CAJ | 학술논문

针对基站配置上百个天线的多用户MISO系统上行信号检测问题, 结合基于变化的最大似然（ML）代价函数判决门限的随机重启策略, 改进主动禁忌搜索（RTS）检测算法性能。仿真实验表明, 相比基本的RTS算法以及似然上升搜索算法（LAS）及其变体, 在相同条件下该算法误符号率性能更优, 尤其是在高阶QAM调制和信道增益矩阵为欠定阵（用户数大于基站天线数）时, 其他算法存在严重的性能恶化, 而该算法仍能呈现良好的性能。
침대기참배치상백개천선적다용호MISO계통상행신호검측문제, 결합기우변화적최대사연（ML）대개함수판결문한적수궤중계책략, 개진주동금기수색（RTS）검측산법성능。방진실험표명, 상비기본적RTS산법이급사연상승수색산법（LAS）급기변체, 재상동조건하해산법오부호솔성능경우, 우기시재고계QAM조제화신도증익구진위흠정진（용호수대우기참천선수）시, 기타산법존재엄중적성능악화, 이해산법잉능정현량호적성능。
In order to solve the uplink MISO signal detection problem when the base station was equipped with hundreds of antennas called multi-user massive MISO system, this paper introduced a random restart strategy based on varying maximum likelihood (ML) cost function threshold decision to improve the reactive tabu search (RTS) algorithm. The proposed enhanced algorithm was shown to achieve better symbol error rate performance compared with the basic RTS algorithm and likelihood ascent search algorithm (LAS) and its variants under the same conditions. When the system employs higher-order QAM modulation, other algorithms encountered serious performance deterioration, especially the channel gain matrix is underdetermined as the number of users being greater than the number of base station antenna, while the proposed algorithm still could achieve a fairly good performance.