An improved permutation alignment algorithm for convolutive mixture of radar signals

The convolutive blind source separation (BSS) problem can be solved in frequency domain. To solve the permutation ambiguity problem in frequency domain, this paper presents an improved permutation alignment algorithm. According to features of radar signals, first the frequency domain is divided to some region segmentations. Then the permutation alignment is performed in each region independently. Finally, we fuse all of the regions and estimate the frequency point of the signal with incorrect separation. This algorithm overcomes the problem that the incorrect permutation of a frequency point will influence subsequent frequency points. The simulation results demonstrate the effectiveness and validity of the proposed method.