论文部分内容阅读
风险分析是光纤通信网络的关建技术,而网络风险的影响因子多、因子之间存在相互作用,为了提高光纤通信网络风险分析的准确性,发现存在潜在的网络风险,提出一种因子分析和神经网络的光纤通信网络风险预测模型(FA-RBF)。首先尽可能多的收集光纤通信网络影响因子,并采用因子分析法对影响因子进行处理,选择主要影响因子,然后根据主要因子确定RBF神经网络结构,采用遗传算法对RBF神经网络参数进行选择,最后的应用实例结果表明,FA-RBF可以快速度、准确确定光纤通信网络的影响因子,减少了RBF神经网络向量的维数,提高了光纤通信网络风险分析的精度,加快了风险分析的速度。
Risk analysis is the construction technology of optical fiber communication network. There are many influencing factors of network risk and factors interact with each other. In order to improve the accuracy of the risk analysis of optical fiber communication network and find out the potential network risks, a factor analysis and Neural network fiber communication network risk prediction model (FA-RBF). Firstly, the influencing factors of optical fiber communication network should be collected as much as possible, and the influencing factors should be processed by factor analysis method, the main influencing factors should be selected, then the structure of RBF neural network should be determined according to the main factors, the parameters of RBF neural network should be selected by genetic algorithm, The results show that FA-RBF can quickly and accurately determine the influence factors of optical fiber communication network, reduce the dimension of RBF neural network vector, improve the accuracy of optical fiber communication network risk analysis and speed up the risk analysis.