Autonomous Identification and Optimal Selection of Popular Smart Vehicles for Urban Sensing—An Information-Centric Approach

Today, vehicles are becoming powerful sensor platforms capable of collecting, storing, and sharing large amounts of sensory data by constant monitoring of urban streets. It is quite challenging to upload such data from all vehicles to the infrastructure due to limited bandwidth resources and high cost. This invokes the need to identify the appropriate vehicles, important for different urban sensing tasks based on their natural mobility. This paper address this problem of leveraging the self-decision making ability of a “smart vehicle” to measure its relative importance in the network. To do so, we present InfoRank as an information-centric algorithm for a vehicle to first autonomously rank different location-aware information. It then uses the information importance along its mobility pattern to find its importance in the network. We also present a selection algorithm to find the best ranked vehicles for urban sensing and vicinity monitoring to achieve a desired coverage within a limited budget. Our vehicle ranking system is the first step toward identifying the best information hubs to be used in the network for the efficient collection, storage, and distribution of urban sensory information. We evaluate InfoRank under a scalable simulation environment using realistic vehicular mobility traces. Results show that the proposed ranking system efficiently identified socially important vehicles in comparison to other ranking schemes.