1

Delay in Packet Switched Networks

Required reading:
Kurose § 1.4

CSE 3214, Winter 2016

Instructor: N. Vlajic
 2

Message

Message

Subscriber
A Message

Subscriber
B

Network
nodes

How do we characterize the performance of a network?!

 3
Delay in Packet-Switched Networks

Link/Network Performance Measures: throughput and delay

Link Throughput – capacity potentially available for an application –

generally expressed in “bit per second” [bps]
 effective capacity my be reduced by application multiplexing
and protocol overhead (header bits, acknowledgments, etc.)

effective throughput

Network Delay – average time for a block of data to go from an application on

one system to corresponding application on another system
 4 types of delay contribute to overall delay:

dtotal  dprocessing  dqueueing  dpropagation  dtransmission

 4
Delay in Packet-Switched Networks (cont.)

transmission

A propagation

B nodal
processing queueing

Processing Delay: time required to process a packet – to check for bit errors,
to determine output links, etc.
(a) at source prior to sending,
(b) at any intermediate router, and
(c) at destination prior to delivering to application
 on the order of 10-6·seconds or less – often negligible

Queueing Delay: time spent waiting in a queue at any point along the route
 depends on intensity and nature of traffic arriving at queue(s)
 on the order of 10-6·seconds to 10-3·seconds
 5
Delay in Packet-Switched Networks (cont.)
Propagation Delay: time for one bit to propagate from source to destination
at propagation speed of the link
 depends on physical medium of the link
 on the order of 10-6·seconds
 negligible for two routers on the same LAN; significant for
two geostationary satellites
distance between
source and
d [m] destination [m]
dpropagation [sec] 
s [m/sec] propagation speed
of medium [m/s]

Transmission Delay: time to send out / absorb all of the packet bits
 also known as “store-and-forward” delay
 on the order of 10-6·seconds to 10-3·seconds
 negligible for transmission rates  10 Mbps; significant
for large packets sent over low-speed links

L [bit] packet size [bits]

dtransmission [sec] 
R [bps] link transmission
rate [bps]
 6

distance: d destination

volume: v

How long does it take for the parcels/load

to arrive to the destination ?!

Step 1: load parcels tightly/loosely onto the track (transmission delay)

Step 2: drive truck to the destination (propagtion delay)

 7

[Link]
 8
Delay in Packet-Switched Networks (cont.)
Example [ network delay – one packet, one hop ]
Suppose that a user at one end of Canada sends a 1-Mbit file to a remote server on the
other end over a data link operating at 64 kbps.
Assume that we are using a fiber optic link with a propagation rate of the speed of light,
approximately 3 · 108 m/sec, and that the distance is 4800 km.
Ignore any processing or queueing delays.
What is the overall network delay, i.e. time to transmit the file?

dtotal  dpropagation  dtransmission

d [m] 4800 103 [m]

dpropagation    0.016 [sec]
R=64 [kbps]
s [m/sec] 3 108 [m/sec]
…
A B
d=4800 [km] L [bits] 106 [bits]
dtransmission    15.625 [sec]
R [bps] 64103 [bps]

transmission delay >> propagation delay

A high-speed channel would reduce the overall delay.
 9
Delay in Packet-Switched Networks (cont.)
Example [ network delay – one packet, one hop ]
For the same problem, now suppose that we have a 1-Gbps link.
What is the overall network delay, i.e. time to transmit the file, in this case?

R=64 [kbps]

…
A
d=4800 [km]
B dtotal  dpropagation  dtransmission

Propagation delay still the same: dpropagation  0.016 [sec]

L [bits] 106 [bits]

dtransmission    0.001[sec]
R [bps] 10 9 [bps]

dtotal  0.017 [sec]

Increasing data rate beyond certain value on a long (e.g satellite) link
will not noticeably speed up file delivery.
 10
Packet vs. Virtual Circuit Switching Delay
Example [ network delay – one packet, multiple hops ]
A message needs to be transmitted over a path that involves two intermediate switches.
For simplicity assume that the propagation delay and the bit rate of the transmission lines
are the same, and ignore any queueing delay.
What is the overall end-to-end message delay in case of datagram packet switching?

Draw corresponding diagrams.

Message

Message dpropagation – time for

Router A 1st bit to transit from Router B
router A to router B
Subscriber
A Message

dtransmission – time for router B

to absorb all packet bits

Subscriber
B

Network
nodes
 11
Packet vs. Virtual Circuit Switching Delay (cont.)

(1) delay in case of datagram packet switching

2 intermediate routers  3 hops between source and destination

dtotal = 3*dpropagation + 3*dtransmission + 3*dprocessing

transmission begins last bit last bit packet

(1st bit leaves source) leaves source reaches Switch 1 processed

dtransmission

Source t Switch 1 cannot start

processing packet
Switch 1 until it receives all bits
t
dpropagation dprocessing
Switch 2
t

Destination
t
1-hop delay

overall delay
 Delay Comparison – Multiple Packets 12

Example [ network delay – multiple packets, multiple hops ]

 Delay Comparison – Multiple Packets (cont.) 13

Example [ network delay – multiple packets, multiple hops ]

Assume:
N = number of hops L = message length [bits]
R = data rate [bps] P = packet size [bits] (payload + header)
H = packet overhead
dpropagation = propagation delay per hop [sec]

Compute end-to-end delay for datagram switching, assuming:

N=4, L=3200, R=9600, P=1024, H=16, dpropagation=0.001.
Ignore any queueing or processing delay.

Message Message
message length

Message

Subscriber
A Message
header

Subscriber
B
packet size
Network
nodes
 Delay Comparison – Multiple Packets (cont.) 14

(2) delay in case of datagram packet switching

There are P – H = 1024 – 16 = 1008 data bits per packet.
A message of 3200 bits require four packets: Np = 4.
(3200 bits/1008 bits/packet = 3.17 packets which we round up to 4 packets.)

dtransmission
S
t
P1 P2 P3 P4
R1
R1 t
dpropagation
P1 P2 P3 P4
R2
t
P1 P2 P3 P4
R3
t
P1 P2 P3 P4
D
t
Delay

d = D1 + D2 + D3 + D4, where
D1 = Time to transmit entire 1st packet over all hops
D2 = Time to transmit entire 2nd packet
time to absorb
D3 = Time to transmit entire 3rd packet the rest of the
th
D4 = Time to transmit entire 4 packet message
Delay Comparison – Multiple Packets (cont.) 15

Let:

T = transmission time for one packet = P/R

D1 = N*dpropagation + 4*T =
= 4*dpropagation + 4*P/R =
= 4 x 0.001 + 4*1024/9600 =
= 0.427

D2 = D3 = D4 = T =
= (P/R) =
= (1024/9600) =
= 0.107

d = 0.427 + 3*0.107 =
= 0.748 sec
 Delay Comparison – Multiple Packets (cont.) 16

Relationship between Packet Size – by breaking message into smaller

and Transmission Time packets, nodes can begin transmitting
1st packet as soon as it has arrived,
without waiting for 2nd packet
 this overlap in packet transmission
can result in considerably shorter
overall delay – smaller packet will
sooner be received / absorbed at
the destination
 however, process of using more and
smaller packets eventually results in
increased rather than reduced delay,
since each packet must contain a
header (more overhead)
 in contrast to transmission delay, pro-
cessing and queueing delays always
increase as # of packets increases
 packet-switched network designers
must consider all factors when
attempting to find an optimum packet
size
 17
Queueing Delay and Packet Loss
Queueing Delay – most complex component of network delay

 unlike the other tree delays (dprocessing, dtransmission, dpropagation)

queueing delay can vary from packet to packet!
 example: 10 packets arrive at an empty queue at the same
time  1st packet transmitted will suffer no queue. delay,
last packet transmitted will suffer relatively large queueing
delay
 when characterizing queueing delay, one typically uses
probabilistic and statistical measures, such as: average
queueing delay, average queue size, probability that
queue exceeds some specific size, etc.

queue

Calculation of queueing delay (for known  and w) discussed earlier!!!

 18
Queueing Delay and Packet Loss (cont.)
Traffic Intensity – plays critical role in estimating extent of queueing delay
(Router Utilization)
La aka
traffic intensity  server utilization
R

L = packet size [bits/packet]

arrival bit rate - 
a = average packet arrival rate [packet/sec]
R = output transmission rate [bps] departure bit rate - 

  = La/R  0: average queueing delay small

  = La/R  1: delay becomes large

  = La/R > 1: average rate at which bits arrive exceeds

the rate at which bits can be transmitted
(a) infinite queue  qeueing delay  
(b) finite queue  router drops packets

La [bps] R[bps]
 19

[Link]
 20
Queueing Delay and Packet Loss (cont.)
Example [ average queueing delay ]
Assume La/R=1 (average arrival bit rate = average departure bit rate).
Determine the average queueing delay in the following two cases:
(a) Packets arrive periodically – one packet of size L every L/R seconds.
(b) Packets arrive in bursts, but periodically – N packet every (L/R)*N seconds.

(a) Every packet will arrive at an empty (b) 1st packet: no queueing delay
queue  no queueing delay. 2nd packet: delay = L/R [sec]
3rd packet: delay = 2*L/R [sec]
…
Nth packet: delay = (N-1)*L/R [sec]

average delay = ???

Typically, arrivals do not follow any pattern and packets are spaced apart by random
amount of time  La/R is not usually sufficient to fully characterize delay statistics !
 Chapter 8
Switching

8.1
 Figure 8.1 Switched network

8.2
 Figure 8.2 Taxonomy of switched networks

8.3
 A circuit-switched network is made of a
set of switches connected by physical
links, in which each link is
divided into n channels.

8.4
 Figure 8.3 A trivial circuit-switched network

8.5
 In circuit switching, the resources need
to be reserved during the setup phase;
the resources remain dedicated for the
entire duration of data transfer until the
teardown phase.

8.6
 Figure 8.6 Delay in a circuit-switched network

8.7
 In a packet-switched network, there
is no resource reservation;
resources are allocated on demand.

8.8
 Figure 8.7 A datagram network with four switches (routers)

8.9
 Figure 8.8 Routing table in a datagram network

8.10
 A switch in a datagram network uses a
routing table that is based on the
destination address.

8.11
 The destination address in the header of
a packet in a datagram network
remains the same during the entire
journey of the packet.

8.12
 Figure 8.9 Delay in a datagram network

8.13
 Switching in the Internet is done by
using the datagram approach
to packet switching at
the network layer.

8.14
 VIRTUAL-CIRCUIT NETWORKS

A virtual-circuit network is a cross between a circuit-

switched network and a datagram network. It has
some characteristics of both.

8.15
 Figure 8.10 Virtual-circuit network

8.16
 Figure 8.11 Virtual-circuit identifier

8.17
 Figure 8.12 Switch and tables in a virtual-circuit network

8.18
 Figure 8.13 Source-to-destination data transfer in a virtual-circuit network

8.19
 Figure 8.14 Setup request in a virtual-circuit network

8.20
 Figure 8.15 Setup acknowledgment in a virtual-circuit network

8.21
 In virtual-circuit switching, all packets
belonging to the same source and
destination travel the same path;
but the packets may arrive at the
destination with different delays
if resource allocation is on demand.

8.22
 Figure 8.16 Delay in a virtual-circuit network

8.23
 Switching at the data link layer in a
switched WAN is normally
implemented by using
virtual-circuit techniques.

8.24
 Figure 8.21 Packet switch components

8.25
 Figure 8.22 Input port

8.26
 Figure 8.23 Output port

8.27
 Circuit and Packet Switching

3. A path in a digital circuit-switched network has a data rate of I Mbps. The

exchange of 1.000 bits is required for the setup and teardown phases. The
distance between two parties is 3.000 km. Answer the following questions if
the propagataion speed is 2x108 m/s:

a. What is the total delay if 1.000 bits of data are exchanged during the
data transfer phase?

b. What is the total delay if 100.000 bits of data are exchanged during the
data transfer phase?

c. What is the total delay if 1.000.000 bits of data are exchanged during
the data transfer phase?

d. Find the delay per 1.000 bits of data for each of the above cases and
compare them. What can you infer?

13
 Circuit and Packet Switching

Total delay (t) = delay of setup and tear down (d1) + delay
of data transfer (d2)

Delay of setup and tear down (d1) = ( 3 * propagation

delay) + (3 * transmission delay)
Delay of data transfer = propagation delay + transmission
delay

14
 Circuit and Packet Switching
3. Delay of Setup and tear down (d1)
= (3*propagation delay) + (3*transmission delay)
= 3 (3.000 km / (2*108 m/s)) + 3 ((1.000 bits / 1 Mbps))
= (3X15) ms + (3X1) ms
= 48 ms

Lets assume the data transmission is in one direction:

a. Total delay (t) = d1 + d2
= 48 ms + propagation delay + transmission delay
= 48 ms + 15 ms + 1 ms = 64 ms
b. Total delay (t) = d1 + d2
= 48 ms + propagation delay + transmission delay
= 48 ms + 15 ms + 100 ms = 163 ms
c. Total delay (t) = d1 + d2
= 48 ms + propagation delay + transmission delay
= 48 ms + 15 ms + 1000 ms = 1.053 ms
d. For a, we have 64 ms, for b, we have, = 163 ms, for c, we have, =
1.053 ms. The ratio for the case of 3 is the smallest because of using
one setup and tear down phase for sending more data

15
Datagram Path Length Visited Switches
1 3200 Km 1,3,5
2 11,700 Km 1,2,5
3 12,200 Km 1,2,3,5
4 10,200 Km 1,4,5
5 10,700 Km 1,4,3,5
 Circuit and Packet Switching

Assuming a negligible transmission time (i.e. all datagrams start at

time 0). The arrival times are calculated as:

First: (3.200 km / 2 x 108 m/s) + 3 + 20 + 20 = 59 ms

Switch Delay
(ms)
Second: (11.700 km / 2 x 108 m/s) + 3 + 10 + 20 = 91,5 ms
1 3

2 10
Third: (12.200 km / 2 x 108 m/s) + 3 + 10 + 20 + 20 = 114 ms 3 20

4 7
Fourth: (10.200 km / 2 x 108 m/s) + 3 + 7 + 20 = 81 ms 5 20

Fifth: (10.700 km / 2 x 108 m/s) + 3 + 7 + 20 + 20 = 103,5 ms

The order of arrival is 3 -> 5 -> 2 -> 4 -> 1

17
Switching
Outline

 Switched Network
 Circuit-Switched Network
 Datagram Networks
 Virtual-Circuit Network

Data 2
Communication
Introduction

 In small networks we need some means to allow

one-to-one communication between any two
nodes.
 In LANs this is achieved using one of three
methods:
 Direct point-to-point connection (mesh)
 Via central controller (star)
 Connection to common bus in a multipoint
configuration (bus/hup)

Data 3
Communication
Introduction

 None of the previous works in larger networks

with large physical separation or consisting of a
large number of computers
 The solution is a switching network

Data 4
Communication
Switched Network

 Consists of a series of interlinked nodes called

switches.
 Switches are capable to create temporary
connections between two or more devices

Data 5
Communication
Data 6
Communication
Circuit-Switched Network

 A circuit-switched network consists of a set of

switches connected by physical links.
 A connection between two stations is a
dedicated path made of one or more links
 each connection uses only one dedicated
channel on each link
 Each link is normally divided into n channels by
using FDM or TDM.

Data 7
Communication
Circuit-Switched Network

Data 8
Communication
Circuit-Switched Network

 Switching take place at physical layer

 Resources
 Such as bandwidth in FDM and time slot in TDM
 Switch buffer
 Switch processing time
 Switch I/O ports
 Data transferred are not packetized, continuous
flow
 No addressing involved during data transfer

Data 9
Communication
Example

Data 10
Communication
Transmission phases

 Setup phase
 A dedicated circuit needs to be established
 So create dedicated channel by sending a request

Data 11
Communication
Transmission phases

 Data transfer phase

 Teardown phase
 Signal is sent to each switch to release resources

Data 12
Communication
Delay

Data 13
Communication
Datagram Networks

 Data are transmitted in discrete units called

packets
 Size of the packet depends on the protocol and
network
 Packets switched networks are connectionless,
hence no resource allocation
 Connectionless means the switch does not keep
information about the connection state.
 Datagram switching is done at network layer

Data 14
Communication
Datagram Networks

Data 15
Communication
Routing table & Destination Add.

 A switch in a datagram
network uses a routing table
that is based on the
destination address.
 The destination address in the
header of a packet in a
datagram network remains
the same during the entire
journey of the packet.

Data 16
Communication
Delay

 Total Delay = 3T + 3t+ w1+ w2

Data 17
Communication
Virtual-Circuit Network

 A virtual-circuit network is a cross between a

circuit switched network and a datagram
network. It has some characteristics of both.
 Packets form a single message travel along the
same path.
 Characteristics
 Three phases to transfer data
 Resources can be allocated during setup phase
 Data are packetized and each packet carries an
address in the header
 All packets follow the same path
Data Communication 18
 Implemented in data link layer
Virtual-Circuit Network

Data 19
Communication
Addressing

 Global addressing
 Source and destination needs unique addresses
 Virtual-circuit identifier

Data 20
Communication
 VCI
Virtual Circuit Identifier, VCI is the
label of a VC on a network. This label is
enables the network data to travel between
the two devices on a network. This label will
change as it travels between more than two
devices.

Data 21
Communication
Transmission phases

 Setup phase
 A switch creates an entry for a virtual circuit.
 Request
 acknowledgment

Data 22
Communication
Setup

Data 23
Communication
acknowledgment

Data 24
Communication
Data transfer and teardown phases

After sending all frames, a special frame is send to end the

connection

Data 25
Communication
Efficiency

 In virtual-circuit switching, all packets belonging

to the same source and destination travel the
same path;
• But the packets may arrive at the destination
with different delays if resource allocation is on
demand.

Data 26
Communication
Delay

 Total delay = 3T + 3ζ+setup delay +

teardown delay

Data 27
Communication
Note •
Switching at the data link •
layer in a switched WAN is normally
implemented by using virtual-circuit
techniques.

Data 28
Communication
 Throughput & Delay

2. Suppose a car needs 7 hours for a trip between Kaiserslautern and Berlin
(~655 km). The car has a cargo area of 200 x 100 x 100 cm. It is filled
completely with DVDs, each of them contains full 4.3 GB of data. Each DVD
needs a space of 12,5 x 12,5 x 0,5 cm.
a) How many bits per second are transferred?
Number of DVDs transported: n = 16x8x200 = 25600
Total size of data:
d = 25600 * 4.3 GB = 110080 GB = 110080 * 10003 * 8 bit
= 880640 Gbit
Transfer rate:
r = 880640 Gbit / 7 h = 880640 Gbit / 25200 sec
~ 34,94 Gbit/sec

BUT: “… binary digits that the network is capable of delivering and

accepting per time unit.”

9
Delay

10
 Throughput & Delay

b) Give a (short) statement, based on the definition of end-to-end delay,

why such a „network“ is not used is most cases.

- 7 h for the trip include only the network transit delay, neither the access
delay nor the transmission delay.

- Access delay:
• adding the time to burn all DVDs (suppose 15 min/DVD):
6400 h (~ 267 days)

- Transmission delay:
• Adding the time to transfer the data to the destination host

11
 Transmission Delay

c) Suppose the data is delivered by simply stopping the car at a parking site.
How long is the transmission delay?
- All bits are delivered at the same time Transmission delay = 0 s!

  Number of parallel transmitted bits only

depends on volume of cargo area.

  Application in data networks: Parallel sending over

• multiple frequencies

• more fine grained signal levels (4 levels 2 bits)

12
 Circuit and Packet Switching

3. A path in a digital circuit-switched network has a data rate of I Mbps. The

exchange of 1.000 bits is required for the setup and teardown phases. The
distance between two parties is 3.000 km. Answer the following questions if
the propagataion speed is 2x108 m/s:

a. What is the total delay if 1.000 bits of data are exchanged during the
data transfer phase?

b. What is the total delay if 100.000 bits of data are exchanged during the
data transfer phase?

c. What is the total delay if 1.000.000 bits of data are exchanged during
the data transfer phase?

d. Find the delay per 1.000 bits of data for each of the above cases and
compare them. What can you infer?

13
 Circuit and Packet Switching

} Total delay (t) = delay of setup and tear down (d1) + delay
of data transfer (d2)

} Delay of setup and tear down (d1) = ( 3 * propagation

delay) + (3 * transmission delay)
} Delay of data transfer = propagation delay + transmission
delay

14
 Circuit and Packet Switching
3. Delay of Setup and tear down (d1)
= (3*propagation delay) + (3*transmission delay)
= 3 (3.000 km / (2*108 m/s)) + 3 ((1.000 bits / 1 Mbps))
= (3X15) ms + (3X1) ms
= 48 ms

Lets assume the data transmission is in one direction:

a. Total delay (t) = d1 + d2
= 48 ms + propagation delay + transmission delay
= 48 ms + 15 ms + 1 ms = 64 ms
b. Total delay (t) = d1 + d2
= 48 ms + propagation delay + transmission delay
= 48 ms + 15 ms + 100 ms = 163 ms
c. Total delay (t) = d1 + d2
= 48 ms + propagation delay + transmission delay
= 48 ms + 15 ms + 1000 ms = 1.053 ms
d. For a, we have 64 ms, for b, we have, = 163 ms, for c, we have, =
1.053 ms. The ratio for the case of 3 is the smallest because of using
one setup and tear down phase for sending more data

15
 Circuit and Packet Switching

4. Five equal-size datagrams belonging to the same message leave for the
destination one after another. However, they travel through different paths as
shown in the following table

Datagram Path Length Visited Switches

1 3200 Km 1,3,5
2 11,700 Km 1,2,5
3 12,200 Km 1,2,3,5
4 10,200 Km 1,4,5
5 10,700 Km 1,4,3,5

We assume that the delay for each switch (including waiting and processing)
is 3, 10, 20, 7, and 20 ms respectively. Assuming that the propagation speed
is 2x108 m/s, find the order the datagrams arrive at the destination and the
delay for each. Ignore any other delays in transmission.

16
 Circuit and Packet Switching

Assuming a negligible transmission time (i.e. all datagrams start at

time 0). The arrival times are calculated as:

First: (3.200 km / 2 x 108 m/s) + 3 + 20 + 20 = 59 ms

Switch Delay
(ms)
Second: (11.700 km / 2 x 108 m/s) + 3 + 10 + 20 = 91,5 ms
1 3

2 10
Third: (12.200 km / 2 x 108 m/s) + 3 + 10 + 20 + 20 = 114 ms 3 20

4 7
Fourth: (10.200 km / 2 x 108 m/s) + 3 + 7 + 20 = 81 ms 5 20

Fifth: (10.700 km / 2 x 108 m/s) + 3 + 7 + 20 + 20 = 103,5 ms

The order of arrival is 3 -> 5 -> 2 -> 4 -> 1

17
 Chapter 3
Data and Signals

3.1 Copyright © The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 3-4 TRANSMISSION IMPAIRMENT

Signals travel through transmission media, which are not

perfect. The imperfection causes signal impairment. This
means that the signal at the beginning of the medium is
not the same as the signal at the end of the medium.
What is sent is not what is received. Three causes of
impairment are attenuation, distortion, and noise.

Topics discussed in this section:

 Attenuation
 Distortion
 Noise

3.2
 Figure 3.25 Causes of impairment

3.3
3.4
 Attenuation

 Means loss of energy -> weaker signal

 When a signal travels through a
medium it loses energy overcoming the
resistance of the medium
 Amplifiers are used to compensate for
this loss of energy by amplifying the
signal.

3.5
 Measurement of Attenuation

 To show the loss or gain of energy

the
unit “decibel” is used.

dB = 10log10P2/P1
P1 - input signal
P2 - output signal

3.6
 Figure 3.26 Attenuation

3.7
 Example 3.26

Suppose a signal travels through a transmission medium

and its power is reduced to one-half. This means that P2
is (1/2)P1. In this case, the attenuation (loss of power)
can be calculated as

A loss of 3 dB (–3 dB) is equivalent to losing one-half

the power.
3.8
 Example 3.27

A signal travels through an amplifier, and its power is

increased 10 times. This means that P2 = 10P1 . In this
case, the amplification (gain of power) can be calculated
as

3.9
 Example 3.28

One reason that engineers use the decibel to measure the

changes in the strength of a signal is that decibel
numbers can be added (or subtracted) when we are
measuring several points (cascading) instead of just two.
In Figure 3.27 a signal travels from point 1 to point 4. In
this case, the decibel value can be calculated as

3.10
 Figure 3.27 Decibels for Example 3.28

3.11
 Example 3.29

Sometimes the decibel is used to measure signal power

in milliwatts. In this case, it is referred to as dBm and is
calculated as dBm = 10 log10 Pm , where Pm is the power
in milliwatts. Calculate the power of a signal with dBm =
−30.

Solution
We can calculate the power in the signal as

3.12
 Example 3.30

The loss in a cable is usually defined in decibels per

kilometer (dB/km). If the signal at the beginning of a
cable with −0.3 dB/km has a power of 2 mW, what is the
power of the signal at 5 km?
Solution
The loss in the cable in decibels is 5 × (−0.3) = −1.5 dB.
We can calculate the power as

3.13
3.14
 Distortion
 Means that the signal changes its form or
shape
 Distortion occurs in composite signals
 Each frequency component has its own
propagation speed traveling through a
medium.
 The different components therefore arrive
with different delays at the receiver.
 That means that the signals have different
phases at the receiver than they did at the
source.
3.15
 Figure 3.28 Distortion

3.16
3.17
 Noise
 There are different types of noise
 Thermal - random noise of electrons in the
wire creates an extra signal
 Induced - from motors and appliances,
devices act are transmitter antenna and
medium as receiving antenna.
 Crosstalk - same as above but between
two wires.
 Impulse - Spikes that result from power
lines, lightning, etc.

3.18
 Figure 3.29 Noise

3.19
 Signal to Noise Ratio (SNR)

 To measure the quality of a system the

SNR is often used. It indicates the
strength of the signal wrt the noise
power in the system.
 It is the ratio between two powers.
 It is usually given in dB and referred to
as SNRdB.

3.20
 Example 3.31

The power of a signal is 10 mW and the power of the

noise is 1 μW; what are the values of SNR and SNRdB ?

Solution
The values of SNR and SNRdB can be calculated as
follows:

3.21
 Example 3.32

The values of SNR and SNRdB for a noiseless channel

are

We can never achieve this ratio in real life; it is an ideal.

3.22
 Figure 3.30 Two cases of SNR: a high SNR and a low SNR

3.23