Research On Coordinated Beamforming Technique In Coordinated Multipoint Transmission System

Since the interference between cells in LTE system is so severe that the existinginterference coordination technologies cannot resolve the low spectrum efficiency problem ofedge users, LTE-A introduces the Coordinated Multi-Point transmission/reception (CoMP)technology. CoMP can improve the system capacity and increase the spectrum efficiency ofedge users in the cell by employing Joint Processing (JP) and Coordinated Beamforming (CB),which can reduce interference from the adjacent cells and improve the received signal qualityof users. Beamforming is an important technique to improve the spectrum efficiency in nextgeneration wireless cellular networks, and based on the research of multi-user beamformingtechnique, this paper mainly studies CB technology of CoMP.On the condition that the coordinated base stations has known ideal channel informationof users, coordinated beamforming algorithms based on zero forcing(ZF-CB) andsignal-to-leakage-and-noise ratio (SLNR-CB) are studied. ZF-CB algorithm can make theinterference to edge users in the adjacent cells completely eliminated in the base station side,but the number of transmit antennas of the base station is subjected to certain constraint.SLNR-CB algorithm overcomes the above constraint and is easy to solve. Compared withnon-cooperative case, simulation results show that both algorithms manifest goodperformance, and when the number of transmitting antennas of the base station does not meetthe condition of the ZF-CB algorithm, SLNR-CB algorithm can still achieve goodperformance.When only limited channel information of users has been known by the coordinated basestations, two coordinated beamforming algorithms which are based on extended zero-forcing(EZF-CB) and SLNR (SLNR-CB) are researched by utilizing estimated channel vectorfunction. The channel vector function is estimated by making jointly use of limited feedback channel norm and the index number of the preferred codeword in the channel phase codebookand channel spatial correlation statistics. To this end, based on unitary matrix, a kind ofchannel phase codebook is constructed with link adaptation capability. EZF-CB algorithm hasno limit on the number of transmitter antennas. Simulation results show that the performanceof the EZF-CB algorithm is close to that of the SLNR-CB algorithm with the increase oftransmission power, and as the codebook size rises, the performances of both algorithmsgradually approach that of the ideal channel coefficient.