Design and Development of a Framework for Reliable Data Dissemination in VANET Environment Using Software Defined Networking and Cloud

Abstract

Vehicular networks are one of the cornerstones of an Intelligent Transportation System (ITS). With an increase in the number of vehicles on the road, demand for Intelligent Transportation System is on the rise. The use of ITS is widely accepted and used in many countries for not only traffic congestion and demand control, but also for road safety and efficient infrastructure usage. A high annual rate of road accidents for many years requires improved transportation systems to provide a high level of on-road safety. The Global standing report on road safety, reflective information from a hundred and eighty countries, indicates that worldwide the overall number of deaths by road traffic has reached a quiescent stage with highest on-road traffic death rates. Effective data dissemination is one of the solutions as it is a crucial component of safety as well as infotainment services. In explicit, safety messages, such as, road accident warning, lane change, traffic congestion warning, etc., are the most crucial message types that need to be efficiently disseminated with the minimal delay and high reliability. It is essential in vehicular networks to aid routing and cooperative driving applications etc., apart from road-safety message dissemination through periodic beaconing. Vehicular Ad Hoc Network has high mobility and recurring disconnections due to which reliable and efficient vehicular communications is a difficult task. Traditional information dissemination protocols usually depends upon extra communication to collect information regarding the neighboring regions and fix the dissemination criteria consequently. This strategy incurs significant overhead problems due to which the scalability of data dissemination, a crucial criterion under high-density regions, is affected. Many existing models under I2V environment following static broadcast timer mechanism are not suitable to VANETs. Also, for security and reliable data delivery under V2V communication environment, general forwarding schemes designed for general purpose MANETs cannot be directly applied to VANETs. Also, most of the existing solution for data dissemination in VANETs are unable to provide flexibility and adaptability due to their rigid architecture. These drawbacks introduce an obstacle to meet the Quality of Service (QoS) parameters. Moreover, the existing schemes are unable to provide sustainable communication. Hence, there is a need for a new solution which exhibits flexibility, adaptability to meet the desired QoS parameters and ensures reliable communication. Apart from the data dissemination, data collection from a wide variety of sources that can be processed and organized into many forms for multiple stakeholders, has brought significant change in ITS recently. An existing centralized information processing system in VANETs lacks flexibility, adaptability and also involves a significant initial investment and high operating costs. Moreover, they are affected by problems like the necessity to transmit large amounts of data, the requirement of a more extended period for road network coverage, unsteady transmission, and the demand for an automatic generation and update strategy for the road network information in a large-scale road network. Moreover, cloud-based vehicular network architecture incurs delay due to network delay (Al-Badarneh et al.). This problem motivates us to design a framework for reliable data dissemination among multiple vehicles from RSU under V2X(i.e., V2V or I2V) environment by using Software Defined Networking (SDN) and Cloud. Processing of vehicular network data with cloud and SDN increases accuracy,or coverage of the traffic information, reduction in delay due to preprocessing at smart RSUs, i.e., fog paradigm which is, not attempted full edge, to the best of our knowledge. This thesis work firstly advocates for the generic design of an architecture which exercises the software-defined network concept and cloud computing to support various ITS services. Secondly, we focus on design a new solution for calculating an adaptive broadcast time interval at RSU under I2V (Infrastructure-to-vehicle) environment for disseminating safety alert and infotainment messages among moving vehicles. Thirdly, we propose a network coding based reliable protocol, which uses SDN controller aiming at increasing the success delivery ratio under the V2X environment. At last, to assist a wide variety of VANET services using cloud resources, we develop an SDN based load balancer for the effective utilization of cloud resources and thus, this research work contributes towards an overall performance and QoS enhancement for data dissemination in VANET environment.