Study On Blind Equalization For Communication System

Blind equalization is a new adaptive technology without resorting to a training sequence, which only utilizes the prior information of transmitted signals to equalize the channel character and makes sure that the output sequence approximates the transmitted signals as accurate as possible. Therefore, it can improve channel efficiency, equalize the not ideal character of channel availably, overcome the inter-symbol interference, reduce the error rate and acquire better equalization performance because of not need to send out the training sequence. It currently becomes one of the key technologies in the field of digital communication because of the superior performance, therefore, further study on blind equalization is very important for instructing both the theory and the applications. The main jobs of this thesis are as follows:(1) The basic theory, algorithm form of blind equalization, the basic algorithm and convergence criterion of it are systematically analyzed. The basic theory and property of Bussgang blind equalization algorithm are described and three algorithms including Decision-Directed algorithm,Sato algorithm,Godard algorithm in common used of Bussgang are introduced.(2) CMA blind equalization algorithm of Bussgang algorithms is the emphasis. The CMA algorithm is derived and the convergence performance of traditional CMA algorithm is analyzed. Simulation results proved to show that the convergence speed is quicker but with a larger residual error if bigger step is adopted. In order to minimize the residual error after convergence, the smaller step should be adopted, but at the same time, the convergence speed is slow.(3) A new variable step-size CMA blind equalization algorithm is proposed in order to solve the antinomy between the convergence rate and accuracy because of using the fixed step. The new algorithm makes use of the square of average grads as the parameter to produce the new variable step size. The theoretical analysis and the simulation results proved to show that the new algorithm has improved performance of the convergence speed and residual error than traditional CMA algorithm and can be applied in the self-adaptive blind equalization system availably.