摘要
本文提出了一个新的基于盲信号分离的信道均衡结构和算法。通过对接收信号的过采样,由源序列构成的、长度为N的矢量可以被看成互相独立的N个信号源,与此相应的接收矢量则是该N个独立信号通过线性系统后的输出。为了恢复被传送的序列,我们采用基于神经网络学习的盲信号分离算法,实现信道的盲均衡。模拟结果显示,无论是实信道还是复数信道,该方法都具有较好的均衡效果。
In this paper,we present a new algorithm for channel equalization,which is based on blind signal separation By over sampling,a block of transmitted symbols with length N can be considered as N independent source signals ,and the corresponding received vector is the output of a linear system driven by these N source signals To recover the initial sequence,a robust neural network is used to perform blind separation Simulation results show that no matter the channel is real or complex,the transmitted sequence can be successfully recovered
出处
《通信学报》
EI
CSCD
北大核心
1999年第2期70-74,共5页
Journal on Communications