Distance Based Decision Fusion in a Distributed Wireless Sensor Network

1
Distance Based Decision Fusion in a Distributed
Wireless Sensor Network

• Marco Duarte and Yu-Hen Hu
• Department of Electrical and Computer Engineering

Information Processing in Sensor Networks Palo
Alto, CA April 22-23, 2003
2
Summary

• Sensor networks requires decision fusion
• Majority voting is the most popular decision
  fusion method. It assumes all votes are equally
  accurate.
• Not all sensor decisions are equally accurate.
  Those closer to the target or with higher SNR
  will have better results.

3
Summary (Contd)

• If source location can be estimated, such
  discrepancy can be exploited to improve the
  decision fusion accuracy.
• We formulate three different methods to combine
  sensor decisions based on their distance from the
  target and SNR, and we found encouraging results.

4
Sensor Network Signal Processing Tasks

• Target Detection (CFAR region fusion)
• Target Classification (ML region fusion )
• Target Localization (ML,EBL)
• Target Tracking (Kalman Filter)

D. Li, K.D. Wong, Y.H. Hu, A.M. Sayeed
Detection, Classification and Tracking of
Targets. IEEE Signal Processing Magazine Vol. 19
Issue 2 pp. 17-29
5
Target Classification at Node Level

• Classify acoustic spectral features into AAV, DW,
  or HMMWV on individual sensor nodes
• 50-dimensional spectral feature PSD 0-960 Hz
  from acoustic data (f 4960 Hz)
• One feature vector for every 0.75 sec. of data
• Maximum Likelihood Classifier

AAV DW
HMMWV
6
Not all sensors are equal

• Hypothesis Classification accuracy is a function
  of target-sensor distance and SNR
• Feature is perturbed by background noise
  decision margin
  shrinks
• Each nodes classification rate depends on SNR.
• SNR is also roughly inversely proportional
  vehicle-node distance due to acoustic energy
  attenuation
• Experiment Determine classification rate for
  different levels of distance, SNR

7
Distance vs. Classification Rate
8
SNR vs. Classification Rate
9
Classification Rate as a Function of SNR and
Distance
10
Weighted Decision Fusion

• Optimal (linear) decision fusion perform
  weighted voting of individual results (ei(x) for
  node i)
• Weights (wi for node i) are proportional to
  classification rates

Z. Chair and P. Varshney Optimal Data Fusion
in Multiple Sensor Detection Systems, IEEE
Trans. AES, Vol. 22 No. 1, Jan. 1986, pp. 98-101
11
Distance Based Decision Fusion (DBDF)

• Current system architecture allows detection and
  localization prior to classification, giving
  distance and SNR estimates.
• Accurate localization allows for estimation of
  probability of correct classification based on
  distance.
• Some events may be rejected by fusion algorithm
  as the distance or SNR figures fall outside the
  training data range. The majority voting can be
  used as a backup fusion method.
• Measurements classification rate and acceptance
  rates.

12
DBDF Approach 1 Maximum A Posteriori Decision
Fusion

• Weighting factor as function of distance and SNR,
  determined using CFAR and EBL information.
• We formulate a Maximum A Posteriori (MAP)
  Probability Gating Network, using Bayesian
  estimation
• Parameters SNR, Distance grouping size
• P(xd,s)P(d,s) estimated from experiment data.

13
DBDF Approach 2Distance Truncated Voting

• Simple majority voting performed among nodes
  close enough to the target. Decisions from other
  nodes are discarded.
• Parameter max/threshold distance
• Reduces effect of localization error
• No decision will be made when vehicle is outside
  the distance threshold

14
DBDF Approach 3Nearest Neighbor Fusion

• Node closest to target will also have highest
  SNR, and hence the highest probability of
  correctness
• Region will assign same label as that assigned by
  the node closest to the target
• Lowest computational and communicational burden
• Accuracy of node to target distance critical in
  decision making

15
Weighted Voting Schemes

• This amounts to assigning weights
• MAP Fusion
• Distance Truncated Voting
• Nearest Neighbor
• Baseline simple Majority voting (wi1 for all i)

16
Experiments

• All 4 methods are compared
• Data from SensIT SITEX02, November 2001
• Distance groups by 20m, SNR groups by 5dB
• Distance truncated voting for 50m
• Nearest Neighbor
• Majority Voting
• Due to shortcomings in localization, experiments
  are run with different error levels in location
  estimate s 0m, s 12.5m, s 25m, s 50m.

17
Experiment Results
18
Experiment ResultsClassification Rate
DTV
19
Experiment Results

• Closest node gives highest acceptance,
  classification rates for accurate localization
  estimates
• MAP Fusion has smaller dependence on localization
  error than other methods
• All DBDF methods outperform simple majority
  voting due to the use of distance and SNR
  information.

20
Further Work

• MAP Classifier allows for exclusion of those
  samples with low classification rates (i.e. only
  samples with wi gt 0.5 are allowed).
• This will allow for reduction of communication
  bandwidth used for classification fusion.
• This method can be applied to other signal
  processing tasks.
• Website http//www.ece.wisc.edu/sensit