DETECTION OF THE NUMBER OF SIGNALS IN THE PRESENCE OF COLORED NOISE IN DECENTRALISED PROCESSING

摘要

The problem of estimating the number of signals under a colored noise model is solved when the data are received from many different places that are widely apart by using modified information criterion proposed by the authors. This criterion involves the eigenvalues of the sample canonical correlation matrix obtained from data consisting of both noise and signals as well as from the data consisting of noise only. Some simulation results are also presented.