A collaborative scheduling strategy for IoV computing resources considering location privacy protection in mobile edge computing environment

Abstract

This paper proposes a collaborative scheduling strategy for computing resources of the Internet of vehicles considering location privacy protection in the mobile edge computing environment. Firstly, a multi area multi-user multi MEC server system is designed, in which a MEC server is deployed in each area, and multiple vehicle user equipment in an area can offload computing tasks to MEC servers in different areas by a wireless channel. Then, considering the mobility of users in Internet of vehicles, a vehicle distance prediction based on Kalman filter is proposed to improve the accuracy of vehicle-to-vehicle distance. However, when the vehicle performs the task, it needs to submit the real location, which causes the problem of the location privacy disclosure of vehicle users. Finally, the total cost of communication delay, location privacy of vehicles and energy consumption of all users is formulated as the optimization goal, which take into account the system state, action strategy, reward and punishment function and other factors. Moreover, Double DQN algorithm is used to solve the optimal scheduling strategy for minimizing the total consumption cost of system. Simulation results show that proposed algorithm has the highest computing task completion rate and converges to about 80% after 8000 iterations, and its performance is more ideal compared with other algorithms in terms of system energy cost and task completion rate, which demonstrates the effectiveness of our proposed scheduling strategy.

Resource Allocation Methods in Vanets:a Systematic Literature Review

“…Each step has a reward for each state transition brought on by each action step. [46] AI is great at handling spare and heterogeneous data which makes it a great candidate for VANET resource management as well. [47]…”

Task Scheduling Strategy of Logistics Cloud Robot Based on edge computing

“…During task allocation, the execution strategy corresponding to the solution x needs to be solved to meet the dependencies between tasks and environmental parameters, that is, the solution obtained should be in the feasible space. In the research of task allocation, the existing deep reinforcement learning methods usually regard it as an end-to-end learning task, and have designed different models and training methods [11][12][13][14][15][16][17]. However, by exploring each step of action, the model will not get a reward function value for completing the task until the whole scheduling task is completed, resulting in sparse rewards, large state space, and difficulty in training.…”

Assessment of Innovative Architectures, Challenges and Solutions of Edge Intelligence

“…Fog computing using cloud resources reduced latency in big-data processing, according to the findings. According to Pang, Wang and Fang [18], a comparable architecture was proposed for collecting data from a variety of sensors by employing cloud computing and hierarchical edge strategy. Many sensors provide information to the first layer of collectors, which is called edge level, before it is sent to a generic cloud service provider.…”