Blind Equalization Algorithm Based On Genetic Algorithm And Wavelet Transform

In modern digital communication system, inter symbol interference caused by mulit-path effects and channel distortion of sound propagation degrades the relability severely. The efficient method to overcome inter symbol interference and improve the performance of the system is equalization technique.The traditional equalization technique need to periodically send known training sequence to track the change of the channel. But the known tarining sequence will occupy the limited bandwidth and reduce the communication efficiency. Compared with the traditional equalization technique, blind equalization need not to use training sequence. It equalizes the distortion channels based solely on the statistical features of the received signals sequence and inproves the utilization efficiency of the bandwidth effectively. So it has been widely used in communication area. The main problem ot blind equalization is its contradiction between convergence speed and stead-state error. In addition, it easily immerges inti partial minimum and could rech the optimal solution.This paper introduces the genetic algorithm into blind equalization equalization and research blind equalization algorithm based on genetic algorithm and wavelet transform. Genetic algorithm is a kind of global parallel optimization algorithm. It has strong robustness and golbal convergence ability. So it is suitable for solving multimodal function optimization problem. Applying genetic algorithm in the study of blind equalization problems could accelerate the convergence speed and obtain the global optimal solution.The mainly studies of this paper contents:1) Summarize per-existing blind equalization and find out the primary contradiction. Establish the improving direction of the algorithm.2) Propose new algorithm which combing genetic algorithm and blind equalization in consideration of the specific characteristic of blind equalization.3) Study two kind of improved genetic algorithm and ptimize the fitness function. Bring them into the blind equalization algorithm and improve the performance of the equalizer.