Research On Blind Equalization Algorithm For Shortwave Signals

In the high-speed communication system, in order to eliminate the inter-symbol interferencecaused by the restriction of the limited bandwidth and the transmission of multipath, it’snecessary to introduce the equalization technique to the receiver. The blind equalizationtechnique has been widely used, owing to its independence of training sequence,The main work and innovation of this paper can be summarized as follows:1. The basic definition, mathematical model, and the equalization rule are elaborated. Thecommonly used blind equalization algorithms and the evaluation standard of the equalizationalgorithm are assessed.2. The convergence performance of the CMA algorithm is analyzed, and the variable stepis introduced to the CMA algorithm. According to the design of automatic-adaptive equalization,we propose two new modified CMA algorithms by changing the variable step, which uses theMSE transform and the power output by the equalizer separately as the control factor for thevariable step.3. For the deep fading channel, we design a fractional tap-spacing decision-feedbackequalizer, which can reach lower steady-state error. The forward filter adopts the fractionaltap-spacing structure, which can avoid the frequency aliasing resulted by the subsampling ofsymbol rate equalizer. The decision-feedback structure of the feedback part can eliminate theinter-symbol interference introduced by the forward symbol, reduce the steady-state error of theoutput signal, and avoid the common wrong convergence caused by error diffusion.