Research On NOMA Power Allocation And User Pairing Based On Fairness

The boom development of the Internet has effectively improved people's quality of life,and people's demand for traffic has also increased exponentially.In particular,with the proposal and development of IoT?Internet of Things?,and in order to build a grand blueprint for interconnection of all things,it will be challenging for us to enhance rate in the communication network transmission.The rapid development of new business requires more spectrum resources and higher spectrum utilization,and this will pose a huge challenge to scarce frequency resources.Therefore,the NOMA technology with higher rate and higher spectral efficiency has become the focus of the research,and it is also the focus of this thesis.In many of the current research results,power allocation and user pairing of downlink NOMA systems has become a hot topic,but there are still some shortcomings.Firstly,there are still some shortcomings in the research of user fairness:?1?The unilateral consideration of the maximization of Proportional Fairness?PF?,and it does not involve the constraints of the user's Quality of Service?QoS??2?Considering the user's QoS and rate maximization,or taking into account the maximization of the proportionality fairness factor and the user's lowest data rate,although,which guarantes the fairness of the edge user,but the rates of the system are still below the traditional orthogonal access technologies under the same conditions.Next,there are still some shortcomings in the user pairing:?1?the global optimal algorithm is too complicated;?2?some articles only consider the optimal rate but ignore the pairing condition,which result in two users whose channel conditions are close to each other are paired,so it's more difficult for SIC?Successive Interference Cancellation?to demodulate.The author carries on the thorough research to the above insufficiency,the main work and innovations of this thesis are summarized as follows:?1?In order to ensure the quality of service for users,and to improve the fairness of users in the edge cell,a NOMA proportional fair power allocation strategy is proposed.According to the proportional fairness method,the goal was to maximize the proportionality fairness factor among the users with the least rate constrains in multiplexed users.The nonconvex objective function was transformed into convex function in theoretical analysis,and the optimal solution of problem was obtained by KKT optimal constraint condition.In the condition of promising the sum rate greater than that in the traditional orthogonal multiple access,the power allocation scheme effectively improves the rate of the cell edge users and ensures the fairness of the edge users.?2?In order to reduce the complexity of the algorithm effectively and avoid the pairing of users with similar channel state information,a user pairing algorithm for channel difference and a sub-band selection algorithm based on optimal sum rate are proposed.In downlink,if the total number of users y is greater than the threshold x?x indicates the upper bound of the number of users of the KM algorithm?,a dichotomy is used to divide the user set.Next,the sub-sets are sorted according to the descending order of the channel conditions,and each sub-set is divided into four groups,and then the channel difference matrix is constructed by cross-combined.The KM algorithm is used to solve the matching result.The complexity of the algorithm is?₄?⁴?7)2)₂?.In the case of sub-band selection,the bilateral mutual rules are adopted.Each pair of matched users selects the sub-band based on the priority of channel condition.However,in this case,multiple matched users may select the same sub-band.So,this sub-band has the right to choose the one with the highest rate by the proportional fairness of power allocation method and rejects the other set of users.Simulation results show that this algorithm can effectively reduce the complexity and avoid the matching of users with similar channel conditions.