SENSOR LOCATION CALIBRATION USING SIGNAL SUBSPACE METHODS FOR DIRECTION FINDING TECHNIQUE

The sensor location uncertainty of an array degrades severely its performance of the target-source's distance detection and direction finding. In the case of far field sources, for the direction finding technique based on decompositing the eigenvalue of a matrix, a new method is presented. The method calibrates the sensor location uncertainty utilizing three radiation sources whose directions are not known accurately. In this paper a new iterative algorithm using the extracted signal subspace is developed, and the iterative procedure is analysed in detail to show why this algorithm converges to the global optimal point. It is the very key point that a technique for selecting directions of calibrating sources is given. Simulation results illustrate that the new method presented here is successful and practicable.