Efficient data dissemination protocols for V2X communication networks

With the rapid development of intelligent transportation system (ITS) technology and applications, the real-time wireless connectivity of moving vehicles has attracted increasing attention in the academia and industries. Relying on vehicle-to-everything (V2X) communication, vehicles are enabled to c...

Full description

Saved in:
Bibliographic Details
Main Author: Gao, Yumeng
Other Authors: Guan Yong Liang
Format: Theses and Dissertations
Language:English
Published: 2019
Subjects:
Online Access:https://hdl.handle.net/10356/90009
http://hdl.handle.net/10220/49892
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:With the rapid development of intelligent transportation system (ITS) technology and applications, the real-time wireless connectivity of moving vehicles has attracted increasing attention in the academia and industries. Relying on vehicle-to-everything (V2X) communication, vehicles are enabled to communicate with vehicles directly, as well as with roadside infrastructures. V2X communication is able to support a variety of applications to address road safety concern and support commercial or regulatory ITS services. Confronted by dynamic vehicular environment and stringent latency requirements, the design of efficient V2X communication protocols are of great relevance. In this thesis, we aim to design efficient data dissemination schemes for a number of realistic V2X use-cases. To this end, we utilize techniques such as network coding and Markov chain modeling, and focus on media access control (MAC) layer enhancement and scheduling optimization. Firstly, the impact of the contention-window size of the IEEE 802.11p standard for periodic broadcast of short messages using dedicated short-range communication (DSRC) among vehicles is investigated. We apply Markov chain to model the periodic broadcasting behavior (such as broadcasting basic safety message) of the vehicles under the IEEE 802.11p MAC layer standard for both saturated and unsaturated network conditions. We attain enhancements by analytically optimizing the window size for maximizing the communication throughput, which are then verified by computer simulations. Secondly, a reliable basic safety message (BSM) dissemination scheme is proposed for the non-line-of-sight (NLOS) road intersection scenario. The periodic BSM exchanges by all vehicles are particularly crucial for preventing car collisions at road intersections without traffic signals. Large concrete buildings and obstacles may create a NLOS condition, leading to severe signal blockage. In order to address this problem and meet the timeliness requirement of BSM dissemination, we exploit random linear network coding and introduce controlled repetitive transmission patterns. Two promising network-coded relaying schemes are proposed to improve the reliability and scalability of the system. Thirdly, an efficient V2X content distribution scheme based on batched sparse (BATS) code (a two-layer network code) with distributed scheduling is studied, in which digital maps or entertainment content are distributed from a roadside unit (RSU) to a group of vehicles passing by. To improve the transmission reliability in the lossy vehicular communication environment and possibly disconnected RSU cells, the original file is encoded with a BATS code for broadcast by a RSU, then a cooperative vehicle-to-vehicle (V2V) data sharing phase follows to repair the packets lost in the first I2V phase. To reduce the total content-distribution delay, a utility-based distributed algorithm is proposed to efficiently schedule the cooperative V2V data sharing phase. The proposed content distribution scheme is evaluated by extensive computer simulations, considering multi-lane road and realistic vehicular traffic settings, and shown to significantly reduce the transmission delay and communication overhead compared with known protocols such as CodeOn and CodeTorrent. Finally, the problem of optimal scheduling for multi-hop V2V video streaming with network coding (applied to mitigate the well-known curse of exponential multi-hop throughput drop) in a Cellular-V2X setting is studied. The proposed network-coded scheduling algorithm considers not just the direct V2V communication link, but also the RF overhearing link, in the multi-hop V2V network. To further improve the system performance, channel reuse is applied for long multi-hop networks. Simulation results show that the proposed network-coded scheduling scheme achieves substantial end-to-end throughput (achievable normalized generation rate) gain over conventional packet erasure correction coding schemes.