Abstract:

The sensor position errors have great negative influence on the directionfinding performance of the multiple signal classification (MUSIC) algorithm. Therefore, a cumulantsbased instrumental sensors method for selfcalibration of sensor position errors is presented. Firstly, a maximal set of nonredundant fourthorder cumulants matrix is constructed. Then, the joint estimation of direction of arrival (DOA) and the sensor position errors can be carried out based on the fourthorder cumulants matrix constructed above, therein the DOA estimations are obtained by onedimension search, and the sensor position errors can be offset by solving a linear least squares problem. The theoretical analysis on the consistency of the novel algorithms is fulfilled under certain conditions. The results of computer simulations indicate that the novel algorithm can reduce requirement for the number of the instrumental sensors and restrain the negative impact of the Gauss colored noise.