SVM-Based Multi-Hop Video Packet Delay Bound Violation Modeling over Mesh Networks

It is crucial to develop feasible methodologies and techniques for accurately analyzing, predicting end-to-end delay performance over multi-hop network communication paths. In this work, we develop a multi-hop packet delay bound violation model to predict the packet loss probability and end-to-end video distortion for video streaming over multi-hop networks. To this end, we extract salient features to characterize the input source and network conditions of links along the transmission path and construct a statistical learning-based model using Support Vector Machines (SVM) for nonlinear regression. Our extensive simulation results with NS-2 demonstrate that the proposed model achieves accurate and reliable prediction of the packet delay bound violation probability and end-to-end video distortion under various network settings and transmission conditions. Compared with genetic neural network (GNN) prediction model supposed in our previous research, It shows that SVM has a higher accuracy and greater generalization ability of prediction than GNN, which proved the validity and practicality of the model.