4/10/2002

Name: Qishi Wu

Committee:
Dr. Iyengar (Chair)
Dr. J Bush Jones
Dr. Aiichiro Nakano

Date: 2:00 PM, Thursday, April 18, 2002
Location: 297 Coates Hall

Title: On Optimal Deployment of Probabilistic Detectors in the Plane

Abstract:
One of the essential tasks in the design of distributed sensor systems is the deployment of sensors for an optimal surveillance of a target region while ensuring robustness and reliability. The sensors with probabilistic detection capabilities with different costs are considered here. A sensor deployment problem for a planar grid region is formulated as a combinatorial optimization problem to maximize the overall detection probability within a given deployment cost. This sensor placement problem is shown to be NP-complete and an approximate solution is proposed based on the genetic algorithm method. The solution is obtained by the specific choices of genetic encoding, fitness function, selection mechanism, and also genetic operators such as crossover, mutation, translocation etc. for this problem. Simulation results are presented to show the benefits of this method as well as its comparative performance with a greedy sensor placement method. Both systems based on genetic algorithm and uniform placement have been implemented in C++.

All are welcome.