Research On Energy-Efficiency And Privacy-Preserving Task Offloading In MEC-Based Healthcare System

With the rapid development of information and communication technology and Internet of Things(Io T)technology,healthcare has become one of the most attractive application areas of Io T.In order to better adapt to the needs of mobile computing,Mobile Edge Computing(MEC)has become an effective supplement and improvement to the mobile healthcare system supported by Mobile Cloud Computing(MCC).Although the energy consumption of healthcare devices has been reduced to a certain extent through task offloading,the battery capacity limitation of healthcare devices is still a key bottleneck in the development of Io T applications.Furthermore,there are few researches on the security and privacy issues of task offloading.The attacker can infer the wireless channel information by monitoring the task offload rate received by the MEC server to reveal the location information of the mobile user.Therefore,this thesis mainly conducts the following research work on the problems of insufficient energy in healthcare systems and privacy leakage in the process of task offloading:(1)Aiming at the problem of insufficient battery energy in healthcare devices,with the optimization goal of minimizing the task computation cost of the healthcare system,an energy-aware task offloading scheme based on Reinforcement Learning(RL)is proposed by modeling the goal problem as a Markov Decision Process(MDP).The proposed offloading scheme optimizes the task offloading strategy according to the current battery level,the current task volume and the energy harvesting level,thereby minimizing the task computation cost and improving the computation performance of the healthcare system.Simulation experiments show that the proposed task offloading scheme can achieve better results than the comparison schemes under different tasks and different numbers of users.It can achieve the purpose of reducing the computation cost of the healthcare system.(2)Aiming at the problems that the green energy harvesting technology is usually difficult to ensure the stability of energy supply and the location privacy leakage during task offloading,an energy-aware and privacy-preserving task offloading scheme based on Deep Reinforcement Learning(DRL)is proposed.The proposed offloading scheme optimizes the task offloading rate by analyzing the wireless channel gain,the level of battery power,the amounts of harvested energy and the generated tasks,which improves the computation performance of healthcare system and protects the user location privacy.Moreover,we consider the impact of RF-enabled Wireless Energy Transfer Access Points(WET-APs)on location privacy leakage for the first time.Simulation results show that our scheme can reduce the computation latency and improve the privacy protection level of users.