Very recently, signed quadrature spatial modulation (sQSM) is developed as a competent technique that expands the spatial constellation diagram of QSM system by adding a bipolar dimension. Despite the enhanced spectral efficiency, the existing optimal maximum likelihood (ML) presents a serious computational challenge for large scale sQSM systems. Therefore, developing a reduced complexity detector for sQSM schemes is of significant importance to enable implementing and enjoying the inherent advantages of this promising system. Toward this end, a Tree Search (TS) optimal low complexity detector, called TSopt, for sQSM Multiple Input Multiple Output (MIMO) system is proposed and analyzed in this paper. The proposed detector expands the computationally complex ML detector for sQSM into a tree-structure representation. The idea of the suggested algorithm is to employ an efficient searching strategy that can expeditiously find the branch corresponding to the minimum error without tracing the entire nodes as in the ML case. It is reported that the proposed TSopt algorithm achieves the exact error performance as ML detector but with substantial reduction in computational complexity. Besides, complexity analysis in terms of the number of visited nodes of the TSopt algorithm is analyzed and a closed-form expression for the expected complexity at high SNR values is derived. Reported results disclose agreement between simulation and expected analytical complexity with substantial gains of around 60–80% in complexity reduction for different system parameters.

中文翻译：

---

有符号正交空间调制 MIMO 系统的最优低复杂度检测器


最近，有符号正交空间调制（sQSM）被开发为一种有效的技术，通过添加双极维度来扩展 QSM 系统的空间星座图。尽管频谱效率有所提高，但现有的最佳最大似然 (ML) 对大规模 sQSM 系统提出了严峻的计算挑战。因此，为 sQSM 方案开发一种降低复杂度的检测器对于实现和享受这一有前途的系统的固有优势具有重要意义。为此，本文提出并分析了用于 sQSM 多输入多输出（MIMO）系统的树搜索（TS）最优低复杂度检测器（称为 TSopt）。所提出的检测器将 sQSM 的计算复杂的 ML 检测器扩展为树结构表示。该算法的思想是采用一种有效的搜索策略，可以快速找到与最小误差相对应的分支，而无需像机器学习情况那样跟踪整个节点。据报道，所提出的 TSopt 算法实现了与 ML 检测器一样的精确误差性能，但计算复杂度大幅降低。此外，还分析了 TSopt 算法的访问节点数量方面的复杂性分析，并导出了高 SNR 值下预期复杂性的封闭式表达式。报告的结果揭示了仿真与预期分析复杂性之间的一致性，不同系统参数的复杂性降低了 60-80% 左右。