针对双选择性衰落信道的联合期望传播与高斯近似迭代均衡算法

Joint expectation propagation and Gaussian approximation iterative equalization algorithm for doubly selective fading channel

针对双选择性衰落信道,为提升信息传输的可靠性,设计了一种联合期望传播与高斯近似的迭代均衡算法。该算法借鉴高斯近似的思想,将接收信号中由多径传播效应导致的符号间串扰近似为高斯分布随机变量,进而计算数据比特的对数似然比信息,以较低的复杂度实现译码器与均衡器之间的信息传递。为了进一步提升高斯近似的准确性并改善迭代算法的收敛性,创新性地利用期望传播的处理准则对译码器输出的后验信息进行去相关处理。仿真结果显示,期望传播处理准则可以显著改善迭代算法的收敛性和均衡的准确性,与线性最小均方误差和NDDE等经典均衡算法相比,所提出的联合期望传播与高斯近似的迭代均衡算法可以获得1~1.5 dB的性能增益,有效提升了双选择性衰落信道下信息传输的可靠性。

In order to improve the reliability of information transmission over the doubly selective fading channel,a joint expectation propagation and Gaussian approximation(EP-GA)iterative equalization algorithm is proposed.Based on the principle of Gaussian approximation,the inter-symbol interference caused by the multi-path propagation is approximated as a Gaussian-distributed random variable.Then,the log-likelihood ratio information of the data bit is calculated.In this way,the information exchange between the decoder and the equalizer is accomplished in low complexity.To further improve the accuracy of the Gaussian approximation and convergence of the iterative algorithm,the processing criterion of expectation propagation is employed to perform decorrelation processing on the posterior information output by the decoder.The simulation results show that the processing criterion of expectation propagation could significantly accelerate the iteration convergence and improve the equalization accuracy.In comparison with the classical linear minimum mean square error and the NDDE equalization algorithms,the proposed EP-GA iterative equalization algorithm could achieve a performance gain of 1~1.5 dB.Therefore,the information transmission reliability can be improved over the doubly selective fading channel.