RBF神经网络在视频业务建模及预测中的应用

ＡＴＭ技术在关键在于业务控制，能否有效地实施业务控制则取决于对业务特征的了解和预测能力。传统的解析方法对视频业务进行预测的局限性较为明显。本文采用径向基函数神经网络对视频业务进行建模和预测，并提出了分别采用ＬＢＧ算法和Ｈｅｓｔｅｎｅｓ奇异值解进行隐含层神经元中心选择和输出神经腱权值计算的改进方法。

An application of RBF neural network in video traffic modeling and predicting

One of the key technologies of ATM is traffic control Whether a traffic control solution is effective depending on the through understanding of the characteristics and the ability of prediction The traditional analytical methods are obviously intractable for video traffic As a settlement,a radial basis function(RBF)neural networks has been adopted in this paper for video traffic modeling and prediction In addition,a proposal concerning the LBG basis algorithm and Hestenes singular value decomposition(SVD)method are involved,to be able to improve the selection of the centers of hidden layer neuron and the calculation method of the weights of the output layer synapse in the RBF network A series of practical video data has been used to evaluate the proposed approach