Research For Max Flow Routing In Vehicular Delay-Tolerant Networks

In Delay-Tolerant Networks (DTNs), transmissions are always restricted by bandwidth, node buffer, mobility model and many other factors. This makes it very hard to calculate the whole network capacity, which can truly imply us to make a well schedule for packets transmission. Many existing works uses nodes'buffer management to solve this problem, which cannot actually optimize throughput performance.As a special kind of DTNs, Vehicle Ad-hoc Network (VANET) has some different features comparing with others. The biggest difference is that the topology of VANET has great variability, which is caused by the rapid movement of the vehicle nodes. This makes not only the static routing algorithm cannot be applied in this network, but also some of the algorithms designed for DTNs cannot be shown in outstanding performance in vehicle network.For solving rapid movement of vehicles, we use the Vehicular Mobility Patterns (VMP) in this work to predict the future trajectory of the vehicles. Acts as a vehicle for the movement of historical reference data for compliance with the obvious, this based on previous literature and data generated by a taxi in Shanghai has a very high precision VMP, will be of use in the routing path can be a good predictive vehicle to help Improve overall network throughput performance.In this work, we explore an alternative way to improve DTN routing performance, especially in the whole network throughput. Each transmission between a pair of source and destination can be taken as a max flow problem. We propose a reasonable solution of max flow problem in DTN. And use the result to direct node of how to transmitting packets. The simulation shows that our solution can perform a good throughput with small overhead.