Multi-Hop Propagation-Based Computation Offloading Scheme in Vehicular Communication Environment

This paper studies computation offloading in vehicular communication environment. With the ever-increasing number of vehicles on the roads and the advent of the vehicle-to-vehicle (V2V) communications, vehicles are allowed to offload their computation-intensive and time-sensitive tasks to other neighboring vehicles, which are service vehicles (SV) that offer their underutilized computing resources as a service, and the vehicles that request to offload tasks are request vehicles (RVs), computation offloading enables RVs to overcome their computing resource limits and make full use of the nearby computing resources. In this paper, we propose a multi-hop propagation-based scheme for vehicular computation offloading, where both the requirements of the tasks and mobility of the vehicles are considered, the objective is to maximize the utilization of available computing resources of vehicles around RV in a dynamic vehicular communication environment. Instead of RV only to offload tasks to SVs within one-hop communication range, the scheme we proposed allows RV to offload tasks outside of one-hop communication range by using SVs as relay nodes (RNs). Due to high mobility of vehicles, we establish a set of conditions to ensure that communication link is alive during the computation offloading process, and a multi-hop propagation search (MPS) algorithm is proposed to find a propagation path for minimizing the total energy consumption while delay constraints of tasks are satisfied. Simulation results are shown the performance of the algorithm MPS.