Optimal Passive Source Localization

Abstract

In order to optimize the estimation of an object's position, this paper proposes a procedure for placing acoustical sensors in 3D space, using passive source localization. A standard performance measure in estimation theory is the Cramer-Rao Lower Bound (CRLB), which describes the lower bound of the variance of unbiased estimators. In the case of passive source localization, this bound depends on the sensor and source positions, as well as on the propagation speed and the assumptions made about the disturbance noise. A procedure for an optimal sensor placement, using this CRLB as the objective function, is described. In order to assure optimal coverage of a surveillance area, the average CRLB of multiple source positions (in this area) is minimized.