A SA-BP neural network algorithm based channel quality evaluation method in optical access network

Abstract

So far,existing works related to quality evaluation of channel mainly focused on equipment layer and network layer in optical access network.Most of researches lack synthesis evaluation methods covering physical layer,network layer and business layer.As a result,it is hard for administrators to judge the actual quality of channel by network monitoring completely and accurately.To solve this problem,key influence factors of channel quality in optical access network was analyzed firstly,and a comprehensive multi-layer-multi-index evaluation model of channel quality was proposed.Then a SA-BP neural network algorithm was designed to train the parameters of the proposed evaluation model to make the result more accurate.The simulation results show that the proposed method has higher accuracy and stability,and can improve the quality and efficiency of network operation and maintenance.

Keywords

• optical access network;channel quality;evaluation model;simulated annealing-back propagation neural network algorithm