Corrected mold decision multi-mold blind equalization method

Abstract

The invention discloses a corrected mold decision multi-mode blind equalization method MMDMMA. The method comprises the following steps: re-defining a cost function, and acquiring a weight vector update formula of a forward equalizer and a feedback filter through the defined cost function according to a stochastic gradient descent method; replacing a fixed decision factor of the conventional mold decision multi-mold blind equalization method MMDMMA by using a leakage factor with time-variant characteristics, so that a time-varying signal can be accurately tracked. The replacement has the effects that a decision error is large, a time-varying leakage factor is large, and a convergence rate is high before convergence; the decision error is small, the time-varying leakage factor is small, and the convergence rate is low after the convergence; due to the adoption of the time-varying leakage factor, divergence of the blind equalization process in the MDMMA method is avoided, and misadjustment is avoided. The method is low in mean square error, high in convergence speed, and high in capacities of inhibiting intersymbol interference and correcting phase rotation. Compared with a mold decision multi-mode blind equalization method MDMMA and a norm blind equalization method CMA, the method MMDMMA is obviously improved.

Description

technical field [0001] The invention belongs to the technical field of satellite communication, and in particular relates to an improved judgment multi-mode blind equalization method. Background technique [0002] In the mobile satellite communication system, intersymbol interference (ISI) and phase deflection caused by multipath fading and Doppler effect cause severe distortion of the transmitted signal and distortion at the receiving end, seriously affecting the communication quality. In order to improve the communication quality, the researchers proposed a blind equalization method CMA based on data reliability decision guidance. This type of method mainly uses the forward equalizer to eliminate the leading distortion of channel equalization, and uses the feedback filter to cancel the tail distortion. However, as modern communication technology requires higher and higher power utilization of modulated signals, high-order amplitude modulation signal QAM is often used to im...

AI Technical Summary

Problems solved by technology

[0015] In view of the constellation characteristics of high-order QAM signals and the defect that the fixed decision factor in the existing mode-decision multi-mode blind equalization method MDMMA does not have dynamic tracking performance, the present invention discloses a modified mode-decision multi-mode blind equalization method MMDMMA, through De...

Images