Scheduling Algorithm Research On Coordinated Communication Technology

As the evolution technology of LTE, LTE-A provides higher data rate, higher spectral efficiency, and broader coverage. As performance improvement of cell edge users become one of the major research field in LTE-A, Multi-point Collaborative Communication (CoMP) is the key technology to solve this problem. CoMP techniques utilize the coordination of multiple transition points, to perform transmission and reception for one or more users, thus significantly enhance cell throughput and cell-edge users’data rate. The growing scarcity of radio spectrum puts forward higher requirements on channel resource utilization, and scheduling is the key point of resource allocation in CoMP system and the core of realizing high spectrum efficiency.This thesis firstly elaborates CoMP technology in LTE-A, including the emergence and development of CoMP, LTE resource scheduler, and high light is throw upon scheduling algorithms at last. Based on analysis of the existing conventional scheduling algorithms, this thesis researches on the scheduling algorithm in uplink coordinated communication scenario.First a joint scheduling (JS) algorithm based on projection matrix is proposed. According to the defined scheduling factor, the scheduling algorithm extends the UE set by sequentially admitting UE interferes the least and suffers from low average throughput to the group, therefore a fair scheduling approach in which users in a cluster interfere slightly is obtained. Simulation results shows that the proposed scheme achieves a better tradeoff between system capacity and throughput fairness while reducing the performance-complexity significantly. A scheduling algorithm for uplink MU-MIMO system is put forward. The prosed scheduling algorithm is a hierarchical scheduling scheme based on improved determinant pairing scheduling and joint reception. The proposed algorithm introduces an adjustable threshold to distinguish center UEs and cell-edge UEs. An improved pairing scheduling is performed to UEs with fine channel condition. Joint reception is performed to cell-edge UEs by employing semi-static cluster. The simulation results demonstrate that the proposed hierarchical scheduling algorithm precedes DPS and JR in system throughput and user throughput fairness with better flexibility.The last part of this thesis reviews and summarizes the research work and discusses the further research directions.