Localization of a stationary source within a uniform circular sensor array

The problem of source localization and angle of arrival estimation using linear and planar antenna arrays has been addressed broadly. Techniques such as MUSIC and ESPRIT have been developed to solve these types of problems. In this paper we develop a technique to locate a stationary source within a uniformly spaced circular sensor array. With this arrangement, source signals emitting far from one sensor can still be effectively detected by an opposing sensor in the array. We first develop a mathematical model for this type of source-sensor arrangement followed by simulation using the Polak-Ribiere Conjugate Gradient optimization technique. It is observed that for well-correlated sensors, the Polak-Ribiere Conjugate Gradient approach with strict Wolfe-Powell conditions for convergence can accurately solve the source localization problem.