Joint Resource Allocation For Mobile Edge Computing Networks With Massive MIMO And Wireless Power Transfer

The mobile edge computing(MEC)network creates a high-performance,lowlatency and high-bandwidth service environment by deploying edge cloud servers on the nearest wireless access network,providing users with information processing and cloud computing capabilities.The combination of MEC and massive MIMO technology can significantly improve spectrum efficiency while enhancing user calculating ability;the combination of MEC and wireless energy transfer technology can optimize the energy consumption of MEC network based on ensuring green energy supply for users.However,due to the coupling of communication,computation and power supply,system performance depends on the joint optimization of the communication process,the offloading process,and the energy transfer process.In order to improve the delay of MEC system,improve spectrum efficiency and reduce system energy consumption,this paper studies the communication and computing resource allocation and power control in MEC networks based on massive MIMO and MEC networks based on wireless energy transmission.The main contents summarized as follows:(1)The joint resource allocation and power control strategy of MEC networks based on massive MIMO is studied.For msaaive MIMO-MEC networks,the spatial division multiple access(SDMA)protocol is adopted,and the computing tasks of each user need to be offloaded to the edge server for processing,and the task offloading and processing must be completed within the task deadline.Massive MIMO systems employ pilot-assisted channel estimation to obtain channel state information for offloading decisions.The joint pilot transmission power,data transmission power,and computational resource allocation algorithm that minimize the maximum delay of all users are studied.In view of the non-convex and non-linearity of the optimized model,an improved fruit fly optimization algorithm based on the penalty function steepest descent method is proposed.The algorithm uses the approximate optimal solution obtained by the penalty function steepest descent method as the local search initial point of the fruit fly optimization algorithm,reduces the population size and iteration number of the fruit fly optimization algorithm,and reduces the algorithm complexity.The relationship between the performance of the algorithm and the number of users,the amount of data transferred by the computation task,the computing resources,and the number of antennas is simulated and analyzed,and the complexity of the algorithm is evaluated.The superiority of the proposed algorithm is verified.(2)The joint resource allocation and power control strategy of MEC networks based on wireless energy transfer is studied.For the WPT-MEC network,the time division duplex orthogonal frequency division multiple access(TDD-OFDMA)protocol is adopted.The user energy and computing resources all comes from the base station,the computing tasks of each user need to be offloaded to the edge server for processing,and the task offloading and processing must be completed within the task deadline.Research on joint channel allocation,uplink and downlink time allocation and power control algorithms that minimize user energy consumption.In view of the fact that the optimized model is a mixed integer nonlinear programming(MINLP)problem,it belongs to the typical NP-hard,and the heuristic algorithm is proposed.The decomposition thought is used to relax the original problem to the energy efficiency power allocation under a given channel allocation,and further relax the energy efficiency power distribution to the energy efficiency power allocation under the given uplink and downlink time allocation.Firstly,the dual conversion is implemented based on the water injection algorithm,and then the convex optimization theory is used to find the optimal solution,and then use the fixed-step one-dimensional search to solve the optimal uplink and downlink time allocation.The relationship between algorithm performance and average user energy consumption,system energy consumption,user energy consumption and channel utilization is simulated.The proposed algorithm can significantly reduce energy consumption.