球形解码( Sphere Decoder，SD) 算法能以较低的复杂度实现多输入多输出( Multiple Input Multiple Output，MIMO) 系统的最优检测，是当前受到普遍关注的 MIMO 检测算法。对当前球形解码 的 主 要 研究 成 果 进行 综 述，根 据 搜索策略进行分类，重点分析基于深度优先策略的 VB、CL 和基于宽度优先策略的 K-Best、FSD 算法，并且讨论了几种初始半径的选择方法，最后在准静态平坦瑞利衰落环境下对上述算法进行了性能仿真比较-Sphere decoding (Sphere Decoder, SD) algorithm can achieve lower complexity MIMO (Multiple Input Multiple Output, MIMO) system, optimal detection, are currently subject to widespread concern MIMO detection algorithm. The main findings of the current sphere decoding were reviewed, classified according to the search strategy, the depth of focus of the analysis based on the priority strategies VB, CL and K-Best, FSD algorithm based on breadth-first strategy, and discuss the selection of the initial radius of several methods Finally, in the quasi-static flat Rayleigh fading environment on the performance of these algorithms are simulated and compared