Millimeter Wave Massive MIMO Beamforming And Its Security

In this paper,millimeter wave(mm Wave)massive multiple input multiple output beamforming and physical layer security are studied.The detailed work is concluded as follows.A new hierarchical codebook is designed for beam training in mm Wave communication,and the beam training algorithm with limited feedback is proposed based on the designed codebook.Different from the existing hierarchical beam training algorithms which typically require a feedback in each layer of the codebook to indicate the best codeword,beam training in this paper only needs two feedbacks in total.The proposed algorithm divides the hierarchical beam training into two stages and requires only one feedback in each stage by using the designed codebook.In the first stage,beam training is performed by using the codewords in the top layer of our designed codebook,which narrows the search range of the channel angle of departure(Ao D).In the second stage,based on the feedback information from the first stage,codewords in the other layers of our designed codebook are used for beam training,and the transmitter can obtain an estimate of the channel Ao D based on the two feedbacks.The simulation results show that,compared with the existing hierarchical beam training algorithms,the proposed algorithm can significantly reduce the feedback overhead while the success rate and multiuser averaged sum-rate remain almost unchanged.Mm Wave multicast wireless communication and physical layer security in the presence of a eavesdrop-per are also studied by designing and optimizing beamforming for confidential signal and artificial noise.(a)As for confidential signal,beam training is used to find the best codeword to match the corresponding user and its two neighboring codewords to form a composite beam.By minimizing the variance of beam gain within the composite beam coverage,the composite beam is optimized offline,and the channel gain is esti-mated based on the optimized composite beam.Then,the antenna array is divided into multiple sub-arrays according to the max-min fair criterion to reduce the computational complexity of beamforming algorithm.Specifically,the size of each sub-array is determined by the estimated channel gain,and the corresponding phase factor for each sub-array is optimized.(b)As for artificial noise,beamforming is designed and opti-mized by minimizing the interference to legitimate users under the constant envelope constrain and limited phase resolution of phase shifter.In addition to the alternately phase search algorithm,a beamforming algo-rithm based on the null space of legitimate users' channel matrix is proposed.The simulation results show that the proposed confidential signal beamforming algorithm can greatly reduce the computational complex-ity with limited performance loss compared with the existing algorithms? furthermore,the proposed artificial noise beamforming algorithm can effectively improve the secrecy rate of wireless communication system under the hardware constrains.