A Research And Validation Of Mu-Mimo Scheduling Strategy Based On Limited Feedback

With the development and application of mobile Internet, data traffic in mobile cellular systems is in exponential growth, the efficiency of single-link has been close to the Shannon limit. But multi-antenna technology can effectively improve network capacity, making multi-user multiple-input multiple-output (MU-MIMO) has become one of the core physical layer technology of next generation mobile communication system standard (IMT-Advanced).Taking the actual deployment into consideration, the channel quality of users is different. Maximizing system throughput, assuring the fairness of user enjoying service under different channel quality and effectively reducing the interference between users of MU-MIMO are the concerned problems of scheduling strategy. Meanwhile, the complexity, system overhead and performance gains need to be balanced. Based on the above background, this paper focuses on the user selection and multi-user pairing algorithm based on limited channel state information feedback in LTE-A framework, proposes a solution assuring the fairness of resource allocation and taking the system overhead and implementation complexity into consideration. The main content includes:First, a research of precoding technology and channel state information feedback under the assumption of SU-MIMO and MU-MIMO is made. This paper makes analysis of the existing MU-MIMO scheduling strategy and through system-level simulation, compares the performance of different algorithms.Secondly, this paper proposes a long-term average user throughput algorithm based on proportional fair scheduling strategy and carries out derivation and analysis under SU-MIMO and MU-MIMO scenario. Proving that the cumulative distribution function of user data rate and the number of users within a cell can be used to effectively estimate the long-term average user throughput, meanwhile, the MU-CSI pairing algorithm based on limited feedback were studied. This paper proposes two MU-MIMO pairing algorithms based on MU-CSI (CQI) orthogonal PMI&RI set, respectively, exhaustive search (MOPS), and pairing cluster search (MOPC).Finally, based on the above research on user selection and multi-user pairing, this paper proposes a joint scheduling strategy under heterogeneous networks of macro and small cell, summing up the computing complexity and system overhead of different pairing strategies. Through system-level simulation, verify the balance of proposed algorithm among MU-MIMO performance, user scheduling complexity and system overhead. The simulation results are given to demonstrate the superiority of the algorithm.