Wireless sensor network survey

Wireless sensor network survey

Author: Jennifer Yick, Biswanath Mukherjee, Dipak GhosalReported by Jiang

2

Introduction

• Subject: wireless sensor network (WSN)

• WSN consists of spatially distributed autonomous sensors to monitor physical or environmental conditions, such as temperature, sound, pressure, etc. and to cooperatively pass their data through the network to a main location.

What?

3

Overall image

A typically WSN consists of generic and gateway nodes.

4

Typically WSN

• A WSN typically has little or no infrastructure. There are two types of WSNs.

Structured WSN Unstructured WSN

5

Structured WSN

• Deployed in a pre-planned manner• Fewer nodes• Lower network maintenance• Lower cost• No uncovered regions

6

Unstructured WSN

• Densely deployed (many node)• Randomly Deployed • Can have uncovered regions• Left unattended to perform the task• Maintenance is difficult

a. managing connectivityb. detecting failures

7

generic nodes

gateway nodes

8

Requirement

WSN

military target tracking hazardous environment exploration

natural disaster relief

surveillance

biomedical health monitoring

seismic sensing

9

Why we select WSN ? Not traditional networks.• Power and environment constraints

determine us to design a lower power, feasible, and smart network.

• Sensors that are smaller, cheaper, and intelligent.

10

About this paper

• Goal: present a comprehensive review of the recent literature.

① An overview of the key issues in a WSN② Compare different types of sensor networks③ Applications on WSN④ Internal sensor system⑤ Network services⑥ Communication protocol

11

Key issues

• Energy constraint • Quality of service (QoS)

Each sensor node is an individual system. Development should satisfy current requirement.

Self-organizingConsume less power

Total number and placement

Address network dynamicsOptimize communication and

be energy efficiency

12

Types of sensor networks

1. terrestrial WSN• Ad Hoc (unstructured)• Preplanned (structured)

2. underground WSN• Preplanned, with additional sink nodes to relay data.• more expensive equipment, deployment, maintenance

3. underwater WSN • fewer sensor nodes( sparse deployment)• more expensive than terrestrial• acoustic wave communication

– Limited bandwidth– long propagation delay– signal fading

13

Types of sensor networks(cont.)

4. multi-media WSN • sensor nodes equipped with cameras and microphones• pre-planned to guarantee coverage• High bandwidth/low energy, QoS, filtering, data

processing and compressing techniques

5. mobile WSN• ability to reposition and organize itself in the network• Start with Initial deployment and spread out to gather

information• deployment, localization, self-organization, navigation

and control, coverage, energy, maintenance, data process

14

WSN applications

15

Example: CenWits• Berkeley Mica2 sensor mote• GPS receivers• radio frequencies (RF)-based

sensors• storage and processing devices

16

Internal sensor system

17

Internal sensor system

•sensor platform –radio components–processors–Storage–sensors (multiple)

•OS–OS must support these sensor platforms.

It’s hard to design a general platform to be applied to all applications due to requirements vary in terms computation, storage and user interface.

18

Internal sensor system Standards• IEEE 802.15.4, ZigBee• WirelessHART • ISA100.11 • IETF 6LoWPAN • IEEE 802.15.3 • Wibree

19

Internal sensor system Standard example: ZigBee• IEEE 802.15.4:

–standard for low rate wireless personal area networks (LR-WPAN)

–low cost deployment, low complexity, power consumption

–topology :star and peer-to-peer

–MAC layer: CSMA-CA mechanism

• ZigBee–simple, low cost, and low

power–embedded applications–can form mesh networks

20

Internal sensor system Storage• problems

–storage space is limited –Communication is expensive

• Solutions–Aggregation and compression –query-and-collect (selective gathering)–a storage model to satisfy storage constraints and query

requirements

21

Internal sensor system Testbeds• Provides researchers a way to test their protocols,

algorithms, network issues and applications in real world setting

• Controlled environment to deploy, configure, run, and monitoring of sensor remotely

22

Internal sensor system Testbeds example: Orbit• a two-dimensional grid of 400 802.11 radio nodes.• dynamically interconnected into specified topologies

with reproducible wireless channel models.

23

Internal sensor system Diagnostics and debugging support• Measure and monitor the sensor node

performance of the overall network• To guarantee the success of the sensor

network in the real environment

24

Network services

① Localization② Synchronization③ Coverage④ Compression and

aggregation⑤ Security

25

Network services Localization• Problem:

–determining the node’s location (position) • Solutions:

–global positioning system (GPS)•Simple•Expensive•outdoor–beacon (or anchor) nodes•does not scale well in large networks•problems may arise due to environmental conditions–proximity-based•Make use of neighbor nodes to determine their position •then act as beacons for other nodes

• Other solutions

26

Network services Synchronization• Time synchronization is important for

–routing –power conservation–Lifetime–Cooperation–Scheduling

27

Network services Coverage• Is important in evaluating effectiveness• Degree of coverage is application dependent• Impacts on energy conservation• Techniques:

–selecting minimal set of active nodes to be awake to maintain coverage

–sensor deployment strategies

28

Network services Compression and aggregation• Both of them

–reduce communication cost –increase reliability of data transfer

• Data-compression–compressing data before transmission to base –Decompression occurs at the base station–no information should be lost

• data aggregation–data is collected from multiple sensors–combined together to transmit to base station–Is used in cluster base architectures

29

Network services Security• Constraints in incorporating security into a

WSN –limitations in storage–limitations in communication–limitations in computation–limitations in processing capabilities

30

Network services Open research issues• localization

–efficient algorithms–minimum energy–minimum cost–minimum localization errors

• Coverage: optimizing for better energy conservation• time synchronization: minimizing uncertainty errors over long periods of

time and dealing with precision• compression and aggregation: Development of various scheme

–event-based data collection–continuous data collection

• Secure monitoring: protocols have to monitor, detect, and respond to attacks

–It has done for network and data-link layer (can be improved)–Should be done for different layers of the protocol stack –Cross-layer secure monitoring is another research area

31

Communication protocol

① Transport layer② Network layer③ Data-link layer④ Physical layer

32

Communication protocol Transport layer• Packet loss

–may be due to •bad radio communication, •congestion, •packet collision, •memory full, •node failures–Detection and recovering•Improve throughput •Energy expenditure

33

Communication protocol Transport layer• Congestion control/packet recovery

–hop-by-hop•intermediate cache•more energy efficient (shorter retransmission)•higher reliability–end-to-end•source caches the packet•Variable reliability

34

Communication protocol Transport layer(Open research issues)• cross-layer optimization

–selecting better paths for retransmission–getting error reports from the link layer

• Fairness–assign packets with priority–frequently-changing topology

• Congestion control with active queue management

35

Communication protocol Network layer• Important:

–energy efficiency –traffic flows

• Routing protocols–location-based: considers node location to route

data–cluster-based: employs cluster heads to do data

aggregation and relay to base station

36

Communication protocol Network layer (Open research issues)

• Future research issues should address –Security•Experimental studies regarding security applied to different routing protocols in WSNs should be examined–QoS•guarantees end-to-end delay and energy efficient routing–node mobility•handle frequent topology changes and reliable delivery

37

Communication protocol Data-link layer (Open research issues)

• system performance optimization• Cross-layer optimization

–Cross-layer interaction can •reduce packet overhead on each layer•reduce energy consumption–Interaction with the MAC layer provide •congestion control information•enhance route selection–Comparing performance of existing protocols of static

network in a mobile network–improve communication reliability and energy efficiency

38

Communication protocol Physical layer• Minimizing the energy consumption

–Optimizing of circuitry energy•reduction of wakeup and startup times–Optimizing of transmission energy•Modulation schemes

• Future work–new innovations in low power radio design with emerging

technologies–exploring ultra-wideband techniques as an alternative for

communication–creating simple modulation schemes to reduce synchronization and

transmission power–building more energy-efficient protocols and algorithms

39

Communication protocol Cross-layer interactions (Open research issues)• Collaboration between all the layers to

achieve higher –energy saving–network performance–network lifetime

40

Thanks! That’s all for today.