Research On Joint Uplink And Downlink Resource Allocation And Traffic Access Technology Based On F-OFDM

The 5G is approaching,people's requirements for service types and service quality are getting higher and higher.Therefore,it is very necessary to research that how to allocate the limited spectrum resources to meet the business needs of individual users.MIMO technology can increase the system capacity without the additional system bandwidth.However,mobile equipments are usually equipped with very few antennas,making it difficult to construct a MIMO system.Thus,we can consider multiple mobile equipments as one entity with multiple antennas to form a MU-MIMO system.The traditional OFDM technology sets the same parameters for the entire frequency band,and the out-of-band leakage is severe,which makes it difficult to support diverse service requirements in 5G.Based on the OFDM technology,the F-OFDM technology divides the system frequency band into multiple sub-bands and adds sub-band filters at each end of each sub-band.Then,the purpose of setting different parameters according to different service requirements on each sub-band is realized,and the spectrum utilization rate is improved.In addition,the cellular network includes uplink and downlink,and they have their own service requirements and physical resources.It is obviously limited to study only one link.Therefore,we studied the joint uplink and downlink resource allocation and traffic access problems based on F-OFDM.First,when the initial channel state information is known in one scheduling period,the channel state information of each subframe is predicted according to the Gauss-Markov channel model.Further,the mean channel capacity of the uplink and downlink in the scheduling period can be calculated by the statistical method.Then,the concept of Qo S exponent is introduced.At the physical layer,according to the effective bandwidth requirement of each user,and introducing uplink and downlink subframe configuration factor,we establish the resource allocation model based on the principle of on-demand allocation.At the MAC layer,the relationship between the price of a single item and the demand of goods in economics is used to give the utility function of the system,and then the effective capacity constraint of each user is combined to establish the traffic access model.Finally,for the two models of the physical layer and the MAC layer,this paper uses a cross-layer interactive iterative algorithm to transfer the results of the two models until they reach a stable state.For the resource allocation model at the physical layer,this paper firstly uses the Fast Unfolding community detection algorithm to classify the network topology map composed of users and resources,and then uses the iterative Hungarian algorithm to detect each community and obtain the optimal resource allocation strategy.For the traffic access model at the MAC layer,because the utility function is convex for the traffic access variable,and the traffic access variable is subject to the effective capacity,this paper uses the gradient iterative method based on the effective capacity constraint to solve the problem,and obtains the optimal traffic access strategy.Furthermore,this paper uses MATLAB simulation to verify the superiority and stability of the proposed algorithm in terms of convergence,access rate and complexcity,and obtain the optimal uplink and downlink subframe configuration factor under the parameter setting.