Optimal superimposed training design for MIMO/OFDM systems over rapidly time-varying multipath channels

In this paper, we address the problems of super-imposed training (ST)-aided channel estimation for OFDM systems over rapidly time-varying (RTV) multi-path channels. Based on the piece-wise linear assumption, the superimposed training sequences are optimized with regard to minimizing the channel mean square error. Meanwhile, we derive a closed-form expression of channel estimation analysis. Simulation results show that the proposed design of training sequence outperforms that of the existing ST for RTV channel estimation.