Performance Evaluation of Local Area Network based on Support Vector Machine
Qi Liu, Yin Liu, Yiyong Lin, Ling Song He, Yunzhi Huang · Advances in engineering research/Advances in Engineering Research · 2016
This work presents an IP performance evaluation method, based on the Support Vector Machine (SVM).In this work, eight network parameters are collected: CPU utilization of switchboard, utilization of memory, network link-off, delay, delay jitter, bandwidth, bit rate of transmission and bit rate of reception.The performance of the local area network is classified into three grades: excellent, good and failed.In this work, the collected network parameters are processed using SVM classifier.The average classification accuracy is achieved using k-fold cross validation method.The experiment results indicate that the classification accuracy of the proposed method is over 90%.This proposed evaluation system could be applied to a real-time network performance evaluation application effectively.