Satellite channel complex-valued neural polynomial network blind equalization system and method

Abstract

The invention discloses a satellite channel complex-valued neural polynomial network blind equalization system and a method. According to the invention, a complex-valued neural polynomial network is adopted as a blind equalization module, wherein a single-layer neural network is used to compensate the linear characteristic of a satellite channel, and a nonlinear memoryless processor is used to compensate the nonlinear characteristic of the satellite channel. A six-layer fuzzy neural network controller is designed by building fuzzy control rules, and a sixth-layer weight vector is adjusted by a fixed-step constant modulus approach. The six-layer fuzzy neural network controller controls the iteration step length of the weight vectors of the single-layer neural network and the nonlinear memoryless processor at high precision. The satellite channel complex-valued neural polynomial network blind equalization system has the advantages of simple structure, high convergence rate and small steady-state error. The problem of high complex-valuedity is well solved. The contradiction between convergence rate and mean square error is overcome effectively.

Description

Technical field [0001] The invention relates to a satellite channel complex neural polynomial network blind equalization system and method, and belongs to the technical field of complex neural polynomial network blind equalization methods in satellite communications. Background technique [0002] With the increase in demand for satellite broadband services, spectrum resources tend to be in short supply. How to invent effective methods to efficiently use satellite frequency band resources and obtain high-quality satellite communications is a hot topic for scientists and engineers in various countries. In the satellite communication system, in order to use the power on the satellite as effectively as possible, the High Power Amplifier (HPA) mostly works near the saturation point, and the resulting nonlinear distortion seriously restricts the transmission power and transmission efficiency. [0003] At present, the effective method to solve this problem is to use Volterra normal mode e...

AI Technical Summary

Problems solved by technology

However, there is still a lack of satellite channel blind equalization method combining complex neural polynomial network and fuzzy neural network (FNN, fuzzyneuralnetwork) controller in the prior art

Images