Adaptive filter based blind residual self-interference cancellation in full-duplex MIMO transceivers

Abstract

In this paper, we propose an adaptive filter based architecture for blind residual self-interference (SI) cancellation in a full-duplex (FD) multiple-input-multiple-output (MIMO) transceiver. Correspondingly, we present and compare the performances of three different SI cancellation algorithms employing adaptive filters based on the least mean squares (LMS), normalized LMS (NLMS) and recursive least squares (RLS) techniques. Simulation results obtained for a base-station in an FD MIMO multiple access/broadcast channel system show that the proposed RLS based SI cancellation algorithm offers the best performance in terms of convergence rate, bit-error-rate (BER) and signal-to-interference-plus-noise ratio (SINR). We also compare the usage of parallel adaptive filters at each receiver chain of an FD MIMO transceiver versus the usage of a single two-dimensional adaptive filter. It is observed that the usage of parallel adaptive filters offers relatively faster convergence, but at the cost of slightly inferior BER and SINR performance.