Multi-Destination Computation Offloading In Vehicular Networks

Mobile computation offloading enables resources-constrained mobile devices to offload their computation intensive tasks to other available computing resources for local energy savings. In this paper, we study the offloading decision and task scheduling issues when multiple serving vehicles (SVs) can be utilized in vehicular network. The overall energy consumption of users with Dynamic Voltage Scaling (DVS) technology is minimized subject to the delay constraint of each task. For the ideal cases, the assignment and scheduling in an offline style are firstly formulated as a mixed-integer nonlinear programming (MINLP) problem, and the optimal solution is derived based on dynamic programming. After that, two online strategies, Energy Consumption Minimization (ECM) based low-complexity assignment, and Resource Reservation (RR) assignment are also proposed. Simulation results demonstrate the improvements in energy saving when the proposed strategies incorporated with the DVS technology are adopted.