Advanced Data Communication Networks - IAUNresearch.iaun.ac.ir/pd/faghani/pdfs/UploadFile_7917.pdf · Wireless Sensor Networks (WSN) ... Routing can be multihop. Topology is dynamic.

Session 10:Mobile Ad Hoc and Wireless Sensor

Networks

Advanced Data CommunicationNetworks

Dr. Farhad Faghani Assistant Professor, Department of Electrical Engineering,Najafabad Branch, Islamic Azad University.

Outline

I. Classification of Communication NetworksII. Introduction to Ad Hoc and sensor NetworksIII. Wireless sensor Network Design issuesIV. Crosslayer designV. Wireless application and QOSVI. Ad Hoc Routing Protocol

Classification of CommunicationNetworks

�Wired Networks�LAN, MAN, WAN, and Internet

�Wireless Networks�Infrastructured networks (cellular networks)�Infrastructureless networks (Ad Hoc wireless

networks)�Wireless Sensor Networks (WSN)

Infrastructured networks (cellular networks)�Frequencies, time slots, or codes reused at spatially-

separated locations�Decreasing the cell size increases capacity�Resources:

�Channels�Bandwidth�Power�Rate�Base stations�Access

�Dynamic Resource AllocationAllocate resources as user and network conditions change

BASESTATION

Infrastructureless networks(Ad Hoc wireless networks)

�Peer-to-peercommunications.

�No backboneinfrastructure.

�Routing can bemultihop.

�Topology is dynamic.�Fully connected with

different link SINRs

Different channels, distances, mobility, energy, and rate requirements.

Wireless Sensor Networks (WSN)Energy-Constrained Nodes

�Limited nodeprocessing/communication capabilities

�Nodes can cooperate in transmissionand reception.

�Intelligence must be “in the network”�Data flows to centralized location.�Low per-node rates but 10s to 1000s

of nodes�Data highly correlated in time and

space.

dotmag

acoustic

WSN, Energy-Constrained Nodes, (Cont’d)�Sensor modes

�Active mode� Idle mode�Sleep mode

�Major sources of energy waste� Idle listening when no sensing events�Collisions�Control overhead�Overhearing

Solution: Node schedulingPeriodic listen and sleep

Design Principles�Sleep – majority of the time, >99%�Wakeup – quickly start processing�Active – minimize work & return to sleep

processingdata acquisitioncommunication

sleep

Pow

er

Time

WSN, Energy-Constrained Nodes, Cont’d�Each node can only send a finite number of bits.�Short-range networks must consider both transmit

and processing energy.�Sleep modes save energy but complicate networking.

�Changes everything about the network design:�Bit allocation must be optimized across all protocols.�Delay vs. throughput vs. node/network lifetime tradeoffs.�Optimization of node cooperation.

Applications (Ad Hoc / WSN)�Battlefield communications�Wireless LANs�Emergency infrastructures�Communication infrastructure

for automated vehicles�Sensor networks�Medical applications (on-body)�Smart Buildings/Agriculture�Wide area (Environment

monitoring, counter-terrorism,etc.)

Heterogeneous Wireless Sensor Network

PDA

3d ultrasonicanemometer

Temperature,humidity

HPWREN

AnimalMonitoring

Notebook CellularPhone PC

Ship Monitoring

DataDistribution

Network

Precipitation

Solar radiation

In-flight camera

Weather station

Mobile and Stationary Operations

Stationary camera

Seismic

Storage

DataAcquisition

Network

Wireless Applications and QoSWireless Internet accessNth generation CellularWireless Ad Hoc NetworksSensor NetworksWireless EntertainmentSmart Homes/SpacesAutomated HighwaysAll this and more…

Applications have hard delay constraints, rate requirements,and energy constraints that must be met

These requirements are collectively called QoS

Cross Layer DesignChallenges to meeting QoS

� Wireless channels are a difficultand capacity-limited broadcastcommunications medium

� Traffic patterns, user locations,and network conditions areconstantly changing

� No single layer in the protocolstack can guarantee QoS: cross-layer design needed

� It is impossible to guarantee thathard constraints are always met,and average constraints aren’tnecessarily good metrics.

�Hardware�Link�Access�Network�Application

Delay ConstraintsRate RequirementsEnergy Constraints

Mobility

Crosslayer Techniques�Adaptive techniques

�Link, MAC, network, and application adaptation�Resource management and allocation (power control)

�Diversity techniques�Link diversity (antennas, channels, etc.)�Access diversity�Route diversity�Application diversity�Content location/server diversity

�Scheduling�Application scheduling/data prioritization�Resource reservation�Access scheduling

Networking Scenario1(Mesh Network)From Wireless to Wired network Via Multihop

Networking Scenario2 (SkyPilot NeighborNet)

Networking Scenario3(Urban Grid Applications)

Hot Spot

Hot Spot

Hot Spot

Hot Spot

PowerBlackout

STOP

PowerBlackout

STOP

PowerBlackout

STOP

PowerBlackout

STOP

Mobile Ad Hoc Networks (MANET)Routing

�Mechanism for finding paths from source host todestination host.

�Protocols use different metrics to find network paths.�Hop count, bandwidth, delay, power, load, link quality

(wireless).�Most routing protocols for wireless are based on wireline

protocols�Flooding, point-to-point, table-driven, on-demand, etc.

Wireless link

1- Hop 2- Hops 3- Hops

S

Example1: Multi-hop “ad hoc” networking

Carey

Kelly

Example1 (Cont’d)

Carey

Kelly

Example1 (Cont’d)

Carey

Kelly

Example1 (Cont’d)

Carey

Kelly

Example1 (Cont’d)

Carey

Kelly

Example1 (Cont’d)

Carey

Kelly

Example1 (Cont’d)

Carey

Kelly

Example1 (Cont’d)

Carey

Kelly

Example1 (Cont’d)

Carey

Kelly

Example1 (Cont’d)

Carey

Kelly

Example2: Sensor aggregate routingTTDD (Two Tier Data Dissemination)

�Data dissemination to mobile sinks�Two-tier query & data forwarding�Objectives

�Source proactively builds a grid structure to support dataavailability for mobile sinks� Mobility pattern is unknown a priori

�Localize impacts of sink mobility on data forwarding�Only a small set of sensor nodes maintain forwarding

state

Example2 (Cont’d) TTDD Basics

Source

Dissemination Node

Sink

Data Announcement

Query

Data

ImmediateDisseminationNode

Example2 (Cont’d) TTDD Mobile Sinks

Source

Dissemination Node

Sink

Data

ImmediateDisseminationNode

ImmediateDisseminationNode

TrajectoryForwarding

TrajectoryForwarding

Ad Hoc Routing Protocols�Proactive Routing Protocol

�continuously evaluate the routes�attempt to maintain consistent, up-to-date routing

information� when a route is needed, one may be ready immediately

�when the network topology changes� the protocol responds by propagating updates throughout the

network to maintain a consistent view

�Reactive Routing Protocol�on-demand�Ex: DSR, AODV

Ad Hoc Routing Protocols

AD-HOC MOBILEROUTING PROTOCOLS

ON-DEMAND-DRIVENREACTIVE

HYBRIDDSDV

CGSR

TABLE DRIVEN/PROACTIVE

DSR

AODV

ZRP

DSDV�Destination Sequenced Distance Vector�Table-driven�Each node maintains a routing table�Routing hops to each destination�Sequence number

�Problem�A lot of control traffic in the network

Clustering Protocol�Cluster Gateway Switch Routing (CGSR)

�Table-driven for inter-cluster routing�Uses DSDV for intra-cluster routing

C3

M2

C2

C1

AODV�Ad Hoc On-demand Distance Vector

�On-demand driven�Nodes that are not on the selected path do not maintain

routing information�Route discovery

� The source node broadcasts a route request packet (RREQ)� The destination or an intermediate node with “fresh enough” route

to the destination replies a route reply packet (RREP)

AODV� Problem

�A node along the routemoves

� Solution�Upstream neighbor notices

the move�Propagates a link failure

notification message toeach of its active upstreamneighbors

�The source node receivesthe message and re-initiateroute discovery

N2

N4N1

N3

N5

N6

N7

N8

Source

Destination

N2

N4N1

N3

N5

N6

N7

N8

Source

Destination

ZRP

�Zone Routing Protocol�Hybrid protocol

� On-demand� Proactive

Zone Radius =

r Hops

Zone of Node Y

Zone of Node Y

Node X

Zone of Node X

Node Z

Zone of Node Z

Border Node

Border Node

Bor