Research On Resource Collaborative Optimization Based On NOMA-MEC System

The development of 5G mobile communication system has put forward higher requirements on ultra-reliability,low latency and massive communication,which push a series of new critical technologies emerged.Non-orthogonal multiple access(NOMA)as one new multiple access technology,allows multiple users to share the same frequency band resources simultaneously,which effectively improves the utilization efficiency of frequency resources.Mobile edge computing(MEC)supports mobile users to offload computation-intensive tasks to the edge servers for computing,which not only relieves the pressure of limited computing resources for mobile users,but also makes up for the disadvantage of long offloading latency in cloud computing.The introduction of NOMA provides new research ideas for the development of MEC,combining NOMA with MEC not only adapts the high demand of delay and spectrum utilization of future communication networks,but also provides new ideas for some problems and challenges of existing communication networks.Therefore,based on the NOMA-MEC system,this paper conducts some research for improving the offloading experience of MEC network users and the information freshness of real-time application devices respectively.Firstly,in order to improve the offloading experience of edge users,and to utilize effectively the uneven distribution of computing and communication resources of the MEC network,a system model based on NOMA-MEC system with near-far MEC server cooperative offloading is proposed,which improves the reliability of the user's remote offloading.Aim to minimize the system total energy consumption,an optimization model is established to jointly optimize task division,time management,local CPU frequency and power allocation.The proposed problem is a non-convex problem,so a joint communication and computing resource iterative optimization algorithm based on Taylor approximation and alternation is proposed.The algorithm deduces the closed-form optimal solutions of four optimization variables in three steps.Simulation results show that the performance of the proposed scheme based on collaborative NOMA is better than other benchmark schemes,and the proposed algorithm has lower complexity and fast convergence speed.Secondly,in order to promote the information freshness of real-time application which has multi-class status updates,maintaining the fairness of all class of status updates information to improve the decision-making accuracy of the control receiver,this paper propose a NOMA-MEC status updates system for the real-time equipment users with two classes status updates information,and defines the fair equilibrium function to measure the information age equality of the two classes of update data.Taking the system average peak age as the optimization objective,an optimization model on power allocation and offloading management as the joint optimization variables,and the information age equity as the constraint is established.The proposed problem is a nonconvex problem and continuous randomly,then a fairness update-offloading strategy algorithm based on discrete method is proposed.The simulation results show that the proposed model and update-offloading strategy algorithm can effectively improve the freshness of the system and the age fairness of two status updates.