Complex Baseband Myriad Filtering and Maximum Likelihood MSK Demodulation under Symmetric alpha-stable Noise

Symmetric cr-stable (ScxS) distribution noise is widely used to model co-channel and network interference in wireless communication systems. As robust and adaptive techniques, myriad filtering (MyF) and spherically symmetric vector MyF have been applied to suppress univariate and spherically symmetric multivariate SaS distribution noise, respectively. At a communication receiver, the received band-pass noisy signals are usually down-converted to complex baseband, and the resulted complex baseband SaS noise has been demonstrated not to be circularly symmetric. In this paper, we proposed a complex baseband MyF (CBMyF) to suppress the non-circularly symmetric complex baseband SaS noise. Besides, there are few researches with respect to the demodulation of memory modulation signals, e.g., minimum shift keying (MSK) signal, under SaS noise. Thus, based on CBMyF, we proposed coherent and non-coherent MSK demodulation algorithms under SaS noise in this paper. Furthermore, maximum likelihood (ML) MSK demodulation under SaS noise also been proposed. Simulation results show that the bit error rate (BER) performance of the proposed CBMyF based MSK demodulation can closely approach that of ML demodulation. Meanwhile, the proposed CBMyF is compared with the common used clipper, and the results validate its advantage of robustness and adaptivity.