EEE 497: Telecommunication Networks

Telecommunication Switching System

Dr. Md. Forkan Uddin
Professor, Dept. of EEE, BUET, Dhaka 1000

1
 Types of Switching
• Space division switching
 Blocking vs non-blocking
 Single stage vs multi-stage
• Time division switching
• Time-space switching
• Time-space-time switching
• Space-time-space switching

2
2
Space Division Switching: Crossbar

3
 Crossbar Switching: Rectangular
• Contains an array of horizontal and vertical wires
• Both wires are connected to initially separated contact points
of switches
• Horizontal and vertical bars (shown as dotted lines) are
mechanically connected to these contact points and attached
to the electromagnets
• When both horizontal and vertical bars connected to the
electromagnet are activated, the contact of the intersection of
the two bars will close together
• Thus the contact is made and continues to hold
• When the electromagnets are de-energized both horizontal
bar and vertical bars are released from the contact

4
 Crossbar Switching: Rectangular
• The crossbar switch is known as a non-blocking crossbar
configuration
• It requires N*N switching elements for N subscribers
• The paths in the circuit are separated from each other
spatially
• A crosspoint switch is referred to as a space division switch
because it moves a bit stream from one circuit/bus to another

5
Problems in Single Stage X-bar Switching
• The number of switches in single stage O(N2)
• The large number of cross-points on each inlet and outlet line
imply a large amount of capacitive loading on the message
paths
• Another fundamental deficiency of single-stage switches is that
one specific cross-point is needed for each specific connection
• If that cross-point fails, the associated connection cannot be
established
• Analysis of a large single-stage switch reveals that the cross-
points are very inefficiently utilized

6
 Multi-stage (X-bar) Switching
 How to increase the utilization efficiency of the cross-points?
Any particular cross-point be usable for more than one
potential connection (Blocking may occur)
• How to avoid blocking?
 More than one path be available for any potential
connection
• The alternate paths serve to eliminate or reduce blocking and
also to provide protection against failures
• The sharing of cross-points for potential paths through the
switch is accomplished by multiple-stage switching

7
Multi-stage X-bar Switching

8
 Multi-stage X-bar Switching
• k possible paths through the switch for any particular
connection between inlets and outlets
• Each of the k paths utilizes a separate center stage array
• Provides alternate paths through the switch to circumvent
failures
• Each switching link is connected a limited number of cross-
points and hence, capacitive loading is minimized
• Total number of cross-points, Nx
N/n input switches × nk
crosspoints/input switch

 k intermediate switches×(N/n)2
crosspoints/intermediate switch

 N/n output switches × nk

crosspoints/output switch 9
 Non-blocking Multi-Stage

• Since each first-stage array has n inlets, only n - I of these inlets can be
busy
• Similarly, at most n - I links to the appropriate third-stage array can be
busy
•Thus, k=(n-1)+(n-1)+1 =2n -1 to avoid blocking fully 10
 Non-blocking Multi-Stage Switching
• Thus,

• The optimum value of n is

• The optimum value of number of switches

11
 Multi-Stage Switching with Blocking
• Strictly non-blocking switches are rarely needed
• Equipment for the public telephone network is designed to
provide a certain maximum probability of blocking for the
busiest hour of the day
• A typical residential telephone is busy 5-l0% of the time
during the busy hour
• A certain percentage of blocking reduces the number of
switches significantly

12
 Multi-Stage Switching: Blocking
Probability
Notations:
p: fraction of time that a particular link is in use, i.e., the
probability that a link is busy
q=1-p: the probability that a link is idle
• Assumption: The probabilities in different links are
independent
 When n parallel links are used for one connection, the
probability that all the links are busy, i.e., blocking probability
B=pn

 When n series links are used for one connection, the

probability that all the links are free is qn . Hence, blocking
probability
B=1-qn 13
 Multi-Stage Switching: Probability
Graph (Lee Graph)
• p’= probability that an interstage link is busy

14
Multi-Stage Switching: Blocking
Probability

15
Performance with Blocking

16
Five Stage Switching Network

17
Probability Graph and Blocking
Probability

18
Impact of increasing Stage

19
Limitations of Lee Graph Based Models
• Assumption: The busy/idle probabilities in the different links
are assumed to be independent
• Facts:
 The probabilities are not independent particularly when
significant amounts of expansion are present
 Consider a switching matrix with k=2n -1 in three stage
switching, the expression of blocking probability provide finite
blocking probability even though the switch is known to be
strictly non blocking (Why?????)
 The inaccuracy is due to the fact the when n-1+n-1=2n-2
paths are busy, the remaining paths are assumed also busy
with probability though those are idle

20
Limitations of Lee Graph Based Models
• When space expansion exists, the Lee model provide
erroneously high value of blocking probability
• A more accurate but not exact analysis of multi-stage
switching matrices was presented by C. Jacobaeus in 1950

21
Jacobaeus model

22
Comparison

23
 Time Division Switching (TDS)
• Multiple-stage switching save cross-points as well as
implementation cost in space division switching
• In space division switching, a cross-point assigned to a particular
connection is dedicated to that connection for its duration
• Time division switching involves the sharing of cross-points for
short periods of time
• This paves way for the reassign of cross points and its associated
circuits for other needed connections
• Greater savings in cross points can be achieved
• By using a dynamic control mechanisms, a switching element can
be assigned to many inlet-outlet pairs for few microseconds
• Time division switching uses time division multiplexing to achieve
switching
• Two types: analog TDS and digital TDS 24
 Analog TDS
• Analog time division switching has become obsolete

25
 Analog TDS
• A single switching bus supports a multiple number of
connections by interleaving PAM samples from receive line
interfaces to transmit line interfaces
• Two cyclic control stores are used
• First control store controls gating of inputs onto the bus one
sample at a time
• The second control store operates in synchronising the first
and selects the appropriate output line for each input sample

26
 Digital TDS
• The digital time division multiplexed signals requires (i)
switching between time slots and (ii) switching between
physical lines
• The switching between time slots are usually referred as time
switching
• Thus, time switching and line switching are required
• Time switching is done by time slot interchange (TSI)
• Line switching is done by controlling input and output gates

27
 Digital TDS: Time Switching
• The basic requirement of time switching is that the transfer of
information arriving at in a time slot of one input link to other
time slot of any one of output link
• A time switch operates by writing data into and reading data
out of a single memory and uses time slot interchange (TSI)

28
 Digital TDS: Time Switching
• In TSI operation, inputs are sequentially controlled and
outputs are selectively controlled
• The RAM have several memory locations, each size is the
same as of single time slot
• S/P and P/S converters are used to write the data into the
memory and read the data out of memory

Memory Data Register (MDR) 29

 Digital TDS: Line Switching
• The input and output lines are connected to a high speed bus
through input and output gates
• Each input gate is closed during one of the available time slots
• During the same time slot, only one output gate closed
• This pair of gates allows a burst of data to be transferred from
one input line to a specific output line through the bus
• The contol unit opens and closes the gates according to
switching need

30
 Digital TDS: Example
• Channel 3 of the first TDM link is connected to channel l7 of
the last TDM link
• The indicated connection implies that information arriving in
time slot 3 of the first input link is transferred to time slot l7 of
the last output link
• Return connection is required and realized by transferring
information from time slot 17 of the last input link to time slot
3 of the first output link

31
 Two Dimensional Switching
• Combination of the time and space switches leads to a
configuration that achieved both time slot interchange and
sample switching across trunks
• These structures also permit a large number of simultaneous
connections to be supported for a given technology
• Larger digital switching system requires both time and space
switching

32
 Two Dimensional Switching
• Time- space (TS) switching: use time and space switching
• Time switching: Time slot interchange (TSI)
• Space switching: cross-bar switching

33
 Multi-dimensional Switching
• The multiple stages overcomes the limitations of the
individual switches and cost savings can also be achieved
• TST, STS, TSST, TSSSST and TSTSTSTSTSTSTS are the switching
system configurations used in digital switching system
• TST structure is the most common

34
STS

35
TST

36