Blind Source Separation with a Time-Varying Mixing Matrix

Abstract

Abstract-Blind source separation (BSS) is the process of using multiple sensors to separate multiple random signals from a received linear combination. In this paper, we apply blind source separation techniques to mixtures of digital communications signals. We predict the achievable symbol error rate when the signals are received on the same carrier frequency and blindly separated. In a wireless environment where the sources are mobile or the environment is changing, the mixing matrix will vary with time. Our primary contribution is that we identify the major source of error in separation. We extend an adaptive step size algorithm to the complex-valued case to mitigate the errors in a dynamic environment.

A Khatri–Rao subspace approach to blind identification of mixtures of quasi-stationary sources

“…PAPA remains computationally competitive, except for K ¼23 where BGWEDGE yields the fastest runtime. overly ill-conditioned mixing matrices [12,17,35], we constrain the columns of A to satisfy ja H i a j j o 0:8; ∀i≠j. The algorithms under comparison are NAP-initialized subspace ACDC (Algorithm 3), SOBIUM [8], TALS [6] and UDSEP [12].…”

Variable learning rate EASI-based adaptive blind source separation in situation of nonstationary source and linear time-varying systems

“…Chambers et al presented a method for dealing with abrupt changes in the mixing matrix [23], and an adaptive algorithm that can separate noisy time varying mixtures has been reported by Enescu et al [24]. DeYoung et al [25] applied BSS techniques to mixtures of digital communication signals in which the sources are mobile or the environment is changing, and the mixing matrix will vary with time. Their results indicate that the main difficulty in the separation phase is the ill-conditioned nature of the channel matrix.…”