A Low Complexity Channel Estimator for LTE Uplink in Non-sample-spaced Multipath Channels

Abstract

The discrete Fourier transform (DFT) based channel estimation has been widely used in multiple wireless orthogonal frequency division multiplexing (OFDM) system owing to the good performance-complexity trade-off. However, in practical multipath channels with non-sample-spaced path delay, this method suffers significant performance loss caused by energy leakage. In this paper, we use a mirror extension DFT (ME-DFT) algorithm by extending channel frequency response (CFR) on border subcarriers of the least squares (LS) channel estimation to mitigate the leakage effect. The length of the extended CFR fulfills the power of two to facilitate efficient fast Fourier transform (FFT) implementation. Simulation results shows that the ME-DFT based method exhibits superior energy concentration property and implementation efficiency as compared to the traditional DFT-based method.