Bayesian joint detections for 60GHz millimeter-wave communications with the power amplifier nonlinearity

For the emerging 60GHz millimeter-wave communications, the nonlinearity is usually inevitable due to RF power amplifiers operating in the ultra-high frequency and enormous bandwidth, which, in collusion with frequency-selective propagations, poses great challenges to signal detections. In contrast to classical schemes calibrating nonlinear distortions in transmitters, a blind detection algorithm is presented in this investigation, with which both the multipath response and symbols contaminated by nonlinear distortions and multipath interferences are estimated in receiver-ends. The Monte-Carlo sequential importance sampling based particle filtering is used, and the non-analytical distribution is approximated numerically by a group of random measures with evolving weights. By applying the Taylor´s series expansion techniques, a local linearization model is further constructed to facilitate the practical design of a sequential detector. Simulation results validate the proposed blind detection scheme. By excluding the transmitting predistorter with complex computations and implementations, the presented algorithm provides a promising signal detection framework in 60GHz systems.