Low Rank Adaptive Filtering with the Fractional Fourier Transform for Severe, Non-stationary, Co-Channel Interference Suppression

This technical note compares performance of the correlations subtraction architecture of the multistage Wiener filter (CSA-MWF) in the Fractional Fourier Transform (FrFT) domain, MMSE-MWF-FrFT, to the minimum mean-square error, MMSE-FrFT, algorithm and to the conventional MMSE-FFT algorithm in severe cochannel interference. Using four equal power binary phase shift keying (BPSK) signals and a short N = 4 sample training sequence, we show that the MMSE-MWF-FrFT algorithm is able to provide bit error rates (BERs) from 10 −4 to 10 −6 at just 10−12 dB Eb/N0 where the other two algorithms fail.