Wireless Sensor Networks

1
Wireless Sensor Networks

• Lecturer Michael O'Grady
• Course MSc Ubiquitous Multimedia Systems
• Unit Context Sensitive Service Delivery
• Lecture

2
Objectives

• Sensor Components
• WSN Components
• Design issues
• Applications

3
Some Thoughts

• Recall Weisers vision for Ubiq. Computing
• Sense physical phenomena
• Two Objectives
• Ubiquity inject computation into the physical
  environment with a high spatial density
• Invisibility- computational nodes operate
  autonomously
• Recall Ambient Intelligence..
• How will this happen?
• One potential Solution Wireless Sensor Networks
  (WSNs)

4
Classification of Sensor Networks

• Event Detection
• Report specified event
• Periodic Measurements
• Report measurements at certain time intervals
• Function approximation
• Value approximation using a sample temperature
• Isothermal points identify edge of forest fire
• Tracking
• Surveillance

5
Architecture of a WSN

• Three Components (for this discussion!)
• Node
• Sink
• Task Node Manager

6
Sensor Node Tasks

• Computation
• Storage
• Communication
• Sensing/Actuation

7
Sensor Node Characteristics

• Small Size
• Battery
• Limited CPU
• Wireless communications capability
• One or more sensors (temperature etc)

8
Sensor Node Components

• Sensing Unit
• Processing Unit
• Power Unit
• Transceiver

9
Node - Sensing Unit

• Senses or measures certain physical phenomenon
• Light
• Sound
• Et c etc
• Analogue signal generated by Sensor
• Signal digitised and sent to the Processing Unit
• ADC Analogue-to-Digital Converter (ADC)
• Note alternative Sensor architectures available

10
Node - Processing Unit

• Processes sensed data
• Collaborates with other Nodes to achieve WSNs
  objectives.
• Components
• Microcontrollers/Microprocessors
• FPGA (Field programmable Gate Arrays)
• Reprogrammable / reconfigurable
• Memory
• Flash (low cost / storage capacity)

11
Node - Power Unit

• Batteries
• Alkaline
• wide voltage range
• Large physical size
• Lithium
• Constant voltage supply
• Low currents
• Major Challenge How to effectively preserve and
  capture power in a WSN?????

12
Node - Transceiver

• Three popular communication schemes
• Optical Communication (Laser)
• Low energy requirement (no antenna)
• secure
• Line of Sight required
• Sensitive to atmospheric conditions
• Infra-Red (IR)
• No antenna
• Limited in its broadcasting capacity
• Radio Freuency (RF)
• Antenna required
• Easy to use

13
WSN - Sink

• Sink
• A node that data should be delivered to
• Base station
• Cluster head
• Gateway Node
• Internet
• Cellular network
• PSTN

14
WSN- Task Node Manager

• Management
• Administration
• Interface
• WSN health
• WSN status
• Alarms

15
Design Issues I

• Fault Tolerance
• Sustain WSN operation despite individual node
  failure
• Scalability
• Maintain performance regardless of WSN size
• Production Costs
• Directly proportional to the cost of the node
• Ideally ltlt 1 USD
• Hardware Constraints
• Node should be less than size of matchbox!!!!!

16
Design Issues II

• Sensor Network Topology
• Deployment Strategies
• Pre-deployment
• How? Plane?, one by one?
• Post-deployment
• Topology changes
• Position
• Reachability
• Energy
• Malfunctionality
• Task
• Redeployment
• how?

17
Design Issues III

• Environment
• Close to observed phenomenon
• Transmission Media
• RF
• Infrared
• optical
• Power Consumption
• Strategy for power management/conversation
• Three domains
• Sensing
• Communication
• Data processing

18
Applications of WSNs

• Disaster Relief
• Environmental monitoring
• Intelligent Buildings
• Facility management
• Machine monitoring
• Precision agriculture
• Medicine
• Logistics
• Telematics

19
Unconventional Power Sources

• Power - an Achilles heel for WSNs?
• Recall Energy Scavenging
• Photovoltaics (solar)
• Temperature gradients
• Vibrations
• Mechanical energy
• Pressure Variation
• Piezoelectric generator
• Shoe
• Air/liquid flow
• Miniature gas turbines

20
Autonomic Computing

• What is Autonomic Computing?
• What was the motivation for its development?
• What are its 4 essential characteristics?
• Why integrate Autonomic computing concepts with
  Wireless Sensor Networks?

21
Some Readings

• Culler, D., Estrin, D., Srivastava, M. Overview
  of Sensor Networks
• Akyildiz, I. F. Su, W. Sankarasubramaniam, Y.
  and Cayirci, E. A. 2002. A survey on sensor
  networks. IEEE Communications Magazine
  40(8)102-114.
• Kephart, J. O. and Chess, D. M. 2003. The Vision
  of Autonomic Computing. Computer 36, 1 (Jan.
  2003), 41-50.

22
The End
23
Project Demonstration

• Wednesday 28th November 2006
• Format
• Short overview of the project
• Demo of the project
• Individual Contributions
• Final Report
• Due this coming Monday 6.00 PM
• Remember - names Students Numbers

24
Provisional Exam Timetable

• COMP40320 Adaptive Personalisation
• Tuesday 12th December 2006
• RDS 12.0013.45
• COMP40300 - Context Sensitive Service Delivery
• Saturday 16th December 2006
• RDS 12.00-13.45

25
Good Luck