Improving the Performance of Emergency Message Dissemination Scheme for VANET using Selective Forwarding Algorithm in Multi Directional Road Scenario

Abstract

A kind of mobile ad hoc network called VANET is created for vehicular communication. The performance and efficiency of transportation systems are being greatly enhanced by VANET, which is assisting in reshaping Intelligent Transportation Systems. In a VANET, vehicles can share and exchange information in terms of Vehicle-to-Vehicle or Vehicle-to-Infrastructure communication. V2V is communication among vehicles and V2I is communication between vehicles with roadside infrastructures. Dissemination of safety messages in VANET highly relies on the broadcasting mechanism. Since VANET is characterized by its dynamic nature and high mobility, flooding and simple broadcasting cause the broadcast storm problem. Because of broadcast storm issues, message broadcasting processes are slowed down in ways like higher overhead, broadcast collisions, and delayed dissemination. This thesis aims to provide a VANET emergency message dissemination scheme with a high delivery ratio, little end-to-end delay, and minimal overhead. A selective forwarding scheme based on the SIR epidemic spreading model is designed to solve the broadcast storm and network overhead problems that are caused due to the redundant selection of forwarding nodes in a multi directional road scenario. An algorithm for selecting stable forwarding nodes and safety message dissemination is proposed. In the proposed approach the forwarding node is selected by considering the passive acknowledgment, distance, direction, and link lifetime as a parameter. We used the SUMO simulator for generating mobility and OMNET++ for simulating our network with the Veins framework. From the simulation, we get 160ms end-to end delay, 40sec duration of forwarding node, and 83% emergency message delivery ratio in the proposed approach, whereas the existing scheme results in 187ms end-to-end delay, 35sec duration of forwarding node, and 80% Emergency message delivery ratio with 100 node density. The result of this simulation showed that the proposed approach improves the emergency message delivery ratio by minimizing the end-to-end-delay and overhead as compared to the previous approach.