FAULT TOLERANT TECHNIQUES FOR WIRELESS AD HOC SENSOR NETWORKS

1
FAULT
TOLERANT TECHNIQUES FOR WIRELESS AD HOC SENSOR
NETWORKS

• RAM MOHAN BOPPUDI
• EE 585 CASE STUDY 2

2
WIRELESS AD HOC SENSOR NETWORK
3
INTRODUCTION

• A wireless ad hoc sensor network.
• 
  
• Some examples of wireless ad hoc sensor networks
  are the following
• Military sensor networks .  
• Sensor networks to detect and characterize
  Chemical, Biological, Radiological, Nuclear, and
  Explosive (CBRNE) attacks and material.  
• Sensor networks to detect and monitor
  environmental changes.  
• Wireless traffic sensor networks.  
• Wireless surveillance sensor networks.  
• Wireless parking lot sensor networks.

4
INTRODUCTION(CONT..)

• Wireless ad hoc sensor network requirements
  include the
• following
• Large number of (mostly stationary) sensors.  
• Low energy use. 
• Network self-organization. 
• Collaborative signal processing. 
• Querying ability.

5
GOALS

• The basic goals of a wireless ad hoc sensor
  network
• generally depend upon the application, but the
  following
• tasks are common to many networks
• Determine the value of some parameter at a given
  location.  
• Detect the occurrence of events of interest and
  estimate parameters of the detected event or
  events . 
• Classify a detected object . 
• Track an object.

6
SENSOR

• We focus our attention on how to back-up
  one type of sensor with another.
• There are two main reasons for this decision.
  The first is that technology trends
  indicate that sensing has by far the highest
  fault rates.
• The second is that there is a wide consensus
  that multi-modal sensor fusion is the key
  for successful and widespread use of
  embedded sensor networks.
• Two fault models
• -sensors that produce binary outputs.
• -sensors with multilevel digital
  outputs.

7
SENSOR NODE
8
MULTIMODAL SENSOR FUSION

• Problem of multi modal sensor fusion for digital
  binary sensors.
• Sensor resource allocation and assignment
  problems ask to identify the minimal number of
  sensors.

9
MULTI MODAL SENSOR NETWORK
10
FAULT TOLERANT SOLUTION
11
TRADITIONAL VS HETEROGENEOUS

• Traditional methods double redundancy.
• - triple
  redundancy.
• Importance of heterogeneous over traditional.
• -low energy.
• -cost.

12
EXPERIMENTAL RESULTS
13
ALGORITHMIC ENGINES TO SRA PROBLEM

• SRA problem.
• Two different techniques to solve the allocation
  problem
• ILP-based and
• Simulated annealing based.
• -move.
• -objective function.
• -cooling schedule.
• -stop ping criteria.

14
FAULT TOLERANCE IN
MULTIMODAL SENSORY SYSTEMS

• A number of sensors, some of them different with
  modalities.
• The goal is to extract as accurately as possible
  the information requested by a user from noisy
  measurements.
• Four types of sensors in following example for
  introduced approach.

15
(No Transcript)
16
MULTI MODAL SENSOR SYSTEMS
17
FAULT TOLERANCE IN MULTIMODAL SENSOR SYSTEMS
18
MULTIMODAL SENSOR SYSTEMS
19
CONCLUSION

• New approach to multimodal sensor fusion.
• Key idea is to use one type of sensor to back up
  sensors of different types by exploiting
  flexibility during multimodal sensor data fusion.
• Formulated the problem for 2 diff. type of
  sensors binary and multi-level sensors.
• Developed algorithms to solve them .

20
REFERENCES

• Fault tolerance techniques for wireless ad hoc
  sensor networks by
• Farinaz
  Kounshafar
• Miodrag
  Potkonjak
• Alberto
  Vincentilli.
• http//w3.antd.nist.gov/wahn_ssn.shtml.