio (VSWR) is defined as the following ways VSWR(ρ) =
Z1 Vmax Z = or 0 which one is greater. Z0 Vmin Z1
Where, Vmax and Vmin are the maximum and minimum values of voltage signal at any distance x from source. And reflection coefficient (Γ) =
|Vref | |Vinc |
Where, Vref and Vinc are the reflected and incidence values of voltage signal at any distance x from source From the above case II, Γ = Therefore ρ =
1 2
|Vinc +Vref | Vmax (1 + | Γ |) = = Vmin (1 – | Γ |) |Vinc – Vref |
From the above case II, ρ =
(1 + | Γ |) =3 (1 – | Γ |)
Actually VSWR is directly proportional to the mismatch between load impedance (Zl) and Z0. From the VSWR definition it is clear that ∞ ≥ ρ ≥ 1, and 1 ≥ Γ ≥ 0. From the standing wave patterns, we got some distance relations between voltage node (Vmin), voltage anti-node (Vmax ), current node (Imin) and current anti-node (Imax). Distance between Vmin and Vmax = distance between Vmin and Imin = distance between Vmax and Imax = distance between Imax and Imin =
λ and = distance between Vmin and next Vmin = distance between Imin and next Imin = distance 4
between Vmax and next Vmax =
λ . 2
96
LINE COMMUNICATION SYSTEM
This is the proof of the above statement. From the Fig.(10.6) we got 1st Vmax at point P and next Vmax at point Q. to traverse the distance from P to Q signal needs a time of
T . 2
Signal takes T time to traverse a distance = λ Therefore, it takes
T λ time to traverse a distance = = distance between two nearest Vmax s. 2 2
Similarly we can say that the distance between Vmin and Vmax =
Fig. 10.7
λ . 4
97
TRANSMISSION LINE
➢ GROUP VELOCITY AND PHASE VELOCITY Let us assume that the equation of the stationary or standing wave (formed by adding incidence and reflected wave) is given by the equation as follows: Vy = Vinc cos (ω0t – βx) Now, if the signal actually holds two different frequencies instead of only one frequency, then say the frequencies are (ω0 + ∆ω) and (ω0 – ∆ω) . The corresponding β values are (β + ∆β) corresponding to (ω0 + ∆ω) and (β – ∆β) corresponding to (ω0 – ∆ω) Then the equation with 1st frequency component is Vy1 = Vinc cos [(ω0 + ∆ω)t – (β + ∆β)x] ...(20) And the equation with 2nd frequency component is Vy2 = Vinc cos [(ω0 – ∆ω)t – (β – ∆β)x] ...(21) Now adding equation (20) and (21) we can get the exact value of Vy with two different frequencies without the loss of generality. Vy = Vy1 + Vy2 = 2 Vinc cos (ω0t – βx) cos (∆ωt – ∆ βx) ...(22) The equation (22) shows clearly that the resultant signal consisting of a superposition of slow varying signal over a more rapid one, i.e., it gives birth to beats. As the resultant signal consisting of two different frequency components (ω0 + ∆ω) and (ω0 – ∆ω), it is therefore very similar to that of the double side band amplitude modulated signal with carrier frequency ω0 and modulating signal ∆ω with the carrier suppressed (DSBSC). Now if the 1st cosine term angle, i.e. (ω0t – βx) = constant for constant phase point (P), then ω dx = 0 = ν = Phase Velocity β dt
...(23)
From figure (4) and (5) it is clear that if the standing wave is plotted with respect to time axis, the constant phase point (P) will propagate along the positive x axis and the velocity of that particular propagation must be equal to that of the change of frequency ω0, i.e.,
ω0 . Therefore the phase velocity β
can be defined by the above equation. Similarly, if the 2nd cosine term angle, i.e. (∆ωt – ∆ βx) = constant, then ∆ω0 dx = = υ = Group Velocity ∆β dt
...(24)
98
LINE COMMUNICATION SYSTEM
Fig. 10.8 Standing wave showing one group.
From, the figure above the group velocity concept will be discussed here. Though the signal consists of two different frequency signals, hence it shows a repeating pattern. One group is repeating through the entire positive x direction from source to destination. Now the pattern is apparently stationary, but it is obvious that if we plot the signal with respect to time axis, the signal group will propagate and it is very clear that the velocity of that particular propagation must be equal to the low frequency signal amplitude change, i.e.
∆ω0 . That’s why the group velocity can be defined by the above equation. ∆β
11 Telephone Network 1. Concept 2. Subscriber Loop System 3. Switching Hierarchy 4. Transmission System : Radio Wave Propagation 5. Numbering Scheme 6. Charging Plan 7. Signalling Technique ➢
Concept
The concept of telephone network is completely based upon the operating principle of circuit switching. The 1st step of telephone communication is the establishment of network. 2nd step is the completion of communication. And finally demolition of the physical circuit/ network. Moreover, in case of telephone communication we need two different signals, speech and control signal, and alongwith power. We will learn here the network part. This chapter will follow how the network is designed using hierarchical manner, how charging planes are made and how the telephone network can be designated using the basis communicating channels supported for speech signal and control signal. In another chapter we will see how the routing is done and what are its supporting algorithms. ➢ SUBSCRIBER LOOP SYSTEM
The connection between every telephone subscriber with the near by switching office is maintained by a dedicated pair of wires. Apparently it looks very simple connection but there are a lot of stages to be connected for the establishment between exchange & subscriber. The stages are shown below in tabular form. As for an example, if a subscriber moves his house to a nearby area served by same exchange but a different distribution point, he can be permitted to retain same telephone number by a suitable jumper connection at MDF. Distribution & feeder points should have flexible cross point connection capability. Three problems may arise due to above occurrence: 99
100
LINE COMMUNICATION SYSTEM
(a) cost of drop wires increases as length increases (b) introduction of delay due to drop wire length (c) probability of frequently disconnection increases
Fig. 11.1 MDF : main distribution frame; MF : main feeder; FP : feeder point; BF : branch feeder; DP : distribution point; DC : distribution cable; DW : drop wires
Fig. 11.2 Cable hierarchy for subscriber loops
101
TELEPHONE NETWORK
Technical Specification for Subscriber Line GAUGE NO
DIAMETER
D.C RESISTANCE
ATTENUATION
[ AWG ]
[ mm ]
[ Ohm/Km ]
[ Db/Km ]
19
0.91
26.40
0.71
22
0.64
52.95
1.01
24
0.51
84.22
1.27
26
0.41
133.89
1.61
❖ Subscriber Loop Interface. Several functions are performed by the exchange for two different transmission path (voice & signaling) through subscriber line. An interface which performs these activities in Exchange end is called as Subscriber Loop Interface. The complete functions are designated by a single word as BORSCHT B ---- battery feed O ---- overvoltage protection R ---- ringing S ---- supervision C ---- coding H ---- hybrid T ----- test ❖ Overvoltage protection is for protection from lightning & open circuit condition of transmission line. Supervision is needed for detection of off-hook condition. Coding is necessary for digital exchange (EWSD, OCB- 283, CDOT, E10B etc) for analog to digital conversion, similarly for digital to analog decoding is needed. Hybrid is for conversion of two wires to four wires. Subscriber side is two wire connections. Again digital exchange - exchange needs four wire connections, a transformer based hybrid circuit is employed to make the entire signal coupling free.
Fig. 11.3
102
LINE COMMUNICATION SYSTEM
Two-wire connection between subscriber and exchange, four-wire between exchanges. Nowadays, the junction between two-wire and four-wire resides in the local exchange or in a point between the subscriber and the local exchange; that is, in a remote subscriber stage or multiplexer, where the signal is converted from an analog to a digital signal. The hybrid equipment is used to separate the two voice directions of the two-wire connection so that they can be applied to the two pairs of a four-wire connection. Earlier equipment was generally composed of two differential transformers and a line balance, as shown below:
Fig. 11.4
Today with the invention of VLSI technology it is very much simpler & economical to use a single chip to perform all the interfacing functions instead of individual circuitry like A-D, D-A converter, hybrid network circuit etc. The single chip is called Subscriber Loop Interface Circuit (SLIC). ➢ SWITCHING HIERARCHY
Fig. 11.5
103
TELEPHONE NETWORK
Looking towards the increasing number of subscribers, it is very easy to realize that the entire switching arrangement cannot be controlled by only a single switching point. That's why we need a switching hierarchy. Very firstly, there are local switching points or end points. Each end point is directly connected to the subscribers' local to that corresponding exchange using subscriber looping arrangement. Next the end points are to be connected internally also. This connection is done by the trunk between two end offices. A number of end offices are there under each toll office. It means in hierarchical order, toll office is on the upper step of the end office. End offices transfer their subscribers' calls to the toll office. Charging plan is done exclusively in the toll offices. By TMSS (time multiplexed space switching) and TMTS (time multiplexed time switching) the signals/calls transferred from one toll office to other. A very populous area may have a large number of end offices. That's why much of the traffic will involve calls to different end offices. To facilitate this end office to end office traffic a tandem switching office is provided. Call between subscribers under different end offices in same city will go by way of a tandem office. ➢ TRANSMISSION SYSTEM: RADIO WAVE PROPAGATION
In an earth environment, electromagnetic waves propagate in ways that depend not only on their own properties but also on those of the environment itself. Waves travel in straight lines, except where the earth and the atmosphere alter their path. Waves with frequencies above HF travel in a straight line. They propagate by means of so-called space waves. They are sometimes called troposphereic waves, since they travel in the troposphere, the portion of the atmosphere closest to the ground. Frequencies below the HF range travel around the curvature of the earth. They are called ground waves or surfaces waves. All broadcast radio signals received in daytime propagate by means of surface waves. Waves in the HF range are reflected by the ionized layers of the atmosphere and are called sky waves. Such signals are beamed into the sky and come down again after reflection, returning to earth well beyond the horizon.
Ground Waves Ground waves progress along the surface of the earth and must be vertically polarised to prevent short circuiting the electric component. A wave induces currents in the ground over which it passes and thus loses some energy by absorption.
Fig. 11.6 Ground Waves
104
LINE COMMUNICATION SYSTEM
The surface wave is also attenuated by diffraction as shown above. The wave front gradually tilts over as the wave propagates. The increasing tilt causes greater short circuiting of the electric field component of the wave and hence field strength reduction.
Sky-Wave Propagation - The Ionosphere The ionosphere is the upper portion of the atmosphere, which absorbs large quantities of radiant energy from the sun, becoming heated and ionized. This results in four main layers D, E, F1 and F2, in ascending order as shown. The last two combine at night to form one single layer.
Terms and Definitions The virtual height of an ionospheric layer is best understood with the aid of the Fig. 11.7.
Fig. 11.7 Virtual Height of Ionosphere
Critical frequency for a given layer is the highest frequency that will be returned down to earth by that layer after having been beamed straight up at it. Skip distance is the shortest distance from the transmitter at which a sky wave of fixed frequency will be returned to earth. In detail we will discuss this wave propagation part in "wave propagation" chapter.
Tropospheric Scattering When a radio wave passing through the troposphere meets turbulence, it makes an abrupt change in velocity. This causes a small amount of the energy to be scattered in a forward direction and returned to Earth at distances beyond the horizon. This phenomenon is repeated as the radio wave meets other turbulences in its path. The total received signal is an accumulation of the energy received from each of the turbulences. This scattering mode of propagation enables vhf and uhf signals to be transmitted far beyond the normal line-of-sight. To better understand how these signals are transmitted over greater distances, you must first consider the propagation characteristics of the space wave used in vhf and uhf line-of-sight communications. When the space wave is transmitted, it undergoes very little attenuation within the line-of-sight horizon. When it reaches the horizon, the wave is diffracted and follows the Earth’s curvature. Beyond the horizon, the rate of attenuation increases very rapidly and signals soon
105
TELEPHONE NETWORK
become very weak and unusable. Tropospheric scattering, on the other hand, provides a usable signal at distances beyond the point where the diffracted space wave drops to an unusable level. This is because of the height at which scattering takes place. The turbulence that causes the scattering can be visualized as a relay station located above the horizon; it receives the transmitted energy and then reradiates it in a forward direction to some point beyond the line-of-sight distance. A high gain receiving antenna aimed toward this scattered energy can then capture it. The magnitude of the received signal depends on the number of turbulences causing scatter in the desired direction and the gain of the receiving antenna. The scatter area used for tropospheric scatter is known as the scatter volume. The angle at which the receiving antenna must be aimed to capture the scattered energy is called the scatter angle. The scatter volume and scatter angle are shown in Fig. 11.8.
Fig. 11.8 Tropospheric scattering propagation
➢ NUMBERING SCHEME
The main object of numbering plan to make uniformity for the identification of individual subscriber connected to a telecommunication network. In earlier days exchanges were identified by the name of cities or towns, but as the subscriber density increases number of exchange also rises . Then it becomes very much difficult to maintain all the subscribers. It needs different identification which highlights the introduction of numbering plan all over the world. The common numbering plan is sometimes called as Linked Numbering Scheme. Previously centrally located large Exchange in a City or Town serving all the main business centers is known as Main Exchange. Different small Exchanges serving the needs of locality residents are Satellite Exchange. The area containing both the Main Exchange & Satellite Exchange is called as Multiexchange Area.
106
LINE COMMUNICATION SYSTEM
Fig. 11.9
International Dialing Codes As per CCITT recommendations the entire world has been divided into 9 different zones on the basis of International Subscriber Dialing (ISD) 1. North America 2. Africa 3 & 4. Europe 5. South America 6. Australia 7. Russia and East Europe 8. Far East 9. Middle East In zone 9 India has country code 91, Pakistan 92, Bahrain 973, UAE 971 like that in zone 6 Australia 61, New Zealand 64, Fizi island 679.
Structure of Full International Number
Fig. 11.10
107
TELEPHONE NETWORK
Any country can be subdivided into several telecom zones on the basis of Subscriber Trunk Dialing (STD) or Direct Distance Dialing (DDD). INDIA is divided into 8 telecom region, In zone 3 : Kolkata code 33;Asansol code 341;Izol code 3832 In zone 8 : Bangalore code 80;Mysore code 821;Warrangal code 8712.
Four Possible Approaches to Dialing Procedures NUMBER
APPROACH PROCEDURE
Procedure 1
Single uniform procedure for all calls like local, national (STD), international (ISD)
Procedure 2
Two different procedures, one for ISD calls & another for local & STD calls
Procedure 3
Three different procedures one for ISD calls, one for STD calls, one for local calls
Procedure 4
Four procedures, three as above & fourth is calls for adjacent numbering areas
➢ CHARGING PLAN
The charging plan provides for recovering both capital costs involved in buying the resources like switching equipments, optical fibre communication network elements etc, and the operating costs from subscribers.The cost of dedicated resources like telephone instrument and subscriber line from nearby switching offices must be recovered from individual customers by monthly or bimonthly or quarterly or half yearly or annual basis along with call charges. Mainly following are the different stages for charging a subscriber: (a) an initial charge for providing a network connection (b) a rental or leasing charge (c) charges for individual calls made (a) No. of local calls over a duration (b) No. of STD calls [ if any ] (c) No. of ISD calls [ if any ] (d) overall service tax @ some percentage as per Govt. rules
Tariff Variation in INDIA PERIOD OF DAY
METER PULSE REPETITION RATE
08.00 Hrs ---------- 19.00 Hrs
X
19.00 Hrs ---------- 22.00 Hrs
X/2
22.00 Hrs ---------- 06.00 Hrs
X/4
06.00 Hrs ---------- 08.00 Hrs
X/2
108
LINE COMMUNICATION SYSTEM
Distance vs Metering Pulse Rate in INDIA DISTANCE
METERING PULSE RATE
[ Km ]
[ pulse/minute ]
20 -------------- 50
1.67
50 -------------- 100
5.00
100 ------------- 200
7.50
200 ------------- 500
15.00
500 ------------- 1000
20.00
> 1000
30.00
➢ SIGNALLING TECHNIQUE
The signaling functions of a telephone network refer to the means for transferring related control information between the various terminals, switching nodes and users of the network Signaling functions can be broadly categorised as belonging to one or two types: Supervisory or information bearing. Supervisory signals convey status or control of network elements. e.g. request for service (off-hook), ready to receive (dial tone), call alerting (ringing), call termination (on-hook), and busy tones. Information bearing signals include: called party address, calling party address, toll charges. In addition to call related signaling functions, switching nodes communicate between themselves and network control centres to provide certain functions related to network management. Network related signals may convey status such as maintenance test signals, all lines busy, equipment failures or they may contain information related to routing and flow control.
Subscriber Loop Signalling It depends upon the type of telephone instrument used like : DTMF telephone set or ROTARY DIAL telephone set. There are a lot of upgraded facilities available with DTMF telephone like: answering machine service, ISDN etc.
Intraexchange or Register Signalling It depends upon entirely the type & design of switching system used by the local exchange of subscriber. The signaling system differs for analog switching system like : STROWGER & CROSSBAR and digital switching system like : E10B, OCB - 283.
Interexchange or Interregister Signalling It takes place between several exchanges with common control subsystem. Here address digits transfer takes place from originating exchange to terminating exchange.
109
TELEPHONE NETWORK
Classification Signaling is achieved using one of two basic techniques: (a) In-channel signaling (b) Common channel signaling In-channel signaling uses the same transmission facilities or channel for signaling as for voice. In-band systems transmit the signaling information in the same band of frequencies used by the voice signal. The advantage is that it can be used on any transmission medium. The disadvantage is that mutual interference between signaling and voice information is possible (speech might activate signaling). Example is multi-frequency (MF) signaling between exchanges. Common Channel Signaling uses one channel for all signaling functions of a group of voice channels. It uses a dedicated data link between the stored program control elements (processors) of the switching systems. The data link sends messages that identify channels (trunks) and events related to that channel. Some advantages of common channel signaling are : • Only one set of signaling facilities needed for each associated trunk group. • Separate channels used for voice and signaling thus no interference • Control channel inaccessible to user, source of fraud eliminated • Multiple connections can be set up rapidly as forwarding of control information can overlap a circuit set up through the node • Common channel signaling (see Fig. 11.11).
Fig. 11.11 Common Channel Signaling
110
LINE COMMUNICATION SYSTEM
Fig. 11.12
INCHANNEL SIGNALLING
COMMON CHANNEL SIGNALLING
1. speechtrunk required during control signal
1. no requirements of speech trunk
2. interference of voice & control signal
2. no chance of interference
3. expensive due to requirement of separate
3. economical due to requirement of set of
signaling trunk for each group.
signaling equipment for whole group.
4. potential misuse by subscriber
4. no potential misuse by subscriber
5. relatively slow
5. relatively fast
6. speech circuit reliability is assured
6. no automatic test of speech circuit
7. difficult to change or add signal
7. more flexibility
8. difficult to handle signal during speech
8. easy to handle
9. relatively not critical
9. relatively critical
111
TELEPHONE NETWORK
CHANNEL ASSOCIATED SIGNALLING
CHANNEL NON ASSOCIATED SIGNALLING
Channel Associated Signalling Speech path & corresponding signalling path same e.g. : in above fig. speech path : A-B, A-C-B, B-D signalling path : A-B, A-C-B, B-D
Channel Nonassociated Signalling Speech path & corresponding signalling path different e.g. in above fig speech path : A-B, B-C Signalling path : A-C-D-B, B-D-C
Quasiassociated Signalling is the combination of both channel associated & channel non associated signalling path for a single speech path.
Fig. 11.13
12 Routing in Telephony 1. Concept 2. Classification of Routing 3. Centralized Dynamic Routing 4. Distributed Dynamic Routing 5. Alternate Routing 6. Alternate Routing Diagram 7. Routing Strategies 8. Tabular Reprepsentation of Fixed Routing 9. Example of Flooding 10. Least Cost Algorithm 11. Comparison Between Two Algorithms ➢
Concept Routing Routing is a process through which message can traverse from source to destination. Routes Routes are the selected paths through which messages traverse. Router Router is the device by which routing procedure takes place.
112
113
ROUTING IN TELEPHONY
➢ CLASSIFICATION OF ROUTING
Fig. 12.1
Routing in Circuit Switched Network • Many connections will need paths through more than one switch • Need to find a route • Efficiency • Resilience • Public telephone switches are a tree structure • Static routing uses the same approach all the time • Dynamic routing allows for changes in routing depending on traffic • Uses a peer structure for nodes
Routing in Packet Switched Network • Complex, crucial aspect of packet switched networks • Characteristics required • Correctness • Simplicity • Robustness • Stability • Fairness • Optimality • Efficiency
114
LINE COMMUNICATION SYSTEM
➢ CENTRALIZED DYNAMIC ROUTING
Controller will decide the free path for message traversal from source to destination.
Fig. 12.2 Centralized Dynamic Routing
Problems: (i) link between source and controller fails (ii) controller itself fails (iii) any intermediate node or concentrator fails ➢ DISTRIBUTED DYNAMIC ROUTING
Fig. 12.3 Distributed Dynamic Routing
Own exchange routing or distributed routing allows alternative routes to be chosen at intermediate nodes or concentrator decided by any intermediate controller. When new exchanges are introduced little modifications are required in the switch, this is advantageous. Moreover if one intermediate node or link fails the message will traverse through a newly established route coming out from predefined route. If one controller fails other in the group shares the load of faulty controller.
115
ROUTING IN TELEPHONY
➢ ALTERNATE ROUTING
• Possible routes between end offices predefined • Originating switch selects appropriate route • Routes listed in preference order Different sets of routes may be used at different times. ➢ ALTERNATE ROUTING DIAGRAM
Fig. 12.4(a) Topology Time Period
First route
Second route
Third route
Fourth and final route
Morning
a
b
c
d
Afternooon
a
d
b
c
Evening
a
d
c
b
Weekend
a
c
b
d
Fig. 12.4(b) Routing Table.
Performance Criteria (i) Used for selection of route (ii) Minimum hop (iii) Least cost Optimalitity & Fairness are opposite in nature; If priority is compared w.r.t distance this is unfair.
116
LINE COMMUNICATION SYSTEM
Network Information of Routing • Routing decisions usually based on knowledge of network (not always) • Distributed routing • Nodes use local knowledge • May collect info from adjacent nodes • May collect info from all nodes on a potential route • Central routing • Collect info from all nodes • Update timing • When is network info held by nodes updated • Fixed-never updated • Adaptive-regular updates ➢ ROUTING STRATEGIES
• Fixed • Flooding • Random • Adaptive
Fixed Routing Criteria • Single permanent route for each source to destination pair • Determine routes using a least cost algorithm • Route fixed, at least until a change in network topology
Flooding Routing Criteria • No network info required • Packet sent by node to every neighbor • Incoming packets retransmitted on every link except incoming link • Eventually a number of copies will arrive at destination • Each packet is uniquely numbered so duplicates can be discarded • Nodes can remember packets already forwarded to keep network load in bounds • Can include a hop count in packets • All possible routes are tried • Very robust • At least one packet will have taken minimum hop count route • Can be used to set up virtual circuit • All nodes are visited • Useful to distribute information (e.g. routing)
ROUTING IN TELEPHONY
Random Routing Criteria • Node selects one outgoing path for retransmission of incoming packet • Selection can be random or round robin • Can select outgoing path based on probability calculation • No network info needed • Route is typically not least cost nor minimum hop
Adaptive Routing Criteria • Used by almost all packet switching networks • Routing decisions change as conditions on the network change • Failure • Congestion • Requires info about network • Decisions more complex • Tradeoff between quality of network info and overhead • Reacting too quickly can cause oscillation • Too slowly to be relevant ➢ TABULAR REPRESENTATION OF FIXED ROUTING
117
118
LINE COMMUNICATION SYSTEM
Node 1 Directory Destination Next Node
Node 2 Directory Destination Next Node
Node 3 Directory Destination Next Node
Node 4 Directory Destination Next Node
Node 5 Directory Destination Next Node
Node 6 Directory Destination Next Node
EXAMPLE OF FLOODING
119
ROUTING IN TELEPHONY
Fig. 12.5
➢ LEAST COST ALGORITHM
• Basis for routing decisions • Can minimize hop with each link cost 1 • Can have link value inversely proportional to capacity • Given network of nodes connected by bi-directional links • Each link has a cost in each direction • Define cost of path between two nodes as sum of costs of links traversed • For each pair of nodes, find a path with the least cost • Link costs in different directions may be different • E.g. length of packet queue
(A) Dijkstra’s Algorithm • Find shortest paths from given source node to all other nodes, by developing paths in order of increasing path length • N = set of nodes in the network • s = source node • T = set of nodes so far incorporated by the algorithm • w(i, j) = link cost from node i to node j • w(i, i) = 0 • w(i, j) = ∞ if the two nodes are not directly connected • w(i, j) ≥ 0 if the two nodes are directly connected • L(n) = cost of least-cost path from node s to node n currently known • At termination, L(n) is cost of least-cost path from s to n Procedure: • Step 1 [Initialization] • T = {s} Set of nodes so far incorporated consists of only source node • L(n) = w(s, n) for n ≠ s
120
LINE COMMUNICATION SYSTEM
• Initial path costs to neighboring nodes are simply link costs • Step 2 [Get Next Node] • Find neighboring node not in T with least-cost path from s • Incorporate node into T • Also incorporate the edge that is incident on that node and a node in T that contributes to the path • Step 3 [Update Least-Cost Paths] • L(n) = min[L(n), L(x) + w(x, n)] for all n ∉ T • If latter term is minimum, path from s to n is path from s to x concatenated with edge from x to n • Algorithm terminates when all nodes have been added to T • At termination, value L(x) associated with each node x is cost (length) of least-cost path from s to x. • In addition, T defines least-cost path from s to each other node • One iteration of steps 2 and 3 adds one new node to T • Defines least cost path from s to that node
Example of Dijkstra's Algorithm
121
ROUTING IN TELEPHONY
No
T
L(2)
path
L(3)
path
L(4)
path
L(5)
path
L(6)
path
1
(1)
2
1-2
5
1-3
1
1-4
none
none
none
none
2
(1,4)
2
1-2
4
1-43
1
1-4
2
1-45
none
none
3
(1,2,4)
2
1-2
4 3
1-4-
1
1-4
2 5
1-4-
none
none
4
(1,2,4,5)
2
1-2
3
1-45-3
1
1-4
2
1-45
4
1-45-6
5
(1,2,3,4,5)
2
1-2
3
1-45-3
1
1-4
2
1-45
4
1-45-6
6
(1,2,3,4,5,6)
2
1-2
3
1-45-3
1
1-4
2
1-45
4
1-45-6
(B) Bellman-Ford Algorithm • Find shortest paths from given node subject to constraint that paths contain at most one link • Find the shortest paths with a constraint of paths of at most two links • And so on • s = source node • w(i, j) = link cost from node i to node j • w(i, i) = 0 • w(i, j) = ∞ if the two nodes are not directly connected • w(i, j) ≥ 0 if the two nodes are directly connected • h = maximum number of links in path at current stage of the algorithm • Lh(n) = cost of least-cost path from s to n under constraint of no more than h links
Bellman-Ford Algorithm Method • Step 1 [Initialization] • L0(n) = ∞, for all n ≠ s • Lh(s) = 0, for all h. • Step 2 [Update] . • For each successive h ≥ 0 • For each n ≠ s, compute • Lh + 1(n) = min j[Lh(j) + w(j, n)]. • Connect n with predecessor node j that achieves minimum. • Eliminate any connection of n with different predecessor node formed during an earlier iteration. • Path from s to n terminates with link from j to n.
122
LINE COMMUNICATION SYSTEM
• For each iteration of step 2 with h = K and for each destination node n, algorithm compares paths from s to n of length K = 1 with path from previous iteration. • If previous path shorter it is retained. • Otherwise new path is defined.
Example of Bellman-Ford Algorithm
h
Lh(2)
path
Lh(3)
path
Lh(4)
path
Lh(5)
path
Lh(6)
path
0
None
None
None
None
None
None
None
None
None
None
1
2
1-2
5
1-3
1
1-4
None
None
None
None
2
2
1-2
4
1-4-3
1
1-4
2
1-4-5
10
1-3-6
3
2
1-2
3
1-45-3
1
1-4
2
1-4-5
4
1-45-6
4
2
1-2
3
1-45-3
1
1-4
2
1-4-5
4
1-45-6
ROUTING IN TELEPHONY
123
➢ COMPARISON BETWEEN TWO ALGORITHMS
• Results from two algorithms agree • Information gathered • Bellman-Ford • Calculation for node n involves knowledge of link cost to all neighboring nodes plus total cost to each neighbor from s • Each node can maintain set of costs and paths for every other node • Can exchange information with direct neighbors • Can update costs and paths based on information from neighbors and knowledge of link costs • Dijkstra • Each node needs complete topology • Must know link costs of all links in network • Must exchange information with all other nodes
13 Wave Propagation 1. Concept 2. Troposphere 3. Stratosphere 4. Ionosphere 5. Atmospheric Propagation 6. Atmospheric Effects on Propagation
➢
Concept
While radio waves traveling in free space have little outside influence to affect them, radio waves traveling in the earth's atmosphere have many influences that affect them. We have all experienced problems with radio waves, caused by certain atmospheric conditions complicating what at first seemed to be a relatively simple electronic problem. These problem-causing conditions result from a lack of uniformity in the earth’s atmosphere. Many factors can affect atmospheric conditions, either positively or negatively. Three of these are variations in geographic height, differences in geographic location, and changes in time (day, night, season and year). To understand wave propagation, you must have at least a basic understanding of the earth’s atmosphere. The earth’s atmosphere is divided into three separate regions, or layers. They are the troposphere, the stratosphere, and the ionosphere. These layers are illustrated in Fig. 13.1. ➢ TROPOSPHERE
Almost all weather phenomena take place in the troposphere. The temperature in this region decreases rapidly with altitude. Clouds form, and there may be a lot of turbulence because of variations in the temperature, pressure, and density. These conditions have a profound effect on the propagation of radio waves, as we will explain later in this chapter.
124
125
WAVE PROPAGATION
Fig. 13.1 Atmospheric layers.
➢ STRATOSPHERE
The stratosphere is located between the troposphere and the ionosphere. The temperature throughout this region is almost constant and there is little water vapor present. Because it is a relatively calm region with little or no temperature change, the stratosphere has almost no effect on radio waves. ➢ IONOSPHERE
This is the most important region of the earth’s atmosphere for long distance, point-to-point communications. Because the existence of the ionosphere is directly related to radiation emitted from the sun, the movement of the earth about the sun or changes in the sun’s activity will result in variations in the ionosphere. These variations are of two general types : (1) those that more or less occur in cycles and therefore, can be predicted with reasonable accuracy;and (2) those that are irregular as a result of abnormal behavior of the sun and, therefore, cannot be predicted. Both regular and irregular variations have important effects on radio-wave propagation. Since irregular variations cannot be predicted, we will concentrate on regular variations.
Regular Variations The regular variations can be divided into four main classes: daily, 27-day, seasonal, and 11-year. We will concentrate our discussion on daily variations, since they have the greatest effect on your job. Daily variations in the ionosphere produce four cloud-like layers of electrically-charged gas atoms called ions, which enable radio waves to be propagated great distances around the earth. Ions are formed by a process called ionization.
Ionization In ionization, high-energy ultraviolet light waves from the sun periodically enter the ionosphere, strike neutral gas atoms, and knock one or more electrons free from each atom. When the electrons are knocked free, the atoms become positively charged (positive ions) and remain in space, along with the negatively charged free electrons. The free electrons absorb some of the ultraviolet energy that initially set them
126
LINE COMMUNICATION SYSTEM
free and form an ionized layer. Since the atmosphere is bombarded by ultraviolet waves of differing frequencies, several ionized layers are formed at different altitudes. Ultraviolet waves of higher frequencies penetrate the most, so they produce ionized layers in the lower portion of the ionosphere. Conversely, ultraviolet waves of lower frequencies penetrate the least, so they form layers in the upper regions of the ionosphere. An important factor in determining the density of these ionized layers is the elevation angle of the sun. Since this angle changes frequently, the height and thickness of the ionized layers vary, depending on the time of day and the season of the year. Another important factor in determining layer density is known as recombination.
Recombination Recombination is the reverse process of ionization. It occurs when free electrons and positive ions collide, combine, and return the positive ions to their original neutral state. Like ionization, the recombination process depends on the time of day. Between early morning and late afternoon, the rate of ionization exceeds the rate of recombination. During this period the ionized layers reach their greatest density and exert maximum influence on radio waves. However, during the late afternoon and early evening, the rate of recombination exceeds the rate of ionization, causing the densities of the ionized layers to decrease. Throughout the night, density continues to decrease, reaching its lowest point just before sunrise. It is important to understand that this ionization and recombination process varies, depending on the ionospheric layer and the time of day. The following paragraphs provide an explanation of the four ionospheric layers.
Ionospheric Layers The ionosphere is composed of three distinct layers, designated from lowest level to highest level (D, E, and F) as shown in Fig. 13.2. In addition, the F layer is divided into two layers, designated F1 (the lower level) and F2 (the higher level). The presence or absence of these layers in the ionosphere and their height above the earth vary with the position of the sun. At high noon, radiation in the ionosphere above a given point is greatest, while at night it is minimal. When the radiation is removed, many of the particles that were ionized recombine. During the time between these two conditions, the position and number of ionized layers within the ionosphere change. Since the position of the sun varies daily, monthly, and yearly with respect to a specific point on earth, the exact number of layers present is extremely difficult to determine. However, the following eneral statements about these layers can be made.
Fig. 13.2 Layers of the ionosphere.
WAVE PROPAGATION
127
1. D LAYER. The D layer ranges from about 30 to 55 miles above the earth. Ionization in the D layer is low because less ultraviolet light penetrates to this level. At very low frequencies, the D layer and the ground act as a huge waveguide, making communication possible only with large antennas and high-power transmitters. At low and medium frequencies, the D layer becomes highly absorptive, which limits the effective daytime communication range to about 200 miles. At frequencies above about 3 MHz, the D layer begins to lose its absorptive qualities. Long-distance communication is possible at frequencies as high as 30 MHz. Waves at frequencies above this range pass through the D layer but are attenuated. After sunset the D layer disappears because of the rapid recombination of ions. Low frequency and mediumfrequency long-distance communication becomes possible. This is why AM behaves so differently at night. Signals passing through the D layer normally are not absorbed but are propagated by the E and F layers. 2. E LAYER. The E layer ranges from approximately 55 to 90 miles above the earth. The rate of ionospheric recombination in this layer is rather rapid after sunset, causing it to nearly disappear by midnight. The E layer permits medium-range communications on the low-frequency through very high-frequency bands. At frequencies above about 150 MHz, radio waves pass through the E layer. Sometimes a solar flare will cause this layer to ionize at night over specific areas. Propagation in this layer during this time is called SPORADIC-E. The range of communication in sporadic-E often exceeds 1000 miles, but the range is not as great as with F layer propagation. 3. F LAYER. The F layer exists from about 90 to 240 miles above the earth. During daylight hours, the F layer separates into two layers, F1 and F2. During the night, the F1 layer usually disappears, The F layer produces maximum ionization during the afternoon hours, but the effects of the daily cycle are not as pronounced as in the D and E layers. Atoms in the F layer stay ionized for a longer time after sunset, and during maximum sunspot activity, they can stay ionized all night long. Since the F layer is the highest of the ionospheric layers, it also has the longest propagation capability. For horizontal waves, the single-hop F2 distance can reach 3000 miles. For signals to propagate over greater distances, multiple hops are required. The F layer is responsible for most high frequency, long-distance communications. The maximum frequency that the F layer will return depends on the degree of sunspot activity. During maximum sunspot activity, the F layer can return signals at frequencies as high as 100 MHz. During minimum sunspot activity, the maximum usable frequency can drop to as low as 10 MHz. ➢ ATMOSPHERIC PROPAGATION
Within the atmosphere, radio waves can be refracted, reflected, and diffracted. In the following paragraphs, we will discuss these propagation characteristics.
Refraction A radio wave transmitted into ionized layers is always refracted, or bent. This bending of radio waves is called refraction. Notice the radio wave shown in Fig. 13.3, traveling through the earth’s atmosphere at a constant speed. As the wave enters into the denser layer of charged ions, its upper portion starts moving faster than its lower portion. The abrupt speed increase of the upper part of the wave causes it to bend back toward the earth. This bending is always toward the propagation medium where the radio wave’s velocity is the least. The amount of refraction a radio wave undergoes depends on three main factors.
128
LINE COMMUNICATION SYSTEM
1. The ionization density of the layer 2. The frequency of the radio wave 3. The angle at which the radio wave enters the layer.
Fig. 13.3 Radio-wave refraction.
Fig. 13.4 Effects of ionospheric density on radio waves.
Layer Density Fig. 13.4 shows the relationship between radio waves and ionization density. Each ionized layer has a middle region of relatively dense ionization with less intensity above and below. As a radio wave enters a region of increasing ionization, a velocity increase causes it to bend back toward the earth. In the highly dense middle region, refraction occurs more slowly because the ionization density is uniform. As the wave enters the upper less dense region, the velocity of the upper part of the wave decreases and the wave is bent away from the earth.
Frequency The lower the frequency of a radio wave, the more rapidly the wave is refracted by a given degree of ionization. Fig. 13.5 shows three separate waves of differing frequencies entering the ionosphere at
129
WAVE PROPAGATION
the same angle. You can see that the 5-MHz wave is refracted quite sharply, while the 20-MHz wave is refracted less sharply and returns to earth at a greater distance than the 5 MHz wave. Notice that the 100MHz wave is lost into space. For any given ionized layer, there is a frequency, called the escape point, at which energy transmitted directly upward will escape into space. The maximum frequency just below this example, the 100-MHz wave's frequency is greater than the critical frequency for that ionized layer. The critical frequency of a layer depends upon the layer's density. If a wave passes through a particular layer, it may still be refracted by a higher layer if its frequency is lower than the higher layer's critical frequency.
Fig. 13.5 Frequency versus refraction and distance.
Angle of Incidence and Critical Angle When a radio wave encounters a layer of the ionosphere, that wave is returned to earth at the same angle (roughly) as its angle of incidence. Fig. 13.6 shows three radio waves of the same frequency
Fig. 13.6 Incidence angles of radio waves.
entering a layer at different incidence angles. The angle at which wave A strikes the layer is too nearly vertical for the wave to be refracted to earth, However, wave B is refracted back to earth. The angle between wave B and the earth is called the critical angle. Any wave, at a given frequency, that leaves the antenna at an incidence angle greater than the critical angle will be lost into space. This is why wave A was not refracted. Wave C leaves the antenna at the smallest angle that will allow it to be refracted and still return to earth. The critical angle for radio waves depends on the layer density and the wavelength of the signal. As the frequency of a radio wave is increased, the critical angle must be reduced for refraction to occur.
130
LINE COMMUNICATION SYSTEM
Fig. 13.7 Effect of frequency on the critical angle.
Notice in Fig. 13.7 that the 2-MHz wave strikes the ionosphere at the critical angle for that frequency and is refracted. Although the 5-MHz line (broken line) strikes the ionosphere at a less critical angle, it still penetrates the layer and is lost. As the angle is lowered, a critical angle is finally reached for the 5-MHz wave and it is refracted back to earth.
Skip Distance and Zone Recall from your previous studies that a transmitted radio wave separates into two parts, the sky wave and the ground wave. With those two components in mind, we will now briefly discuss skip distance and skip zone. The skip zone is a zone of silence between the point where the ground wave is too weak for reception and the point where the sky wave is first returned to earth. The outer limit of the skip zone varies considerably, depending on the operating frequency, the time of day, the season of the year, sunspot activity, and the direction of transmission.
Skip Distance Look at the relationship between the sky wave skip distance, skip zone, and ground wave coverage shown in Fig. 13.8.
Fig. 13.8 Relationship between skip zone, skip distance and ground wave.
131
WAVE PROPAGATION
The skip distance is the distance from the transmitter to the point where the sky wave first returns to the earth. The skip distance depends on the wave’s frequency and angle of incidence, and the degree of ionization. At very-low, low, and medium frequencies, a skip zone is never present. However, in the high frequency spectrum, a skip zone is often present. As the operating frequency is increased, the skip zone widens to a point where the outer limit of the skip zone might be several thousand miles away. At frequencies above a certain maximum, the outer limit of the skip zone disappears completely, and no F-layer propagation is possible. Occasionally, the first sky wave will return to earth within the range of the ground wave. In this case, severe fading can result from the phase difference between the two waves (the sky wave has a longer path to follow).
Reflection Reflection occurs when radio waves are “bounced” from a flat surface. There are basically two types of reflection that occur in the atmosphere : earth reflection and ionospheric reflection. Fig. 13.9 shows two waves reflected from the earth's surface. Waves A and B bounce off the earth's surface like light off of a mirror. Notice that the positive and negative alternations of radio waves A and B are in phase before they strike the earth’s surface. However, after reflection the radio waves are approximately 180 degrees out of phase. A phase shift has occurred. The amount of phase shift that occurs is not constant. It varies, depending on the wave polarization and the angle at which the wave strikes the surface. Because reflection is not constant, fading occurs. Normally, radio waves reflected in phase produce stronger signals, while those reflected out of phase produce a weak or fading signal. Ionospheric reflection occurs when certain radio waves strike a thin, highly ionized layer in the ionosphere. Although the radio waves are actually refracted, some may be bent back so rapidly that they appear to be reflected. For ionospheric reflection to occur, the highly ionized layer can be approximately no thicker than one wavelength of the wave. Since the ionized layers are often several miles thick, ionospheric reflection mostly occurs at long wavelengths (low frequencies).
Fig. 13.9 Phase shift of reflected radio waves.
Diffraction Diffraction is the ability of radio waves to turn sharp corners and bend around obstacles. As shown in Fig. 13.10, diffraction results in a change of direction of part of the radio-wave energy around the edges of an obstacle. Radio waves with long wavelengths compared to the diameter of an obstruction are easily propagated around the obstruction. However, as the wavelength decreases, the obstruction causes
132
LINE COMMUNICATION SYSTEM
more and more attenuation, until at very-high frequencies a definite shadow zone develops. The shadow zone is basically a blank area on the opposite side of an obstruction in line-of-sight from the transmitter to the receiver. Diffraction can extend the radio range beyond the horizon. By using high power and low-frequencies, radio waves can be made to encircle the earth by diffraction.
Fig. 13.10 Diffraction around an object.
➢ ATMOSPHERIC EFFECTS ON PROPAGATION
As we stated earlier, changes in the ionosphere can produce dramatic changes in the ability to communicate. In some cases, communications distances are greatly extended. In other cases, communications distances are greatly reduced or eliminated. The paragraphs below explain the major problem of reduced communications because of the phenomena of fading and selective fading.
Fading The most troublesome and frustrating problem in receiving radio signals is variations in signal strength, most commonly known as FADING. Several conditions can produce fading. When a radio wave is refracted by the ionosphere or reflected from the earth’s surface, random changes in the polarization of the wave may occur. Vertically and horizontally mounted receiving antennas are designed to receive vertically and horizontally polarized waves, respectively. Therefore, changes in polarization cause changes in the received signal level because of the inability of the antenna to receive polarization changes. Fading also results from absorption of the rf energy in the ionosphere. Most ionospheric absorption occurs in the lower regions of the ionosphere where ionization density is the greatest. As a radio wave passes into the ionosphere, it loses some of its energy to the free electrons and ions present there. Since the amount of absorption of the radio-wave energy varies with the density of the ionospheric layers, there is no fixed relationship between distance and signal strength for ionospheric propagation. Absorption fading occurs for a longer period than other types of fading, since absorption takes place slowly. Under certain conditions, the absorption of energy is so great that communication over any distance beyond the line of sight becomes difficult. Although fading because of absorption is the most serious type of fading, fading on the ionospheric circuits is mainly a result of multipath propagation.
133
WAVE PROPAGATION
Multipath Fading MULTIPATH is simply a term used to describe the multiple paths a radio wave may follow between transmitter and receiver. Such propagation paths include the ground wave, ionospheric refraction, reradiation by the ionospheric layers, reflection from the earth's surface or from more than one ionospheric layer, and so on. Fig. 13.11 shows a few of the paths that a signal can travel between two sites in a typical circuit. One path, XYZ, is the basic ground wave. Another path, XFZ, refracts the wave at the F layer and passes it on to the receiver at point Z. At point Z, the received signal is a combination of the ground wave and the sky wave. These two signals, having traveled different paths, arrive at point Z at different times. Thus, the arriving waves may or may not be in phase with each other. A similar situation may result at point A. Another path, FZFA, results from a greater angle of incidence and two refractions from the F layer. A wave traveling that path and one traveling the XEA path may or may not arrive at point A in phase. Radio waves that are received in phase reinforce each other and produce a stronger signal at the receiving site, while those that are received out of phase produce a weak or fading signal. Small alterations in the transmission path may change the phase relationship of the two signals, causing periodic fading. Multipath fading may be minimized by practices called SPACE DIVERSITY and FREQUENCY DIVERSITY. In space diversity; two or more receiving antennas are spaced some distance apart. Fading does not occur simultaneously at both antennas. Therefore, enough output is almost always available from one of the antennas to provide a useful signal. In frequency diversity, two transmitters and two receivers are used, each pair tuned to a different frequency, with the same information being transmitted simultaneously over both frequencies. One of the two receivers will almost always produce a useful signal.
Fig. 13.11 Multipath transmission
Selective Fading Fading resulting from multipath propagation varies with frequency since each frequency arrives at the receiving point via a different radio path. When a wide band of frequencies is transmitted simultaneously, each frequency will vary in the amount of fading. This variation is called SELECTIVE FADING. When selective fading occurs, all frequencies of the transmitted signal do not retain their original phases and relative amplitudes. This fading causes severe distortion of the signal and limits the total signal transmitted. Frequency shifts and distance changes because of daily variations of the different ionospheric layers are summarized next.
Daily Ionospheric Communications D LAYER. Reflects vlf waves for long-range communications; refracts lf and mf for shortrange communications; has little effect on vhf and above; gone at night. E LAYER. Depends on the angle of the sun: refracts hf waves during the day up to 20 MHz to distances of 1200 miles: greatly reduced at night.
134
LINE COMMUNICATION SYSTEM
F LAYER. Structure and density depend on the time of day and the angle of the sun: consists of one layer at night and splits into two layers during daylight hours. F1 LAYER. Density depends on the angle of the sun; its main effect is to absorb hf waves passing through to the F2 layer. F2 LAYER. Provides long-range hf communications; very variable; height and density change with time of day, season, and sunspot activity.
Other Phenomena that Affect Communications Although daily changes in the ionosphere have the greatest effect on communications, other phenomena also affect communications, both positively and negatively. Those phenomena are discussed briefly in the following paragraphs.
Seasonal Variations in the Ionosphere Seasonal variations are the result of the earth's revolving around the sun, because the relative position of the sun moves from one hemisphere to the other with the changes in seasons. Seasonal variations of the D, E, and F1 layers are directly related to the highest angle of the sun, meaning the ionization density of these layers is greatest during the summer. The F2 layer is just the opposite. Its ionization is greatest during the winter; therefore, operating frequencies for F2 layer propagation are higher in the winter than in the summer.
Sunspots One of the most notable occurrences on the surface of the sun is the appearance and disappearance of dark, irregularly shaped areas known as SUNSPOTS. Sunspots are believed to be caused by violent eruptions on the sun and are characterized by strong magnetic fields. These sunspots cause variations in the ionization level of the ionosphere. Sunspots tend to appear in two cycles, every 27 days and every 11 years.
Irregular Variations Irregular variations are just that, unpredictable changes in the ionosphere that can drastically affect our ability to communicate. The more common variations are sporadic E, ionospheric disturbances, and ionospheric storms. Sporadic E Irregular cloud-like patches of unusually high ionization, called the sporadic E, often format heights near the normal E layer. Their exact cause is not known and their occurrence cannot be predicted. However, sporadic E is known to vary significantly with latitude. In the northern latitudes, it appears to be closely related to the aurora borealis or northern lights. The sporadic E layer can be so thin that radio waves penetrate it easily and are returned to earth by the upper layers, or it can be heavily ionized and extend up to several hundred miles into the ionosphere. This condition may be either harmful or helpful to radio-wave propagation. On the harmful side, sporadic E may blank out the use of higher more favorable layers or cause additional absorption of radio waves at some frequencies. It can also cause additional multipath problems and delay the arrival times of the rays of RF energy. On the helpful side, the critical frequency of the sporadic E can be greater than double the critical frequency of the normal ionospheric layers. This may permit long-distance communications with unusually high frequencies. It may also permit short-distance communications to locations that would
WAVE PROPAGATION
135
normally be in the skip zone. Sporadic E can appear and disappear in a short time during the day or night and usually does not occur at same time for all transmitting or receiving stations. Ionospheric Storms Ionospheric storms are caused by disturbances in the earth's magnetic field. They are associated with both solar eruptions and the 27-day cycle, meaning they are related to the rotation of the sun. The effects of ionospheric storms are a turbulent ionosphere and very erratic skywave propagation. The storms affect mostly the F2 layer, reducing its ion density and causing the critical frequencies to be lower than normal. What this means for communication purposes is that the range of frequencies on a given circuit is smaller than normal and that communications are possible only at lower working frequencies.
Transmission Losses All radio waves propagated over the ionosphere undergo energy losses before arriving at the receiving site. As we discussed earlier, absorption and lower atmospheric levels in the ionosphere account for a large part of these energy losses. There are two other types of losses that also significantly affect propagation. These losses are known as ground reflection losses and freespace loss. The combined effect of absorption ground reflection loss, and freespace loss account for most of the losses of radio transmissions propagated in the ionosphere. GROUND REFLECTION LOSS when propagation is accomplished via multihop refraction, rf energy is lost each time the radio wave is reflected from the earth’s surface. The amount of energy lost depends on the frequency of the wave, the angle of incidence, ground irregularities, and the electrical conductivity of the point of reflection.
Maximum Usable Frequency The higher the frequency of a radio wave, the lower the rate of refraction by the ionosphere. Therefore, for a given angle of incidence and time of day, there is a maximum frequency that can be used for communications between two given locations. This frequency is known as the MAXIMUM USABLE FREQUENCY (muf). Waves at frequencies above the muf are normally refracted so slowly that they return to earth beyond the desired location or pass on through the ionosphere and are lost. Variations in the ionosphere that can raise or lower a predetermined muf may occur at anytime. This is especially true for the highly variable F2 layer.
14 Integrated Services Digital Network [ISDN] 1. Concept 2. IDSN Standards 3. Merits of ISDN System 4. Features of ISDN System 5. New Services Provided by ISDN System 6. Network and Protocol Architecture 7. Functional Grouping (FG) and Reference Points 8. Equipments for ISDN 9. Transmission Channels and Access Arrangement in ISDN System 10. User Network Interfaces 11. Broadband ISDN 12. B-ISDN Switching 13. Operation and Maintenance ➢
Concept
Today’s trunk telephone network and modern telephone exchanges are fully digital. The analogue speech arriving at the exchange from the subscriber’s line is rendered into digital form, passed through the exchanges and trunk lines in digital form, and only converted back to analogue form for sending to the other subscriber's phone line. Network signaling (within the network itself) is also fully digital. However, the telephone network is used nowadays for other things: faxes, computer communications, etc. Many of those things are naturally digital, and it's perverse to convert them to analogue merely to get the data to the telephone exchange (where it will be converted back to digital form: the result of this perversity is quite a reduction in the data carrying capability of the telephone connection). ISDN is a network concept, providing the integration of data, voice and video through digital transmission media. It is based on 64 Kbps digital communication channel. ISDN is a generic term for any network which connects homes and business personalities together with the service companies such as banks, airlines, stock markets using a digital network. 136
INTEGRATED SERVICES DIGITAL NETWORK [ISDN]
137
Starting from late 1970’s the new ISDN system was developed by T1D1 subcommittee of the Exchange Carriers Association in U.S.A and by the CCITT [now ITU-T, international telecommunication union - telephony] study group XVIII. It is the most important development to immerge in the field of Computer Communications. From 1990s ISDN becomes a well conceived and planned area of development in the field of telecommunication. After the evolution of Integrated Digital Network [IDN], within four years ISDN was introduced progressively incorporating additional features and services. ISDN lines are being heavily promoted in Germany nowadays, and are available for little more than the cost of an “ordinary” line. Equipment is also available relatively cheaply (e.g. PC cards). BT, on the other hand, charges an arm and a leg in installation charges and rental, and relatively little equipment has been approved for UK use, most of it very expensive compared with conventional (analogue) telephones. The result is that very few individual UK customers have installed ISDN, and, correspondingly, very few BT staff understand what it is, so you have to go out of your way to get started. In the USA, the EFF (Electronic Frontier Foundation) campaigned strongly in 1993 for the full exploitation of ISDN as a part of an information infrastructure. At that time, the availability of ISDN in the USA was very patchy, being easily available in some areas, and to all intents unavailable in others. Sociological needs, economical necessities and technical developments compelled to marketize the ISDN system. An era of ISDN will be spread out all over the world in the coming twenty years or so.
Fig. 14.1 Block diagram of ISDN network
➢ ISDN STANDARDS
CCITT has been playing a leading role & acting as a coordinating body by issuing ISDN related recommendations & thereby guiding the introduction of ISDN internationally. The first definition of ISDN appeared in CCITT’s recommendation;G.702 in 1972. Then in 1988 ISDN was given a comprehensive set of recommendations known as I-series recommendations. The following figure shows the general structure of I-series recommendations:
138
LINE COMMUNICATION SYSTEM
Fig. 14.2 Structure of CCITT ISDN recommendations – 1999
➢ MERITS OF ISDN SYSTEM
1. AN ISDN user can establish two simultaneous independent telecom calls on the existing pair of telephone wire. 2. The two simultaneous calls may be of any types as speech, data, image or video. 3. Using an ISDN line the data transfer rate with another ISDN system on dial up basis is 64 Kbps and it can go upto 128 Kbps. 4. With an ISDN line the video conferencing can be done with another ISDN subscriber on dial up basis (a) Ordinary video conferencing of 128 Kbps on one ISDN line (b) High quality video conferencing of 384 Kbps on three ISDN line ➢ FEATURES OF ISDN SYSTEM
(a)
Three consecutive public switch telephone network [PSTN] connections can be arranged in a single ISDN line. (b) Two simultaneous tasks can be done on a single ISDN connection such as internet and video conferencing etc. (c) A maximum of 8 terminals can be connected to the network terminator [NT] of ISDN. ➢ NEW SERVICES PROVIDED BY ISDN SYSTEM
ISDN will support variety of services including speech, data communication. Some of them are listed below: (a) Digital Facsimile (b) Electronic mail (c) Videotext
INTEGRATED SERVICES DIGITAL NETWORK [ISDN]
(d) (e) (f) (g) (h) (i)
139
Database access Telex Electronic Fund Transfer Image and Graphics exchange Document storage and transfer Audio and Video conferencing
➢ NETWORK AND PROTOCOL ARCHITECTURE
We already know that ISDN is sufficiently an integrated network. Integration means summation over a fixed or non-fixed limit of a continuous function. But it is next to impossible to change the entire telecommunication equipments to ISDN. Therefore we have to follow a particular chronology to take the advancement with our usable form. Here we use two different transition states named as segregated network and integrated digital network (IDN) before the introduction of complete ISDN.
Fig. 14.3 (a) Segregated network.
Fig. 14.3 (b) Integrated network (IDN).
We all know that the basic four networks used in line communication are as follows: (i) Circuit switched network (ii) Packet switched network (iii) Non-switched network (iv) Signaling network.
140
LINE COMMUNICATION SYSTEM
From the figure above (Fig. 14.3) it is very clear that the first step towards ISDN is segregated network. Here the designer uses one pair of integrator ends (IE) beside the each user. The architecture is like the collaboration of simple single type network and integrated network architecture. Actually the information (any of those four basic types) is marked by the sender IE. Then it is passed through its dedicated type path and finally reaches to the destination user in time sharing basis. This time sharing is done by the destination IE. A very important point is to be noted here that, integrator ends (IE) have two functions, one to multiplex and demultiplex and the other one is the marking or giving the identification to the information that to which type that belongs to. IEs have two more different important functions except the above two;those are security, error detection. Second step is introducing IDN or integrated digital network. The concept of IDN is very much similar to ISDN except the clocking technique. IDN may employ synchronous data transfer, whereas ISDN follows strictly the asynchronous data transfer (ADT). IDN is the proper bridge between segregated network and ISDN. ISDN protocol shows three layer architecture corresponding ISO-OSI seven layers. The three important layers used in ISDN are physical layer, data link layer and network layer. The functions of the three layers are pointed below. The switching theory followed here is a nice compromise between packet switching and circuit switching. The network function also proofs the fact. Layer 1:Physical layer 1. Encoding and decoding of information. 2. Transmission of channel data. 3. Multiplexing to form basic and primary rate. 4. Making and breaking of physical circuit (as circuit switching).
Layer 2:Data link layer 1. Establishing and clearing data links. 2. Error, flow and congestion control. 3. Synchronization: matching the clocking frequency, phase of information at receiver side.
Layer 3:Network layer 1. Addressing and routing. 2. User-to-user signaling. 3. Activation and deactivation of network level connections. 4. Intra-network level multiplexing. 5. Multiplexing between different networks.
INTEGRATED SERVICES DIGITAL NETWORK [ISDN]
141
Fig. 14.4 ISDN protocol architecture.
➢ FUNCTIONAL GROUPING (FG) AND REFERENCE POINTS
Taking into account the regularity factors by different nations, ISDN user network interfaces are functionally grouped and the associated access points at different functional levels are known as reference points. Among the reference points, R, S and T are CCITT reference points. But the other reference point U is not recognized by CCITT. U interface is essentially the loop transmission system carrying digital signals at 192 Kbps rates, which corresponds to BISDN (channel arrangement: 2B+D). Equipment connected to U interface will have to take care of the functions of 7th layers of ISDN. Network termination 1 (NT1) includes the functions of layer 1. Consequently, equipment connected to T reference point will have to concern itself only with the functions of layers 2 to 6 (notice the following figure). NT2 includes the functions of layer 1, 2 and 3. Reference point S includes the functions of layers 4 to 7.
142
LINE COMMUNICATION SYSTEM
Fig. 14.5 Functional grouping and reference points.
It is very interesting to see that every functional grouping must hold the entire 7 layers functions. S, T and U are ISDN compatible points. ISDN equipments are designated as terminal equipment 1 (TE1) whereas non-ISDN equipments are TE2. An adapter (terminal adapter or TA) is used to provide flexibility so that non-ISDN terminals can also be connected with ISDN digital pipe.
Physical Implementation of ISDN Functional Groupings It is not necessary that every reference points must be accessible by the user. But the basic FG property must be satisfied; i.e., each FG must perform the entire functions of 7 layers. Physical implementation of ISDN functional groups are shown in the following figure.
INTEGRATED SERVICES DIGITAL NETWORK [ISDN]
143
Fig. 14.6 Physical implementation of functional grouping.
➢ EQUIPMENTS FOR ISDN
Network Terminator [NT] It looks like a small box with the external cable pair entering at one end with U-inlet & there are maximum 8 terminals but generally 4 terminals are used: 1. ISDN phone set 2. A PC through an ISDN data card or digital modem 3. Digital G4 fax 4. Video conferencing equipments. A single multipurpose communication network with uniform access technology for all services leads to uniform operation and maintenance. Existing customer’s line can be used for providing basic access. No separate networks are required for different services.
144
LINE COMMUNICATION SYSTEM
Fig. 14.7 ISDN customer premises installations
Fig. 14.8 Customer promises equipments (CPE)
Customer Premises Equipments (CPE) • Customer Premises Equipments are classified as NT, TA, TE1 (Terminal Equipment) and TE2.
INTEGRATED SERVICES DIGITAL NETWORK [ISDN]
145
• TE1 are ISDN terminals like Video conferencing equipments, Group 4 Fax, features telephone which are digital and directly connected to NT1 through S-Bus inerface. • TE2 are ISDN terminals like analog phone, PC, G3 Fax, feature telephone which are nondigital can not be directly connected to NT1. They require another interface called adapter (TA) which enables analog to digital conversion and vice versa. ➢ TRANSMISSION CHANNELS AND ACCESS ARRANGEMENT IN ISDN SYSTEM
There are four fundamental channels in ISDN on the basis of entire organization of information transmission. They are as: (a) Basic rate access channel, B channel, 64 Kbps (b) Primary rate access channel, P channel, 128 Kbps (c) Signaling channel, D channel, 16 or 64 Kbps (d) High speed channel, H channel, H0 channel, 384 Kbps H11 channel, 1536 Kbps H12 channel, 1920 Kbps (a) B channel [ Basic rate access ]
Fig. 14.9
2B + D chls B : 64 Kbps data & voice channels D : 16 Kbps control signal B channel is full duplex & used for encoding of digital voice, multiplexing of lower rate data streams. Multiplexing is limited to 8,16,32 Kbps streams. (b) P channel [Primary rate access] 30B + D chls [ European countries ] 23B + D chls [ Japan ] B: 64 Kbps data & voice channels D: 64 Kbps control signal P channel is extensively used in USA, Canada, Japan, Europe, India etc. (c) D channel Its primary function is to carry signaling information for the control of circuit switched connections involving one or more B channels. Otherwise it can be used for following purposes: 1. Transport of user signaling information 2. Low bit rate packet data 3. Telemetry
146
LINE COMMUNICATION SYSTEM
Table 14.1 Operational specification for physical layer of ISDN network
(d) H channel It is used to carry user information at data rate in excess of 64 Kbps in : 1. video teleconferencing 2. high resolution graphics 3. high resolution digital video/audio for transport of television 4. fast FAX services H0 & H11 are used in the North American Version of ISDN, H12 is used in European version. H21 & H22 are intended for broadband ISDN. S : frame size I : information stream rate in bps R : rate in bps to which the information stream is adapted, 8, 16, 32 Kbps D : no. of information bits in each frame RF : frame rate in frames/sec Frame size & frame rate are so choosen that : I = RF * D; R = RF * S ➢ USER NETWORK INTERFACES
Phone Plus Facilities 1. Presentation of calling line identity [CLIP] 2. Restriction of calling line identity [CLIR] 3. Call forwarding services [CF] 4. Call forwarding busy, no reply, unconditional 5. Call holding 6. Three party conference 7. Advice of charge in terms of call units
INTEGRATED SERVICES DIGITAL NETWORK [ISDN]
147
Grade of ISDN Service 1. Reduction of cost and terminal universality and portability. 2. Standard socket to connect 2B + 1D link. 3. Improved reliability and flexibility to add new services. 4. Flexibility of introducing new features to existing services.
Fig. 14.10 Integrated Service Digital Network.
Fig. 14.11 ISDN Categories.
A packet of data transmission is a typical example of bearer service while telephony, telex, videotext and facsimile are some tele-services. Supplementary services are offered in conjunction with either a bearer service or tele-service. LIST OF SUPPLEMENTARY SERVICES 1. abbreviated dialing 2. closed user group 3. call waiting display 4. three party conference 5. city wide Centrex 6. credit card calling
7. don’t disturb 8. reverse charging 9. call bearing 10. direct dialing 11. conference calls 12. call forwarding
148
LINE COMMUNICATION SYSTEM
➢ BROADBAND ISDN The original specifications for the integrated services digital network (ISDN) were based around voice and non-voice telephone-type services: telephony, data, telex, facsimile, as it was hoped that the ISDN would evolve from the (then) emerging digital telephone networks. Indeed, this is one of the reasons that the fundamental element of an ISDN link is the 64 Kb/s B-Channels. However, the planning for ISDN was started around 1976, and as technology evolved, so did the requirements of the users that wanted to use this technology. In 1988, the CCITT released a document that described a new set of Broadband ISDN (B-ISDN) services. To distinguish this new concept from the original ISDN service, we now refer to the latter as Narrowband ISDN (N-ISDN). Since then the CCITT has become the ITU. The B-ISDN work is far from complete, and some of the factors influencing the emergence of the B-ISDN from ITU are: • Demand. Users (both commercial and residential) are showing interest in receiving high speed, reliable services. • Technology. As data processing technologies available to the user have become more sophisticated so has his/her demands, while high speed transmission (based on the use of optical fiber), high speed switching and increased processing power make the realization of these demands possible. • Service integration. There is a need to integrate both circuit switched and packet switched services into one network that can provide interactive and distribution services. • Flexibility. The resulting network must be able to satisfy the needs of the users as well as the network operators in terms of its functionality and usability. It is intended that the B-ISDN will offer both connection oriented (CO) and connectionless (CL) services, however, the CO mode of operation is receiving the greatest attention at the moment. The broadband information transfer is provided by the use of asynchronous transfer mode (ATM), in both cases, using end-to-end logical connections. Each logical connection is accessed as a virtual channel (VC). Many VCs may be used to a single destination and they may be associated by use of a virtual path (VP). There relationship between VCs and VPs with respect to the transmission path is shown in figure. The transmission path is the logical connection between the two end-points, and consists in reality of many links between network exchanges and switches.
Fig. 14.12 Transmission path model for Broadband ISDN.
The VCs are identified at each end of the connection by a virtual channel identifier (VCI) and user-to-user data VCs are unidirectional. Similarly, the VP is identified by a virtual path identifier (VPI). The VCI/VPI pair uniquely identifies a user-to-user information flow and is carried in each ATM cell header. Both VCIs and VPIs in general have only local significance. The concept of the link
INTEGRATED SERVICES DIGITAL NETWORK [ISDN]
149
can be applied to both VCs and VPs in explaining the use of VCIs and VPIs, and we can say that the VCI/VPI pair identifies a particular link. VCIs and VPIs are used within the network for switching purposes, with virtual channel links and virtual path links being defined as the connection between two points where either the VC or the VP is switched, respectively, i.e. the link is defined to exist between the two points where VCI or VPI value is removed or translated (switched). There will be many virtual channel links comprising a virtual channel connection (VCC) and, similarly, many virtual path links in a virtual path connection (VPC).
Fig. 14.13 Hierarchical layer to layer relationship in the ATM layer and Physical layer
Broadband ISDN Services The need for a Broadband ISDN service sprung from the growing needs of the customers. The planned Broadband ISDN services can broadly be categorized as follows: • Interactive services. These are services allowing information flow between two end users of the network, or between the user and the service provider. Such services can be subdivided: • Conversational services. These are basically end-to-end, real-time communications, between users or between a user and a service provider, e.g. telephone-like services. Indeed,
150
LINE COMMUNICATION SYSTEM
B-ISDN will support N-ISDN type services. Also the additional bandwidth offered will allow such services as video telephony, video conferencing and high volume, high speed data transfer. • Messaging services. This differs from conversational services in that it is mainly a storeand-forward type of service. Applications could include voice and video mail, as well as multi-media mail and traditional electronic mail. • Retrieval services. This service provides access to (public) information stores, and information is sent to the user on demand only. This includes things like tele-shopping, videotext services, still and moving pictures, tele-software and entertainment. • Distribution services. These are mainly broadcast services, are intended for mainly one way interaction from a service provider to a user: • No user control of presentation. This would be for instance, a TV broadcast, where the user can choose simply either to view or not. It is expected that cable TV companies will become interested in Broadband ISDN as a carrier for the high definition TV (HDTV) services that are foreseen for the future. • User controlled presentation. This would apply to broadcast information that the user can partially control, in that the user can decide which part of it he/she accesses, e.g. tele text and news retrieval services. However, many of these services have very high throughput requirements, as shown in table. The burstness is the ratio of the peak bit rate to average bit rate. Table : 14.2 Some Broadband service throughput needs Service Data Document transfer
Bit Rate [Mb/s]
Burstness
1.5 to 130
1—50
1.5 to 45
1—20
Videoconference or videotelephony
1.5 to 130
1—5
Broadband video
1.5 to 130
1—20
TV
30 to 130
1
130
1
HDTV
It is clear that high network capacity is required if this kind of service is to be offered to many user simultaneously. The N-ISDN can currently offer interfaces which aggregate B-Channels to give additional throughput, as shown in tables. However, these are not sufficient for our Broadband service requirements. Table : 14.3 Narrowband ISDN channels Channel
Bit Rate [Kb/s]
Interface
B
64
Basic rate
H0
384
Primary rate
H11
1536
Primary rate
H12
1920
Primary rate
D16
16
Basic rate
D64
64
Primary rate
151
INTEGRATED SERVICES DIGITAL NETWORK [ISDN]
Table : 14.4 Interface Basic rate access Primary rate access
Bit Rate [Kb/s]
Structure
144
2B + D
1544
23B + D64 3H0 + D64 H11 etc.
Primary rate access
2048
30b + D64 5H0 + D64 H12 + D64 etc.
B-ISDN uses dedicated signaling virtual channels (SVCs). There are four types as shown in Table. Table : 14.5 SVCs at the B-ISDN user-to-network interface SVC type
Directionality
SVCs per user interface
Meta-signaling
Bidirectional
1
General broadcast
Unidirectional
1
Selective broadcast
Unidirectional
Several possible
Point-to-point
Bidirectional
1 per signaling end point
There is only meta-signaling channel per interface and it is permanent. It is used to set up instances of the point-to-point SVC as required. It is identified by the use of a specially allocated VCI/ VPI. The broadcast SVCs are directed from network to user. The general broadcast SVC is permanent and provides signaling information to all network users. The selective broadcast SVC is optionally set up by the network provider to provide additional signaling services to only certain terminals. The point-to-point SVC exists only while it is active and is used to establish, maintain, control and release data VCCs between to end points.
➢ B-ISDN SWITCHING The basis B-ISDN service is a compromise between pure packet switched and pure circuit switched network. The service is connection oriented but, is implemented internally by packet switching. Two types of connections are offered, (i) Permanent virtual circuit and (ii) Switched virtual circuit - set up dynamically as needed (as telephone call). In circuit switching (space division switching) there is a connection oriented service. On the other hand in virtual circuit, specially in ATM (Asynchronous Transfer Mode), when a circuit is established, the route is chosen from source to destination, and all the switches (i.e. routers) along the way make table entries so that router can route any packets on the virtual circuit. They also have the opportunity to reserve resources for the new circuit. As the following figure, a virtual circuit is formed between host H1 and host H5 via the switches A, E, C and D. When a packet comes
152
LINE COMMUNICATION SYSTEM
along the virtual circuit to find out which virtual circuit it belongs to, then it looks up for the virtual circuit in its table to determine which communication line to send on.
Fig. 14.14 B-ISDN switching arrangement
Transmission in ATM Network Asynchronous transfer mode can be contrasted with the synchronous T1 system. One T1 frame is generated precisely every 125 µs. This rate is generated by a master clock. Slot k of each frame contains 1 byte of data from the same source and arranged alternately one after another. Here channel one (1) gets exactly one byte at the start of each frame. In ATM, in contrary, has no requirement that cells rigidly alternate among the various sources. Cells arrive randomly from different sources i.e., it does not follow a particular pattern. No requirement of cell ordering is there.
Fig. 14.15 Frame structure of T1 and ATM
ATM link must be situated between two switches or one switch and one computer. The ATM Physical Medium Dependent (PMD) sub-layer is concerned with getting the bits IN and OUT the wire. Different hardwares are needed for different cables and fibers, depending on the
INTEGRATED SERVICES DIGITAL NETWORK [ISDN]
153
speed and line coding technique. The purpose of transmission convergence (TC) sub-layer is to provide a uniform interface to the ATM layer in both directions. Outbound, ATM layer provides a sequence of cells, and the PMD sub-layer encodes according to necessity and pushes them out of the door as a bit stream. Inbound the PMD sub-layer takes the incoming bits from the network and delivers a bit stream to the TC sub-layer. It is upto TC sub-layer to somehow figure out how to tell the ending and beginning on one cell. This job is theoretically impossible. Thus TC sub-layer has its work cut out for it.
➢ OPERATION AND MAINTENANCE The operation and maintenance (OAM) of the ISDN network has five main actions: • Performance monitoring. Managed entities are continuously or periodically monitored in order to generate maintenance event information. • Fault localization. Use of test systems, both internal and external, to determine whether information about faults is complete or sufficient for other actions to take place. • Failure or performance information. Communication of failure and performance information as alarms to other managed entities in the management plane. Also acts as a response service to status report requests. • Defect and failure detection. Errors or malfunctions in the managed entities are detected or predicted resulting in the generation of maintenance event information or service alarms. • System protection. To offer some degree of fault tolerance, the effect of failures in managed entities are minimized by the use of backups, standbys or other resources, and the failed entity is excluded from the normal operation of the system. Other management capabilities such as use of the TMN (Telecommunications Management Network) may also be employed in conjunction with the OAM.
Appendix I Questions from Individual Chapters 1. Introduction 1. What do you mean by simplex, half-duplex and full-duplex communication system? Define with proper example. 2. What is the necessity of introducing switching office or exchange? What are the problems with point-to-point communication system? What is ‘maximum capacity’ of an exchange? 3. What is control function? 4. What are the disadvantages with manual switching exchanges? How automatic exchange can overcome those problems? 5. What do you mean by folded and non-folded network? Describe with proper figure. 6. In basic telephone communication system, what type of analog modulation scheme is employed? Define its modulation index both in earphone and microphone side. 7. What is side-tone ? Is it a problem? If yes, how can we overcome the problem?
2. Advancement of Telecommunication Switching – A Brief History 1. Briefly describe the advancement of telecommunication switching history with proper diagram.
3. Computer Communication – Store and Forward Switching 1. What are the reliable features for successful computer communication? 2. Define with proper: (a) Orphans (b) High and low speed links (c) Remote terminal (d) Local terminal (e) Switching nodes (f) Communication sub-network (g) User sub-network 154
155
APPENDIX I
3. Describe a packet format. 4. What is the basic principle of circuit switched network? 5. What is store and forward switching? How is it different with circuit swtiching? Which one is superior and why? 6. Compare the delay times between the two types of store and forward switching.
4. Strowger Switching System 1. Describe the different types of tones with proper figure in case of modern telephony. 2. What are tone dialing and pulse dialing. 3. Draw the dialing waveform for the numbers 32, 10, 320. 4. Describe various control tones with proper figure and specification. 5. What are the basic elements of Strowger switching? 6. What are the differences between selector hunter and line finder from the operating principle point of view with the help of proper diagram? 7. For a successful call, present the concept of dialing using a flow chart. 8. Describe 1000 line blocking exchange in detail using proper example and diagram. 9. A calling subscriber with id number 325 is calling for another subscriber with id number 208. Describe briefly the steps employed for successful communication. 10. In a pulse dialing arrangement, pulse in-time is 20 µs. Find the total time required to dial the number 207 if the duty cycle (on) is 6% and inter-digit gap time is 125 µs.
5. Crossbar Switching System 1. ‘‘Crossbar switching system is electromechanical or electromagnetic’’ — why? 2. Describe the principle of operation of Crossbar switching system. What is crossbar matrix? 3. What is cross connection problem? How can we overcome the problem of cross communication?
6. Electronics Switching – SPC 1. What do you mean by stored program controlled switching? Why is it called so? 2. What is the basic principle of operation of SPC? 3. How can we classify SPC? What is the basis of this classification? 4. Which type of distributed SPC can be called as the 1st step towards distributed SPC? 5. Briefly describe three modes of CSPC operation. 6. What is three-level processing? Briefly show how one level can take over other according to priority. 7. What is semaphore? Describe deadlock using semaphore concept. 8. Why system software and application software are needed in SPC. Describe with proper
156
LINE COMMUNICATION SYSTEM
example and diagram.
7. Traffic Engineering 1. Define (a) GOS (b) Blocking probability (c) BHCR (d) CCR (e) BHCA (f) Traffic (g) Congestion (h) Service time (i) Availability (j) Reliability 2. Using SXN matrix, describe full availability model. 3. State Erlang-B congestion formula and proof the same using proper example. 4. Proof that BP = GOS, numerically.
8. Time Division Switching 1. What is the basic concept of time division switching? How is it differ from space division switching or SPC? 2. Describe briefly the principle of operation of TDSS. 3. Describe briefly the principle of operation of TDTS. 4. Describe briefly the principle of operation of TMSS. 5. Describe briefly the principle of operation of TMTS. 6. How the time division multiplexing is done? Why is it called so? 7. What do mean by frame? What are bit and frame synchronization? How is it done?
9. Frequency Division Switching 1. How the frequency division multiplexing is done? 2. Compare time division and frequency division multiplexing. 3. What are group, sub-group, and super-group? How the grouping is done?
157
APPENDIX I
10. Transmission Line 1. What is characteristics impedance? 2. What are propagation and phase constant? Why are they called so? 3. ‘‘For infinite long transmission line, the source impedance is the characteristic impedance’’ — proof. 4. ‘‘For finite long transmission line terminated at characteristics impedance, the source impedance is the characteristics impedance’’ — proof. 5. Determine of Z0 and γ using Zs.c. and Zo.c. 6. What is standing wave? 7. What is the range of values of SWR and why? Describe with proper example. 8. Derive the relation between SWR and reflection coefficient. 9. What are group and phase velocities?
11. Telephone Network 1. What is subscriber loop? Describe subscriber loop with proper diagram. 2. What is switching hierarcy? Describe switching hierarchy with proper diagram. 3. What do you mean by trunk? What is STD? 4. What are critical frequency and skip distance? 5. Briefly describe wave propagation through tropospheric wave propagation. 6. How local, national and international numbering scheme is employed? 7. ‘‘Charging plan is a very important in commercial telephony’’ — justify. 8. Compare between inter and intra-exchange signaling. 9. What is control signal? Compare between in-channel and common-channel signaling.
12. Routing 1. Define (a) Routing, (b) Router, (c) Routes. 2. Briefly describe routing in circuit switched and packet switched network. 3. How centralized and distributed routings are operated? 4. What is flooding ? Describe with an example. 5. What is least cost algorithm? Describe Dijkstra’s algorithm with proper example. 6. Describe the procedure of Bellman-Ford alogrithm with proper example.
13. Wave Propagation 1. What are the two general types of variations in the ionosphere? What is the main difference between these two types of variations? 2. What are the four main classes of regular variation which affect the extent of ionization in the ionosphere?
158
LINE COMMUNICATION SYSTEM
3. Which radio frequency bands use the tropospheric scattering principle for propagation of radio waves? 4. Where is the tropospheric region that contributes most strongly to tropospheric scatter propagation? 5. How frequency affects the critical angle?
14. ISDN 1. What are the merits of ISDN system? 2. What are the new services provided by ISDN? 3. What are the two basic transition states from basic telecommunication system to new age ISDN? 4. Briefly describe ISDN protocol architecture. 5. What is functional grouping? Why is it necessary? 6. Name the reference points in ISDN architecture alongwith their corresponding positions. 7. What are the basic equuipments for ISDN? Describe using figure. 8. What are PRI and BRI? 9. What do you mean by BISDN? Describne its switching technique. 10. What is ATM? How ISDN data is transmitted through ATM network?
Appendix II Sampling Theorem ➢ Statement Sampling theorem state that ‘‘If an analog signal is to be reconstructed without error from its sampled values, the sampling frequency must be at least twice the bandwidth of that analog signal’’. ➢ Proof Say an analog signal m(t) has three different frequency components f1, f2 and f3 where f1 < f2 < f3 then the bandwidth of that m(t) signal must be f3. For ease of calculation, we are taking bandwidth of that signal B Hz, in general. Therefore we can say the spectrum M(ω) is band limited by 2πB in ω scale. Now, sampled signal is nothing but the signal obtained by multiplying m(t) by unit impulse train ∆nTs(t).
Fig. 1. Sampling theorem
From the figure the sampled signal is g(t) = m(t) × ∆nTs(t) Time period of the impulse train is Ts, therefore frequency is fs =
1 . Ts
160
LINE COMMUNICATION SYSTEM
Let’s now expand the ∆nTs(t) signal in Fourier series so that we can study the spectrum of g(t) signal. As ∆nTs(t) is even function of time, By Fourier series, ∝
∆nTs(t) = a0 +
∑ (a
n
cos nωs t + bn sin nωs t )
n =1
Here, 1 a0 = Ts
an =
2 Ts
Ts
∫∆
1 Ts
nTs
(t)dt =
nTs
(t) cosnωst dt =
c
Ts
∫∆ c
2 Ts
bn = 0), for even function) Therefore, ∆nTs(t) =
1 (1 + 2 cosωst + 2 cos2ωst + 2 cos 3 ωst + ...........) Ts
Or, g(t) = m(t) × ∆nTs(t) =
=
=
1 [m(t) + 2m(t) cosωst + 2m(t)cos2ωst + 2m(t) cos3ωst + ..........] Ts 1 [m(t ) + 2m(t )(e jwst + e jwst ) 2m(t ) (e j 3 wst + e j3 wst ) + + ..........] Ts 2 + 2m(t ) (e j 2 wst + e j 2 wst ) 2 2 1 [m(t) + m(t)(ejwst + ejwst) + m(t) (ej2wst + ej2wst) + m(t) (ej3wst + ej3wst) + ..........] Ts
From the 1st term of the equation above, it is very clear that spectum of g(t) will be similar to M(ω) except the amplitude. Amplitude will be
1 times of that of G(ω). 2nd term indicates Ts
M(ω) shifted in both sides by the amount of ωs. (using negative frequency concept). Similarly 3rd term indicates M(ω) shifted in both sides by the amount of 2ωs and so on. Now if we can extract the central spectrum by low pass filter from the assembly of spectrum, then we can easily reconstruct the m(t) signal. ❖ Case I :
APPENDIX II
161
From the figure above specific fs i.e., ωs is taken into account. It is very clear that central spectrum can be extracted using LPF. Therefore it is success of reconstruction. Here, ωs – 2πB = 2πB Or, ωs = 4πB Or, fs = 2B ❖ Case II:
From the figure above other fs i.e., ws is taken into account. It is very clear that central spectrum cannot be extracted using LPF because, there is an overlap region of two spectra. Thge error due to this overlap region is called as alliasing error. Therefore it is failure of reconstruction. Here, ωs – 2πB < 2πB Or, ωs < 4πB Or, fs < 2B ❖ Case III:
From the figure above specific fs i.e., ws is taken into account. It is very clear that central spectrum can be extracted using LPF more easily than case I. Therefore it is success of reconstruction. Here, ωs – 2πB > 2πB Or, ωs > 4πB Or, fs > 2B Therefore collectively we can say fs > 2B i.e., sampling theorem is proved. The minimum frequency of sampling for successful reconstruction is Nyquist sampling rate (fs = 2B). For voice frequency, the bandwidth is typically 4kHz. Therefore in telephony, sampling must be done in 8kHz, i.e., 8kByte/second or 64KBPS.
Appendix III Line Communication System—2004 Time Allowed : 3 hours
Full Marks : 70
The questions are of equal value. The figures in the margin indicate full marks. Candidates are required to give their answers in their own words as far as practicable. Answer any seven questions 1. (a) Explain the principle of DTMF dialling. (b) Write down the advantages of DTMF dialling over pulse dialling. (c) Draw the pulse dialling waveform for the number 30. If the pulse dialling rate is 10 pps and the interdigit gap is 400 ms calculate the time taken for dialling. Neglect the time to rotate the dial fingerplate. 2. (a) How do you define a SPC system? What is the usual sampling time and frame time for a speech carrying system of this type using time division switching? How do you calculate each slot time? (b) What are the special services that can be offered to the users in an electronic exchange? (c) In a crossbar exchange, if the number of subscribers is 64, then find the no. of switching elements and the switching capacity? 3. (a) Assuming simultaneously forward and backward wave propagation in a transmission line derive the following general equation for the line; V = Vl cosh γd + Il Z0 sinh γd I = Il cosh γd +
3 3
2+2
4 2 2+2
I1 sinh γd. Z0
Where V and I are the voltage and current respectively at a distance from the load end of the transmission line having characteristic impedance Z0 and propagation constant γ. The load end voltage and current are Vl and Il respectively. (b) A lossless line has a characteristic impedance of 75 ohms and is terminated in a load of 300 ohms. Find out the reflection coefficient and VSWR in the line.
6
2+2
APPENDIX III
4. (a) What is time division space switching? (b) Explain the principle of operation of time multiplexed space switching with diagram. (c) Calculate the number of trunks that can be supported on a time multiplexed space switch, having an 8 kHz sampling rate, given that: (i) 40 channels are multiplexed in each stream. (ii) Control memory access time is 100 ns. (iii) Bus switching and transfer time is 100 ns per transfer. 5. (a) Explain the terms: (i) Call completion rate (CCR) (ii) Busy hour call attempts (BHCA) (iii) Busy hour calling rate. (b) 10,000 subscribers are connected to an exchange. If the exchange is designed to achieve a CCR of 0.8 when the busy hour calling rate is 4.8, calculate BHCA of the exchange. What should be the call processing time for the exchange? 6. (a) Explain the difference between circuit switching and packet switching. (b) Explain channel associated and channel non-associated common channel signaling (CCS). 7. (a) Distinguish between ‘Grade and Service’ and ‘Blocking Probability’ in loss systems. (b) Over a 30 minute observation interval, 90 subscribers initiate calls. Total duration of the calls is 5400 seconds. Calculate the load offered to the network by the subscribers and the average subscriber traffic. (c) Write down Erlang B formula explaining the meaning of each symbol. What does the formula indicate? 8. (a) Describe switching hierarchy and routing. (b) Write down the differences between in channel and common channel signaling. 9. (a) Write down the main advantages of ISDN. (b) Give a brief description of ISDN protocol architecture. (c) What is B-ISDN? 10. Write short notes on any two of the following: (a) Numbering Scheme. (b) Combination Switching. (c) Subscriber Loop System. (d) Group Velocity.
163 2 4 4
2 2 2
3+1 4 6 3
4 2+1 5 5 3 5 2 5×2
164
LINE COMMUNICATION SYSTEM
Hint of answers according to the instruction of head examiner (West Bengal University of Technology) ➢ Subject : Line Communication System ➢ Paper code : EC 501 ➢ Year : 2004
1. (a) Full form of DTMF (Dual Tone Multiple Frequency) is to be mentioned. DTMF dialing arrangements mentioning all the frequencies are to be given. 1 mark is allotted for the diagram. (b) Advantages like (i) faster dialing rate, (ii) end-to-end signaling, (iii) ease of dialing, (iv) enhancement of signaling capability etc. are to be mentioned. (c) Number 30 means (3+10) = 13 pulses Each pulse duration = 0.1 sec. Therefore, duration of 13 pulses = 1.3 sec Interdigit gap = 0.4 sec Total time required for dialing = 1.3 + 0.4 = 1.7sec For figure, refer to the corresponding chapter. 2. (a) Usual sampling time ≈ 5µs Frame time = 125 µs Time for each slot =
1 sec = tS = tm + td + tt + ti (8000 × 24)
= 5.21 µs (b) Special services include (i) Abbreviated dialing, (ii) Recorded number calls, (iii) Call forwarding, (iv) Operator answer, (v) Call back when free, (vi) Calling number record, (vii) Call waiting, (viii) Automatic alarm, (ix) Conference calls, (x) STD barring etc. (c) Number of switching elements = N2 = 64 × 64 = 4096 Switching capacity =
64 = 32 2
3. (a) Start from the relations – d 2V
so that
or,
dx 2
d 2V dx
2
dV dI = (R + jωL)I and – = (G + jωC)V dx dx
= (R + jωL)(G + jωC)V
= γ2V
dI putting the value of dx
165
APPENDIX III
V = A e– γx + B eγx = A (cosh γx – sinh γx) + B (cosh γx + sinh γx) = (A+B) cosh γx – (A – B) sinh γx = M cosh γx – N sinh γx [Taking M = (A + B), N = (A + B)] and
I=
(N cosh γx − M sinh γx ) Z0
V = Vl cosh γd + IlZ0 sinh γd I = Il cosh γd +
V1 sinh γd, where d = (l – x) Z0
** misprinting in question paper (b) VSWR(ρ) =
Z1 300 = =4 Z0 75
Reflection coefficient (Γ) =
3 (ρ – 1) (4 − 1) = = = 0.6 5 (ρ + 1) (4 + 1)
4. (a) Only principle of operation is to be given (Ref. the chapter of Time Division Switching) (b) 2 marks for diagram and 2 marks for explanation to be given (Ref. the chapter of Time Division Switching) (c) N = Number of trunks =
125 . Here, M = 40 (M × tS )
And tS = 100 + 100 = 200 µs Therefore, N =
5. (a) CCR =
125 ≈ 16 (40 × 200)
(Number of successful calls) (Number of call attempts)
BHCA = Number of call attempts in the busy hour BHCR =
(Average busy hour calls) (Total number of subscribers)
(b) Average busy hour calls = 4.8 × 10000 = 48,000 = BHCA × CCR = BHCA × 0.8 Therefore,
BHCA =
Call processing time =
48000 = 60,000 0.8 (60 × 60) = 60 ms 60,000 sec
166
LINE COMMUNICATION SYSTEM
6. (a) Tabular presentation is preferred. For full marks at least 8 points are to be mentioned Circuit switching
Packet switching
(i) Continuous transmission of data
(i) Transmission of packets
(ii) Dedicated transmission path
(ii) No dedicated transmission path
(iii) Operation in real time
(iii) Operation in near real time
(iv) Message not stored
(iv) Message held for short time
(v) Call set-up delay
(v) Packet transmission delay
(vi) No speed or code conversion
(vi) Speed and code conversion
(vii) Blocking may be occurred
(vii) Blocking cannot be occurred
(viii) Fixed bandwidth transmission
(viii) Dynamic use of bandwidth
(b) 3 marks for diagram and 3 marks for explanation are to be given (Ref. the corresponding chapter) 7. (a) GOS = Lost traffic/ Offered traffic. Blocking probability (PB) is the probability that all the servers in the system are busy. GOS is measured from the subscriber point of view whereas PB is a measure from the switching system or network point of view. GOS is obtained by observing the number of rejected subscriber calls. PB is obtained by observing the busy servers in the switching system. GOS is call congestion or loss probability whereas BP is time congestion. (b) Average call arrival rate, C =
90 = 3 calls/min 30
Average holding time per call = th =
5400 = 1 min/call (90 × 60)
Offered load = C × th = 3×1 = 3 E Average subscriber traffic =
3 1 = E 90 30
AR R! (c) PR = 1 + A + A2 AR + ...... + 2! R !
Where A = offered traffic, R = Number of servers in the system PR = Probability that all servers are busy in the system = Blocking probability PB of the system
APPENDIX III
167
8. (a) Three topologies such as mesh, star and hierarchical topologies are to be mentioned with proper diagrams. Details of hierarchical networks mentioning primary area, territory area, and quaternary area are to be given with proper diagrams. For routing, three types of routing such as right through, open exchange and computer controlled routing are to be discussed in brief. (b) Tabular presentation is preferred. For full marks at least 10 correct points are to be mentioned. 9. Ref. the chapter of ISDN. 10. Ref. the corresponding chapters for short notes.
Bibliography 1. Data communication and networks, Tanenbaum (PHI). 2. Automatic exchange systems, Atkinson (The new era publishing Ltd.) 3. Principle of telephony, N.N. Biswan (Radient books) 4. Automatic control system, Kuo (PHI) 5. Satellite communication, Pratt, Bostain 6. Microprocessor Architecture, Programming and Applications with the 8085, Ramesh Gaonkar (Penram international publishing) 7. Computer Networks, W. Stallings (PHI) 8. Telecommunication Switching System and Networks, Thiagararan Viswanthan (PHI) 9. Electronic Communication System, Kennedy, (IInd Edition), Mc. Graw Hill. 10. Structured Computer Organization, Andrews S. Tanenbaum (PHI) 11. Computer Architecture and Logic Design, Bartee (Mc. Graw Hill) 12. Communication System, A. Aruce Carlson/Paul B. Crilly/Janet C. Rutledge (McGraw-Hill) 13. Principles of communication systems, Taub and Schilling (Tata McGraw-Hill) 14. Reference manual for Telecommunication Engineering, R.L. Freeman (Wiley-Interscience, New York, 1985) 15. Review of Digital Communication, J. Das (Wiley Eastern, New Delhi) 16. Digital Transmission system, Bylaski (Peter Peregrinus Ltd.) 17. Introduction to telephony and telegraphy, E.H. Jolley (A.H. Wheeler and Co.) 18. Encyclopedia of Telecommunication, R.A. Mayers (Academic press, San Diego) 19. Principles of communication engineering, Singh and Chhabra (S Chand) 20. Websites for photo and text materials: (A) http://www.ericsson.com/support/telecom (B) www.seg/co.uk/telecomm/automat4.htm (C) http://people.deas/harvard.edu/~jones/cscie129/nu_lectures/lecture11/switching/ Strowger/strowger.html (D) http://web.ukonine.co.uk//freshwater/howauto1.htm (E) http://www.ieee-virtual-museum.org/collection (F) http://www.nr6ca.org/propagation.html (G) http://www.tpub.com/content/neets/14182/index.htm (H) http://www.cs.ucl/ac.uk/staff/S.Bhatti/D51-notes/node24.html (I) http://www.hp.nasa.gov/office/pao/History/satcomhistory.html