基于格基约减辅助的低复杂度MIMO信号检测算法

摘要

In Multiple Input Multiple Output( MIMO) systems,it is difficult to use conventional lattice basis reduction for signal detection in reality due to its high complexity. In order to solve this problem,a low-complexity algorithm based on Brun’ s algorithm is proposed. First,the algorithm uses Singular Value De-composition( SVD) to obtain the integer relationship between channel matrix’s singular vector and conver-ted matrix. Then,it uses Brun’s algorithm to reduce the dual basis of channel matrix. Finally,the algo-rithm uses the new dual basis to perform linear detection. Simulation results indicate that the algorithm’s complexity is one tenth of Lenstra Lenstra Lovasz(LLL) aided signal detection algorithm,and it can a-chieve the performance gains over the conventional linear detection,especially for high signal-to-noise ra-tio( SNR) range. So,the algorithm can get a better tradeoff between detection performance and computa-tional complexity.%在多输入多输出( MIMO)系统中，常规的格基约减辅助信号检测算法由于复杂度高而难以在实际工程中应用。为了解决这一问题，基于Brun算法提出了一种低复杂度的信号检测算法。该算法首先通过奇异值分解( SVD)得到信道矩阵奇异向量和转换矩阵之间的近似整数关系，进而采用Brun算法对信道矩阵的对偶格基进行约减优化，最后将约减后的新对偶格基用于传统线性信号检测。仿真结果表明：该方法的复杂度约为基于常规Lenstra Lenstra Lovasz( LLL)格基约减辅助的MI-MO信号检测算法的0.1倍；同时，与线性检测算法相比，检测性能提升非常明显，特别在较高信噪比( SNR)范围内。因此，该算法能够在检测性能与计算复杂度之间取得较好的折衷。