Effective Low-Complexity Optimization Methods for Joint Phase Noise and Channel Estimation in OFDM

摘要

Phase noise correction is crucial to exploit full advantage of orthogonalfrequency division multiplexing (OFDM) in modern high-data-rate communications.OFDM channel estimation with simultaneous phase noise compensation hastherefore drawn much attention and stimulated continuing efforts. Existingmethods, however, either have not taken into account the fundamental propertiesof phase noise or are only able to provide estimates of limited applicabilityowing to considerable computational complexity. In this paper, we havereformulated the joint estimation problem in the time domain as opposed toexisting frequency-domain approaches, which enables us to develop much moreefficient algorithms using the majorization-minimization technique. Inaddition, we propose a method based on dimensionality reduction and theBayesian Information Criterion (BIC) that can adapt to various phase noiselevels and accomplish much lower mean squared error than the benchmarks withoutincurring much additional computational cost. Several numerical examples withphase noise generated by free-running oscillators or phase-locked loopsdemonstrate that our proposed algorithms outperform existing methods withrespect to both computational efficiency and mean squared error within a largerange of signal-to-noise ratios.