【作者】 孙云山； 张立毅； 李艳琴；

【Author】 Sun Yun-shan, Zhang Li-yi, Li Yan-qin (Department of Information and Engineering ,Taiyuan University of Technology, taiyuan 030024)

【机构】 太原理工大学信息工程学院；

【摘要】 码间干扰是严重影响数字通信质量的重要因素之一,均衡器的作用主要是用来消除码问干扰,盲均衡算法是指在不借助训练序列的情况下,只根据接收信号直接进行均衡。本文提出一种新的基于模糊神经网络分类器的盲均衡算法,它利用盲估计(BE:Blind Estimate)算法与模糊神经网络(FNN:Fuzzy Neuron Network)分类器相结合,先对通信信道进行盲估计,然后利用FNN的分类功能对接收信号进行分类。本算法克服了原有盲均衡算法反解卷积计算量大的缺点,加快了收敛速度。通过仿真,验证了算法的有效性。更多还原

【Abstract】 The inter-symbol interference in digital communication badly affects communication quality. In order to overcome inter - symbol interference, adaptive equalizers generally are added in sink. The conventional e qualizer resorts to a training sequence and it ties up the limited bandwidth; in some special cases, training se quences can not be transmitted. Blind equalization techniques rely on solely the received channel output signal to equalize the channel without a known training sequence available. This paper brings forward a new blind equalization algorithm based on fuzzy neural network. It makes use of a algorithm between blind estimation and fuzzy neuron network, it firstly blindly estimate channel character, and the incepting signals are classified with FNN. This algorithm overcomes the weaknesses of great computation of inverse dispel convolve, and fasten the convergence speed. The validity is validated by simulations.更多还原