Dynamic beam alignment and tracking method and system for terahertz communication networking

Abstract

The application discloses a dynamic beam alignment and tracking method and system for terahertz communication networking, relates to the field of terahertz networking communication, and solves the technical problems of initial acquisition difficulty caused by the extremely narrow terahertz beam, low dynamic tracking precision, link loss and difficulty in maintaining stable connection in mobile platform application; the application comprises the following steps: acquiring attitude information; bidirectional low-frequency beacon initial interaction positioning; initial pointing control based on the low-frequency beacon; bidirectional terahertz signal scanning and reference parameter acquisition; bidirectional low-frequency beacon positioning error calibration; bidirectional continuous low-frequency tracking and quality evaluation; fusion tracking and dynamic angle correction of respective MSAWF-CP algorithm execution; the application can not only exert the advantages of precise terahertz signal pointing, but also compensate for the deficiencies by the stability of low-frequency signals, thereby significantly improving the dynamic tracking precision, smoothness and overall robustness.

Description

Technical Field

[0001] This invention relates to the field of terahertz network communication, and more specifically to a dynamic beam alignment and tracking method and system for terahertz communication networks. Background Technology

[0002] Terahertz (THz) communication is considered a key technology for future 6G and more advanced wireless communications due to its ultra-wide bandwidth potential. However, terahertz waves are characterized by extremely narrow beams, high propagation loss, and susceptibility to blockage, making accurate and fast beam tracking a core challenge for maintaining reliable communication links.

[0003] Existing terahertz beam tracking technologies include:

[0004] Beamforming and steering based on antenna arrays: Highly directional beams are formed using phased array antennas or lens antennas. The beam direction can be controlled by adjusting the phase and amplitude of each antenna element in the array in real time; the tracking method involves calculating the optimal beam direction through algorithms and controlling phase shifters or mechanical devices to align the beam with the target.

[0005] Intelligent Reflector Assistance: Deploying reconfigurable intelligent reflectors (RIS) in communication links. RIS consists of numerous programmable electromagnetic units. By changing the reflection characteristics (phase, amplitude) of these units, the reflection direction of the incident terahertz beam can be dynamically manipulated, even achieving beamforming, focusing, or path diffraction. The tracking method involves the RIS acting as a relay or reflector, dynamically adjusting the direction of its reflected beam to track moving users or compensate for congestion in the base station-user link, effectively "turning" the fixed base station beam to the moving user.

[0006] Sensor-assisted beam tracking: Utilizes additional sensor information to predict or assist in determining the optimal beam direction, reducing the overhead of pure channel detection; the tracking method involves inputting sensor data into the prediction algorithm to estimate the user's future position and orientation, and thereby pre-adjusting the beam direction or narrowing the beam search range.

[0007] Closed-loop feedback based on channel sounding: Both communicating parties periodically send known sounding signals, and the receiving end measures channel state information, including received signal strength, angle of arrival, and angle of departure. The tracking method involves comparing continuous CSI measurements to determine channel changes and feeding them back to the transmitting end to adjust the beam direction. Commonly used algorithms include beam scanning, beam differential, and codebook-based beam selection.

[0008] While the above methods can achieve terahertz beam tracking relatively well, they still have the following problems in dynamic environments:

[0009] 1. The narrow beam and limited scanning range of terahertz waves result in long initial target acquisition time and a high failure rate;

[0010] 2. Facing terahertz signal fluctuations or even interruptions caused by relative motion between the communicating parties, platform jitter, or changes in the channel environment;

[0011] 3. The link is prone to loss of lock and it is difficult to maintain a stable connection in mobile platform applications.

[0012] For example, the Chinese patent "An Automatic Calibration System, Method, Device and Medium for Millimeter-Wave Radar" (Patent Application No.: CN202410753692.0, Publication No.: CN118501830A). This patent utilizes a reconfigurable smart surface (RIS) to automatically calibrate millimeter-wave radar, optimizing the reflection characteristics of the RIS to improve the quality of millimeter-wave signals.

[0013] However, this patented technology focuses more on the performance optimization of the radar system itself and its adaptation to specific environments. It is mainly aimed at the open-loop or semi-open-loop optimization of static or quasi-static radar system calibration, and does not solve the most critical problem in terahertz communication: dynamic, fast, and accurate beam alignment and tracking. Summary of the Invention

[0014] To address the problems existing in the prior art, this invention provides a dynamic beam alignment and tracking method and system for terahertz communication networking, which solves the technical problems of initial acquisition difficulties, low dynamic tracking accuracy, easy link loss, and difficulty in maintaining stable connection in mobile platform applications caused by the extremely narrow terahertz beam.

[0015] A dynamic beam alignment and tracking method for terahertz communication networks, where the two communicating parties are each other's local and remote ends, including:

[0016] Step S1: The local end collects inertial measurement data and calculates the attitude information;

[0017] Step S2: The other end sends a low-frequency pulse signal, and the local end uses a multi-antenna low-frequency equipment receiving array to receive the low-frequency pulse signal and calculates the measured phase difference of different antennas to obtain the low-frequency directional angle of the other end relative to the local end.

[0018] The steps S1 and S2 are not in any particular order;

[0019] Step S3: Based on the attitude information of this end and the low-frequency azimuth angle of the other end relative to this end, generate a gimbal control command and adjust the terahertz antenna to roughly point towards the other end.

[0020] Step S4: This end acts as the scanning party, actively transmitting terahertz signals when adjusting the terahertz antenna. The other end acts as the detection party, receiving the terahertz signals and measuring the signal strength, and feeding back the signal strength information to the scanning party until optimal alignment is achieved, i.e., the signal strength reaches its maximum. Record the pointing parameters and maximum power value at this time.

[0021] Step S5: Repeat steps S1-S2 to obtain the low-frequency directional angle of the other end relative to the local end, and compare it with the pointing parameters at the optimal alignment determined in step S4 to obtain the inherent deviation of low-frequency alignment as a calibration coefficient; the pointing parameters at the optimal alignment are determined by the local antenna encoder reading combined with the local attitude information.

[0022] Step S6: Repeat step S2 to obtain the real-time low-frequency azimuth angle and update the local end's estimate of the azimuth angle to the opposite end in combination with the calibration parameters. Output the local end's estimate of the azimuth angle to the opposite end as the preprocessed low-frequency angle and low-frequency signal quality index simultaneously.

[0023] Step S7: Obtain the processing results of the aforementioned process, calculate the adaptive weights using the MSAWF-CP algorithm, fuse the preprocessed low-frequency angle and terahertz power angle deviation to obtain the fused weighted angle of this end to the opposite end direction, and perform rotation alignment based on the fused weighted angle and attitude information.

[0024] Further, step S1 includes: real-time acquisition of the local device's own three-axis acceleration and three-axis angular velocity data, processing them through an attitude calculation algorithm to determine the local device's real-time attitude information representing the local device's attitude at a certain moment, including pitch angle, yaw angle, and roll angle.

[0025] Furthermore, the multiple antennas mentioned in step S2 are at least three antennas.

[0026] Further, step S4 includes: the scanning device performs a fine scan at a small angle near its current pointing direction, and the detection device measures the RSSI in real time; the scanning device adjusts its pointing direction based on the feedback from the detection device until the RSSI received by the detection device reaches its maximum; each device records the pointing parameters when its own antenna achieves optimal alignment during this bidirectional optimization process; and records the maximum power value RSSI measured at this time. THz,max Based on the RSSI variation data during the scanning process, the sensitivity / slope factor of the RSSI-angle curve near the peak is initially estimated. k After this step is completed, the terahertz link between the two parties is established, and the system enters tracking mode.

[0027] Furthermore, the low-frequency signal quality indicators mentioned in step S6 include the signal-to-noise ratio obtained by comparing the intensity of the received low-frequency pulse signal with the background noise power, and the short-term variance of the angle estimate obtained by statistically calculating the preprocessed low-frequency angles continuously output within a time window. The algorithm confidence level (Conf) is derived from the consistency of phase difference measurements, the success rate or uniqueness of ambiguity resolution, and the evaluation of potential multipath effects. AOA,k , .

[0028] Furthermore, the process of obtaining the processing result of the aforementioned process in step S7 includes obtaining the local preprocessed low-frequency angle θ′ output in S6. LF,Remote←Local,k and its quality index SNR LF,Re mote←Local,k (Signal-to-noise ratio of the low-frequency signal received from the other end), short-term variance of the recent AoA estimate from this end to the other end. Confidence of low-frequency AOA algorithm AOA,k ; Obtain the real-time RSSI of the signal power from the other end using the local terahertz receiver. Remote←Local,k Calculate its short-term variance Using signal power RSSI Remote←local,k Maximum power value RSSI THz,max And the predicted sensitivity / slope factor. k The terahertz power angle deviation Δθ was calculated. Thz,Re mote←Locak,k (i.e., the deviation of the antenna at this end from the direction of the antenna at the other end).

[0029] Further, the calculation of adaptive weights in step S7 includes:

[0030]

[0031] In the formula, ε is a small positive constant to prevent the denominator from being 0; T LF T THz ω is an intermediate parameter. LF For low-frequency weights, ω THz For high-frequency weights, f1 and f2 are functions that map their respective quality indicators to trust scores.

[0032] Furthermore, the fusion weighted angle obtained in step S7 by fusing the low-frequency angle and terahertz power angle deviation during preprocessing includes:

[0033]

[0034] In the formula, θ′ LF,Re mote←Local,k For local low-frequency angle preprocessing, Δθ THz,Re mote←Locak,k This refers to the terahertz power angle deviation. To integrate weighted perspectives.

[0035] A dynamic beam alignment and tracking system for terahertz communication networks is set up on both communicating parties, with each party acting as the local end and the peer end. It is used to execute the dynamic beam alignment and tracking method for terahertz communication networks and includes: an attitude sensing module, a low-frequency auxiliary direction finding module, a high-frequency transceiver and sensing module, a beam pointing execution module, and a core control and fusion module.

[0036] The attitude perception module is used to process the collected inertial data and calculate the real-time attitude information of the local carrier.

[0037] The low-frequency auxiliary direction finding module is used to synchronously receive low-frequency pulse signals from the other end, digitize them first, and then process them accordingly.

[0038] The high-frequency transceiver and sensing module is used to generate, amplify and transmit terahertz communication signals via a terahertz antenna, while simultaneously receiving terahertz signals from the other end and measuring the output signal strength.

[0039] A beam pointing execution module is used to radiate and receive terahertz beams with high directivity, and the beam pointing module includes a pan-tilt unit;

[0040] The core control and fusion module receives and processes real-time data from the attitude perception module, the low-frequency auxiliary direction finding module, and the high-frequency transceiver and perception module. Based on the fusion or prediction results and combined with real-time attitude information, it generates the final control command for the gimbal in the drive beam pointing execution module.

[0041] Furthermore, the low-frequency auxiliary direction finding module includes a low-frequency radio frequency beacon transmitter, a low-frequency receiver array, a multi-channel receiving and digitizing unit, and a low-frequency signal processing unit. The low-frequency radio frequency beacon transmitter is used to transmit a known low-frequency pulse signal to the other end according to a preset protocol. The low-frequency receiver array is composed of multiple antenna elements arranged in a predetermined geometric configuration and is used to synchronously receive low-frequency pulse signals from the other end. The multi-channel receiving and digitizing unit is used to amplify, filter, and digitize the analog signals received by each antenna. The low-frequency signal processing unit is connected to the multi-channel receiving and digitizing unit and is used to process the digitized signal.

[0042] The high-frequency transceiver and sensing module includes a terahertz transceiver and an RF power detection unit. The terahertz transceiver is used to generate, amplify, and transmit terahertz communication signals via a terahertz antenna, while simultaneously receiving terahertz signals from the other end. The RF power detection unit is used to measure the intensity of the received terahertz signal in real time and output the intensity value.

[0043] The beneficial effects of this invention include:

[0044] Firstly, addressing the challenges of difficult and time-consuming initial acquisition in terahertz communication systems due to their extremely narrow beamwidth, existing technologies often rely on terahertz scanning over a wide area to search for targets. This is not only inefficient but also prone to failure when the target is moving rapidly or its initial position is highly uncertain. This invention introduces low-frequency assistance, utilizing its wide beamwidth and broad coverage to quickly and reliably acquire the approximate location information of the other end. This transforms the previously blind search into targeted guidance, significantly shortening the initial link establishment time and improving the acquisition success rate. Especially in non-line-of-sight (NLOS) scenarios or those with partial obstruction, the diffraction capability of low-frequency signals allows them to still provide effective initial guidance—an advantage that is difficult to match by simply relying on terahertz scanning.

[0045] Secondly, terahertz beams are extremely sensitive to minute platform jitter, rapid target maneuvers, and atmospheric disturbances. Traditional tracking mechanisms based on terahertz signal feedback experience a sharp decline in performance when signal quality deteriorates. This invention introduces low-frequency assistance, which can continuously provide relatively stable end-angle information. Even when the terahertz signal momentarily fades or is interfered with, the system can still maintain approximate tracking of the target by relying on low-frequency guidance, preventing complete loss of lock. Furthermore, the "Mode Switching-Based Adaptive Weighted Fusion and Compensation Prediction Algorithm (MSAWF-CP)" proposed in this invention constitutes the core innovation of high-low frequency fusion. This algorithm can evaluate the signal quality from the low-frequency AoA system and terahertz RSSI in real time and dynamically adjust the weights of their respective information in the final pointing decision. This intelligent, adaptive fusion mechanism ensures that the system always prioritizes the most reliable sensor information, thereby leveraging the advantages of precise terahertz signal orientation in complex and ever-changing real-world environments while compensating for the shortcomings of low-frequency signals with stability. This significantly improves the accuracy, smoothness, and overall robustness of dynamic tracking, far surpassing schemes that rely solely on high-frequency or simple fixed-weight fusion.

[0046] Furthermore, the symmetrical design of this invention for bidirectional communication links enables both communicating parties to possess complete assisted alignment and fusion tracking capabilities, which is crucial for establishing and maintaining stable and efficient bidirectional terahertz communication. Each end can actively sense the orientation of the other end and intelligently adjust its own pointing, forming a collaborative closed-loop system. Attached Figure Description

[0047] Figure 1 This is a flowchart of a dynamic beam alignment and tracking method for terahertz communication networking, as described in an embodiment of this application.

[0048] Figure 2 These are simulation effect diagrams related to the embodiments of this application. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0050] Example 1

[0051] The following is in conjunction with the appendix Figure 1 Specific embodiments of the present invention will be described in detail;

[0052] A dynamic beam alignment and tracking method for terahertz communication networks, where the two communicating parties are each other's local and remote ends, including:

[0053] Step S1: Obtain attitude information. The local device collects inertial measurement data and calculates the attitude information.

[0054] Step S2: Initial interactive positioning of the two-way low-frequency beacon. The other end sends a low-frequency pulse signal. The local end uses a low-frequency equipment receiving array with multiple antennas to receive the low-frequency pulse signal and calculates the phase difference of the different antennas to obtain the low-frequency azimuth angle of the other end relative to the local end.

[0055] The steps S1 and S2 are not in any particular order;

[0056] Step S3: Based on the low-frequency beacon control initial pointing, generate gimbal control commands based on the attitude information of this end and the low-frequency azimuth angle of the other end relative to this end, and adjust the terahertz antenna to roughly point to the other end.

[0057] Step S4: Bidirectional terahertz signal scanning and pointing parameter acquisition. This end acts as the scanning party and actively transmits terahertz signals when adjusting the terahertz antenna. The other end acts as the detection party, receives the terahertz signals and measures the signal strength, and feeds back the signal strength information to the scanning party until the optimal alignment is achieved, that is, the signal strength reaches the maximum. Record the pointing parameters and maximum power value at this time.

[0058] Step S5: Two-way low-frequency beacon positioning error calibration. Repeat steps S1-S2 to obtain the low-frequency azimuth angle of the other end relative to the local end. Compare it with the pointing parameters at the optimal alignment determined in step S4 to obtain the inherent deviation of low-frequency alignment as the calibration coefficient. The pointing parameters at the optimal alignment are determined by the local antenna encoder reading combined with the local attitude information.

[0059] Step S6: Bidirectional continuous low-frequency tracking and quality assessment. Repeat step S2 to obtain the real-time low-frequency azimuth angle and update the local end's estimate of the azimuth angle to the remote end in combination with the calibration parameters. Output the local end's estimate of the azimuth angle to the remote end as the preprocessed low-frequency angle and low-frequency signal quality index simultaneously.

[0060] Step S7: Each performs fusion tracking and dynamic angle correction using the MSAWF-CP algorithm to obtain the processing results of the aforementioned process. The MSAWF-CP algorithm is used to calculate adaptive weights, and the preprocessed low-frequency angle and terahertz power angle deviation are fused to obtain the fused weighted angle of the local end to the opposite end direction. Rotation alignment is performed based on the fused weighted angle and attitude information.

[0061] In another embodiment, step S1 includes: real-time acquisition of the local device's own three-axis acceleration and three-axis angular velocity data, processing them through an attitude calculation algorithm to determine the local device's real-time attitude information representing the local device's attitude at a certain moment, including pitch angle, yaw angle, and roll angle. This attitude information will be input to the local device's control module and subsequent fusion algorithm module as a reference for coordinate transformation, motion compensation, and pointing control.

[0062] In another embodiment, the multiple antennas mentioned in step S2 are at least three antennas. When the low-frequency beacon signal (433MHz pulse signal) from the other end arrives at the array, the path length of the signal to each antenna element is different due to the difference in the spatial position of each antenna element, thereby generating a measurable phase difference.

[0063] The signal processing module calculates the angle of arrival (Angle of Arrival) of the beacon signal by comparing the phase information between the received signals from each antenna with high precision and combining this with the known array geometry (antenna spacing and arrangement). This angle is the initial azimuth and elevation angle (θ) of the beacon signal relative to the low-frequency array coordinate system of the local end. LF,Re mote←Local,initial,az ,θ LF,Re mote←Local,initial,el ).

[0064] Simultaneously, the received low-frequency signal command from the other end is evaluated, i.e., the initial signal-to-noise ratio (SNR). LF,Remote←Local,initial .

[0065] Using multiple antennas instead of just two helps to eliminate the angular ambiguity that may exist in single-baseline interferometric direction finding through multi-baseline processing, and can improve the angle measurement accuracy and robustness to noise through signal processing.

[0066] Specifically, regarding the number of antennas: Minimum requirement: In principle, a minimum of two antennas are needed to achieve phase difference direction finding. However, considering the resolution of two-dimensional direction finding and angular ambiguity, a fully functional system of this invention typically requires at least three non-collinearly arranged antennas.

[0067] Number Limits: Increasing the number of antennas typically leads to performance improvements, such as higher angular resolution, stronger multipath resistance, and better interference immunity. However, increasing the number of antennas also directly results in higher system costs, larger physical dimensions (especially for low-frequency antennas), and a significant increase in signal processing computational complexity. This is a major limiting factor for systems that need to be deployed on compact devices such as unmanned platforms.

[0068] In another embodiment, step S4 includes: the scanning device performs a fine scan at a small angle near its current pointing direction, and the detection device measures the RSSI in real time; the scanning device adjusts its pointing direction according to the feedback from the detection device until the RSSI received by the detection device reaches its maximum; each device records the pointing parameters when its own antenna achieves optimal alignment during this bidirectional optimization process; and records the maximum power value RSSI measured at this time. THz,max Based on the RSSI variation data during the scanning process, the sensitivity / slope factor of the RSSI-angle curve near the peak is initially estimated. k After this step is completed, the terahertz link between the two parties is established, and the system enters tracking mode.

[0069] In another embodiment, the low-frequency signal quality indicators in step S6 include the signal-to-noise ratio obtained by comparing the intensity of the received low-frequency pulse signal with the background noise power, and the short-term variance of the angle estimate obtained by statistically calculating the preprocessed low-frequency angles continuously output within a time window. The algorithm confidence score (Conf) is derived from evaluating the consistency of phase difference measurements, the success rate or uniqueness of ambiguity resolution, and potential multipath effects. AOA,k .

[0070] Low-frequency signal quality metrics include signal-to-noise ratio (SNR, obtained by comparing the strength of the received low-frequency pulse signal with the background noise power, directly reflecting the clarity and detectability of the low-frequency signal) and short-term variance of angle estimation. (This is obtained by statistically calculating the preprocessed low-frequency angles continuously output within a short time window, reflecting the stability and jitter of the low-frequency angle estimation. The smaller the variance, the higher the stability.) Algorithm Confidence AOA,k (This is a comprehensive indicator, which may be derived from the consistency of phase difference measurements, the success rate or uniqueness of ambiguity resolution, and the assessment results of potential multipath effects. The higher the confidence level, the more reliable the current low-frequency AoA results.)

[0071] In another embodiment, obtaining the processing result of the aforementioned process in step S7 includes obtaining the local preprocessed low-frequency angle θ′ output in S6. LF,Re mote←Local,k and its quality index SNR LF,Re mote←Local,k(Signal-to-noise ratio of the low-frequency signal received from the other end), short-term variance of the recent AoA estimate from this end to the other end. Confidence of low-frequency AOA algorithm AOA,k ; Obtain the real-time RSSI of the signal power from the other end using the local terahertz receiver. Re mote←Local,k Calculate its short-term variance Using signal power RSSI Re mote←local,k Maximum power value RSSI THz,max And the predicted sensitivity / slope factor. k The terahertz power angle deviation Δθ was calculated. THz,Remote←Locak,k (i.e., the deviation of the antenna at this end from the direction of the antenna at the other end).

[0072] The calculation of adaptive weights includes:

[0073]

[0074] In the formula, ε is a small positive constant to prevent the denominator from being 0; T LF T THz ω is an intermediate parameter. LF For low-frequency weights, ω THz For high-frequency weights, f1 and f2 are functions that map their respective quality indicators to trust scores.

[0075] The fusion weighted angle obtained by the preprocessing of low-frequency angle and terahertz power angle deviation includes:

[0076]

[0077] In the formula, θ′ LF,Re mote←Local,k For local low-frequency angle preprocessing, Δθ THz,Re mote←Locak,k This refers to the terahertz power angle deviation. To integrate weighted perspectives.

[0078] The specific simulation results of this embodiment are as follows: Figure 2The figure illustrates the simulation results of the high-low frequency fusion-assisted alignment and dynamic tracking algorithm in the bidirectional terahertz communication system proposed in this embodiment. It visually compares the error of angle estimation relying solely on low-frequency signals during system operation with the estimation error fused using the MSAWF-CP (Mode Switching Adaptive Weighted Fusion and Compensation Prediction) algorithm. In this simulation, the simulation duration was 10 seconds, the sampling time interval was 0.1 seconds, and the target moved at a constant angular velocity of 10 degrees / second. The figure shows the instantaneous angle error (in degrees) produced by three different angle estimation methods throughout the simulation period: using only the low-frequency beacon (blue curve), using only high-frequency RSSI estimation (green curve), and fusing both using the MSAWF-CP algorithm (orange curve). The root mean square error (RMSE) values ​​for each method are also directly labeled in the legend.

[0079] Throughout the simulation, the low-frequency angle estimation error (blue curve) exhibited continuous and significant fluctuations. Its error value varied between 0.7 and 0.95 degrees, demonstrating low accuracy and high instability. The root mean square error (RMSE) for angle estimation using only low-frequency beacons reached 0.792393 degrees.

[0080] The high-frequency RSSI estimation exhibits dynamic characteristics related to signal quality throughout the simulation. While its error may be relatively small at certain times, it can become very large at other times, even exceeding the error of low-frequency estimation, due to deterioration in simulated signal quality (e.g., target deviation from the beam center leading to a decrease in RSSI, or a decrease in the sensitivity of the RSSI-estimated angle). Overall, the RMSE for angle estimation relying solely on high-frequency RSSI estimation is 0.638602 degrees. Although this value is better than pure low-frequency estimation, its error variability is significant, indicating a strong dependence on signal quality.

[0081] In contrast, the angle estimation error (orange curve) after adopting the MSAWF-CP fusion algorithm is significantly reduced and exhibits higher stability. Although the fusion error fluctuates slightly in the initial stage (approximately 0-2 seconds) due to algorithm convergence and initial information uncertainty, it remains at a low level for most of the subsequent time, with fluctuations far smaller than those of pure low-frequency estimation. Particularly noteworthy is that even though the high-frequency signal quality naturally degrades as the target deviates from the beam center during simulation (leading to fluctuations in high-frequency estimation accuracy), the MSAWF-CP algorithm effectively suppresses the overall error through intelligent mode switching and adaptive weight adjustment. As shown in the figure, the root mean square error (RMSE) of angle estimation after adopting the MSAWF-CP fusion algorithm is significantly reduced to 0.208355 degrees. Compared to using only low-precision low-frequency beacons, the MSAWF-CP fusion algorithm reduces the root mean square error of angle tracking by approximately 73.7%. Compared to using only high-frequency RSSI estimation, the fusion algorithm reduces the RMSE by approximately 55.13%. This significant performance improvement clearly demonstrates the effectiveness of the high-low frequency fusion strategy proposed in this invention. By dynamically evaluating and integrating sensor information with varying accuracies and characteristics, the MSAWF-CP algorithm successfully overcomes the limitations of a single sensor, especially when the accuracy of low-frequency sensors is inherently limited. It effectively utilizes the high-precision potential of high-frequency sensors under good signal conditions and intelligently balances the contributions of both sensors under different signal qualities, ultimately achieving stable and accurate angle tracking far superior to any single signal source. In summary, the simulation results strongly support the core idea of ​​this invention: even when using a low-frequency system with lower accuracy as an auxiliary, intelligent fusion with a high-frequency system with high-precision potential can significantly improve the dynamic alignment and tracking performance of terahertz communication systems, providing an effective technical approach for achieving robust and efficient terahertz links.

[0082] In another embodiment, a dynamic beam alignment and tracking system for terahertz communication networking is provided, which is respectively set on both communication parties, with each party being the local end and the peer end, and is used to execute the dynamic beam alignment and tracking method for terahertz communication networking described in the foregoing embodiment. The system includes: an attitude sensing module, a low-frequency auxiliary direction finding module, a high-frequency transceiver and sensing module, a beam pointing execution module, and a core control and fusion module.

[0083] The attitude perception module is used to process the collected inertial data and calculate the real-time attitude information of the local carrier.

[0084] The low-frequency auxiliary direction finding module is used to synchronously receive low-frequency pulse signals from the other end, digitize them first, and then process them accordingly.

[0085] The high-frequency transceiver and sensing module is used to generate, amplify and transmit terahertz communication signals via a terahertz antenna, while simultaneously receiving terahertz signals from the other end and measuring the output signal strength.

[0086] A beam pointing execution module is used to radiate and receive terahertz beams with high directivity, and the beam pointing module includes a pan-tilt unit;

[0087] The core control and fusion module receives and processes real-time data from the attitude perception module, the low-frequency auxiliary direction finding module, and the high-frequency transceiver and perception module. Based on the fusion or prediction results and combined with real-time attitude information, it generates the final control command for the gimbal in the drive beam pointing execution module.

[0088] Specifically, the attitude perception module includes: an inertial measurement unit (IMU): one or more sensors used to acquire real-time triaxial acceleration and triaxial angular velocity data of the local carrier; and an attitude calculation unit connected to the IMU, which can be a software module running in a processor or dedicated hardware. This unit processes the acquired inertial data and calculates the real-time attitude information of the local carrier. This attitude information provides a spatial reference for subsequent coordinate transformation, motion compensation, and control command generation.

[0089] The low-frequency auxiliary direction finding module includes: a low-frequency radio frequency beacon transmitter for transmitting known low-frequency pulse signals to the other end according to a preset protocol; a low-frequency receiver array composed of multiple (e.g., 4) antenna elements arranged in a predetermined geometric configuration for synchronously receiving low-frequency pulse signals from the other end; a multi-channel receiving and digitizing unit containing a radio frequency front-end and an analog-to-digital converter (ADC) corresponding to each antenna element for amplifying, filtering, and digitizing the analog signals received by each antenna; and a low-frequency signal processing unit connected to the multi-channel receiving and digitizing unit for processing the digitized signals.

[0090] The high-frequency transceiver and sensing module includes: a terahertz transceiver: used to generate, amplify, and transmit terahertz communication signals via a terahertz antenna, while simultaneously receiving terahertz signals from the other end; and an RF power detection unit: located in the terahertz receiver link and connected to the terahertz transceiver. This unit is used to measure the intensity (RSSI) of the received terahertz signal in real time and output the intensity value (usually voltage or a digital quantity).

[0091] The beam pointing execution module includes: a terahertz antenna connected to the terahertz transceiver for radiating and receiving a highly directional terahertz beam; and a precision pan-tilt unit, a two-axis (or multi-axis) mechanical rotating platform for supporting the terahertz antenna.

[0092] Core control and fusion module (the aforementioned "embedded control board" and part of the "signal processing module"): This module acts as the "brain" of the system, typically implemented by a high-performance embedded processor (such as an FPGA, DSP, or multi-core CPU), and connects to all the aforementioned modules. It is responsible for executing the core logic of the method of this invention, including...

[0093] Data aggregation and preprocessing: Receive real-time data from the attitude sensing module, low-frequency auxiliary direction finding module, and high-frequency transceiver and sensing module.

[0094] Execute the MSAWF-CP algorithm: Implement the dynamic mode switching, adaptive weight calculation, high- and low-frequency information fusion, and optional prediction compensation described in step S7.

[0095] System process control and command generation: Manage the overall process from S1 to S7, and generate the final control command for the drive beam pointing execution module (gimbal) based on the fusion or prediction results and real-time attitude information.

[0096] The embodiments described above merely illustrate specific implementation methods of this application, and while the descriptions are detailed and specific, they should not be construed as limiting the scope of protection of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the technical solution of this application, and these modifications and improvements all fall within the scope of protection of this application.

Claims

1. A dynamic beam alignment and tracking method for terahertz communication networking, characterized in that, The two communicating parties are each other's local end and peer end, including: Step S1: The local end collects inertial measurement data and calculates the attitude information; Step S2: The other end sends a low-frequency pulse signal, and the local end uses a multi-antenna low-frequency equipment receiving array to receive the low-frequency pulse signal and calculates the measured phase difference of different antennas to obtain the low-frequency directional angle of the other end relative to the local end. The steps S1 and S2 are not in any particular order; Step S3: Based on the attitude information of this end and the low-frequency azimuth angle of the other end relative to this end, generate a gimbal control command and adjust the terahertz antenna to roughly point towards the other end. Step S4: This end acts as the scanning party, actively transmitting terahertz signals when adjusting the terahertz antenna. The other end acts as the detection party, receiving the terahertz signals and measuring the signal strength, and feeding back the signal strength information to the scanning party until optimal alignment is achieved, i.e., the signal strength reaches its maximum. Record the pointing parameters and maximum power value at this time. Step S5: Repeat steps S1-S2 to obtain the low-frequency directional angle of the other end relative to the local end, and compare it with the pointing parameters at the optimal alignment determined in step S4 to obtain the inherent deviation of low-frequency alignment as a calibration coefficient; the pointing parameters at the optimal alignment are determined by the local antenna encoder reading combined with the local attitude information. Step S6: Repeat step S2 to obtain the real-time low-frequency azimuth angle and update the local end's estimate of the azimuth angle to the opposite end in combination with the calibration parameters. Output the local end's estimate of the azimuth angle to the opposite end as the preprocessed low-frequency angle and low-frequency signal quality index simultaneously. Step S7: Obtain the processing results of the aforementioned process, calculate the adaptive weights using the MSAWF-CP algorithm, fuse the preprocessed low-frequency angle and terahertz power angle deviation to obtain the fused weighted angle of this end to the opposite end direction, and perform rotation alignment based on the fused weighted angle and attitude information; The MSAWF-CP algorithm refers to an adaptive weighted fusion and compensation prediction algorithm based on mode switching; The process of obtaining the results of the aforementioned processing includes obtaining the local preprocessed low-frequency angle output by S6. And its quality indicators include the signal-to-noise ratio of the low-frequency signal received at the local end from the remote end. The short-run variance of the recent AoA estimates from this end. Confidence of low-frequency AOA algorithm ; Obtain the real-time signal power from the other end using the local terahertz receiver. Calculate its short-term variance Utilizing signal power Maximum power value And the predicted sensitivity / slope characteristics The terahertz power angle deviation was calculated. ; The calculation of adaptive weights includes: ； ； ； In the formula, It is a small positive number to prevent the denominator from being 0; , For intermediate parameters, For low-frequency weights, For high-frequency weights, It is a function that maps each quality indicator to a trust score; The fusion weighted angle obtained by the deviation between the low-frequency angle and the terahertz power angle in the fusion preprocessing includes: ； In the formula, For local low-frequency angle preprocessing, This refers to the terahertz power angle deviation. To integrate weighted perspectives.

2. The dynamic beam alignment and tracking method for terahertz communication networking according to claim 1, characterized in that, Step S1 includes: real-time acquisition of the three-axis acceleration and three-axis angular velocity data of the local device, processing them through an attitude calculation algorithm to determine the local device's real-time attitude information, which represents the local device's attitude at a certain moment, including pitch angle, yaw angle and roll angle.

3. The dynamic beam alignment and tracking method for terahertz communication networking according to claim 1, characterized in that, The multiple antennas mentioned in step S2 are at least 3 antennas.

4. The dynamic beam alignment and tracking method for terahertz communication networking according to claim 1, characterized in that, Step S4 includes: the scanning device performs a fine scan at a small angle near its current pointing direction, and the detection device measures the RSSI in real time; the scanning device adjusts its pointing direction based on the feedback from the detection device until the RSSI received by the detection device reaches its maximum; each device records the pointing parameters when its own antenna achieves optimal alignment during this bidirectional optimization process; and records the maximum power value measured at this time. Based on the RSSI variation data during the scanning process, the sensitivity / slope characteristics of the RSSI-angle curve near the peak are initially estimated. After this step is completed, the terahertz link between the two parties is established, and the system enters tracking mode.

5. The dynamic beam alignment and tracking method for terahertz communication networking according to claim 1, characterized in that, The low-frequency signal quality indicators mentioned in step S6 include the signal-to-noise ratio obtained by comparing the intensity of the received low-frequency pulse signal with the background noise power, and the short-term variance of the angle estimate obtained by statistically calculating the preprocessed low-frequency angles continuously output within a time window. The algorithm confidence level is derived from the evaluation of the consistency of phase difference measurements, the success rate or uniqueness of ambiguity resolution, and potential multipath effects. .

6. A dynamic beam alignment and tracking system for terahertz communication networking, characterized in that, Each component is set on one of the two communicating parties, which are each other's local and remote ends, and is used to execute the dynamic beam alignment and tracking method for terahertz communication networking as described in any one of claims 1-5, including: an attitude sensing module, a low-frequency auxiliary direction finding module, a high-frequency transceiver and sensing module, a beam pointing execution module, and a core control and fusion module; The attitude perception module is used to process the collected inertial data and calculate the real-time attitude information of the local carrier. The low-frequency auxiliary direction finding module is used to synchronously receive low-frequency pulse signals from the other end, digitize them first, and then process them accordingly. The high-frequency transceiver and sensing module is used to generate, amplify and transmit terahertz communication signals via a terahertz antenna, while simultaneously receiving terahertz signals from the other end and measuring the output signal strength. A beam pointing execution module is used to radiate and receive terahertz beams with high directivity, and the beam pointing execution module includes a pan-tilt unit; The core control and fusion module receives and processes real-time data from the attitude perception module, the low-frequency auxiliary direction finding module, and the high-frequency transceiver and perception module. Based on the fusion or prediction results and combined with real-time attitude information, it generates the final control command for the gimbal in the drive beam pointing execution module.

7. The dynamic beam alignment and tracking system for terahertz communication networking according to claim 6, characterized in that, The low-frequency auxiliary direction finding module includes a low-frequency radio frequency beacon transmitter, a low-frequency receiver array, a multi-channel receiving and digitizing unit, and a low-frequency signal processing unit. The low-frequency radio frequency beacon transmitter is used to transmit a known low-frequency pulse signal to the other end according to a preset protocol. The low-frequency receiver array consists of multiple antenna elements arranged in a predetermined geometric configuration, used to synchronously receive low-frequency pulse signals from the other end; the multi-channel receiving and digitizing unit is used to amplify, filter, and digitize the analog signals received by each antenna; the low-frequency signal processing unit is connected to the multi-channel receiving and digitizing unit and is used to process the digitized signal. The high-frequency transceiver and sensing module includes a terahertz transceiver and an RF power detection unit. The terahertz transceiver is used to generate, amplify, and transmit terahertz communication signals via a terahertz antenna, while simultaneously receiving terahertz signals from the other end. The RF power detection unit is used to measure the intensity of the received terahertz signal in real time and output the intensity value.

Images