Research On Interference Coordination Schemes With Limited Cooperation For4G Network

Orthogonal Frequency Division Multiplexing (OFDM) and Multiple Input Multiple Output (MIMO) are both adopted to increase the spectrum efficiency in Fourth Generation (4G) mobile cellular system. However, with dense frequency reuse scheme (reuse factor of1), Inter-cell Interference (ICI) will degrade the achievable spectrum efficiency of OFDM and MIMO seriously.In order to improve the achievable spectrum efficiency, especially for the cell edge user spectrum efficiency, interference coordination is seen as an important technique for4G cellular systems. Generally, interference coordination schemes needs the coorperation of several adjacent Base Stations (BSs). According to the cooperative level of BSs, interference coordination schemes can be divided into two categories: interference coordination with full cooperation and interference coordination with limited cooperation. For interference coordination with full cooperation, the data to a User Equipment (UE) is transmitted from several BSs simultaneously, and the data and Channel State Informaton (CSI) of every UE should be shared among multiple BSs; while for interference coordination with limited cooperation, the data to a UE is transmitted form a single BS, and only the CSI and scheduling information is needed to be shared. Although interference coordination with full cooperation can transfer the interference signal seen with single cell processing to useful signal, the full sharing of data and CSI will bring large amount of signalling overhead. Since the capacity of backhaul link among BSs is limited in realistic systems, interference coordination with limited cooperation gains much more attention from standardizing forum and research institute. We also focus on interference coordination with limited cooperation in this thesis.Since Orthorgonal Frequency Division Multiple Access (OFDMA) and Spatial Division Multiple Access (SDMA) are adopted based on OFDM and MIMO in4G system, we study the interference coordination with limited cooperation in frequency domain and spatial domain separately. The problems that are investigated in this thesis are:dynamic sub-band allocation in frequency domain, transmit mode selection in spatial domain, and linear precoding matrix designing in spatial domain. We propose four innovation ideas in this thesis, and the main research work and achievements of this thesis are listed as follows:(1)An Uplink Coordinated Scheduling scheme based on Resource Sorting (UCSRS) is proposed to deal with the problem of dynamic sub-band allocation in frequency domain. Firstly, we analyze the reason for uplink interference fluctuation and its effect on the throughput. Based on the analysis, the resource sorting and transmit power allocation is used in UCSRS to suppress the effect of interference fluctuation on the users’ throughput. Moreover, the characteristic of dispatity in the users’ability to bear interference is used to avoid the users’Signal to Interference plus Noise Ratio (SINR) being too low to be served. Simulation results show that UCSRS can suppress the ICI and uplink interference fluctuation effectively. Meanwhile, UCSRS can effectively improve the average cell spectrum efficiency and cell edge user spectrum efficiency.(2) A downlink transmission mode selection scheme with eigenvalues of interfering channels (EVITMS) is proposed. Firstly, we deduce lower bound on the capacity based on the eigenvalues of interfering channels, and find that the tendency of the lower bound on the capacity changing with cell edge SNR and users’location agrees with that of the average capacity. With the lower bound on the capacity, the global optimizing problem of downlink transmission mode selection is transferred into a local optimizing problem of receive antenna selection, and a heuristic algorithm, named as EVITMS, is proposed. With EVITMS, the receive antennas that suffer high interference will not be activated. Simulation results show that EVITMS can effectively increase the SINR of active antennas. Meanwhile, the average cell spectrum efficiency and cell edge user sepetrum efficiency can be improved too. Moreover, compared with the transmission mode selection with full interferring information feedback, EVITMS can get96%achievable average cell spectrum efficiency of the scheme with full interferring information feedback. However, the signaling overhead brought by scheme with full interfering information feedback times that of EVITMS by T2M, in which T is the iteration times, and M is the number of receive antennas of a single UE.(3) A downlink distributed precoding method based on Minimum Interference Leakage (PMIL) and a downlink adaptive freedom allocation scheme with PMIL (PMIL-AFA) is proposed. Firstly, the problem of downlink decentralized precoding matrix designing is formulated as a joint precoding and freedom allocation problem. In order to solve the joint optimization problem, we divide it into two sub-problems: precoding matrix designing with fixed freedom allocation and adaptive freedom allocation. With fixed freedom allocation, PMIL can nullify the Inter-User interference and minimize the sum of interference leakage to users served by adjacent cells. Thus, the SINR of every interfered user can be improved. With PMIL, we deduce the lower bound on the capacity variation with the freedom allocation varies. Based on the lower bound on the capacity variation, PMIL-AFA is used to adjust the freedom utilized to mitigate the ICI dynamically. Simulation results show that PMIL-AFA can effectively increase the average cell spectrum efficiency and cell edge user spectrum efficiency.