SWITCHING TECHNIQUES

Switching techniques
• Telecommunication Switching Systems and Networks employ Switching
techniques for routing data
• In large networks, there may be more than one path for transmitting data
from the sender to the receiver
• Selecting a path that data must take out of the available options can be
understood as Switching
• The information may be switched while it travels between various
communication channels

• Three typical switching techniques available for digital traffic are −

• Circuit Switching
• Message Switching
• Packet Switching
 Circuit Switching
• Circuit switching is a technique that directly connects the sender and
the receiver in an unbroken path.
• Telephone switching equipment, for example, establishes a path that
connects the caller's telephone to the receiver's telephone by making
a physical connection.
• With this type of switching technique, once a connection is
established, a dedicated path exists between both ends until the
connection is terminated.
• Routing decisions must be made when the circuit is first established,
but there are no decisions made after that time.
 Circuit Switching

• Circuit switching in a network operates almost the same way as the

telephone system works.
• A complete end-to-end path must exist before communication can
take place.
• The computer initiating the data transfer must ask for a connection to
the destination.
• Once the connection has been initiated and completed to the
destination device, the destination device must acknowledge that it is
ready and willing to carry on a transfer.
Circuit Switching
• In Circuit switching, two nodes communicate with each other over a
dedicated communication path
• A circuit is established to transfer the data
• These circuits may be permanent or temporary
• Applications that use circuit switching may have to go through three
phases
• Different phases are −

• Establishing a circuit
• Transferring the data
• Disconnecting the circuit
• figure below shows the pattern of Circuit switching
 Circuit switching

Advantages:
• The communication channel (once established) is dedicated.

Disadvantages:
• Possible long wait to establish a connection, (10 seconds,
more on long- distance or international calls.) during which
no data can be transmitted.
• More expensive than any other switching techniques,
because a dedicated path is required for each connection.
• Inefficient use of the communication channel, because the
channel is not used when the connected systems are not
using it.
• Circuit switching was designed for voice applications
• Telephone is the best suitable example of circuit switching
• Before a user can make a call, a virtual path between the called
subscriber and the calling subscriber is established over the network
• Waiting time lasts long, and there is no data transfer.
• Each connection has a dedicated path, and this gets expensive.
• When connected systems do not use the channel, it is kept idle.
• The circuit pattern is made once the connection is established, using
the dedicated path which is intended for data transfer, in the circuit
switching
• The telephone system is a common example of Circuit Switching
technique
 Message Switching

• With message switching there is no need to establish a dedicated path

between two stations.
• When a station sends a message, the destination address is appended to
the message.
• The message is then transmitted through the network, in its entirety,
from node to node.
• Each node receives the entire message, stores it in its entirety on disk,
and then transmits the message to the next node.
• This type of network is called a store-and-forward network.
Message Switching

• In message switching, the whole message is treated as a data unit

• The data is transferred in its entire circuitry
• A switch working on message switching, first receives the whole
message and buffers it until there are resources available to transfer it
to the next hop
• If the next hop is not having enough resource to accommodate large
size message, the message is stored and the switch waits
 Message Switching

A message-switching node is typically a general-purpose computer. The device needs

sufficient secondary-storage capacity to store the incoming messages, which could be
long. A time delay is introduced using this type of scheme due to store- and-forward
time, plus the time required to find the next node in the transmission path.
• figure shows the pattern of Message switching.
 Message Switching

Advantages:
• Channel efficiency can be greater compared to circuit-
switched systems, because more devices are sharing the
channel.
• Traffic congestion can be reduced, because messages may be
temporarily stored in route.
• Message priorities can be established due to store-and-forward
technique.
• Message broadcasting can be achieved with the use of
broadcast address appended in the message.
Message Switching

Disadvantages
• Message switching is not compatible with interactive
applications.
• Store-and-forward devices are expensive, because they
must have large disks to hold potentially long messages.
• In this technique, the data is stored and forwarded
• Store-and-Forward technique
• This technique was considered a substitute to circuit switching
• But the transmission delay that followed the end to end delay of message
transmission added to the propagation delay and slowed down the entire
process

• Every switch in the transit path needs enough storage to accommodate the entire
message

• Because of the waiting included until resources are available, message switching is
very slow

• Message switching was not a solution for streaming media and real-time applications

• The data packets are accepted even when the network is busy; this slows down the
delivery
• Hence, this is not recommended for real time applications like voice and
video
 Packet Switching
• The packet switching technique is derived from message switching
• Here the message is broken down into smaller chunks called Packets
• The header of each packet contains the switching information which is
then transmitted independently
• The header contains details such as source, destination and intermediate
node address information
• Intermediate networking devices can store small size packets and don’t
take many resources either on the carrier path or in the internal memory of
switches

• Individual routing of packets is done where a total set of packets need not
be sent in the same route
• As the data is split, bandwidth is reduced
• This switching is used for performing data rate conversion
 Packet Switching
• Packet switching can be seen as a solution that tries to combine the
advantages of message and circuit switching and to minimize the
disadvantages of both.
• There are two methods of packet switching: Datagram and virtual circuit.
• Figure below shows the pattern of Packet switching
 Packet Switching
• In both packet switching methods, a message is broken into small parts,
called packets.
• Each packet is tagged with appropriate source and destination addresses.
• Since packets have a strictly defined maximum length, they can be stored
in main memory instead of disk, therefore access delay and cost are
minimized.
• Also the transmission speeds, between nodes, are optimized.
• With current technology, packets are generally accepted onto the network
on a first-come, first-served basis.
• If the network becomes overloaded, packets are delayed or discarded
(``dropped'').
 Packet size

• The size of the packet can vary from 180 bits, the size for the
Datakit® virtual circuit switch designed by Bell Labs for
communications and business applications; to 1,024 or
2,048 bits for the 1PSS® switch, also designed by Bell Labs
for public data networking; to 53 bytes for ATM switching,
such as Lucent Technologies' packet switches.
 Packet switching
• In packet switching, the analog signal from your phone is converted into a digital
data stream.
• That series of digital bits is then divided into relatively tiny clusters of bits, called
packets.
• Each packet has at its beginning the digital address -- a long number -- to which it
is being sent.
• The system blasts out all those tiny packets, as fast as it can, and they travel
across the nation's digital backbone systems to their destination
• The telephone, or rather the telephone system, of the person you're calling.
• They do not necessarily travel together; they do not travel sequentially.
• They don't even all travel via the same route.
• But eventually they arrive at the right point -- that digital address added to the
front of each string of digital data
• At their destination these are reassembled into the correct order, then converted
to analog form, so your friend can understand what you're saying.
 Packet Switching: Datagram
• Datagram packet switching is similar to message switching in
that each packet is a self-contained unit with complete
addressing information attached.
• This fact allows packets to take a variety of possible paths
through the network.
• So the packets, each with the same destination address, do not
follow the same route, and they may arrive out of sequence at
the exit point node (or the destination).
• Reordering is done at the destination point based on the
sequence number of the packets.
• It is possible for a packet to be destroyed if one of the nodes on
its way is crashed momentarily.
Thus all its queued packets may be lost.
 Packet Switching: Virtual Circuit
• In the virtual circuit approach, a preplanned route is established
before any data packets are sent.
• A logical connection is established when
 a sender send a "call request packet" to the receiver and
 the receiver send back an acknowledge packet "call accepted
packet" to the sender if the receiver agrees on conversational
parameters.
• The conversational parameters can be maximum packet sizes,
path to be taken, and other variables necessary to establish and
maintain the conversation.
• Virtual circuits imply acknowledgements, flow control, and error
control, so virtual circuits are reliable.
• That is, they have the capability to inform upper-protocol layers
if a transmission problem occurs.
Packet Switching:Virtual Circuit

• In virtual circuit, the route between stations does not mean that
this is a dedicated path, as in circuit switching.
• A packet is still buffered at each node and queued for output over
a line.

Difference between virtual circuit and datagram approaches:

 With virtual circuit, the node does not need to make a routing
decision for each packet.
 It is made only once for all packets using that virtual circuit.
Packet Switching: Virtual Circuit

VC's offer guarantees that

the packets sent arrive in the order sent

with no duplicates or omissions
with no errors (with high probability)
regardless of how they are implemented internally.
Advantages of packet switching
Advantages:
• Packet switching is cost effective, because switching
devices do not need massive amount of secondary
storage.
• Packet switching offers improved delay characteristics,
because there are no long messages in the queue
(maximum packet size is fixed).
• Packet can be rerouted if there is any problem, such as,
busy or disabled links.
• The advantage of packet switching is that many
network users can share the same channel at the same
time.
• Packet switching can maximize link efficiency by
making optimal use of link bandwidth.
Disadvantages of packet switching

Disadvantages:
• Protocols for packet switching are typically more complex.
• It can add some initial costs in implementation.
• If packet is lost, sender needs to retransmit the data.
• Another disadvantage is that packet-switched systems still
can’t deliver the same quality as dedicated circuits in
applications requiring very little delay - like voice
conversations or moving images.
• The line efficiency of packet switching can be enhanced by
multiplexing the packets from multiple applications over the carrier
• The internet which uses this packet switching enables the user to
differentiate data streams based on priorities
• Depending upon the priority list, these packets are forwarded after
storing to provide quality of service
• The packet switching technique was proved to be an efficient
technique and is being widely used in both voice and data transfer
• The transmission resources are allocated using different techniques
such as Statistical Multiplexing or Dynamic Bandwidth allocation
Statistical Multiplexing
• Statistical multiplexing is a communication link sharing technique, which is used
in packet switching
• The shared linking is variable in statistical multiplexing, whereas it is fixed in TDM
or FDM
• This is a strategic application for maximizing the utilization of bandwidth
• This can increase the efficiency of network, as well

• By allocating the bandwidth for channels with valid data packets, statistical
multiplexing technique combines the input traffic to maximize channel efficiency
• Each stream is divided into packets, and delivered on a first-come, first-served
basis
• The increase in priority levels allow to allocate more bandwidth
• The time slots are taken care not to be wasted in the statistical multiplexing
whereas they are wasted in time division multiplexing
Network Traffic
• Network traffic is simply the data that moves along the network in a given time
• The data transmission is done in the form of packets, where the number of
packets transmitted per unit time is considered as load
• Controlling of this network traffic includes managing, prioritizing, controlling or
reducing the network traffic
• The amount and type of traffic on a network can also be measured with the help
of a few techniques
• Network traffic needs to be monitored as this helps in network security; high data
rate might cause damage to the network
• A measure of the total work done by a resource or facility, over a period (usually
24 hours) is understood as Traffic Volume and is measured in Erlang-hours
• The traffic volume is defined as the product of the average traffic intensity and
the period of

• $$Traffic \:\: volume = Traffic \: Intensity \times Time\: period$$

Congestion
• Congestion in a network is said to have occurred when load on the
network is greater than the capacity of the network
• When the buffer size of the node exceeds the data received, then the
traffic will be high
• This further leads to congestion
• Amount of data moved from a node to the other can be called as
Throughput
• The following figure shows congestion
• In the above figure, when the data packets arrive at Node from the
senders A, B and C then the node cannot transmit the data to the
receiver at a faster rate
• There occurs a delay in transmission or may be data loss due to heavy
congestion
• When too many packets arrive at the port in a packet switched
network, then the performance degrades and such a situation is
called Congestion
• The data waits in the queue line for transmission
• When the queue line is utilized more than 80%, then the queue line is
said to be congested
• The Congestion control techniques help in controlling the congestion
• The graph, drawn between throughput and packet send shows the
difference between congestion controlled transmission and uncontrolled
transmission
• The techniques used for congestion control are of two types – open loop and
closed loop
• The loops differ by the protocols they issue

Open Loop
• The open loop congestion control mechanism produces protocols to avoid
congestion
• These protocols are sent to the source and the destination

Closed Loop
• The closed loop congestion control mechanism produces protocols that allow the
system to enter the congested state and then detect and remove the congestion
• The explicit and implicit feedback methods help in the running of the mechanism
TIME DIVISION SWITCHING
Time Division Switching- working
• Electronic switching systems may be either
• Space Division Switching or
• Time Division Switching

• In space division switching, a dedicated path is established between

the calling and the called subscribers for the entire duration of the
call

• In time division switching, sampled values of speech signals are

transferred at fixed intervals
• Time division switching may be
• analog or
• digital

• In analog switching, the sampled voltage levels are transmitted as

they are

• In binary switching, they are binary coded and transmitted

• If the coded values are transferred during the same time interval from
input to output, the technique is called Space Switching
• If the values are stored and transferred to the output at a late time
interval, the technique is called as Time Switching
 Space Division Switching
• In space division switching, crosspoints are used to establish a specific
connection between two subscribers
• Crosspoints of multistage space switches assigned to a particular
connection is dedicated to that connection for its duration
• Thus the crosspoints can not be shared.
• The paths in a circuit are separated from each other, spatially in space
division switching
• Initially designed for analog networks, but it is being used for both
analog and digital switching
• A Crosspoint switch is mostly referred to as a space division switch
because it moves a bit stream from one circuit or bus to another
• The switching system where any channel of one of its incoming PCM
highway is connected to any channel of an outgoing PCM highway,
where both of them are spatially separated is called the Space
Division Switching
• The Crosspoint matrix connects the incoming and outgoing PCM
highways, where different channels of an incoming PCM frame may
need to be switched by different Crosspoints in order to reach
different destinations
• Space division switching was developed for the analog environment
• It has been carried over to digital communication as well
• This requires separate physical path for each signal connection, and uses
metallic or semiconductor gates

Advantage of Space Division Switching

• It is instantaneous.

Disadvantage of Space Division Switching

• Number of Crosspoints required to make space-division switching are
acceptable in terms of blocking
 Time Division Switching
• Time division switching involves the sharing of crosspoints for shorter periods of time
• This paves way for the reassignment of crosspoints and its associated circuits for
other needed connections
• Therefore, in time division switching, greater savings in crosspoints can be achieved
• By using a dynamic control mechanisms, a switching element can be assigned to
many inlet-outlet pairs for few microseconds
• This is the principle of time division switching
• Time division switching comes under digital switching techniques,
• Pulse Code Modulated signals are mostly present at the input and the output ports
• A digital Switching system is one, where the inputs of any PCM
highway can be connected to the outputs of any PCM highway, to
establish a call
• The incoming and outgoing signals when received and re-transmitted
in a different time slot, is Time Division Switching
• The digitized speech information is sliced into a sequence of time
intervals or slots
• Additional voice circuit slots, corresponding to other users are
inserted into this bit stream of data
• Hence, the data is sent in time frames
• The main difference between space division multiplexing and time
division multiplexing is sharing of Crosspoints
• Crosspoints are not shared in space division switching,
• They can be shared in time division multiplexing, for shorter periods
• This helps in reassigning the Crosspoints and its associated circuitry
for other connections as well
• Time division switches use time division multiplexing, in switching
• The two popular methods of TDM are TSI (Time and Slot Interchange)
and TDM bus
• Data sent at the transmitter reaches the receiver in the same order, in
an ordinary time division multiplexing
• In TSI mechanism, the data sent is changed according to the ordering
of slots based on the desired connections
• It consists of RAM with several memory locations such as input,
output locations and control unit
• TSI changes the ordering of slots based on the desired connections
• The demultiplexer separates the stots and passes them to the proper
outputs
• Both of the techniques are used in digital transmission
• In ordinary time division mutliplexing, the data reaches the output in the
same order as they sent
• TDM bus utilizes multiplexing to place all the signals on a common
transmission path
• The bus must have higher data rate than individual I/O lines
• The TDM uses a control unit
• The control unit opens and closes the gates according to the switching
need
• Main advantage of time division multiplexing is that, there is no need of
Crosspoints
• However, processing each connection creates delay as each time slot must
be stored by RAM, then retrieved and then passed on
Time Division Multiplexing
• When the transmission of data or signals is done in digital means,
using the limited number of resources available, then the Time
Division Multiplexing is used for the transmission of such data
• Multiplexing is the process in communication, which merges two or
more signals at its input into a single output, which when de-
multiplexed, offers all those signals separately as they were
• Multiplexers are broadly classified as Analog and Digital
• Time Division Multiplexing comes under Digital Multiplexing
• There are two types of TDM called as Synchronous and Asynchronous
TDM
Time Division Space Switching
• Time division switches may also employ space division switching
techniques
• An appropriate mixture of both time and space division switching is
advantageous in various circumstances
• A Time division space switch takes outputs of several time-division
switches (say, TSI switches) which are then given as inputs to space
division switches
• This means that one of the two similar outputs produced by a TDM
switch, can be selected by space switch to deliver to another output
path which reduces the number of Crosspoints
• The model of time division space switch is as shown in the following
figure
• The interchange of time slots is not possible in time division switching
• Incoming time slot transfers the data to its dedicated output time slot
only
• Hence, time multiplexed switches do not provide full availability
• A time multiplexed Time Division Space Switch can be configured
around a space array, which has M input horizontals and N output
verticals
• If both inputs and outputs are equal, M=N the switch leads to non-
blocking
• If inputs are greater than outputs; for concentrating switch we have
M>N
• If the outputs are higher, the switch expands gathering one more
connection
• In every time slot, one logic gate per vertical if M>N, or one logic per
horizontal if M>N is enabled for one-to-one connections
• In every time slot, up to N or M samples are switched simultaneously
• Because of the parallel transfer of N or M data samples in each time
slot, a large number of channels can be multiplexed per input line
• If along with multiplexing for N control memory modules, full
availability has to be achieved, one should opt for time division time
multiplexing technique
Time Division Time Switching
• The main advantage of time division time multiplexing technique is
that, unlike time division space switching, it allows time slot
interchange (TSI) of sample values
• In TSI, a speech sample input during a time slot may be sent to the
output during a different time slot
• Which implies a delay between reception and transmission of a
sample
• The time slot counter increments by one, at the end of each clock
pulse, the contents of which provide location addresses for data
memory and control memory
• The input sample is read at the beginning of the time slot and it is
clocked at the end of the clock pulse
• Because of the storage action, the sample is delayed at least by a
single time slot in passing from the input to the output, even if there
is no time slot interchange.
• A TSI which can be expanding or concentrating
• Has different number of time slots per frame at input and output also
• For an expanding switch, the output bit rate is higher
• For a concentrating switch, the input bit rate is higher
• The handling of input and output subscribers in this technique can be
done in four ways, such as serial-in/serial-out, parallel-in/parallel-out,
serial-in/parallel-out,parallel-in/serial-out
TIME MULTIPLEXED SPACE
SWITCHING & TIME
MULTIPLEXED TIME
SWITCHING
a
Time Multiplexed Space Switch
• Time division switches where an outlet or an inlet corresponded to a
single subscriber line with one speech sample appearing every 125 µs
on the line.
• Such switches are used in local exchanges.
• We now consider switches which are essential in transit exchanges.
• Here inlets and outlets are trunks that carry time division multiplexed
data streams.
• We call these switches time multiplexed switches.
• A time multiplexed time division space switch is displayed in Figure.
• There are N incoming trunks and N outgoing trunks, each one
carrying a time division multiplexed stream of M samples per frame.
• Every frame is of 125-µs time duration.
• In one frame time, a total of MN speech samples have to be switched.
• One sample duration, 125µs microseconds, is generally referred to as
a time slot.
• In one time slot, N samples are switched.
• Figureillustrates an output-
controlled switch.
• Output is cyclically scanned.
• There is a 1-to-M relationship
between outlets and control
memory locations
• It implies that there are M
locations in the control
memory corresponding to
each outlet.
• Control memory has MN words.
• If we view control memory as M blocks of N words each, a location
address can be specified in a two dimensional form (i,j)where i is the
block address and j is the word within the block.
• We have 1 < i < M and 1 < j < N. The block address i corresponds to
the time slot i and the word address j to the outlet j
• The first N locations of the control memory correspond to the first
time slot, the next N locations, i.e. locations N + 1 to 2 N when
addressed linearly,
• or locations (2,1) to (2, N) when addressed in a two dimensional form,
correspond to the time slot 2 and so on
• Therefore, if the location ( i, j) contains an inlet address k, it implies
that inlet k is connected to the outlet j during the time slot I
• The number of trunks that can be supported on this switch is given by
N = 125/(Mts)
where ts is the switching time including memory access time per inlet-
outlet pair
• The cost of the switch is estimated as
C = No. of switches + No. of memory words = 2N + MN
• The cost of an equivalent single-stage space division network is
(MN)^2 .
EXAMPLE : Calculate the number of trunks that can be supported on a time
multiplexed space switch, given that
• (a) 32 channels are multiplexed in each stream
• (b) Control memory access time is 100 ns
• (c) Bus switching and transfer time is 100 ns per transfer.

Solution
• We apply EqN = 125/(Mts) to determine N. For M = 32, the total switching time ts
is given by
• ts = 100 + 100 = 200 ns.

• Therefore, N =125/ 32 x 200 x 10^- 3 ≈ 20

• The number of trunks supported can be increased considerably by
split ting the control memory into N independent modules of M
words each and letting each module service one input or output line
• If the control memory modules are arranged such that each one
services one input line and switches the input data to the appropriate
output during every time slot, the switch is input-controlled
• If the modules serve the output lines by switching the appropriate
inputs, the switch is output-controlled.
• An output-controlled configuration is shown in Fig
• There is one control memory module for each output line
• The control memory modules contain the addresses of inlets that
should be switched to the respective outputs in each of the M time
slots. Location 1 of every module corresponds to the time slot 1,
location 2 to time slot 2 and so on
• All locations corresponding to a particular time slot are read out in
parallel
• As a result, there is no constraint due to time on the number of
trunks that can be supported by the switch. The only time constraint
that needs to be satisfied is M = 125/tm
where tm is the control memory access time
• In writing previous Eq. we assume that the time required for
decoding addresses from the control memory and switching the gates
are negligible when compared to the memory access time
• If this is not the case, t m may be substituted by ts which includes
these times
• We now estimate the cost of the switch
• The switch has N switching matrices of size N x 1 each at the inputs.
• There are NM memory words
• Therefore, the cost is
C = N(N x 1) + NM = N^2 + NM
Time Multiplexed Time Switch

• Unlike time multiplexed space switches, it permits time slot

interchange (TSI) of sample values.
• Such an operation necessarily implies a delay between the reception
and the transmission of a sample.
• Illustration in next graph.
TIME SLOT INTERCHANGE SWITCH
• M channels are multiplexed on each
trunk
• The switch is organized in sequential
write/ random read fashion
• Time slot duration, tTS= 125/M
• Time slot clock runs at the time slot
rate
• Time slot counter is incremented by
1 at the end of each time slot.
COMPONENTS

• Contents of counter provides location addresses for the data memory

and the control memory.
• Data memory and control memory access take place simultaneously
at the starting of the time slot.
• Contents of the control memory are used as the address of the data
memory and the data read out to the output trunk
• Even if there is no time slot interchange, a sample is delayed by a
minimum of one time slot in action.
• Depending on the output time slot, delay range is tTS to MtTS
microsec.
• In the example given in figure 1st location in CM contains value 1
which implies that the contents of input time slot 1 is switched to
output time slot 1.
• Delay for this sample is tTS microsec
• Location 2 contains 7. Therefore, input time slot 7 is switched to
output time slot 2.
Delay for this sample: (M – (7-2) + 1)tTS μ s =(M – 4)tTS μ s
• Location 3 contains 4. Therefore, input time slot 4 is switched to
output time slot 3.
Delay for this sample: (M – (4-3) + 1)tTS μ s =MtTS μ s = 125 μ s
There are two sequential memory access per time slot.
• So, time constraint may be stated as tTS = 2tm, 125 = 2Mtm
Cost estimation
• No switching elements!
• Cost is equal to the number of memory locations.
• There are M locations each in the control and data memory
• So, total cost is given by
C = 2M units
EXAMPLE
EXAMPLE 2 Calculate the maximum access time that can be permitted for the data
and control memories in a TSI switch with a single input and single output trunk
multiplexing 2500 channels. Also, estimate the cost of the switch and compare it
with that of a single stage space division switch.

Solution:
tm =125µs/2M=(125*10^-6)/(2500*2) = 25 ns and C = 2 * 2500 = 5000 units
This switch is non-blocking and fully available.
• An equivalent single stage Space Division Switch uses a matrix of 2500 X 2500. So,
cost is 6.25 million units!
• Cost advantage of time switch = (6.25*10^6)/5000 = 1250 Thus a TSI switch is a
lot cheaper.