Data Aided Iterative Channel Estimation in OFDM Systems Using a Controlled Superimposition of Training Sequences

摘要

This work pertains to the use of superimposed training for channel estimation in orthogonal frequency division multiplexing (OFDM) based systems. An iterative time domain Least Squares based channel estimator is proposed. The estimator is generalized to provide scope for exploiting the coherence time and the coherence bandwidth of the channel. By exploiting the periodicity of the training sequences in the time domain and inserting zeros instead of data at some of the training sequence subcarrier locations depending on the desired estimation accuracy, a controlled superimposition technique is proposed. This method includes the flexibility to trade off between bandwidth efficiency and performance without any change in the structure of the channel estimator. The mean squared estimation error (MSEE) performance of such a system is mathematically analyzed and a training sequence selection criterion optimizing the same is proposed. The simulation performance of the scheme is presented in terms of the MSEE and also its impact on the bit error rate is shown. Such a scheme is attractive in high data rate scenarios in closed loop OFDM systems.