Minimization of sensor usage for target tracking in a network of irregularly spaced sensors

Authors:
  • TST.J. Stevens
    T.J. Stevens
  • DMD. Morrell
    D. Morrell

Abstract
We address the following scenario: a single target moves through a eld of stationary sensors with known locations. At each time epoch, each sensor is either active or not; each active sensor outputs either target detected or not detected. The probability of target detection is a decreasing function of the distance from a sensor to the target. A particle lter is used to track the target through the sensor eld using all active sensor outputs. A heuristic conguration strategy is used to determine which sensors should be activated; Monte Carlo simulations show that the conguration strategy leads to a signicant reduction in required active sensors with lit- tle degradation in the tracker performance.