一种新的球形解码初始半径确定方法

New Method for Initial Radius Selection of Sphere Decoding

球形解码(Sphere Decoding,SD)是多输入多输出系统(MIMO)系统中一种非常高效的解码算法.它的初始搜索半径对算法的计算复杂度有很大的影响.提出一种名为IR-ZF-OSUC的初始搜索半径确定方法.它利用了SD内嵌的QR分解结果来得到初始半径.相比其他类似方法,因半径更为接近最优值,且无需专门的处理步骤,使SD具有更低的计算复杂度.仿真表明,使用IR-ZF-OSUC来确定初始半径的SD相比基于其他策略的SD,在很宽的SNR范围内具有最低的计算复杂度.

Sphere Decoding is a very promising decoding strategy for MIMO system. The initial radius is one of important parameters that have significant impacts on the computational complexity. In this paper, a new method for initial radius selection called IR-ZF-OSUC of Sphere Decoding for MIMO system is proposed. It utilizes the result of QR decomposition which is inherent in Sphere Decoding to obtain the initial radius. Unlike other approaches, since the obtained radius is nearer to the optimal one, and no extra processing is required, lower computational complexity is achieved. The simulation results show that the computational complexity of Sphere Decoding with IR-ZF-OSUC is lower than the Ones using other strategies over a wide range of SNRs.