SDN (Software Defined Network) flow control method based on fuzzy C-means and hybrid kernel least square support vector machine

Abstract

The invention belongs to the field of network flow prediction, and particularly relates to an SDN (Software Defined Network) flow control method based on a fuzzy C mean value and a mixed kernel least square support vector machine, which comprises the following steps of: converting non-stationary SDN flow data into a stationary time sequence component by adopting discrete wavelet transform; processing the stationary time sequence component to obtain amplitude signals of a high frequency band and a low frequency band; clustering the amplitude signals of the high and low frequency bands by adopting a fuzzy C-means algorithm; an optimized self-adaptive mixed kernel least square support vector machine prediction model is adopted to predict the clustered components respectively; reconstructing the prediction results of all the components to obtain a prediction result of the SDN network data traffic; according to the method, a fuzzy C-means algorithm is utilized, a membership mechanism is introduced, time sequence components are divided into a high-frequency low-amplitude component, an intermediate-frequency middle-amplitude component and a low-frequency high-amplitude component according to amplitude-frequency characteristics of the time sequence components, and accurate prediction is provided for subsequent classification prediction.

Description

technical field [0001] The invention belongs to the field of network flow forecasting, and in particular relates to an SDN network process control method based on fuzzy C-means and hybrid kernel least squares support vector machines. Background technique [0002] Software-defined networking (SDN) has gradually become an emerging industry in the current network industry. Its main idea is to separate the control plane and data plane that originally belonged to network switches and routers, so as to realize the real separation of forwarding and data. Compared with the complexity of the traditional SNMP network distributed measurement system, the SDN network can realize centralized monitoring of network traffic data, and network traffic prediction is also one of the important ways to improve its service quality and ensure its service security. The traditional traffic forecasting method mainly integrates the traffic data into a traffic time series, that is, the traffic forecastin...

AI Technical Summary

Problems solved by technology

In LSSVM, although the sample data is originally complex and has different dimensions, it is easy to separate the data and map the corresponding data to a high-dimensional space through the kernel function, so it is widely used in time series forecasting; Zhu Qianyu et al. A traffic forecasting model combining Empirical Mode Decomposition (EMD) and Particle Swarm Optimization (PSO) to optimize the least squares support vector machine is proposed. The traffic sequence is decomposed and stabilized by EMD, and then PSO is used to optimize the parameters of the LSSVM forecasting model. Finally, Effectively improve the prediction accuracy of the model, but only one type of kernel function is used for prediction analysis, and the adaptability of each decomposition component to different kernel functions is not considered

Images