Doppler Shift Estimation for Millimeter-Wave Communication Systems on High-Speed Railways

Abstract

The millimeter-wave (mmWave) technology is one possible solution to addressing the explosive requirements for mobile data access on high-speed railways (HSRs). However, utilizing the mmWave technology on HSRs will result in large Doppler shifts, which should be estimated and compensated. Traditional Doppler shift estimators, such as the cyclic prefix (CP)-based estimator (CPBE), the two-training-symbol-based estimator (TBE), and the best linear unbiased estimator (BLUE), have inferior estimation accuracy especially at low signal-to-noise ratio (SNR). In this paper, we, therefore, propose three new Doppler shift estimators for mmWave communication systems on HSRs: a radio environment map (REM)-based estimator (RBE) derived from the \textit {maximum a posteriori} algorithm; an equally-divided structure-based estimator (ESBE) that divides the effective orthogonal frequency-division multiplexing symbol (OFDM) into multiple equal fragments; and an enhanced ESBE (EESBE) that takes the results of RBE as \textit {a priori} knowledge. Simulation results show that these three algorithms outperform existing Doppler shift estimators in mmWave communication systems on HSRs.