An efficient alternating optimization algorithm for detecting the number of coherent signals

Summary form only given. Wax and Ziskind have proposed an approach for detecting the number of coherent signals based on the MDL (minimum description length) principle. Since this approach includes a multivariate nonlinear optimization problem to estimate bearing parameters, huge computational costs are required to solve the problem. This paper propose an efficient alternative optimization algorithm for this problem. We can successfully separate a component depending on one variable parameter from the MDL criterion. It makes the algorithm efficient. Computer simulations are shown to demonstrate the validity of the proposed algorithm.