Adaptive Context Mediation in Dynamic and Large Scale Vehicular Networks using Relevance Backpropagation

摘要

Adaptive context mediation in large scale vehicle networks leads towards telematics which refers to the concept of vehicles equipped with context-aware embedded smart computing devices with communication capabilities over certain networks. With the use of telematics we can make use of wide range of smart inter-vehicle communication applications like emergency message transmission, collision avoidance, congestion monitoring and intelligent parking space locator. In this paper we present certain requirements for adaptive context-aware information mediation in large scale vehicle networks. We make use of quality attributes for context information and network properties of large scale vehicle networks and experiment with real time data collected using a car simulator. We simulate flooding and other dissemination based communication techniques with relevance backpropagation for large scale vehicle networks using OM-NET++ to analyze the flow of context information. Our simulation results show that our context-aware and adaptive directed diffusion of information using relevance backpropagation increases the performance of the nodes in a large scale vehicle network with less communication overhead.