Tabu-Search-Based Pilot Assignment for Cell-Free Massive MIMO Systems

Abstract

In this correspondence, an efficient pilot assignment scheme based on tabu search (TS) is proposed to mitigate the severe pilot contamination effect caused by pilot reuse in cell-free massive multiple-input multiple-output systems. By exploiting the local neighborhood search, we first define the neighborhood and search objective function. Then, the tabu list is introduced to record the prohibited solutions and to avoid being trapped in the local optimum. In addition, we consider a fair policy for the pilot reused times to further improve the system performance. We propose a TS-based pilot assignment iterative algorithm to search the suboptimal pilot assignment result with low complexity. Numerical results verify that the proposed TS-based pilot assignment scheme can significantly improve the 95%-likely per-user throughput with similar complexity compared to the conventional random and greedy schemes.