Terahertz Beam Tracking Technology Based On Antenna Array

Because of its ultra-large bandwidth,the terahertz frequency band can effectively alleviate the increasingly serious pressure on spectrum resources of modern wireless communications and provide ultra-high data transmission rates.However,compared to low frequency bands,the propagation of terahertz waves will experience more severe path loss,which limits the coverage of terahertz communications.Terahertz beamforming technology uses beams with high directional gain to increase the coverage distance of terahertz communications.However,the higher the directional gain,the narrower the beam,which puts forward higher requirements for beam alignment and tracking.Especially,when the user is in motion,it is more difficult to align and track the beam.How to quickly and accurately perform beam pairing and tracking is one of the main problems that need to be explored in terahertz communication.For this reason,this paper studies the problem of extremely narrow beam tracking in terahertz communication.Existing beam tracking technologies can be roughly divided into two categories: the first category is beam tracking based on a priori information of user motion,and the second category is beam tracking based on codebook search.The first type of method models user mobility,and uses methods such as Kalman filtering for beam tracking.However,the kinematic model assumed by the existing literature is not accurate enough.The second type of method uses a codebook designed in advance to search for the best codeword to perform beam tracking.The accuracy of estimation and pilot overhead are two key issues that need to be considered in the beam tracking method.This article proposes corresponding methods for these problems,aiming to further improve beam tracking accuracy or reduce pilot overhead.First,this paper studies the tracking of extremely narrow beams in vibration scenarios.In emergency communication under strong winds,the communication terminal will swing within a certain range when it is interfered by external forces.The displacement caused by this rocking motion will cause the beam direction to shift.In order to deal with this problem,we introduced a vibration model,combined with the Kalman filter method,and improved the state model and observation model.Specifically,a discrete-time recurrence model is established based on the differential equation of motion,and this model is used as a linear observation model in Kalman filtering,and a very narrow beam tracking algorithm based on Kalman filtering is proposed.Compared with the traditional Kalman filter algorithm,the algorithm proposed in this paper has higher beam tracking accuracy,and the narrower the beam,the greater the gain brought by the algorithm proposed in this paper.Second,the paper studies the tracking of extremely narrow beams in parallel moving scenarios,such as pedestrians and vehicles on the road.In a translational scenario,the user's movement trajectory is mainly a straight line or a smooth curve,and within the time interval of beam tracking,its trajectory can be approximated by a line segment.In this paper,the local linear motion law is used,and the most suitable scheme is selected from three codebook schemes.Combined with the channel rotation,a translation-oriented extremely narrow beam tracking algorithm is proposed.Simulation results show that the algorithm has higher beam tracking accuracy through codebook optimization and channel rotation.Finally,this paper studies the problem of terahertz ultra-wideband and extremely narrow beam tracking.When the signal bandwidth reaches a certain level,the different frequency component signals will deviate from the beam direction after passing through the frequency flat phase shifter,causing beam splitting.This beam splitting phenomenon makes effective beam tracking impossible according to the traditional narrowband method.Aiming at the beam splitting problem of wideband signals,Professor Dai Linglong's team from Tsinghua University proposed a transceiver structure based on a true time delay circuit.The true time delay circuit compensates for the deviated phases of different subcarrier signals and overcomes the problem of beam splitting..Based on the structure of the true time delay circuit transceiver,this paper analyzes the relationship between beam splitting and real-time delay circuit parameters,and proposes a multi-level beam training scheme.Specifically,the true time delay circuit controls the beam splitting range,dynamically adjusts the beam direction,and finds the optimal beam at a faster speed.The method in this paper has higher beam tracking accuracy when the signal-to-noise ratio is low.