Download Collaborative Signal Processing in Wireless Sensor Systems: Detection & Tracking and more Study notes Computer Science in PDF only on Docsity!
CS 603 - Wireless Sensor Systems, Western Michigan University
Collaborative Signal Processing
Zille Huma Kamal
CS 603 - Wireless Sensor Systems, Western Michigan University
Detection, Classification and
Tracking
CSP in a detection,
classification and
tracking
application.
CS 603 - Wireless Sensor Systems, Western Michigan University
Need for CSP
Collaboration increases accuracy of results
Does not exhaust any single mote in the
environment
Towards Fault Tolerance
Energy conserved, accuracy achieved and
network life increased
CS 603 - Wireless Sensor Systems, Western Michigan University
Focus on CSP
Training Phase
Testing Phase
Classification and Fusion Phase
Data Fusion
Decision Fusion
CS 603 - Wireless Sensor Systems, Western Michigan University
Flow Chart of CSP process
CS 603 - Wireless Sensor Systems, Western Michigan University
Notations used
CS 603 - Wireless Sensor Systems, Western Michigan University
Classification
Bayes Rule can be transformed into a maximum a posterior rule,
formulated as:
The resulting MAP matrix can be reduced by normalization,
formulated as:
- where, a,x, y represents the x,y element of the d x d MAP , matrix.
CS 603 - Wireless Sensor Systems, Western Michigan University
References
[DS] A. D’Costa, A.M Sayeed, “ Collaborative
Signal Processign for Distributed
Classification in Sensor Networks, ”
Electrical and Computer Engineering,
University of Wisconsin-Madison.
[LWHS] D. Li, K.D Wong, Y.H Hu, A.M
Sayeed, “ Detection, Classification and
Tracking of Targets in Distributed
Sensor Networks, ” Electrical and
Computer Engineering, University of
Wisconsin-Madison.
