Time Domain Channel Estimation for OQAM-OFDM Systems: Algorithms and Performance Bounds

In this paper, we first present a general time domain model for the channel estimation in the orthogonal frequency division multiplexing system with offset quadrature amplitude modulation (OQAM-OFDM), and utilize the frequency domain pilots to estimate the time domain channel impulse responses. Different form the conventional methods, there is no specific requirement for the length of the symbol interval compared to the the maximum channel delay spread in the proposed scheme. Furthermore, with the proposed time domain model, the channel statistic information could be utilized to improve the performance of the channel estimation. Then, we propose two channel estimation schemes, i.e., linear minimum mean square error (LMMSE) and weighted least square (WLS), and we also derive their corresponding Bayesian Cramér-Rao Bound (BCRB) and Cramér-Rao Bound (CRB) bounds, respectively. Simulation results demonstrate that the BCRB and CRB bounds could be achieved by the proposed LMMSE and WLS methods, respectively. Moreover, simulation results show that the proposed methods are much robust to the time synchronization error compared to the conventional frequency domain methods, and imply that the pulse shaping filter with waveforms concentrated in the time domain could be employed in OQAM-OFDM systems to improve the channel estimation performance and spectral efficiency.