Mạng không dây 2 pot

Omar Alfandi Media Access Protocol Physical Channel Radio Outline • Multiple Access Technique • Designing Issues of MAC protocols • Classification of MAC protocolsClassification of MAC p

Trang 1

Wireless Ad Hoc & Sensor Networks

• Repetition – Physical Layer

• Issues

• Summary

2

Terms of Lecture

• Weekly lecture (2 SWS, 5 Credits)

• Problem Sheets (self solution)

• Written Exam: 90 minutes at end of semester

• Target audience: AI BSc (5++ sem ); AI MSc (1++ sem );Target audience: AI BSc (5++ sem.); AI MSc (1++ sem.);

Trang 2

Terms of Lecture

• Literature

– Ad Hoc Wireless Networks: Architectures and Protocols; C

Murthy & B Manoj, Prentice Hall, 2004 ISBN: 013147023X

(First is the basic text book used for the lecture structure)

– Protocols and Architectures for Wireless Sensor Networks; H

Karl & A.Willig; 2005; Wiley & Sons; ISBN 0470095102

(Second is also used for Sensor Networks)

– Further Literature and papers, will be announced in lecture

• Issues

• Summary

6

Lecture Overview – OSI Reference Model

Application • A layer is a collection of

conceptually similar functions that

provideservices to the layer above

it andreceivesservice from the

Transport Protocol

it and receivesservice from the

layer below it

Network Protocol

• Conceptually two instances at one

layer are connected by a horizontal

Media Access Protocol layer are connected by a horizontal

protocol connection on that layerPhysical Channel

(Radio)

Lecture Overview

• Introduction - Today

• External Invited Talk (1 lecture, CS Colloquium)

• Medium Access Schemes (1 lecture)

• Routing and Secure Routing (2 lectures)

• Energy Management (1 lecture)

• Transport Layer Protocols & QoS (1 lecture)p y ( )

• Security (2 lectures)

• Sensor Networks (3 lectures)( )

• Final written Exam (last lecture date) ( )

Trang 3

• Issues

• Summary

Definition – Ad Hoc Network

• Ad hoc is a Latin phrase which means "for this [purpose]"

• The purpose is to interconnect computational nodes for

• Nodes participate in routing of packets, deciding

dynamically based on connectivity to neighbour nodes dynamically, based on connectivity to neighbour nodes.

Trang 4

Definition – Sensor Network

• Sensor from the Latin word sentire which means “to

feel” or “to perceive”

• A Sensor measures a physical quantity and converts it

B

D C

14

Cellular vs Ad Hoc

Cellular Networks Ad Hoc Wireless Networks

Fixed infrastructure based Infrastructure less

Single-hop wireless links Multi-hop wireless links

High reliability Frequent path breaks due to mobility

Low complexity mobile hosts Mobile hosts also routers

Geographical reuse of spectrum Carrier sense bases reuse of

spectrum Widely deployed, currently 4G Remaining issues, low commercial

• Issues

• Summary

Trang 5

• Wireless Mesh Networks

17

Applications - Military

• Why? Establish communication among a group of

soldiers/vehicles for tactical operations

• Where? Areas with impossible infrastructure set up

• Security is crucial, eavesdropping and other attacks can compromise information and personel safety.

18

Applications – Transportation (C2C)

• Why? Primary reduce number of lethal accidents

Secondary enable new kinds of services

• How? Enable Car to Car (C2C) and Car to Infrastructure

(C2I) communication for road safety messages

Applications – Sensor Networks

• Why? Monitoring of physical parameters and

transmitting to a sensor sink

• Where? Health care, home security, military, environmental monitoring

• Issues: mobility, network size, deployment density,

t i tpower constraints

Trang 6

Applications - Animal Monitoring

1 Biologists put sensors in underground nests of storm petrel

2 And on 10cm stilts

3 Devices record data about birds

4 Transmit to research station

5 And from there via satellite to lab

21

Applications – Collaboration

• Why? Required instant communication

• Where? Conference (file exchange), Lecture (notes

distribution) using laptops/smart phones

• Properties: Lower security than military, energy

constraints, uni- and multicast.

22

Applications – Emergency Operations

• Why? Required communication for rescue, crowd

control, commando operations activities

• Where? Areas with no/destroyed infrastructure due to

natural calamities, war, etc

• Properties: self configurable, decentralized, capable of

voice communication

Applications – Wireless Mesh Networks

• Why? Provision of alternate communication capability to mobile/fixed nodes, opposed to cellular networks

• Where? Areas with no/low cable coverage or cost

constraints or quick deployment needs.

• Properties: Simple expandability, high availability

Trang 7

Applications – Hybrid Wireless Networks

• What? Multi-hop cellular networks

• Why? Exponential growth in subscriber base of cellular

networks, over 4 bn in 2008

• Properties: mobile nodes are involved in routing, High

capacity/coverage, centric routing topology maintenance

25

Applications – Hybrid Wireless Networks

Cellular Multi-Hop Networks

Trang 8

Physical Layer - Spectrum

Physical Layer - Frequencies and Regulations

Europe (CEPT/ETSI) USA (FCC) Japan Mobile

Physical Layer - Signal Propagation Ranges

• Straight line propagation

i t f

– no signal detection

– Signal part of background noise

interference

Physical Layer – Signal Propagation

Free space propagation

P rr = P tt G tt G rr ( λ / 4πd ) 22

• Simplest path loss model, a direct-path signal

• The following definitions are assumed:

– Pr- The received signal power.

– Pt- The transmitted signal power.

G Th i f h i i – Gr- The gain of the receiving antenna.

– Gt- The gain of the transmitting antenna.

– λ - The wavelength of the carrier (i e the center frequency of the λ - The wavelength of the carrier (i.e., the center frequency of the radiated signal)

– d - The distance between the transmitting and receiving antennas.

Trang 9

Physical Layer - Antennas

side (xz)/top (yz) views

• reflection at large obstacles

• scattering at small obstacles

Physical Layer - Effects of Mobility

• Channel characteristics change over time and location

– signal paths change

– distance to sender changes

– obstacles position changes

power

• Fading

short term

short term fading

long term fading

– short term

– long term

t

• Doppler shift:

change/shift in the frequency

Physical Layer - Modulation and Demodulation

digital

analog baseband digital

synchronization decision

digital data analog

demodulation

baseband signal

101101001 radio receiverradio

carrier

Trang 10

Physical Layer - Digital Modulation1 0 1

• Modulation of digital signals

– very simple

– low bandwidth requirements

tibl t i t f

– very susceptible to interference

• Frequency Shift Keying (FSK):

t

– needs larger bandwidth

• Phase Shift Keying (PSK):

• Issues

• Summary

Trang 11

• Ad Hoc networks serve the purpose of connecting nodes

instantly, without infrastructure

• Ancient use in natural societies with voice, drums,

trumpets for high speed communicationp g p

• Sensors use ad hoc networks to communicate registered

physical parameters to a monitoring sink

• Compared to cellular networks ad hoc nodes are more

complex and deal with dynamic topology and resource

• Physical Layer has highly special characteristics

• Issues exist in all layers due to distribution and dynamicsy y

• Outlook: Next lecture will tackle properties and issues of MAC layer and possible solutions

Trang 12

Medium Access Control

Media Access Protocol

Prof Dr Dieter Hogrefe

Dr Omar Alfandi

Physical Channel (Radio)

Outline

• Multiple Access Technique

• Designing Issues of MAC protocols

• Classification of MAC protocolsClassification of MAC protocols

• Protocols examples

• Characteristics of Link layer protocols

• The lower layers in detail

• Summary

2

Media Access Control (Intro.)

• Wireless medium is shared

• Many nodes may need to access the wireless medium to

send or receive messages

• Concurrent message transmissions may interfere with

each other  collisions  message drops

Multiple Access Technique

• Reservation-based (Recall: mobile communication 1)

– FDMA : Frequency Division Multiple Access – TDMA : Time Division Multiple Access – CDMA : Code Division Multiple Access SDMA : Space Division Multiple Access – SDMA : Space Division Multiple Access

• Random

– ALOHA : University of Hawaii Protocol ALOHA : University of Hawaii Protocol – CSMA : Carrier Sense Multiple Access – MACA : Multiple Access with Collision Avoidance

• Random with reservation

– DAMA : Demand Assigned Multiple Access – PRMA : Packet Reservation Multiple Access

Trang 13

• FDMA (Frequency Division Multiple Access)

– assign a certain frequency to a transmission channel

– permanent (radio broadcast), slow hopping (GSM), fast hopping

(FHSS, Frequency Hopping Spread Spectrum)

• TDMA (Time Division Multiple Access)

– assign a fixed sending frequency for a certain amount of time

• CDMA (Code Division Multiple Access)CDMA (Code Division Multiple Access)

• SDMA (Space Division Multiple Access)

– segment space into sectors, use directed antennas g p ,

– Use cells to reuse frequencies

– Frequency division duplex (FDD)

• Combination of two simplex channels with different carrier frequencies

– Time division duplex (TDD)

• Time sharing of a single channel achieves quasi-simultaneous Time sharing of a single channel achieves quasi simultaneous duplex transmission

6

Random Access

• However, wireless communication is often much more

ad-hoc

– New terminals have to register with the network

– Terminals request access to the medium spontaneously

In many cases there is no central control

– In many cases there is no central control

Other access methods such as distributed and

non-arbitrated = random access

Multiple Access

Characteristics:

• Shared medium : radio channel is shared by an priori unknown number of stations

• Broadcast medium: all stations within transmission range

of a sender receive the signal

Trang 14

Wired vs Wireless

• Ethernet uses 1-persistent CSMA/CD

– carrier sense multiple access with collision detection

• Sense if the medium is free and start sending as soon as it

becomes free

• While sending listen to the medium to detect other senders

• In case of a collision immediately stop sending and wait for the

random amount of time

• Problems in wireless networks

– signal strength decreases quickly with distance

– senders apply CS and CD, but the collisions happen at receivers pp y , pp

– Energy efficiency: having the radio turned on costs almost as

much energy as transmitting, so to seriously save energy one

needs to turn the radio off!

9

Outline

• Designing Issues of MAC protocolsg g p

• Classification of MAC protocolsClassification of MAC protocols

• Summary

10

Need for MAC Protocols ?

• Popular CSMA/CD (Carrier Sense Multiple

Access/Collision Detection) scheme is not applicable to

wireless networks

• CSMA suffers hidden terminal & exposed terminal

problems

Collision Detection is impossible in wireless

• Collision Detection is impossible in wireless

communication

Specific MAC protocols for the access to the

physical layer

Hidden Terminal Problem

• A sends to B, C cannot receive A

• C wants to send to B, C senses a “free” medium (CS fails)

• collision at B, A cannot receive the collision (CD fails)

• A is “hidden” for C

B

Trang 15

Exposed Terminal Problem

• B sends to A, C wants to send to D

• C has to wait, CS signals a medium in use

• since A is outside the radio range of C waiting is not

Near and Far Terminals

• Terminals A and B send, C receives

– the signal of terminal B hides A’s signal – C cannot receive A

• Classification of MAC protocols

• Summary

Classification of MAC protocols

Trang 16

In general (1/2)

• Contention-based protocols:

– A node does not make any resource reservation a priori.

– Whenever a node receives a packet to be transmitted, it

contends with its neighbour nodes for access

– Can not provide QoS (Quality of Service) guarantees to session Can not provide QoS (Quality of Service) guarantees to session

since nodes not guaranteed regular access to the channel

• Contention-based with reservation

– Wireless networks may need to support real-time traffic

– Reservation mechanisms for reserving bandwidth a priori

– Such protocols can provide QoS support to time-sensitive traffic

sessions

17

In general (2/2)

• Contention-based with scheduling

– These protocols focus on packet scheduling at nodes, and also

scheduling nodes for access to the channel – Used for enforcing priorities among flows whose packets are queued at nodes

q – Some of them take into consideration battery characteristics (remaining battery power)

• Other protocols

18

Outline

• Protocols examplesProtocols examples

• Summary

Multiple Access with Collision Avoidance (MACA)

• MACA uses a two step signaling procedure to address the hidden

RTS

and exposed terminal problems

• Use short signaling packets for collision avoidance

from a receiver with a short RTS packet before it sends a data packet Clear to send CTS: the receiver

Data

s

u s y

– Clear to send CTS: the receiver grants the right to send as soon as it

y

Trang 17

• Network allocation vector (NAV)

• Duration during which other sender have to keep quiet to avoid a

collision

• If control (RTS-CTS) messages collide with each other

or with data packets, a backoff procedure is activated

(backoff is binary exponential)

• Example: Wireless LAN (IEEE 802.11)

• MACA avoids the problem of exposed terminals

– B wants to send to A,

d C t D and C to D – now C does not have

to wait as C cannot

RTS CTS

• MACAW extends MACA : RTS-CTS-DS-DATA-ACK

– DLL (Data Link Layer) acknowledgements

– An improved backoff mechanism

– DS (Data Sending) message:

• Say that a neighbour of the sender overhears an RTS but not a CTS

(from the receiver)

• In this case it can not tell if RTS-CTS was successful or not

• When it overhears the DS, it realizes that the RTS-CTS was

successful, and it defers its own transmission

MACA extensions (2/2)

• MACA –by invitation (MACA-BI) : RTR-DATA

– Is a receiver-initiated MAC protocol, the receiver node initiate

d t t i i data transmission – It reduces the number of control packets used in the MACA protocol

p – MACA-BI eliminate the need for the RTS packet, it uses RTR (ready to receive) control packet to the sender.

RTR k t i i f ti b t th ti i t l d i – RTR packets carries information about the time interval during which the DATA packet would be transmitted

– The efficiency of the MAC-BI scheme is mainly dependent on the y y p ability of the receiver node to predict accurately the arrival rates

of the traffic at the sender nodes.

Trang 18

Media Access with Reduced Handshake (MARCH)

• MARCH is receiver-initiated protocol

• Unlike MACA-BI does not require any traffic prediction

mechanism

• In MARCH the RTS packet is used only for the first

packet of the stream From the second packet onward,

only the CTS packet is used

• The protocol exploits the broadcast nature of the traffic

to reduce the number of the handshakes involved in data

transmission

25

Reservation-based MAC protocol - DAMA

• Demand Assigned Multiple Access (DAMA)

• Practical systems therefore use reservation whenever possible

– But: Every scalable system needs an Aloha style component.

• DAMA allows a sender to reserve timeslots Two phase approach

• Reservation phase:

– a sender reserves a future time-slot

– sending within this reserved time-slot is possible without collision sending within this reserved time-slot is possible without collision – reservation also causes higher delays

• Termination phase: collision-free transmission using p greserved timeslots

26

DAMA: Explicit Reservation

• Aloha mode for reservation: competition for small

reservation slots, collisions possible

• Reserved mode for data transmission within successful

reserved slots (no collisions possible)

• It is important for all stations to keep the reservation list

consistent at any point in time and, therefore, all stations

have to synchronize from time to time

collisions

t reserved reserved reserved reserved

PRMA: Implicit Reservation

• Packet Reservation Multiple Access (PRMA)

• A certain number of slots form a frame, frames are repeated.

• Competition for this slots starts again as soon as the slot was empty

in the last frame reservation 1 2 3 4 5 6 7 8 time-slot

in the last frame

2

frame3frame 4

collision at reservation attempts

A -BAFD

AC-ABAF-frame5

attempts

A C E E B A F D

t ACEEBAFD

Trang 19

Distributed PRMA

• Every frame consists of n mini-slots and x data-slots

• Every station has its own mini-slot and can reserve up to

k data-slots using this mini-slot (i.e x = nk).

• Other stations can send data in unused data-slots

according to a round-robin sending scheme (best-effort

– Case 1: Node A and B are both not connected

• Node A sends invitation message

• Node B answers that it is not connected to any other node

• Node A tells B to pick slot/frequency for the link p q y

• Node B returns the link specification

– Case 2: Node A has neighbours and node B does not

• Node A creates the link specification and instructs Node B to use it

– Case 3: Node A has no neighbours, but node B has some

• Node B creates the link specification and instructs node A to use it p

– Case 4: Both nodes have links to neighbours

• Nodes exchange their schedules and pick free slots/frequencies in

mutual agreement

Schedule-based MAC protocols – TRAMA

• TRAMA: Traffic-adaptive medium access protocol

• Nodes are synchronised

• Time is divided into cycles that consists of

– Random access periods – Scheduled access periods

• Nodes exchange neighbourhood information

– Learning about their two-hop neighbourhood by using the

‘neighbourhood exchange protocol’

• In random access period send small incremental neighbourhood update information in randomly selected time slots

• Nodes exchange schedules

– Using the ‘schedule exchange protocol’

• Similar to neighbourhood information exchange

Trang 20

Schedule-based MAC protocols – TRAMA II

• As a result: Each node knows its two-hop neighbourhood and

the schedule

• Problem

– How to decide which slot (in scheduled access period) to use?

• Solution: ‘Adaptive Election’

– Use node identifier x and globally known hash function h

– For time slot t, compute priority p as follows: p = h(x  t)

– Compute this priority for next k time slots for the node itself and all Compute this priority for next k time slots for the node itself and all

• Summary

34

Link Layer Protocols

• Link Layer protocols cover the following topics

– Error Control

• Make sure that the sent bits arrive and no other

 forward and backward error control

• Discovery and management of links to neighbouring nodes

Goal: Create a reliable communication link

Error control

• Error control has to ensure that data transport is

– Error-free  deliver exactly the sent bits/packets – In-Sequence  deliver them in the original order – Duplicate-free  and at most once

Loss free  and at least once – Loss-free  and at least once

• Causes: fading, interferences, loss of bit synchronisation

– Results in bit errors packet losses Results in bit errors, packet losses

• Mostly occurring in bursts

– In wireless networks high average bit error rates: 10 -2 10 -4

• Approaches

– Backward error control: ARQ (Automatic Repeat Request) – Forward error control: FEC (Forward Error Correction)

Trang 21

Error control – ARQ

• Idea of ARQ

– Transmitting node’s link layer accepts a data packet, creates a

link-layer packet by adding a header and a checksum and

transmits this packet to the receiver

– Receiver checks packet’s integrity with the help of the checksum

and provides feedback; on negative feedback  retransmission

• Standard ARQ Protocols

– Alternating bit g

• Transmitter buffers one packet; single bit sequence number

– Go-back N

• Buffer up to N packets; packets that were not ack are retransmitted Buffer up to N packets; packets that were not ack are retransmitted

– Selective Repeat/ Selective Reject

• Sender and Receiver can buffer up to N packets; timer exceeds

missing packets are re requested

Channel Encoder (FEC)

leaver Modulator

Inter-Information source

Source symbols

Channel symbols

Digital waveform

Tx antenna

Channel

Channel decoder (FEC)

leaver Demodulator

Deinter-Information sink

Rxantenna

Source symbols

Channel symbols

g waveform

– K bits of user data are mapped to n channel symbols; however,

coding of two successive k-bit blocks is not independent

• Also hybrid schemes i e combination of ARQ and FEC

• Also hybrid schemes, i.e combination of ARQ and FEC

• Optimal packet size depends on

– Overhead – payload size – and bit error rate (BER)

• For known BER optimal frame length is easy to determine

• Problem: How to estimate BER? ( adaptive schemes)

Collect channel state information at the receiver (RSSI FEC ) – Collect channel state information at the receiver (RSSI, FEC, …)

• Second problem: How long are observations valid? (aging)

– Only recent past is credible y p

Trang 22

Link management

• Goal

– Decide to which neighbours a link should be established

• Problem: Link quality

– is not binary (good vs bad), i.e link quality has several

characteristics

– is time variable due to mobility, interferences, etc

– has to be estimated, actively by sending probe packets and , y y g p p

evaluating Reponses or passively by overhearing

• Establish a ‘neighbourhood table’ to store neighbouring

nodes and their associated link qualities

– Can be automatically constructed as part of MAC protocols

41

Link management – Link Quality Characteristics

• Experiments show that the simple circular shape for the region of communication is not realistic

– Instead

• Irregular shape of the region of communication

• Correlation between distance and loss rate is weak Correlation between distance and loss rate is weak

• Asymmetric links are rather frequent

• Packet loss rate is time variable even when neighbours are stationary  Significant short term variations

stationary  Significant short-term variations

• Regions of communication

– Effective region: Effective region: Link quality should be

consistently >= 90% of packets arrive – Poor region: packet loss rates beyond 90%

y understood in a statistical and time- varying sense

– Transitional region: anything in between

42

Link quality estimation

• How to estimate the quality of a link in the field?

• Reduce unnecessary link quality estimation effort to save energy

• Passive vs active estimators

– Active: node sends out special packets and collects responses

– Passive: node observers transmissions in its neighbourhood Passive: node observers transmissions in its neighbourhood

Outline

• Summary

Trang 23

802.11 – The lower layers in detail

• PMD ( Physical Medium Dependent) • MAC

• PMD ( Physical Medium Dependent)

– channel selection, PHY-MIB

• Station Management

– coordination of all management functions

– roaming – power management – MIB (management information

– Asynchronous Data Service (mandatory)

• exchange of data packets based on “best-effort”

• support of broadcast and multicast – Time-Bounded Service (optional)

• implemented using PCF (Point Coordination Function)

• Access methods

– DFWMAC-DCF CSMA/CA (mandatory)

• collision avoidance via binary exponential back-off mechanism y p

• minimum distance between consecutive packets

• ACK packet for acknowledgements (not used for broadcasts) – DFWMAC-DCF w/ RTS/CTS (optional) DFWMAC DCF w/ RTS/CTS (optional)

• avoids hidden terminal problem – DFWMAC-PCF (optional)

• access point polls terminals according to a list

46

MAC layer

• defined through different inter frame spaces

• no guaranteed, hard priorities

SIFS (Sh t I t F S i )

• SIFS (Short Inter Frame Spacing)

– highest priority, for ACK, CTS, polling response

• PIFS (PCF IFS) ( )

– medium priority, for time-bounded service using PCF

• DIFS (DCF, Distributed Coordination Function IFS)

– lowest priority, for asynchronous data service

PIFS DIFS DIFS

t medium busy SIFS

PIFS

next frame contention

direct access if medium is free  DIFS

CSMA/CA

contention window (randomized back off

medium busy

DIFS DIFS

next frame

(randomized back-off mechanism)

t

slot time

direct access if medium is free  DIFS

• station ready to send starts sensing the medium (Carrier Sense based on CCA, Clear Channel Assessment)

• if the medium is free for the duration of an Inter-Frame Space (IFS), the station can start sending (IFS depends on service type)

• if the medium is busy, the station has to wait for a free IFS, then the station must additionally wait a random back-off time (collision y ( avoidance, multiple of slot-time)

• if another station occupies the medium during the back-off time of the station, the back-off timer stops (fairness) , p ( )

Trang 24

CSMA/CA 2

• Sending unicast packets

– station has to wait for DIFS before sending data

– receivers acknowledge at once (after waiting for SIFS) if the

packet was received correctly (CRC)

– automatic retransmission of data packets in case of transmission automatic retransmission of data packets in case of transmission

errors

DIFS

SIFS DIFS

ACK receiver

t

data other

stations

receiver

DIFS

t waiting time

stations

contention

49

Outline

• Summary

50

Summary

• The most important design goal of a MAC protocol is to

enable shared access to the common wireless medium

• The issues associated with the design of the MAC

protocol of wireless ad hoc networks are:

– Error-prone shared broadcast channel

– Hidden and exposed problems

– Distributed nature / lack of central coordination

Summary

• The Ad Hoc wireless networks MAC protocols have been classified into different categories

• Some protocols were discussed as examples

• In the MAC layer of the sensor networks lectures other MAC protocols design issues will be discussed

• Outlook: Next lecture will talk about routing protocols for

ad hoc networks

Trang 25

Summary – Next Session

Application

Transport Protocol

Network Protocol

53

Trang 26

Prof Dr Dieter Hogrefe

Dr Omar Alfandi

Outline

• Issues in Designing Routing Protocols

• Classification of Routing Protocols

– Proactive Routing Protocols g – Reactive Routing Protocols – Hybrid Routing Protocols

• Summary

2

Mobility

• Dynamic network topologyy p gy

• Session disruption by Movement

– End node movement

• Wired: Fiber + WDM → abundant bandwidth

• Wireless: Limited radio band → low bandwidth

• Keeping routing overhead low, is key

• Complete topology information not suitable forAdHoc wireless

AdHoc wireless

Trang 27

Error-Prone Shared Broadcast Radio

Channel

• Link capacity varies

• Link-error probability variesp y

• MAC layer interaction necessary (Cross Layer)

– Using better-quality links for route determination g q y

• Broadcast nature is a challenge

– Packet collision → Terminal Problems

• But wireless is not transitive!

– A hears B – A hears C A hears C – B and C may not hear each other – B and C can only sense their channel, but need to know if A’s channel is clear

• A and C can’t hear each other B can hear both

• A and C sense a clear channel, transmit, and

Trang 28

Resource Constraints

• Limitations for mobile nodes

– Battery life (1500 mAh mobile; 35 Ah portable) Battery life (1500 mAh mobile; 35 Ah portable)

– Processing power (2 MIPS RISC mobile; 20MIPS

CISC portable)

– Weight constraints (150g mobile; 2500g portable)

– Size constraints (100 cm³ mobile; 2000 cm³ portable)

• Routing protocol needs to adress

– Energy effcienct route detection Energy effcienct route detection

– Processing effort minimization

9

Ideal AdHoc Routing Protocol

• 1 Fully distributed, no central control node

– Lower control overhead, increased scalability Lower control overhead, increased scalability

• 2 Adaptive to frequent topology changes(mobility)

• 3 Route computation: Minimal node involvement

– Quick access to routes for each node – Minimum connection set up time

• 4 Localization of route discovery/maintenance

– No global state maintenance -> high control overhead

10

Ideal AdHoc Routing Protocol (2)

• 5 Loop free/stale route free routes

• 6 Minimal packet collisions

– Reduced broadcasts by each node

• 7 Convergence to optimal routes7 Convergence to optimal routes

– In case topology becomes stable

– Quick convergence

• 8 Low resource comsumption

– Bandwidth, cpu/ram, energy

• 9 Local topology updates only

• 10 QoS support as demanded by applications

Outline

• Summary

Trang 29

• Path finding processes and routing metrics

deviate from traditional wired protocols

1 Update Mechanism 2 Temporal Info . 3 Topology Info. 4 ResourceUtilization

Table Driven (Proactive) History based Flat Routing Power Aware Routing

On-Demand (Reactive) Prediction based Hierarchical Routing Georouting

14

Classification

1 Update Mechanism Table Driven

(Proactive)

DSDV

On-Demand (Reactive) DSR

Hybrid

CEDAR DSDV

History Prediction

3 Topology Info.

Hierarchical History

based

DSDV

Prediction based

FORP

Flat Routing

DSR

Hierarchical Routing

CGSR DSDV

Trang 30

4 Resource Utilization

Power Aware

Routing Georouting FloodingEfficient

17

Outline

• Summary

18

Proactive routing protocols: DSDV

Ad Hoc Protocols

(Reactive) Predictionbased HierarchicalRouting Georouting

DSDV: Routing Updates

• Routes to all destinations are readily available at every node’s routing table at all times

• Each routing table entry at every given node maintains:

<dest-addr, dest-seq#, next-hop, hop-count>

• Seq#s (created by destinations) are used to distinguish stale routes and avoid formation of route loops

• The tables are exchanged between neighbours at regular intervals to keep an up-to-date view of the network topology (can happen time based and eventnetwork topology (can happen time-based and event-based)

Trang 31

DSDV: Routing Updates

• Each node sends a periodic update to its neighbors

– The update contains its own dest-seq# and other routing

i f ti f th d ti ti

information for the destination

< dest-addr, dest-seq#, hop-count >

• Also each node sends routing table updates caused by

important changes in the local topology (e.g link failure)

• When a node receives two routes to the sameWhen a node receives two routes to the same

destination from two neighbors it would:

– Choose the route with the largest sequence number

– If the same: Choose the smalest hop-count

4 Node B receives the information and adds

<A, 101, 1> to its own routing table, , g

Trang 32

DSDV: Adding a link

5 Node B propagates the new routing information to its p p g g

neighbors <A, 101, 1>

6 The neighbors complement their routing tables

<A, 101, B, 2> and propagate their information to their

neighbors

25

DSDV: Failure of a link

1 A link between B and D fails

2 Node B notices the failure:

- Sets hop-counts for D and E on ∞

- increment seq# for D and E

• in order to obtain information about a particular pdestination node, a node has to wait for a table update message initiated by the same destination node

Trang 33

– Proactive Routing Protocols g

– Reactive Routing Protocols

– Hybrid Routing Protocols

• Summary

29

Reactive routing protocols: AODV

Ad Hoc Protocols

1 Update Mechanism 2 Temporal Info . 3 TopologyInfo. 4 ResourceUtilization

Table Driven (Proactive) Historybased Flat Routing Power Aware Routing

On-Demand (Reactive) Predictionbased HierarchicalRouting Georouting

30

AODV

• Managing only active routes

• Every node maintains two counters: node seq#, node

messages that each node broadcasts at set intervals

• All intermediate nodes having valid routes to the

destination or the destination node itself are allowed to

destination, or the destination node itself, are allowed to

send RouteReply packets to the source.

AODV: Route Discovery

1 Node S needs to find a route to D

2 S initiates a Route Request (RREQ) message

<D‘s-IP-addr., D‘s-Seq#, S‘s-IP-addr., S‘s-Seq#, hop-count(=0)> q q p ( ) The pair <S‘s-IP-addr., S‘s-Seq#> is the RREQ unique identifier helping the nodes droping duplicate RREQs.

3 S sends the RREQ message to its neighbors

Trang 34

4 Node A receives RREQ

- makes route reverse entry for S

Des=S, nexthop=S, hopcount=1

- A has no direct route to D, So sends RREQ to its neighbors

5 Node C receives the RREQ

- makes route reverse entry for S

Des=S, nexthop=A, hopcount=2

- C has a direct route to D, and the seq# for route to D is S‘s

seq# in RREQ

33

6 Node C receives RREQ

- C initiates a route reply (RREP) RREP=<D‘s IP addr, D‘s-seq#, S‘s IP addr, _ _ , q , _ _ ,hopcount_to_D(=1)>

- C sends the RREP to A

34

7 Node A receives RREP

- A makes a route forward entry for Dy

Dest=D, nexthop=C, hopcount=2

- A sends RREP to S

8 Node S receives RREP

- S makes a route forward entry for DyDest=D, nexthop=A, hopcount=3

Trang 35

9 Node S receives RREP

- S makes a forward entry route for D

Dest=D, nexthop=A, hopcount=3, p , p

- S sends the data packets on the new route to D

37

AODV: Route Management

1 The route between C and D fails

2 Node C invalidates the route to D in its routing table

3 Node C generates Route Error message (RERR)g g ( )

- Lists all of the destinations which are not accessible now

- Sends RERR to the upstream neighbor

38

AODV: Route Management

4 Node A receives RERR message

- Checks whether C is the next hop on the route to D

- Deletes the route to D

RERR b d t ti t S

- RERR broadcast continues to S

5 Node S receives RERR message

- Examines if A is the next hop on the route to D

- Deletes the route to D

- Finds a new route to D if necessary

AODV Advantages:

• routes are established on demand and destination sequence numbers are used to find the latest route to

th d ti tithe destination

AODV Disadvantages:

• multiple RouteReply packets in response to a single RouteRequest packet can lead to heavy control

h doverhead

consumption

Trang 36

Reactive routing protocols: DSR

D ynamic Source Routingy g

Ad Hoc Protocols

41

DSR: Route Discovery

DSR Route Discovery:

Node S needs a route to D

1 S sends RREQ message

2 A receives the message, but it has no route to D

- A adds its own address and sends the message to its neighbors

3 C receives RREQ message but has no route to D

3 C receives RREQ message, but has no route to D

- adds its own address and sends the packet to its neighbors

42

4 Node C receives RREQ message, but has no route to

D

- adds its own address and sends the packet to its neighbors

5 Node D receives RREQ and sends a RREP back to C

- The RREP contains the route from S to D

6 Node C receives RREP

- RREP is broadcasted continuesly to A

7 Node A receives the RREP message

- sends RREP to S

8 Node S receives the RREP message

- S will use the discovered route to forward data

• DSR nodes may learn more than one route during route discovery

Trang 37

DSR Advantages:

• A route is established only when it is required

• The intermediate nodes also utilize the route cache

information efficiently to reduce the control overhead

DSR Disadvantages:

• The route maintenance mechanism does not locally

repair a broken link

• The connection setup delay is higher than in table-driven

protocols

• Considerable routing overhead is involved due to the

source-routing mechanism This routing overhead is

directly proportional to the path length

45

AODV and DSR differences:

• DSR uses source routing in which a data packet carries the complete path to be traversed, but AODV uses next

hop record in which the source node and the intermediate nodes store the next-hop information corresponding to each flow for data packet transmissioncorresponding to each flow for data packet transmission

• DSR route uses cache, AODV uses routing table

• DSR route cach entries have no lifetimesAODV routing table entries have lifetimes

• DSR has alternate route available when one breaksAODV nodes do not alternate route when one breaks

46

Reactive routing protocols: LAR

Ad Hoc Protocols

LAR

• A location based reactive routing protocol

– Location information of the nodes is used (e.g Obtained from GPS)

– All nodes should know their current location

• Expected Zone: The area that probably contains the

desired destination

– EZ is estimated based on any previous location information and y p the destination‘s velocity

• D= last known location of node D at time t0

• D‘= location of node D at present time t1

• D = location of node D at present time t1 – D‘ is unknown to the sender

• r=(t1- t0)*Vd

• Vd= estimated speed of D

Trang 38

LAR: Request Zone

• Request Zone contains the expected zone and the

location of the sender and the RREQ is limited in that

– Only nodes within the request zone forward the RREQ

• Request zone explicitly is specified in RREQ

• Each node must know its location to determine whether

it is within the request zone or not

49

LAR: Request Zone

• If route discovery with the smallest request zone fails, the sender would try to discover the route with a larger

request zone (after a time out)

– The larger zone can cover the entire network

The rest of the route discovery is similar to DSR

• The rest of the route discovery is similar to DSR

50

LAR: Advantages and Disadvantages

• Advantages

– reduced RREQ flood (only in the request zone)

– reduced route discovery overhead

• Disadvantages

– Nodes need to know their locations

– The possible existence of obstructions for radio

transmissions is not taken into account

Outline

• Summary

Trang 39

Hybrid routing protocols: ZRP

Z one Routing Protocolg

Ad Hoc Protocols

53

ZRP: Zone Routing Protocol

• Combining the advantages of both reactive and proactive approaches

• Each node maintains an up-to-date map of a zone centered at every node (r-hops)

• Within a zone the routes are immidiately available

(intra-zone routing protocol)

• For destinations out of the zone a reactive route

discovery protocol is used (inter-zone routing protocol)

• In MANETs it can be assumed that the most traffic is

• In MANETs it can be assumed that the most traffic is directed to nearby nodes, so ZRP reduces the proactive scope to a limited zonep

54

ZRP: Route Discovery

• The larger the routing zone, the higher the update control

traffic

• When a node s needs a route to a destination d, it checks

whether node d is within its zone

If so it delivers the packet directly Otherwise node s

• If so, it delivers the packet directly Otherwise, node s

directly routes the RREQ to the border nodes to its

peripheral nodes

p p

• If any peripheral node finds node d to be located within its

routing zone, it sends a RREP back to s; otherwise, the

node rebordercasts the RREQ packet to the peripheral

nodes

• This process continues until node d is located

• In the figure node 8 needs a route to node 16 node

• node 8 bordercasts RouteRequests to nodes 2, 3, 5, 7, 9, 10,

13 14 d 15

13, 14 and 15

• Nodes 10 and 14 find the information about node 16 to be available in their intrazone routing tables and hence they available in their intrazone routing tables, and hence they originate RouteReply packets back to node 8

• During RouteRequest propagation, g g every node that forwards the RREQ appends its address to it

• This information is used for delivering the RREP packet back

to the source.

Trang 40

• When an intermediate node in an active path detects a

broken link in the path, it performs

• A local path reconfiguration in which the broken link is

bypassed by means of a short alternate path connecting

the ends of the broken link

• A path update message is then sent to the sender node

to inform it about the change in path

57

ZRP: Advantages and Disadvantages

• ZRP reduces the control overhead compared to the RREQ flooding mechanism employed in on demand

approaches and the periodic flooding of routing information packets in table driven approaches

But in the absence of a query control ZRP tends to

• But in the absence of a query control, ZRP tends to produce higher control overhead than the

aforementioned schemes This can happen due to theaforementioned schemes This can happen due to the large overlapping of nodes' routing zones The query control must ensure that redundant or duplicate RREQs are not forwarded Also, the decision on the zone radius has a significant impact on the performance of the protocol

Outline

• Summary

Tiêu đề	Mạng không dây 2 pot
Tác giả	Dieter Hogrefe, Omar Alfandi
Trường học	Dai Hoc Cong Nghiep Ha Noi
Chuyên ngành	Wireless Networks
Thể loại	Giáo trình
Năm xuất bản	2010/2011
Thành phố	Hà Nội

Định dạng
Số trang	234
Dung lượng	8,02 MB