The Discrete Probability Density Method for Emitter Geolocation

Abstract

The discrete probability density (DPD) method is a novel technique for fusing sensor data by numerically combining the probability density function (PDF) representing the error estimate of each measured value. This paper describes the sampling of multiple PDFs at common discrete intervals to calculate a joint discrete probability density distribution. This concept is extended into multi-dimensional space to combine sensor measurements taken from different locations by projecting each PDF onto an array of sample points. The joint product is calculated at each point to produce a discrete probability density distribution representing the fused result. The DPD method is applied to emitter geolocation using line-of-bearing (LOB) measurements and compared to Stansfield's method and the Cramer-Rao lower bound using Monte-Carlo simulation. It is found that the DPD method performs better than Stansfield's method in the presence of large bias errors, especially for large number of LOBs