Intelligent wireless vehicular sensor networks

Abstract

The increasing number of registered vehicles, traffic density, congestion and limited road space has had an adverse effect on the number of accidents and the resultant costs, injuries and fatalities. Inefficiencies in the transportation system are causing major risks and degradation in the quality of service offered to road users. The area of Intelligent Transportation Systems is the key effort to improve and develop the current transport scheme. Information technology and communication capabilities are applied to the infrastructure in an effort to manage and enhance traffic safety, reduce congestion times, lower the fuel consumption and decrease motor vehicle emissions, thus improving the quality of life. This dissertation proposes a conceptual ad-hoc routing algorithm that guarantees efficient use of data gathered and made available to the network with active and timely vehicle detection and data delivery. The development process of the routing protocol involved the implementation of an appropriate simulation platform comprising road traffic modelling bi-directionally coupled to a network simulator, the design and configuration of a realistic environment, and the study and realization of the modular algorithm. A test bed was set up to analyse each module, and finally, with the architecture completely in place, the system was simulated and results generated. Our simulation results show that the developed algorithm outperformed existing standard non-overlay geographic GPSR routing algorithm in terms of routing protocol overhead, data packet delivery ratio and transmission bit error rate.