Top Banner
CSE-342 B.Eng. (Electrical Engineering) Telecommunication Network Lecture 5 Telecommunication Switching Muniba Q. Zaman
86
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: lec5

CSE-342

B.Eng. (Electrical Engineering)

TelecommunicationNetwork

Lecture 5

Telecommunication Switching

Muniba Q. Zaman

Page 2: lec5

EVOLUTION

CSE-342 2

Page 3: lec5

Switching

If we could have point to point lines between all senders and all receivers, then we would not need switching

3CSE-342

Page 4: lec5

Switching

Imagine point to point lines between all pairs of people:-

# of people # of lines2 13 3

10 45 100 4950 1000 0.5 million

In general, for n people ?Advantage ?Disadvantage ?

4CSE-342

Page 5: lec5

Switching

A switch provides temporary path between end users in a communication network

Traffic engineering is science of designing switching network so that circuits are optimally used & are highly available, most important consideration is cost

Shehryar

MoezUmar

Arsalan

Rohan

Umair

Usama

5CSE-342

Page 6: lec5

Introduction Definition

Switching the process of directing a signal from its source to its destination the establishing, on demand, of an individual connection from a desired inlet to a

desired outlet within a set of inlets and outlets for as long as is required for the transfer of information

Directing Data From Source to Destination involves: involves: Dedicated lineDedicated line – a telephone line leased by a firm and used exclusively by that firm SwitchSwitch – a special purpose computer that directs incoming messages along a path Private branch exchange (PBX)Private branch exchange (PBX) – a special-purpose computer that distributes calls

within a customer’s site Switch Switch a special purpose computer that directs incoming messages along a path

Switching example – PSTN, PDN, PSPDN

6CSE-342

Page 7: lec5

Exchange Structure

Elements of an exchange include following :-1. Concentrator

Concentrates calls from terminations to links2. Expander

Expands terminating calls to terminations & trunks3. Distributor or group switch

For interconnecting switches, # of links decided on statistical basis Types of calls could be :-

Own exchange calls Outgoing junction calls Incoming junction calls Transit calls

Signaling System To pass call connection / disconnection information between termination &

exchange and between exchanges Channel associated or common channel signaling

Control System To interpret signaling information and take appropriate measures

7CSE-342

Page 8: lec5

Switching Types

Point-to-Point Network Switching Circuit Switching, Message Switching, Packet Switching, Cell Switching Connection-Oriented versus Connectionless Virtual Circuit versus Datagram Networks Internal/External Abstractions

Point-to-Point Switching Circuit Switching Store-and -Forward Networks

Message Switching Packet Switching

connection-oriented vs connectionless virtual circuit vs datagram

Cell Switching

CSE-342 8

Page 9: lec5

Switch System Classification

9CSE-342

Page 10: lec5

Message Switching

Message is first stored in buffer and then sent forward as and when resources become available or sufficient messages have been collected

Also called store and forward It is example of queuing system Example is telegraph message Advantage

Better utilization of bandwidth

Disadvantage Delay

10CSE-342

Page 11: lec5

Circuit Switching

A path is established between caller and destination for that call

If all the ccts are busy then call is lost therefore it is an example of lost call system

Advantage No delay , real time communication

Disadvantage Low bandwidth efficiency, ccts used for fraction of time

speech contains pauses and silent interval

Example is PSTN

11CSE-342

Page 12: lec5

CSE-342 12

Circuit Switching

Circuit Switching (Telephone network) One physical link is established for each connection

Seeking out and establishing a physical copper path end-to-end [historic definition] .

Circuit switching implies the need to first set up a dedicated, end-to-end path for the connection before the information transfer takes place.

Once the connection is made the only delay is propagation time. Advantage

No delay , real time communication Disadvantage

Low bandwidth efficiency, ccts used

for fraction of time speech contains

pauses and silent interval

Page 13: lec5

Packet Switching

Message is broken down into packets and then each packet is sent separately (possibly by different routes)

If outgoing routes are busy then they are kept in queue therefore packet switch is an example of queuing system

Switches should be intelligent to choose best routing path

Example is internet Advantage

Better bandwidth efficiency Disadvantage

Packet switching introduces latency 13CSE-342

Page 14: lec5

Packet Switching

CSE-342 14

Page 15: lec5

Comparison

Characteristics Circuit Packet

Origin Voice tel Data NW

Connection less or oriented

Connection- Oriented

Both

Delay Low High

NW intelligence

centralized De-centralized

BW efficiency Low High

Information Loss

Low High15CSE-342

Page 16: lec5

Circuit Switching vs. Packet Switching

CIRCUIT SWITCHINGCIRCUIT SWITCHING: The method used in telephone networks Sets up a temporary circuit between the source and the destination Resources are reserved for the duration of the session (call)

PACKET SWITCHINGPACKET SWITCHING: Appropriate when data are transmitted infrequently from a large

number of nodes Used on the Internet

The message is divided into packets containing control information No circuit is established

Provides better sharing of resources Multiple users share the same resources

A store-and-forward network where the block of transfer is a complete packet. A packet is a variable length block of data with a tight upper bound.

CSE-342 16

Page 17: lec5

Circuit Switching vs Packet Switching

(a) Circuit switching. (b) Packet switching.

CSE-342 17

Page 18: lec5

CSE-342 18 Circuit switching Message switching Packet switching

Event Timing

Page 19: lec5

Event Timing

CSE-342 19

DCC 6th Ed., W. Stallings, Figure 10.3 Circuit switching Virtual circuit packet switching Datagram Packet switching

Page 20: lec5

Connection-Oriented

A setup stage is used to determine the end-to-end path before a connection is established.

Data flow streams are identified by some type of connection indicator (e.g. OSI, X.25, SNA).

CSE-342 20

Page 21: lec5

Connectionless Oriented

No set up is needed Each packet contains information which allows the packet to be

individually routed hop-by-hop through the network.

CSE-342 21

Page 22: lec5

Datagram vs Virtual Circuit

Datagram Each datagram packet may be individually routed.

Virtual Circuit Virtual circuit set up is required. All packets in a virtual circuit follow the same path.

CSE-342 22

Page 23: lec5

External Virtual Circuit/Datagram

CSE-342 23

Page 24: lec5

Internal Virtual Circuit

And Datagram

Operation

CSE-342 24

Page 25: lec5

Connection Strategies

Circuit switching. A permanent physical link is established for the duration of the communication (i.e., telephone system).

Message switching. A temporary link is established for the duration of one message transfer (i.e., post-office mailing system).

Packet switching. Messages of variable length are divided into fixed-length packets which are sent to the destination. Each packet may take a different path through the network. The packets must be reassembled into messages as they arrive.

Circuit switching requires setup time, but incurs less overhead for shipping each message, and may waste network bandwidth. Message and packet switching require less setup time, but incur more overhead per message.

CSE-342 25

Page 26: lec5

CSE-342 26

Circuit switched and packet-switched networks

Page 27: lec5

CSE-342

B.Eng. (Electrical Engineering)

TelecommunicationNetwork

Switching Technology

Page 28: lec5

Categorizing switching

CSE-342 28

SPC: Stored program control

Page 29: lec5

Switch System Classification

29CSE-342

Page 30: lec5

Manual Switching 1876 A. G. Bell telephone patent 1878 The first exchange constructed

in La Porte, the US connect any two of the 21

subscribers using manual switching A ringing signal sounded at the

operator's switchboard when any of the subscribers turned the crank of his telephone

Upon answering the signal, the operator was asked to connect the call

End-to-end physical path or connection

Sequencing preserved: ABCDE in implies ABCDE out

Nothing is lost or inserted

CSE-342 30

switchboard

Plug-cord

Page 31: lec5

Manual Switching

A simple manual switchboard has subscribers’ terminations connected to jacks (sockets)

The “links” consist of a pair of plugs on flexible cords (wires) which can be used to interconnect any pair of jacks

Signaling consists of calling indicators (e.g. relays and lamps) and clearing indicators

Call routing instructions are passed verbally

31CSE-342

Page 32: lec5

Manual Switching

32CSE-342

Page 33: lec5

Manual Switching

Manual exchange demonstrate following features: Central-battery operation Loop/disconnect signaling The multiple Busy testing Concentration Metering and ticketing Classes of service Common control Scanning Stored-program control Common-channel signaling

33CSE-342

Page 34: lec5

Manual Switching

Advantages Human control Services such as advice of duration and charge, transfer of calls

when absent, wake up calls etc., which are so complex to provide automatically, pose no problem on manual exchanges

Disadvantage The weakness of the manual exchange, which has resulted in its

almost complete disappearance, was essentially its slowness Long distance calling involved chain of operators

34CSE-342

Page 35: lec5

Functions of a Switching System

1. Attending

2. Information receiving (address signal)

3. Information processing

4. Busy testing

5. Interconnection

6. Alerting

7. Supervision

8. Information sending

CSE-342 35

Page 36: lec5

Switch System Classification

36CSE-342

Page 37: lec5

Strowger

In 1891, first automatic telephony system was introduced by Strowger

Strowger’s business was being taken away by his competitor’s wife who was working as telephone operator

Stepping relay was used to allow a subscriber dialing a telephone to select one of ten lines

When the subscriber dials the telephone a series of electrical pulses are generated on the line (at a maximum rate of ten per second)

Each pulse causes the uni-selector (which starts at the 'home' position) to be advanced by one step 37CSE-342

Page 38: lec5

Strowger (Two motion Selector)

Two motion selectors typically have 10 rows with 10 contacts each

It can move vertically and horizontally It can therefore accept two dialed digits from

a subscriber and route the call to any of 100 numbers

The first digit moves the selector vertically up to the corresponding level

The second digit moves the wipers around the contacts of that level

38CSE-342

Page 39: lec5

Strowger’s Step By Step Switch

CSE-342 39

Rotating shaft

Slidingarm

10 X 10 Array of contacts

Page 40: lec5

Two motion Selector

40CSE-342

Page 41: lec5

Two motion Selector

41CSE-342

Page 42: lec5

Strowger

In an 100 line exchange each user has to be provided with his own 100 outlet 2 motion selector

This can be avoided by using stage of uni selectors at the beginning

If we want to increase the size of exchange then we can use more banks of 2 motion selectors

Director is used to translate the digits dialed by user into actual routing digits for calls outside exchange

42CSE-342

Page 43: lec5

Strowger Switch

CSE-342 43

Strowger switch

Page 44: lec5

Uniselector & Group selector

Uniselector one which has a single rotary switch with a bank of contacts Depending upon the number of switching contacts, uniselectors

are identified as 10 outlet or 24 outlet uniselectors Several uniselectors can be graded together so that multiple

incoming circuits can be connected to multiple outgoing circuits cheaper act as a concentrator to enable a large number of lines to share

a smaller number of two-motion selectors. Operation: self-drive mode of operation performed by a drive

mechanism of a rotary switch. This mechanism contains an armature, electromagnet, Pawl, and Ratchet wheel. The wiper is attached to the ratchet wheel. When the line relay detects a calling signal, the magnet is energized and operates the ratchet wheel, pawl and its associated wiper. When the electromagnet is de-energised the armature is released and returns to its rest position.

CSE-342 44

Page 45: lec5

CSE-342 45

Page 46: lec5

Uni-Selector

46CSE-342

Page 47: lec5

Strowger Switch

CSE-342 47

Uniselector

0

9

8

7

6

5

4

3

2

1

0

9

8

7

6

5

4

3

2

1

0

9

8

7

6

5

4

3

2

1

0

9

8

7

6

5

4

3

2

1

1st 2nd 3rd Final

A

B

C

Group Selector

Page 48: lec5

Step by Step Switching

In general, the Strowger switching system consists of: subscriber’s line circuit, line finder & alloter circuit Group selector and final selector

CSE-342 48

Page 49: lec5

Subscriber line circuit (SLC) Every subscriber is connected to his local exchange by one

pair of wires This single pair carries the voice in both directions and the

ring current to ring the bell when a call is received At the exchange, every subscriber line terminates into its

own SLC This consists of a pair of relays dedicated to that subscriber Remaining switching circuits are shared by all the

subscribers. When the subscriber lifts his handset, current starts to flow

on the line. This is detected by the SLC.

CSE-342 49

Page 50: lec5

Line Finder & Alloter Mechanism to find a free selector and to connect the calling

subscriber to that free selector

Group selector used in the exchange that has more than one rank of two-

motion selector. The selector each step vertically in response to a dialed digit. However, during the inter-digit pause they operate in a self-drive mode like a uniselector. Thus, the wiper hunt for a free outlet on the selected level, which leads to a selector in the next switching stage. The selector in the last stage respond to two dialed digits and is called final selectors.

CSE-342 50

Page 51: lec5

Routing of a Local call in Strowger System

CSE-342 51

Page 52: lec5

Why step by step? Disadvantages:

regular maintenance by the skilled technicians are necessary It is not feasible to select an alternate route for interoffice calls, if all the

trunks are busy limited to dial pulses If calling rate is high, heavy operation is performed by the system and

the life time of the system is less Congestion could arise when the switching system is heavily loaded capacity of switching system reduces because of fire service, police

ambulance, fault reports, directory enquiry, operator assistance etc can accept only 7 to 9 pulses in 1 second hence not suitable for fast

dialing Large in size Noise

CSE-342 52

Page 53: lec5

Circuit . . .

CSE-342 53

SxSlinklink

SxSlink

Could be 10 or more links

Page 54: lec5

Example

Construct a switch with 50 outlets of 3 wires each from a 25-way uniselector with 6 wipers

54CSE-342

Page 55: lec5

Linked Numbering Scheme (LNS)

For junction calls, dialed digits are used to route a call through each exchange

This requires different dialing codes to reach the same destination from different originating exchanges

Example: call from exchange A to A, B to A, C to A, D to A via C

This is Disadvantageous Hence use Linked Numbering scheme – numbering

scheme of an area and the trunking of its exchange are arranged so that a customer can be reached by dialing the same digits from any exchange in the area

Example: 10,000-line main exchange and several 1000-line satellite exchange can use 5 digit linked numbering scheme. This require an extra rank of selectors

CSE-342 55

Page 56: lec5

Register-translator-senders

Although LNS used widely but larger cities require a more general solution

Ideally, route for call should be completely separate from the digits dialled to establish the connection

Solution: Register-translator-sender equipment Number dialled by the customer is stored (registered) in

the register Stored digits are then analyzed to determine the routing If necessary, part of the number is translated into a

different number which is sent out to establish the connection

By using different translations all customer can dial same digits to reach a terminating exchange

CSE-342 56

Page 57: lec5

Register used only at the beginning of call Time taken to complete the whole process is only fraction of a

minute Therefore, registers are provided as common controls,

connected to a trunk by an auxiliary switch In electromechanical systems, register function is performed by

relays or switches Modern systems use more rapid signaling methods in order to

reduce the post-dialing delay Translator function consists of looking up a table of data

Dialed digits stored in the register are used to access an address in a store and the number read out from that address is the translation

Since charge for a call depends on its destination, translation may also contain the charging rate for the call

CSE-342 57

Register-translator-senders

Page 58: lec5

Registers were added to step-by-step exchange for use in large cities

Uniform 7-digit numbering scheme is used throughout the area

The first three digits (ABC) are an exchange code (which is translated) and the last four digits are the customer’s number on that exchange (which is not translated)

These registers are called Directors

CSE-342 58

Page 59: lec5

Switch System Classification

59CSE-342

Page 60: lec5

Crossbar Strowger switches require regular maintenance; banks

need cleaning, mechanisms need lubrication and adjustment, and wipers & cords wear out

Replace manually operated switch by a matrix of telephone relays, with their contacts multiplied together horizontally and vertically.

This means a switch with N inlets and N outlets will require N2 relays for its crosspoint Uneconomic solution

More economic solution was provided by G. A. Betulander in 1917 by the invention of crossbar switch

Crossbar switch retains a set of contacts at each point but these are operated through horizontal (select magnets) and vertical (hold or bridge) magnets at the sides of the switch

60 CSE-342

Page 61: lec5

CSE-342 61

Page 62: lec5

Crossbar

A crossbar switch connects multiple inputs to multiple outputs in a matrix manner

If the switch has M inputs and N outputs, then a crossbar has a matrix with M x N cross-points

At each cross-point some form of switch contact is needed to complete the connection between horizontal and vertical lines

Any of the M inlets can be connected to any of the N outlets by closing the appropriate switch contacts

Maximum number of simultaneous connections that can be carried by a matrix switch is given by which ever of the number of inlets or outlets is smaller

62CSE-342

Page 63: lec5

CSE-342 63

Page 64: lec5

Inlet 1 to outlet  2  by  closing  contact ? Inlet 4 to outlet  3  by  closing  contact  ?

64CSE-342

Page 65: lec5

Crossbar

Source:M. P. Clark, Networks and Telecommunications Design and Operation – 2nd Edition, John Wiley & Sons Ltd, pp. 96, 1997.

Crossbar switch

65CSE-342

Page 66: lec5

Crossbar Switches

Crossbar exchanges are example of common/register controlled systems i.e. control circuit is shared

Basic building blocks are :- Link frame (consist of number of crossbar) Marker (Controls connection between inlets and outlets) Register (decodes dialed number)

Crossbar has no ‘intelligence’ A Marker must decide which magnets to operate a single marker can control many switches and serve

many registers Thus even a large exchange needs few markers This is called centralized control

66CSE-342

Page 67: lec5

Crossbar Switches

Customer line circuit detects calling condition Signals relevant marker to establish

connection through concentration stage to free transmission bridge

Transmission bridge uses another marker to select free register

Register will receive and analyze number and then seize Marker to set appropriate switches in group selector or concentration stage

67CSE-342

Page 68: lec5

Crossbar Switches

Once connection has been established then marker and register are released

Less numbers of markers and registers are needed as they are being used during setting up of connection and dialing only

Supervision of call is then provided by transmission bridge

Dialed number and number allocated to particular line need not to have any relation due to use of register for translation

Number of switching stages need not to be dependent on exchange numbering scheme

68CSE-342

Page 69: lec5

Crossbar

CSE-342 69

horizontals

Crosspointdetail

verticals

Page 70: lec5

Crossbar

CSE-342 70

Page 71: lec5

Multistage switching

Three possible situation for crossbar could be :- m=n (non blocking), m>n (concentrating), m<n (expanding)

As N increases the number of required cross points can be excessively high but fewer than 25% of the cross points are used at a given time.

In order to save cross points, crossbar can be arranged in stages

It can suffer internal blocking if sufficient number of stages are not provided

71CSE-342

Page 72: lec5

Multistage switching

How does it work?

Divide the 16 inlets into groups of 4. 1st outlet of each Stage 1 block is connected to an inlet of

the 1st Stage 2 block. 2nd outlet of each Stage 1 block is connected to an inlet of

the 2nd Stage 2 block. 3rd outlet of each Stage 1 block is connected to an inlet of

the 3rd Stage 2 block… ith outlet of each Stage 1 block is connected to an inlet of

the ith Stage 2 block.

1

4 x 4

2

4 x 4

3

4 x 4

4

4 x 4

1

4 x 4

2

4 x 4

3

4 x 4

4

4 x 4

72CSE-342

Page 73: lec5

Multistage switching

100 x 100 using 10 by 10 matrices

10 x 10

1

10

1

10

10 x 10

1

10

1

10

(10)

10 x 10

1

10

1

10

10 x 10

1

10

1

10

(10)100 Inlets 100 Outlets

73CSE-342

Page 74: lec5

It may not always be possible to make a connection even when the required outgoing trunk is free???

Blocking – there are free outgoing trunks and free links available but they cannot be used together

Traffic capacity is therefore affected by internal blocking as well as congestion of the external trunks

When marker is instructed to set up a connection, this also defines the link to be used and the select and bridge magnets to be operated

The connection is not made until the marker has checked the busy/free condition of the outgoing trunk and the relevant link

The connection is made only if both are found to be free This is called conditional selection

CSE-342 74

Page 75: lec5

Multistage Switching

75CSE-342

Page 76: lec5

Multistage Switching

Let N = number of inlets and outlets Let n = size of each inlet-outlet group Let k = number of center-stage groups There are N/n groups in the first and last stages There are (N/n) (nk) cross points in the first and

last stages There are k middle groups with (N/n)2 cross

points in each middle array There are Nx = 2Nk+k(N/n)2 cross points total

which is much smaller than the number of cross points in a single-stage switch (N2)

76CSE-342

Page 77: lec5

Multistage Switching

Example: For N = 100, The square matrix will require 100 x 100 = 10,000 cross points Assuming n = K = 10; # of cross points are: = 10(100/10)2 + 2 x 100 x 10 = 2200 Saving = 78%. The saving will increase as N

increases Advantage: The number of cross points Disadvantage: blocking Clos criterion: condition of non blocking

n = (N/2)1/2

k > 2n – 1 Cross points ≥ 4N [(2N)1/2 – 1]

77CSE-342

Page 78: lec5

Switch System Classification

78CSE-342

Page 79: lec5

Stored Program control

In electromechanical exchanges, common control mainly use switches and relays

Since they are operated much more frequently therefore wear out earlier

Also, Crossbar switches were slow in processing calls Therefore, relays were replaced by vacuum tubes,

transistors, CRT etc to improve speed, whose life is almost independent of its frequency of operation

After invention of modern digital computers they decided to use them for controlling switches also

Computer use the store program concept i.e. instructions are stored in memory and are executed one by one

SPC also carries out exchange control function through program stored in memory

79CSE-342

Page 80: lec5

Stored Program control

SPC allowed several features to be extended to users like: Call barring Repeat last call Reminder calls e.g. wake-up call Call diversion Three way calls Charge advice

Computer has to be highly tolerant to faults In 1965, first SPC exchange was installed in USA by

AT&T at New Jersey There are two approaches of implementing SPC

Centralized Distributed

80CSE-342

Page 81: lec5

Central Control

In Central control, central processor controls every thing and performs different functions with the same hardware by executing different programs

Central control uses reed relay for cross points Reed relay consist of sealed contacts inside an

operating coil and glass They are faster and more reliable as there are

sealed and there are no external moving parts Each cross point has its own reed relay This system is similar to manual exchange i.e.

single processor is controlling every thing

81CSE-342

Page 82: lec5

Reed Relay

82CSE-342

Page 83: lec5

Reed Relay with Central Control

A scanner is continuously monitoring all the connections

This record is accessible to processor Processor uses marker to establish path

between calling line and register Register sends data to processor for action Processor establishes connection through reed

relay switching network using marker More then one processors are employed in

actual exchanges

83CSE-342

Page 84: lec5

Reed Relay with Central Control

Registers

Processors Scanner

Markers

LC Reed Relay

SW NW

Junctions

84CSE-342

Page 85: lec5

Reed Relay with Central ControlTXE-4 (Telephone Exchange Electronics)

85CSE-342

Page 86: lec5

Distributed Control

Control functions are shared by many processors in an exchange

Offers better availability and reliability and was possible because of low cost of processors

Control functions can be divided horizontally or vertically

Exchange control functions are divided among number of processors and each processor is responsible to do all the tasks

In horizontal decomposition, each processor perform only one or some function

86CSE-342