Pilot Contamination Reduction by Shifted Frame Structure in Massive MIMO TDD Wireless System

摘要

Pilot contamination limits the performance of massive multiple-input multiple-output(MIMO) system severely due to the degraded channel estimation. An efficient way to solve this problem in time division duplex(TDD) wireless system is shifting the location of pilots in time frames used in neighboring cells. However, the shifted frame structure has only been studied in MIMO system with the ideal independent and identically distributed(i.i.d.) channel coefficients. In this paper, the shifted frame structure is studied in a measured channel with a large number of antennas for a certain class of channel fading statistics. To deal with the high inter-cell interference caused by the shifted frame structure in such a measured channel condition, we propose a scenario with a covariance-aided channel estimator. This scenario shows that pilot contamination can be solved and the high inter-cell interference can be made to vanish asymptotically with the number of antennas. The key of the interference rejection is obtaining the precise condition on the second order statistics for the desired and interference signals. The most significant information of the second order statistics comes from two parts, one is the channel state information(CSI) among the base stations(BSs), the other comes from the estimated information exchanged by the BSs, which depicts the channel between the BSs and the users. The simulations give powerful results of the interference rejection and the achievable rate promotion.