The actual traffic prediction method based on glowworm swarm optimisation. (2nd April 2021)

• Record Type: Journal Article
• Authors: Chen, Ke; Li, Qirui; Zhu, Xingtong; Zuo, Liyun
• Is Part Of: International journal of wireless and mobile computing. Volume 20:Number 2(2021)
• Page Start: 153
• Page End: 158
• Publication Date: 2021-04-02
• Subjects: congestion -- prediction -- accuracy -- glowworm swarm; Mobile computing -- Periodicals; Wireless communication systems -- Periodicals; 004.6
• Languages: English
• ISSNs: 1741-1084

Abstract

In order to mitigate congestion caused by the rapid growth of computer networks, a novel traffic prediction algorithm PGS (Prediction method based on Glowworm Swarm) is proposed using the glowworm swarm optimisation method. In this algorithm, the arrival flow is regarded as glowworm swarm and the node service rate is regarded as attractiveness at first, and in order to improve the prediction accuracy, the optimal position and attractiveness are obtained with the individuals' moving operation and random flying operations. Then, a simulation with OPENT and MATLAB is conducted to research on the key factors of prediction error for PGS. Compared to the wavelet transform prediction method, the prediction error is decreased 1.08%. The result shows that PGS has better adaptability.