Resource Allocation Algorithm Based On System Utility In MEC Networks

In recent years,with the rapid development of computationally intensive applications and the massive growth of data traffic,mobile devices are not able to meet the user's demand for computing power and storage capacity due to volume and battery capacity limitations.Mobile Edge Computing(MEC)is considered to be a key technology to improve the computing power of mobile devices.By deploying computing resources to the edge of the mobile network,users can quickly and steadily use the deployed computing resources to process computing tasks.Mobile edge computing technology can effectively reduce the delay and energy consumption of users processing computing tasks.Resource allocation is one of the key technologies in mobile edge computing networks.As the number of access devices increases and users'demand for resources increases,MEC networks need to continuously optimize resource allocation algorithms.Lots of works about resource allocation in MEC have been done,but there are still problems like inadequate utilization of resources and complex computation.In view of the above problems,this paper conducts research on the algorithm based on system utility optimization in the MEC network to obtain a superior solution.Firstly,a distributed resource allocation algorithm based on alternating direction multiplier method is proposed in MEC considering cognitive technology.Firstly,the system considers the attachment of secondary users,the calculation of unloading decisions and the allocation of resources jointly,and plans the problem as a non-con vex problem to maximize the system benefit function.Secondly,the problem is transformed into a solvable convex problem by dilution of binary variables and substitution of product terms.In order to reduce the complexity of the algorithm,a distributed iterative algorithm based on alternating direction multiplier method is proposed to optimize the system efficiency.Finally,the simulation results show that the proposed algorithm reduces the complexity of the algorithm and improves the system efficiency.Secondly,in mobile edge computing networks,a resource allocation algorithm based on deep reinforcement learning is proposed.Firstly,the network is virtualized,based on which communication model,network service model and computational unloading model are established.Secondly,a resource allocation algorithm based on deep reinforcement learning is proposed.In the proposed algorithm,the system's benefits and resource consumption through users are designed as rewards and losses in deep reinforcement learning.Through the training of neural network,the algorithm can obtain the highest reward for resource allocation in exploring learning.Finally,through simulation and comparison of various neural network parameters and training models,it is proved that the proposed model can achieve higher system benefits.