Research on intelligent task offloading algorithm based on compressed sensing in internet of vehicles
Wenbo Zhang, Jie Zhang, Qi Liu, Shuai Li · 2023
In this paper, a vehicle-side-cloud collaborative offloading architecture for vehicular networking tasks is proposed and an optimization model is constructed to minimize the load under the conditions of computational task latency and communication link stability constraints. Secondly, a compressed sensing technique is introduced to reduce the amount of data for offloading tasks by compressing the sampling to eliminate a large amount of redundant data. Finally, a deep Q-network is introduced to solve the optimization problem, and the terminal executes different offloading strategies according to the task attributes, which can update itself in the environment under time-varying channel conditions. The simulation results show that the intelligent task offloading algorithm based on compressed sensing (CSITO) algorithm has certain superiority.