Title: Optimisation method of MAC protocol based on SVM neural network in VANET

Authors: Yucai Zhou; Xiaoya Xu; Caihong Liu; Yuelin Li

Addresses: School of Energy and Power, Changsha University of Science and Technology, Changsha, 410076, China ' Luohe Vocational Technology College, Luohe 462000, Henan, China ' Luohe Vocational Technology College, Luohe 462000, Henan, China ' The College of Automotive and Mechanical Engineering, Changsha University of Science and Technology, Wangxin Rd, Yuelu, Changsha, Hunan, China

Abstract: This paper analyses the function expression of the optimal minimum competition window which integrates the network node number, average collision data frame length and sending rate of data frame. At the same time, the proposed optimised method of MAC protocol combines with the SVN neural network which can memory communication environment which include node density and mobile velocity. Each terminal node in the network runs proposed MAC protocol optimisation algorithm based on this function expression to adaptive adjust their minimum competition window and back off the optimal value to improve the network performance. The simulation results show that the effort of optimised algorithm in the Ad Hoc system is limited for unsaturated business VANET; but high accuracy and effect of the optimised algorithm in aspects of throughput and transmission delay has improved significantly for saturated business of VANET.

Keywords: VANET; IEEE 802.11; MAC protocol; SVM neural network.

DOI: 10.1504/IJIPT.2020.108001

International Journal of Internet Protocol Technology, 2020 Vol.13 No.3, pp.158 - 166

Received: 01 Mar 2019
Accepted: 02 Jun 2019
Published online: 01 Jul 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article