Bandwidth Efficiency and Service Adaptiveness Oriented Data Dissemination in Heterogeneous Vehicular Networks

Heterogeneous network resources are expected to cooperate with each other to support efficient data services in vehicular networks. However, current data scheduling methods cannot efficiently exploit the benefit of heterogeneous wireless communication interfaces in vehicular networks. In this paper, we propose a software-defined network based service architecture, which enables the scheduler to manage heterogeneous network resources in a centralized way. We analyze the heterogeneity of both data items and networks in terms of data sizes and network features (e.g., cost, transmission rate, coverage, etc.), respectively. On this basis, we formulate a data broadcast and network interface selection (DBNIS) problem, which aims to minimize both the service delay and the network access cost. To tackle the problem efficiently, we propose a coding-assisted multiobjective evolutionary algorithm (CMOEA), which consists of two components: packet encoding and network interface selection. Specifically, for packet encoding, we first develop a packet-size based random linear encoding (PRLE) technique to improve bandwidth efficiency. Then, we theoretically analyze the performance bound of PRLE. For network interface selection, we propose a multiobjective algorithm for network interface selection to adaptively satisfy dynamic requirements with respect to service delay and network access cost by deriving a set of pareto-solutions. Finally, we build the simulation model and implement CMOEA for performance evaluation. The comprehensive simulation results demonstrate the superiority of CMOEA under a wide range of scenarios.