Interconnecting smart cities by vehicles: How feasible is it?

In future smart cities cars will play an important role not only by transporting people and goods, but also as information hubs. Cars will be equipped with various communication technologies and will be able to offer their own resources like data storage, processing power, and sensor data (e.g., camera pictures or temperature). In Car4ICT, cars in smart cities connect service producers and consumers by offering service discovery for them and transferring the data in-between. But, the proposed architecture cannot be used as-is for interconnecting smart cities or transferring data from and to rural areas. This is due to mobility patterns outside of cities being different and requiring specialized algorithms. Therefore, in this paper we extend Car4ICT to interconnect smart cities and rural areas. The extension still relies on IVC as cellular coverage in rural areas is often sparse. By covering additional areas, the number of available services increases and a redundant communication link between smart cities is added. To support long-distance communication, we changed core parts of the architecture and evaluated the changes with extensive simulations. For this, we used a segment from the freeway between Tokyo and Osaka in Japan and evaluated the transmission delay for various distances. The foundation for these simulations was real world traffic data provided by Japanese authorities. Based on these simulations we can show for which distance Car4ICT in a freeway scenario is still feasible.