Research On Blind Equalization Theory And Algorithm Based On Time Reversal

Wireless channel is a complex channel with time-varying, frequency-dependence and strong interference, the signal is easy to distortion during transmission. The blind equalization technology does not need to send the training sequence cycle, the intersymbol interference was eliminated, the transmission rate of information was improved, but the convergence of blind equalization algorithm was seriously affected by the variability and unpredictability of the channel signal. Time reversal as an emerging technology, it has been widely applied to the underwater acoustic communication, medical diagnosis, target detection of electromagnetic wave. Using the time reversal technology can not only reduce the computational complexity of blind equalization algorithm, but also improve the performance of the receiver equalizer characteristics. This paper studies the research on blind equalization theory and algorithm based on time reversal, the main work is as follows:Firstly, a blind equalization algorithm based on time reversal was proposed under studying the balanced results of the time-reversal in the different channels. And the drawback of decision feedback equalizer for biggest balanced phase components was solved. The simulation experiment of the time reversal of wireless channel and underwater acoustic channel was carried out respectively. By the analysis of the reasons for the differences in different channel equalization simulation results, it is pointed out that time reversal in the application of the wireless channel can achieve very good effect on equilibrium, but equalization effects with varying differences owing to the different characteristics of underwater acoustic communication channel.Secondly, the two blind equalization algorithms based on clustering of time reversal were studied. For the drawbacks of the undesirable traditional dual mode convergence effect and the inaccuracies changeover timing, the time reversal was combined with clustering algorithm in the first. Improve the accuracy of clustering switch timing and the balance ability of the algorithm were improved by that the use of time-reversal properties made cluster which achieved the desired effect switch to DD equalizer algorithm. The second, the clustering recognition of blind equalization algorithm based on time reversal to the axes was proposed by the QAM constellation pointed projection to solve the clustering problem of insufficient data points when The data length was too short and reduce the number of cluster centers. But as a result of clustering recognition algorithm is sensitive to noise, the introduction of time reversal technology can reduce noise interference, thus greatly reduce the complexity of the algorithm and improve the convergence speed of the algorithm.The last one, time reversal in MIMO environment was studied in blind equalization algorithm of orthogonal wavelet transform. For the problems of the signal distortion in the MIMO channel, the complex system structure and the relationship between multiple signal receiving end, two kinds of solution algorithm were proposed. The first algorithm of a single-channel time-reversal was presented, single-channel time-reversal has the advantage of reduce channel estimation requirements and the emission of dealing with the complexity compared with full channel time reversal.