CASQ: Adaptive and cloud-assisted query processing in vehicular sensor networks.

Vehicles in urban cities are equipped with increasing more sensing units. Large amount of data are continuously generated and they bring great potentials to the intelligent and green city traffic management. However, data gathering and query processing remain key and challenging issues due to the huge amount of sensing data, changeable road conditions, rapid network topology and density changes caused by the movement of vehicles. There is great necessity for the cloud and the vehicular sensor networks to integrate and enhance each other on the cooperative urban sensing applications. In this paper we propose an adaptive and cloud-assisted query processing scheme for VANETs, that adopts the concept of edge nodes and integrates the cloud and vehicular networks to facilitate data storage and indexing, so queries could be processed and forwarded along different communication channels according to the cost and time bounds of the queries. Moreover, the cloud calculates result forwarding strategy by solving a Linear Programming problem, where the query results select the best path either through the 4G channel or through the DSRC (Dedicated Short Range Communication). This research is one of the first steps towards the integration of the cloud and the vehicular networks, as well as edge nodes and the 4G channel, to improve the effectiveness and efficiency of the query processing in VANETs. Extensive experiments demonstrate that up to 94% of the queries could be successfully processed in the proposed scheme, much higher than existing query schemes, while at the same time with a relatively low querying cost.