A New Search Algorithm for Allocating Pilots in Asymptotic TDD Massive MIMO Systems

摘要

This paper proposes a novel heuristic algorithm that mitigates pilot contamination in Massive MIMO (multiple-input, multiple-output) systems by optimizing the allocation of pilots, which are a shared resource among base stations (BSs). The goal is to minimize the effect of pilot contamination caused by interference from reusing the same pilot sequence from multiple user equipment (UE) in different cells. Specifically, we aim to maximize the minimum uplink signal to interference ratio (SIR) through the network by assigning the same pilot to the UEs in different cells that cause the least amount of interference. For this purpose, the algorithm starts with two cells that have the worst arithmetic mean of the potential SIR between their users. Then the algorithm continues the assignment procedure for the rest of the cells depending on the arithmetic mean of the worst potential SIR of the users of a cell and the users of the already assigned cells. The simulation results showed that our results are very close to the optimal solution where the exhaustive search is feasible, and outperforms the results obtained from recently published related work which served as a benchmark when the exhaustive search could not be applied to the search space size.