一种新的卫星通信网流量预测算法

A New Traffic Flow Prediction Algorithm for Satellite Communication Network

在通信网络的设计中,使用基于流量预测的网络规划已成为LTE发展的必然趋势。与地面网络不同,卫星网络由于受资源受限和拓扑时变的不利影响,其流量预测算法必须能兼顾精度和效率,这令传统的地面网络预测方法已不再适用。为了解决以上问题,提出了一种新的基于小波回声状态网络的流量预测算法,该算法通过小波多尺度分解的信号处理方法屏蔽了网络流量的噪声,而后结合了无反馈的回声状态网络联合进行预测。仿真证明,新算法相比传统算法能大幅提升网络流量的预测精度和运行效率,为卫星网络的流量规划提供了强有力的决策支持。

In the design of a communication network, it has become an inevitable trend with LET development to use tragic flow prediction method for network planning. Due to suffer from the constrained resource and changeable topology, the traffic flow prediction algorithm of the satellite networks which is different from terrestrial net- works, should fully take into account the accuracy and efficiency, so the traditional prediction methods for ter- restrial networks are no longer fit for application. In order to resolve this problem, a new traffic flow prediction algorithm based on wavelet and echo state networks is proposed in this paper. The new algorithm uses the signal processing method based on multi-scale decomposition of wavelet to shield the noise of the network traffic, and it combines with the non-feedback echo state network to predict. The simulation results show that the new algorithm can greatly improve the prediction accuracy and running efficiency of network traffic compared with traditional algorithms, so it provides a more scientific decision support for satellite communication network traffic planning.