| The traditional adaptive equalization technology because of needing to repeatedly transmit the training sequence, in the underwater acoustic communication, will certainly to take the poor underwater acoustic communication band width, then reduces the communication efficiency. However blind equalization does not need the training sequence, and the theory and the practice proved, blind equalization may under certain traditional, adaptive equalization technology invalid conditions, still has good performance, under specific constraint condition, it can obtain better equalization performance than traditional adaptive equalization technology. With the development of modernized underwater acoustic communication to the high speed information transmission demand, it has the vital significance that the analysis and the research of blind equalization theory algorithm.This paper in view of the underwater acoustic channel characteristic, has made in the theory and the algorithm research and the analysis to a Bussgang blind equalization algorithm and the direct higher order statistic algorithm, and aims at the underwater acoustic channel nonlinear response, has carried on the analysis and the improvement to the forward feed neuron network blind equalization algorithm realization. The research content mainly includes following several aspects:1. From mathematics the blind counter- convolution angle analyzed the blind equalization realization theory basis, the equalization criterion and the constraint condition. According to the BBR formula, explained the statistical property relationship between the input signal and output signal under the premise of the signal with non-Gaussian distribution, and proved the reason of needing higher order cumulant in blind equalization.2. Algorithm rationale and the algorithm realization has been conducted the research to exceptional case CMA of Bussgang blind equalization algorithm. And partial minimum question of the CMA algorithm price function has been...
|
PDF Full Text Request