一种用于电力线通信系统的改进MIMO检测算法

目前多输入多输出(multiple-input multiple-output，MIMO)技术已经被电力线通信(power line communication，PLC)系统采用，但由于MIMO PLC系统噪声呈非高斯分布而且各端口噪声之间存在相关性，故不能直接采用无线系统中的MIMO检测算法。采用了二元Middleton class A分布对MIMO PLC系统中噪声进行建模，提出了基于该噪声分布的最大似然检测改进算法，由于改进最大似然检测算法实现复杂度高，为了便于实现，进一步提出了用近似函数降低复杂度的2种次优的检测算法，优化了算法复杂度。仿真结果表明，与传统的基于高斯噪声分布的最大似然检测算法相比，提出的基于二元Middleton class A类噪声分布的信号检测算法在MIMO PLC系统能获得更好的性能。在性能损失较小的情况下，次优算法的复杂度明显低于最大似然检测改进算法。

An improved MIMO detection algorithm for power line communication system

Currently, the multi-input multi-output (MIMO) technology has been adopted by power line communication system. However, due to the non-Gaussian distribution of MIMO PLC system noise and the correlation between the noise of each port, it can not directly use the MIMO detecting algorithm in wireless system. The binary Middleton class A distribution is employed to model the noise in MIMO PLC system, and then an optimal maximum likelihood detection algorithm for the noise distribution is proposed. Because of the high complexity of the improved ML detection algorithms, in order to facilitate the realization, two suboptimal detection algorithms to reduce complexity with approximate functions are proposed to optimize the algorithm complexity. The simulation results show that compared with the traditional ML detection algorithm based on the Gaussian noise, the signal detection algorithm based on binary Middleton class A noise distribution proposed in this paper can obtain better performance in MIMO PLC system. In the case of less performance loss, the complexity of the sub-optimal algorithm is obviously lower than that of the maximum likelihood detection algorithm.