Basics of lower layers

Dmitri A. Moltchanov
E-mail: [email protected]
http://www.cs.tut.fi/kurssit/TLT-2616/

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

OUTLINE
Mobile ad hoc networks
Classification of wireless networks
The electromagnetic spectrum
Radio propagation
Modulation techniques
Two-way communications
Centralized multiple access schemes
Random multiple access schemes
Error control
Channel adaptation techniques

Lecture: Basics of lower layers

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

1. Mobile ad hoc networks (MANET)

3
1
6
7

8
4
2

Figure 1: Graphical representation of ad hoc network.

Ad hoc networks:
complex wireless distributed system;
nodes can freely and dynamically self-organize into arbitrary and temporary topologies;
What we expect out of ad hoc networks:
provide a multi-hop communication between inherently mobile devises;
provide coverage in areas with low population densities.

Lecture: Basics of lower layers

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Basis of any ad hoc network is the point-to-point wireless link

3
1
6
7

8
4
5

D-L
PHY

D-L
PHY

wireless transmission medium

Figure 2: A wireless link in isolation.

Lecture: Basics of lower layers

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

2. Classification of wireless networks

Based on the coverage areas:
wireless body areas network (WBAN);
wireless personal area network (WPAN);
wireless local area network (WLAN);
wireless wide area network (WWAN).

BAN

PAN

~1 m

~10 m

LAN
WAN

~500 m

WAN

>10 km

Figure 3: Classification of wireless networks based on their coverage area.

Lecture: Basics of lower layers

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

WBAN networks
what: BAN is the network consisting of wearable computers;
aim: is to provide the connectivity between wearable computers: headphones, displays, etc.
WPAN networks
what: PAN is a network in the environment around the person;
aim: PAN connects BAN devices to other mobile and stationary devices.
WLAN networks
what: usually, is a network of laptops;
aim: provide connectivity and Internet access in the high density areas.
WWAN/WMAN networks
what: is a network of arbitrary mobile devices;
aim: is to provide connectivity between remote mobile devices.

Lecture: Basics of lower layers

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

3. The electromagnetic spectrum

Wireless communications: broadcast and reception of electromagnetic waves:
frequency, f :
number of cycles per second of the wave, measured in Hz.
wavelength, :
the length of the cycle, measured in meters.
The speed of propagation waves:
varies from medium to medium;
in vacuum equals to the speed of light.
The following relation holds:
r = f,

(1)

r is the speed of the wave;

r = c in vacuum.

Lecture: Basics of lower layers

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Frequency bands defined by International Telecommunication Union (ITU) is shown in Fig. 4.

mostly usable for communications
Infrared
Frequency (Hz)
100

102

104

106

Ultraviolet
108

1010 1012 1014 1016 1018 1020 1022 1024 1026 1028

100

10-2 10-4 10-6

Radio
108

106

104

102

Wavelength (m)

10-8 10-10 10-12 10-14 10-16 10-18 10-20

Visible
Microwave

X-ray

Gamma ray

not usable:
- affect health
- difficult to modulate
- do not propagate through obstacles

Figure 4: Frequency bands in the electromagnetic spectrum.

Lecture: Basics of lower layers

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Frequency determines properties of the transmission:

low frequency: pass through the obstacles;
higher frequency: more prone to absorbtion by rain or fog;
higher frequency: reflected by obstacles.
Radio waves:
relatively easy to generate and modulate;
have the ability to path through the buildings;
may travel very long distances;
radio transmission is omni-directional.
Microwave:
tend to travel in a straight lines;
can be narrowly focused and concentrated in a small beam;
cannot pass through obstacles.

Lecture: Basics of lower layers

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Infrared waves:
cannot pass through obstacles;
relatively directional and inexpensive to implement;
used in short range communications.
Visible light:
may provide very high bandwidth at a very low cost using optical laser signalling;
hard to focus a very narrow uni-directional laser;
cannot penetrate through rain and fog.
Allocation of waves:
electromagnetic spectrum is a common resource;
international agreements have been drawn to allocate it;
national agreements may override them;
note: remember military usage!

Lecture: Basics of lower layers

10

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

4. Radio propagation
Radio waves experience the following propagation mechanisms:
Reflection:
when: wave hits an objects which is very large compared to its wavelength;
result: phase shift of 180 degrees between the incident and the reflected rays.

Diffraction:
when: wave hits an object that is comparable to its wavelength;
result: wave bends at the edges of the object, propagating in different directions.

Scattering:
when: wave goes through a medium with objects that are small compared to its wavelength;
result: wave gets scattered into several weaker outgoing signals.

Lecture: Basics of lower layers

11

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Figure 5: Illustration of the radio propagation.

Lecture: Basics of lower layers

12

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

4.1. Propagation models

We distinguish between:
large-scale propagation models:
predict the average received signal strength at a given distance from transmitter;
what: capture path loss component;
application: estimation of the radio coverage area around the transmitter.
small-scale propagation models:
characterize the rapid fluctuations of the received signal strength;
what: capture influence of multipath components;
application: performance evaluation of data transmission over the wireless channels.

Historic aspects:
most are designed for applications in cellular networks;
some are applicable to other wireless networks (e.g. ad hoc, vehicular).

Lecture: Basics of lower layers

13

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Classification:
analytical models:
capture path loss based on analytical representation of propagation phenomenons;
+: allows to get predictions very quickly;
: often too complicated;
: limited to the complexity of mathematics.
empirical model:
based on fitting empirical formulas to a set of statistical data;
+: implicitly include all propagation phenomenons;
: cannot be derived without measurements;
: always specific to those environment in which measurements have been carried out.

One more classification:

indoor models;
outdoor models.

Lecture: Basics of lower layers

14

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

4.2. Large-scale propagation (path loss) models

Figure 6: Illustration of the areas with different received local average signal strength.
Examples:
free-space propagation model;
two-ray ground reflection model.
See: T. Rappaport, Wireless communications, 2nd edition, Prentice Hall, 2002.

Lecture: Basics of lower layers

15

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

4.3. Small-scale propagation (fading) models

Used to represent rapid changes of the received signal strength.

Figure 7: Attenuation due to distance and rapid fluctuations of the received signal strength.

Lecture: Basics of lower layers

16

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

4.4. Small-scale (fading) models

Depending on presence of LOS received signal strength has:
LOS: Rician distribution;
No-LOS: Rayleigh distribution
The Rician distribution is given by:
  

2
2
A
( + A )

I
,
p() = 2 exp
0

2 2
2
As A 0 the Rician distribution degenerates to Rayleigh one:


2

p() = 2 exp 2 , 0,

0,

(2)

(3)

Signal-to-noise (SNR) ratio:

received signal strength and
noise (local, interference).

Lecture: Basics of lower layers

17

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Nakagami-m fading:


m
mm m1
exp
p() =
,
(E[])m (m)
E[]

0,

(4)

E[] is the average received SNR

(m) is the gamma function
m is the Nakagami fading parameter
Nakagami-m covers almost all special cases:
m = 0.5 we get the worst possible fading case
m = 1 we get Rayleigh statistical fading model
m > 1 resulting in Rician fading channels
m no fading

Note: how got get symbol (bit) error rate?

Simon, M. and Alouini, M. Digital Communication Over Fading Channels, Wiley, 2005.

Lecture: Basics of lower layers

18

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

5. Modulation techniques
Modulation:
why: information cannot be transferred as is;
what: converting data into electromagnetic waves;
how: altering certain properties of the carrier wave.
Classification based on the nature of the data to be transmitted :
analog modulation techniques:
amplitude modulation;
frequency modulation;
phase modulation.
digital modulation techniques:
amplitude shift keying;
frequency shift keying;
phase shift keying.

Lecture: Basics of lower layers

19

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

5.1. Analog modulation

Characteristics:
used to transmit analog data (e.g. voice);
perform superimposing analog data signal x(t) on a predefined carrier signal c(t).
Amplitude modulation:
frequency and phase of the modulating signal remains the same;
amplitude varies with that of information signal.
Frequency modulation:
amplitude and phase of the modulating signal remains the same;
frequency varies with that of information signal.
Phase modulation:
amplitude and frequency of the modulating signal remains the same;
phase varies with that of information signal.

Lecture: Basics of lower layers

20

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Figure 8: Illustration of the analog amplitude modulation.

Lecture: Basics of lower layers

21

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Figure 9: Illustration of the analog frequency modulations.

Lecture: Basics of lower layers

22

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Figure 10: Illustration of the analog phase modulations.

Lecture: Basics of lower layers

23

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

5.2. Digital modulation

Characteristics:
used to transmit binary data (e.g. PCM voice);
alters certain properties of transmitting data;
difference: changes occur at discrete time instants.
There are a number of digital modulation techniques:
amplitude shift keying (ASK);
frequency shift keying (FSK);
phase shift keying (PSK).

Lecture: Basics of lower layers

24

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Amplitude shift keying

t: transmission time of the symbol;
1: presence of a carrier for t;
0: absence of a carrier t.
Mathematically ASK is represented as:

A cos2f t, for 1,
c
c
s(t) =
0,
for 0.

(5)

Figure 11: Illustration of the amplitude shift keying.

Lecture: Basics of lower layers

25

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Frequency shift keying

Let fc be the carrier frequency and f be the small frequency offset. Then according to FSK:
1: presence of carrier with frequency fc + k for a certain time;
0: presence of carrier with frequency fc k for the same time.
Two-levels: binary FSK (BFSK):

A cos2(f + k)t, for 1,

c
c
s(t) =
Ac cos2(fc k)t, for 0.

(6)

Figure 12: Illustration of the binary frequency shift keying (BFSK).

Lecture: Basics of lower layers

26

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Phase shift keying

Binary PSK (BPSK): change in phase by
0: presence of carrier with frequency fc for a certain time;
1: presence of carrier with a phase difference of .
Mathematically, two-level PSK (binary PSK, BPSK) is given by:

A cos(2f t + ), for 1,
c
c
s(t) =
Ac cos2fc t,
for 0.

(7)

Multiple phase deviation can also be used to encode multiple bits.

Quadrature PSK (QPSK): change in phase by /2:

A
cos(2f
t
+
),

c
c
4

A cos(2f t + 3 ),
c
c
4
s(t) =

Ac cos(2fc t + 5
),

Ac cos(2fc t + 7
),
4

Lecture: Basics of lower layers

for 10,
for 11,

(8)

for 01,
for 00.

27

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

5.3. x-QAM
Characteristics:
BPSK, QPSK, 16-QAM, 64-QAM, 256-QAM, etc.
combines advantages of amplitude and phase keying.

Figure 13: Illustration of 64-QAM signal constellations.

Lecture: Basics of lower layers

28

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

6. Two-way communications
Two approaches:
Time division duplex: uplink and downlink on the same frequency;
Frequency division duplex: uplink and downlink on different frequencies.
downlink

uplink

downlink

uplink

FRAME

FRAME

FRAME

FRAME

time

frequency

Figure 14: Illustration of TDD and FDD principle.

Shortcomings:
: guard bands for synchronization: inefficient use of spectrum;
: assigned slot may not always be in use: inefficient use of spectrum.

Lecture: Basics of lower layers

29

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

7. Centralized multiple access techniques

Why we need multiple access scheme:
bandwidth is a scarce resource at the air interface;
there are a number of users that want to transmit.

Centralized vs. random access:

centralized: master controls assignment (WWAN/WMAN);
random: stations compete for access (WLAN/WPAN/WBAN).

Basically, there are four centralized multiple access schemes:

frequency division multiple access (FDMA);
time division multiple access (TDMA);
code division multiple access (CDMA).

Lecture: Basics of lower layers

30

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

7.1. Frequency division multiple access

What is the basis:
shares available bandwidth in the frequency domain;
available bandwidth is divided into a number of channels;
there should be a guard band between adjacent channels;
each tranbsmitter/receiver pair is assigned the same channel for operation.

F1

F2

Fn
Bandwidth (Hz)

Figure 15: Illustration of FDMA principle.

: guard bands: inefficient use of spectrum;
: assigned slot may not always be in use: inefficient use of spectrum.

Lecture: Basics of lower layers

31

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

7.2. Time division multiple access

What is the basis:
shares the available bandwidth in the time domain;
frequency band is divided into a number of time slots;
a set of periodically repeated time slots is known as TDMA frame;
each node is assigned a slot in each frame and transmits only in this slots.
time slot assigned for a transmitter/receiver pair

...

...

TDMA frame

Figure 16: Illustration of TDMA principle.

Lecture: Basics of lower layers

32

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

7.3. Spread spectrum techniques

What is the basis:
every user uses the entire spectrum;
individual transmission are encoded with a pseudo-random sequences;
assigned codes are orthogonal so that the simultaneous transmissions are possible.
There are two types of spreads spectrum techniques available:
frequency hopping spread spectrum (FHSS);
spectrum is divided into many subchannels;
two communicating systems hop on same frequencies.
direct sequence spread spectrum (DSSS).
stations are assigned orthogonal codes;
use these codes for transmission;
other stations transmissions appears an noise.

Lecture: Basics of lower layers

33

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

8. Random access schemes

What is the basis:
provide access to a channel for multiple concurrent stations;
is not needed when there is a centralized control;
required when the access is distributed;
can be used for decentralized access in TDMA and FDMA channels.

Basic techniques:
ALOHA and slotted ALOHA;
Carrier Sense Multiple Access (CSMA);
Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA);
Carrier Sense Multiple Access with Collision Detection (CSMA/CD).

Lecture: Basics of lower layers

32

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

8.1. ALOHA
Pure ALOHA
a terminal transmits whenever the user data is ready;
if the sender finds that the packet get collided:
it waits for a random period of time;
sends the packet again.
Throughput: low, depends on stations and traffic they generate.
Slotted ALOHA
time is slotted, length on the slot is the time to transmit a packet;
node starts transmission in the beginning of slots only;
if collision occurs:
sender waits for a random number of slots;
transmits packet again.
Throughput: higher than ALOHA but still low.

Lecture: Basics of lower layers

33

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

8.2. Carrier Sense Multiple Access (CSMA)

Why we need:
throughput of ALOHA is very low;
what to do: listen for packet transmissions.
In general, there are three different CSMA schemes:
1-persistent CSMA;
non-persistent CSMA;
p-persistent CSMA.
1-persistent CSMA
when the packet is ready for transmission the sender listens to the channel;
if the channel is free packet is immediately transmitted;
if not the senders continues to listen till the channel becomes free.
Probability of starting transmission when the channel is free: 1.

Lecture: Basics of lower layers

34

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Two bad effects:

Wrong channel free effect;
Synchronization effect
Wrong channel free effect
an arbitrary node starts transmitting;
a node near the destination sense the channel and finds it free since packet has not yet arrived.
Synchronization effect
node 3

start of sensing
t

node 2

start of sensing
t

node 1
t

Figure 17: Illustration of the synchronization effect.

Lecture: Basics of lower layers

35

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

Non-persistent CSMA
This scheme was introduced to combat with the synchronization problem. It works as follows:
when the packet is ready for transmission the sender listens the channel;
if the channel is busy the sender goes in the waiting state for a randomly chosen time;
after this time the sender sense the channel again.
random waiting
node 3
node 2

start of sensing
start of sensing

random waiting

t
t

node 1
t

Figure 18: Illustration of non-persistent CSMA.

advantage: probability of collision is less than for 1-persistent CSMA.

Lecture: Basics of lower layers

36

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

p-persistent CSMA
the channel is slotted;
transmission is a free channel is performed with probability p.
The scheme operates as:
when the packet is ready for transmission the sender listens the channel;
if the channel is busy the sender kepdf listen the channel until it finds the channel idle;
if the channel is idle:
the sender transmits the packet in this slot with probability p;
defers transmission to the next slot with probability q = 1 p.
p1>p

node 2
node 1

both have a packet

p1<p

p1<p
t
t

Lecture: Basics of lower layers

37

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

8.3. Carrier sense multiple access with collision detection

What is new:
capability to detect collisions.
bandwidth is wasted

collision is detected

node 2

node 2
t

node 1

t
node 1

Figure 19: Illustration of collision detection advantages.

The algorithm operates as follows:
if the collision is detected the nodes immediately aborts its current transmission;
then, the node sends a brief jamming signal;
any other transmitting node on hearing the jamming signal abort their transmissions;
after transmitting the jamming signal the node waits for a random time and repeats the CSMA.

Lecture: Basics of lower layers

38

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

9. Error control
BER can be high (around 10E 4 10E 2):
using channel coding (adding additional bits;)
using protocols with retransmission.
We distinguish between following channel coding:
coding using error detecting codes:
cyclic redundancy check (CRC).
coding using error correcting codes:
block codes (BCH);
convolutional codes, turbo codes, etc.
Protocols with retransmissions:
stop-and-wait ARQ (SW-ARQ);
go-Back-N ARQ (GBN-ARQ);
selective-repeat ARQ (SR-ARQ).

Lecture: Basics of lower layers

39

TLT-2616: Ad hoc networks

D.Moltchanov, TUT, 2011

10. Channel adaptation mechanisms

Why do we need something else?:
propagation conditions are too diverse;
classic error control may not be sufficient;
classic error control can be redundant.
Which techniques?
power control
automatic modulation and coding;
hybrid ARQ (HARQ, various types);
spatial diversity (MIMO);
spatial multiplexing (OFDM, MIMO);
source rate adaptation.
Note: these techniques makes wireless systems very complicated.

Lecture: Basics of lower layers

40