| With the development of Internet technology,the demand for wireless data services is dramatically increasing,the future wireless mobile communication system confronts the problem of limited spectrum resources.In order to provide users with faster,more reliable,more energy efficient data transmission services,further exploration of spectrum resources and spatial multiplexing capability of multiple antennas is a key way to solve the above problems.The millimeter wave(mmWave)band has a large number of free resources and has not been fully developed.At the same time,massive multiple-input multiple-output(MIMO)systems can further exploit the spatial multiplexing capability of multiple antennas,and greatly improve spectral efficiency and overcome the severe path loss in mmWave transmission.Therefore,mmWave communication technology,as one of the core technologies for 5G or beyond(Fifth Generation/B5G)wireless communication systems,is the consensus between industry and academia.In 5G or B5G mmWave high speed mobile communication systems,ultra-high mobility,low delay,ultra-dense user distribution and channel time variation will lead to Doppler effect,high penetration loss,increased power consumption of frequent switching,reduced data transmission rate and high computational complexity of traditional channel estimation schemes.Considering the issues including channel physical parameter estimation and other issues in mmWave massive MIMO systems under different time-varying scenarios,and the target of improving spectrum efficiency,high efficiency and low power communication requirements,this dissertation studies channel estimation in different communication scenario,different antenna structures and different movement states by using system model establishment,algorithm design,performance analysis and simulation verification.The main contributions of the dissertation include:(1)In fast time variying scenarios,since mmWave signals are seriously interfered by noise,traditional algorithms cannot extract useful signals from time-varying channels.Especially it is necessary to further eliminate noise for estimating enough paths from time-varying channels with low signal-to-noise ratio when the number of users is increasing,and further denoise for the paths seriously disturbed by noise.The channel estimation algorithm based on noise elimination-based is presented in this paper.Specifically,we firstly use the iterative cancellation method to initially estimate all path parameters.Then,to further improve the estimation accuracy,a decision threshold is set to determine the authenticity of the estimated paths,which can effectively eliminate noise interference.Furthermore,an auxiliary judgment is analyzed to successively select the path with the minimum comparison value until a sufficient number of real paths are selected.Theoretical analysis and simulation results show that the performance of the proposed algorithm is significantly superior than that of the traditional algorithm,especially,the accuracy of channel estimation is improved effectively at low SNR,which is more suitable for multi-user communication scenarios under fast time-varying conditions.(2)The mmWave beamspace channels with special sparse structure cannot be modeled by the one-order Markov process.Especially,the conventional schemes cannot be directly extended to mmWave beamspace massive MIMO systems,and the conventional category of schemes is mainly designed for point-to-point,static or slowly time-variant scenarios.By considering the fast-moving vehicle scenario between a base station and a mobile user,the mathematical model of the fast time-varying channel is first established.The conventional spatial channel can transform into a beamspace channel by employing DLA,which can significantly reduce the dimension of the MIMO systems and the number of required RF chains,and the computational complexity and pilot overhead are significantly reduced.We first excavate a temporal variation law of the physical direction between the base station and each mobile user.Then,based on the temporal variation law and the special sparse structure of mmWave beamspace channels,the obtained beamspace channels is utilized in the previous time slots to predict the prior information(PI)of the beamspace channel in the following time slot.Finally,based on the obtained PI and the sparse structure of beamspace channels,the time-varying beamspace channels can be estimated.Compared with traditional algorithms,the proposed algorithm only needs to estimate the channel parameters of the first three time slots of the time-varying channel,and the whole time-varying channel can be estimated by using the PI of the time-varying channel.Theoretical analysis and simulation results show that the scheme can not only estimate the time-varying channel information of multiple fast users with high accuracy,but also significantly reduce the computational complexity.Therefore,the scheme is suitable for the multi-user communication system of 5G mmWave time-varying channel.(3)The large Doppler effect caused by the user's high-speed movement seriously affected the channel estimation performance.We propose the nested sampling strategy based antenna selection for channel estimation is presented in this paper,and then we extended it to handle the Doppler shift.First,the time-varying channel model for mmWave massive MIMO systems in high-speed mobile scenarios is established.Then,the receiver can only observe a low-dimensional projection of the received signals due to the huge gap between the numbers of radio-frequency(RF)chains and antennas in the hybrid architecture of mmWave.We employ a switch network for analog design,which is equivalent to an antenna selection process.A nested sampling strategy is used to formulate a virtual array with a larger aperture,aiming to reduce the number of RF chains and the complexity of system.Finally,based on the covariance fitting criterion,a joint Doppler and channel estimation method is proposed without need of discretizing the angle space.Compared with the traditional scheme,the proposed scheme can completely eliminate the model mismatch effect and significantly reduces the computational complexity.Theoretical analysis and simulation results show that the proposed algorithm based on nesting sampling can reduce the number of RF chains and power consumption of hardware,and significantly improve the accuracy of time-varying channel estimation,which is suitable for high-speed time-variant mmWave communication system. |
PDF Full Text Request