Joint Phase Noise Estimation and Data Detection in Coded MIMO Systems

摘要

In this paper, the problem of joint oscillator phase noise (PHN) estimationand data detection for multi-input multi-output (MIMO) systems usingbit-interleaved coded modulation (BICM) is analyzed. A new MIMO receiver thatiterates between the estimator and the detector, based on theexpectation-maximization (EM) framework, is proposed. It is shown that at highsignal-to-noise ratios, a maximum a posteriori estimator (MAP) can be used tocarry out the maximization step of the EM algorithm. Moreover, to reduce thecomputational complexity of the proposed EM algorithm, a soft decision-directedextended Kalman filter-smoother (EKFS) is applied instead of the MAP estimatorto track the PHN parameters. Numerical results show that by combining theproposed EKFS based approach with an iterative detector that employs lowdensity parity check (LDPC) codes, PHN can be accurately tracked. Simulationsalso demonstrate that compared to existing algorithms, the proposed iterativereceiver can significantly enhance the performance of MIMO systems in thepresence of PHN.