Multichannel adaptive blind deconvolution using the complex cepstrum of higher order cross-spectra
Dana H. Brooks, Chrysostomos L. Nikias · IEEE Transactions on Signal Processing · 1993
An adaptive cross-spectral blind deconvolution method, cross-TEA, for the case in which data are available from several sensors, is described. Unlike ordinary cross-spectra, higher order cross-spectra are separable, and this fact is used to identify and deconvolve each system. Cross-TEA was tested on four-channel QAM (quadrature amplitude modulation) signals, and the results are compared to those for TEA, an autospectral method.>