TASK OFFLOADING STRATEGY IN INTERNET OF VEHICLES BASED ON NOMA AND IMPROVED ANT COLONY ALGORITHM

In order to further reduce the task offloading latency and energy consumption in the internet of vehicles, this paper proposed a task offloading strategy of internet of vehicles based on NOMA (non-orthogonal multiple access) and improved ant colony algorithm. NOMA was used in this strategy for the communication between nodes, and a system cost model associated with latency and energy cost was constructed with the restriction of the vehicles’ residual time in the serving range of RSU (roadside service unit). An improved ant colony algorithm based on e constraint-handling technique and mixed variables ant colony algorithm was proposed to achieve the approximate minimum system cost. The results of simulation show that the proposed offloading strategy converged more quickly than the traditional mixed variables ant colony algorithm. Compared with the other task offloading strategies, it can achieve lower system cost and the decrease range is up to 38.05%.