An Improved Channel Estimation Based on DFT-MMSE

Traditional DFT-LS algorithm sets channel impulse response as zero beyond cyclic prefix on time domain, which can not only attain better performance than LS, but also reduce the computational complexity. Compared with DFT-LS, MMSE has more complexity in computing, but can attain better performance. This paper proposes a new estimation method based on traditional DFT-MMSE, which sets threshold to effectively filter the noise within the cyclic prefix channel impulse response. Both theoretical analysis and simulation results show that the proposed method has better performance than the traditional DFT- MSSE estimation algorithm. Moreover, its computing complexity is lower than traditional DFT- MSSE and ideal MMSE channel estimation.