基于功率谱密度分析的ATM连接允许控制研究

Research on ATM Connection Admission Control Based on Power spectrum Analyses

本文研究了ＡＴＭ网络中的多媒体实时突发业务流的功率谱密度特性，利用反映业务流相关特性的功率谱密度，设计了一个用于多媒体业务的ＡＴＭ连接允许控制方法，并给出了将人工神经网络技术应用于实现业务流的功率谱密度、负载强度与信元丢失概率之间映射的理论方法，增强了ＡＴＭ连接允许控制的智能特性。

This paper studies the characteristics of real time bursty multimedia traffic flows in ATM networks. Utilizing the power spectrum related to traffic correlation, it designs an ATM connection admission control (CAC) method and gives the theoretical method of the mapping between its power spectrum and traffic intensity of traffic flow and cell loss probability based on artificial neural network. Hence,the intelligence of the ATM CAC will be enhanced.