S. Sawamura et al., MAXIMUM-LIKELIHOOD-ESTIMATION OF MULTIPLE SOURCES USING A LOCAL OPTIMIZATION METHOD, Electronics and communications in Japan. Part 3, Fundamental electronic science, 79(11), 1996, pp. 14-24
An effective method for solving a nonlinear optimizing problem is prop
osed here. This method is based on the maximum likelihood estimation o
f multiple sources by passive sensor array. The method employs the loc
al optimization using the gradient of nonlinear function. Also describ
ed is how to determine the effective initial values leading to the glo
bal minimum but not to the local minimum. The proposed method in this
paper easily obtained the optimized estimate of the maximum likelihood
in multiple sources. This estimation usually requires great processin
g time using conventional methods because of a heavy computing load. S
ome simulations show that the proposed method has the properties of bo
th better accuracy and lower computing time than a conventional method
.