Research On Adaptive Resource Allocation Technology Of Cellular Network Based On OFDMS And MEC

Large capacity and low delay is one of the important features of the Fifth Generation Mobile Communication System(5G).Orthogonal Frequency Division Multiple Access(OFDMA)technology,as a key physical layer technology to support 5G high-capacity communication,can improve wireless spectrum utilization,thus providing users with more high-quality services on the limited resource spectrum.As the key technology of the network layer supporting 5G large-capacity communication,Mobile Edge Computing(MEC)technology can flexibly cope with the problem of mass clustering of network Edge data brought by the explosive growth of user devices,reduce data transmission and processing delay,and improve network service capability.However,there are still some difficulties in how to make full use of OFDMA and MEC technologies to further improve cellular network capacity and reduce service latency.Therefore,this paper focuses on the adaptive and efficient resource scheduling of OFDMA and MEC systems in cellular networks,and discusses the methods to increase the capacity of cellular systems and reduce the network delay under the constraints of power,rate,and number of cache devices.The main research contents and results are as follows:(1)Research on an adaptive resource allocation scheme which can effectively resolve user conflicts and alleviate the interference of up and down links.Aiming at the user conflict when assigning multiple subcarriers among multiple users in the single-cell full-duplex Orthogonal Frequency Division Multiple Access(OFDMA)downlink system,and together considering of the interference between uplink and downlink due to subcarriers multiplexing,an adaptive resource allocation scheme which can resolve above both issues is proposed in this paper.This scheme combines the subcarrier allocation with a greedy algorithm and the power allocation with a iterative water-filling algorithm to solve the optimization problem of subcarrier and power joint distribution:Firstly,the downlink user set for each subcarrier is found through the dichotomy method under the constraint of uplink interference gain,aiming at reducing the complexity of following subcarrier allocation.Then,each sub-carrier is primarily assigned to the user with the highest signal-to-interference-and-noise-ratio(SINR)based on a greedy algorithm under the assumption of equivalent transmitted power.If one user may be matched to more than one subcarrier(user conflict occurs),it selects the one providing with a larger rate.Finally,a feasible optimal power distribution scheme is obtained according to an improved iterative water injection algorithm.Simulation resultsshow that,compared with existing resource allocation schemes,the proposed scheme not only resolves the user conflict in subcarrier allocation,but also improves the efficiency of power distribution and improves the throughput of the downlink system effectively while considering user fairness.(2)Research on the cache replacement and resource allocation scheme of multi-cell multi-mec server.The traditional MEC server cache replacement algorithm only makes a decision on the cache information of a single MEC.Applying it to a multi-mec server system will result in the deletion of useful cache information by mistake,and the user paying more cost to obtain the task result.To solve this problem,this paper introduces the cache information of adjacent MEC in the single MEC server cache replacement algorithm,and proposes an improved Least Frequently Used(LFU)cache replacement scheme.In terms of MEC system resource allocation,the traditional Hungarian algorithm is only applicable to the matching of user tasks and server optimization in the scenario where the user's location is fixed.When the user moves,the distance between the user and the MEC server changes dynamically,the algorithm is difficult to apply.Therefore,an improved Hungarian algorithm is proposed to solve the MEC resource allocation of location mobile users.The simulation results show that the proposed cache replacement scheme can effectively improve the cache hit ratio of users to the MEC server compared with the existing scheme.Compared with the traditional algorithm,the proposed improved Hungarian algorithm can effectively reduce the computation time and has a wider range of application scenarios which is suitable for both user fixed position and user moving.