Research On Resource Allocation Strategy In Non-orthogonal Multiple Access

Non-orthogonal multiple access technology(NOMA)Due to its high performance,low latency,increased user access and compatibility with the existing 4G architecture,it has become a widely studied technology in 5G.Power allocation is a key part of NOMA.User grouping is the precondition for power allocation.However,existing user grouping algorithms and power allocation algorithms have certain limitations,so the research on user grouping algorithms and power allocation algorithms is important Meaning.Existing user grouping algorithms have local'optimize and cannot solve the problem of user dynamic changes.This thesis proposes a dynamic grouping algorithm based on fairness and a user grouping algorithm based on minimum projection according to the principle of user grouping and the characteristics of users' dynamic changes.The fairness-based grouping algorithm takes fairness as a constraint.When the user location or the number of users in the cell changes,users can be dynamically grouped to avoid falling into local optimize and solve the problem of user dynamic changes.The grouping algorithm based on the minimum projection is grouped with the goal of minimizing the user's projection distance,and the matrix calculation of the similarity and gain difference between users is converted into the calculation of the projection distance,which greatly reduces the complexity of the algorithm.The simulation results show that the performance of the proposed algorithm is better than the traditional grouping algorithm.Existing power allocation algorithms have problems such as high complexity,fixed allocation factors that make it unable to adapt to dynamic changes,and resource allocation introduces new interference.This thesis proposes power allocation algorithms suitable for different scenarios according to the 5G channel model standard.The power allocation algorithm based on channel state information is suitable for low-speed mobile scenarios.The algorithm builds an optimization model based on the user's specific channel state information,and solves the power allocated to the user by the Lagrangian multiplier method.The power allocation algorithm based on Markov chain is suitable for the scene of line-of-sight transmission under high-speed movement.The algorithm establishes a directed graph to predict the position of the moving user,predict the channel state based on the position information,and power allocation through the channel state information.The power allocation algorithm based on BP neural network is suitable for non-line-of-sight transmission scenarios under high-speed movement.The algorithm predicts the channel state for a period of time in the future based on the user's historical channel state information,and then allocates power according to the user's needs.The simulation results show that the performance of the proposed algorithm is better than the traditional allocation algorithm.