A new criterion for the determination of the number of signals in high-resolution array processing

A problem central to most high-resolution methods of array processing is the determination of the number of signals K from a finite set of observations. Two common criteria which do not make use of a subjective threshold are the AIC (Akaida information criterion) and MDL (minimum distance length), both of which consist of a likelihood-function term and a penalty term. The idea is to find a number of K such that either of the criteria is minimized. Examination of both criteria reveals that the likelihood function encompasses irrelevant parameters resulting in the relatively inaccurate estimation of the relevant parameters. A criterion is proposed such that a new likelihood function is derived, consisting of only the parameters having some bearing on the determination of K. The improvement of the accuracy in the estimation of the relevant parameters is evaluated. Computer simulations indicate that considerable improvement in performance is attained