Research On The Fairness Of Data Streams Based On Hybrid Beamforming

With the rapid development of smart terminal devices and mobile Internet technology,mobile data services have put forward higher transmission quality and system capacity requirements for wireless communication technologies.Millimeter wave communication has become a research hotspot in the field of wireless communication due to its large channel bandwidth.Large-scale antenna array and beamforming technology,which can effectively offset the huge path loss caused by millimeter wave,become the key technologies of millimeter wave communication.Full digital beamforming is difficult to apply to millimeter wave communication due to its limitations such as high implementation cost and high power consumption.Hybrid beamforming combining digital beamforming and analog beamforming can effectively reduce the number of RF links and the computation complexity of the system,and become the important research direction of the millimeter wave communication.Hybrid block diagonalization beamforming usually uses SVD to obtain the parallel subchannels,which can eliminate inter-user interference well,obtain equivalent parallel subchannels to transmit multi-data streams and effectively improve capacity.However,the singular values of SVD become more and more dispersed with the increasing antenna number.It makes the parallel subchannel gain difference large,which can not guarantee the fairness of the data streams of the system.Hybrid multi-stream diversity beamforming,which is based on hybrid block diagonalization beamforming,can reduce the difference between the parallel subchannel gains,thus ensuring the fairness of data streams of the system by processing the data stream separately.The reminder of thesis is organized as follows.1.The hybrid block diagonalization beamforming and SVD-based hybrid multi-stream diversity beamforming in the multi-user massive MIMO system are studied.The beamforming design of the two algorithms are introduced detailedly in the analog domain and the digital domain in order to realize parallel transmission of multiple data streams.Two low-complexity hybrid multi-stream diversity beamforming algorithms are proposed to improve the system performance while reducing computational complexity.2.The capacity of hybrid block diagonal beamforming and hybrid multi-stream diversity beamforming are analyzed and compared.It is proved that the fairness of data streams is guaranteed by hybrid multi-stream diversity beamforming and can effectively increase the system capacity and improve the system performance.The numerical and simulation results in terms of bit error rate,system capacity and computational complexity are presented.3.In multi-user massive MIMO-OFDM system,hybrid block diagonal beamforming and ZF-based hybrid multi-stream diversity beamforming are studied.Firstly,the hybrid block diagonalization beamforming algorithm based on OFDM is introduced.Then a hybrid multi-stream beamforming based on OFDM is proposed.The system capacity analysis of the two algorithms is present and compared.It is proved that in multi-user massive MIMO-OFDM system the hybrid multi-stream diversity beamforming can effectively guarantee the fairness of data streams,which can increase system capacity and improves system performance.