Improved DFT-based Channel Estimation Algorithm for OFDM Systems

DFT-based channel estimator for OFDM systems has lower computational complexity than MMSE channel estimators,and better performance than LS channel estimators.However,the traditional DFT-based channel estimator only removes the noise in the channel impulse response(CIR) beyond the length of the cyclic prefix(cp),while the noise within the length of the CP is not suppressed at all.Therefore,the performance of the DFT-based channel estimator could be improved further.In this paper,we propose an improved DFT-based channel estimator.In the proposed estimator,we estimate the noise power variance at first,in terms of which we could obtain a threshold.Then we apply this threshold to the CIR within the length of the CP to remove noise further.Simulation results show that our improved channel estimator performs better than the original one.