A high resolution DF algorithm which is suitable for a uniform circular array (UCA) is presented. Here the method is constructed by transforming the UCA from element space to mode space and by reformulating the cost function of ML method to a simple form by means of introducing a new concept of noise manifold matrix which plays a very important role in the practical implementation. Compared with ML method this new algorithm is only a close-form solution rather than a multi-variable optimization one which often becomes useless due to heavy computational burden and local minimum traps. Simulation results strongly verify the high performance of this method.
展开▼
