A Low Complexity Channel Estimation Algorithm for Massive MIMO System

Abstract

Massive MIMO can boost the capacity and increase the coverage, however the pilot overhead of its channel estimation will linearly increase with the antenna increasing. In order to reduce pilot overhead and improve the quality of channel estimation, a multi-dimension Wiener filtering channel estimation algorithm was put forward. First we analyzed Massive MIMO system model and channel model in which Base station configured a two-dimension antenna array; then accomplished channel estimation applying the correlation function of time, frequency, the row and column of the antenna array, and presented a simple closedform solution of the space-time-frequency correlation function to reduce the complexity of the algorithm. The proposed algorithm extends one-dimension space filtering to two-dimension, so doesn’t need to transmit pilots in every raw of the two-dimension antenna array contrast with 3D-Wiener and can reduce beyond 50% pilot overhead. Based on the simulation results, this algorithm can achieve 8dB performance gain over 2D-Winner and 3dB over 3D-Wiener with same pilot overhead and can acquires the optimal throughput with considering both the channel estimation quality and pilot overhead.