Abstract: Massive MIMO technology can effectively improve the spectrum efficiency and reduce the transmission power by increasing the number of antennas at the base station, making it a key technology in 5G mobile communication system. However, with the increase of the number of antennas, the complexity of the uplink signal detection increases significantly, the original detection algorithm cannot be achieved in reality. The Reactive Tabu Search（RTS） algorithm based on machine learning and artificial intelligence has become a hotspot in the industry with the advantage of low complexity. In this paper, RTS signal detection algorithm based on BC-GS （Block Constellations- Gauss Seidel） iterative method is proposed to reduce the computational complexity of initial solution to achieve the original BER performance.

Key words: large-scale MIMO, 5G, Reactive Tabu Search（RTS） algorithm, iterative method, signal detection

摘要： 大规模多输入多输出（MIMO）技术通过基站端天线数量的增加有效提高频谱效率，降低传输功率，使其成为5G移动通信系统的一项关键技术。可是随着天线数量的增加，上行链路信号检测算法的复杂度大幅增加，原有检测算法无法实现。基于机器学习和人工智能的主动禁忌搜索算法（RTS）凭借着复杂度低的优势脱颖而出，成为业内的研究热点。针对RTS算法初始值计算复杂度过高这一问题，提出基于BC-GS（Block Constellations-Gauss Seidel）迭代算法求解初始值的RTS信号检测算法，使其在达到原有算法误码率性能的前提下，从而进一步降低算法复杂度。

关键词: 大规模MIMO, 5G, 主动禁忌搜索算法, 迭代算法, 信号检测