A method and system for dynamic error compensation in millimeter-wave transmission links

Abstract

This invention discloses a dynamic error compensation method and system for millimeter-wave transmission links, relating to the field of radio communication technology. The method includes acquiring real-time status data of the millimeter-wave transmission link and baseband signal data from the transmitting and receiving ends, completing time synchronization alignment and standardization preprocessing of the data; identifying multi-dimensional error sources based on the preprocessed link data; constructing a dynamic error coupling model of the link, quantifying the impact weights of various error sources on the transmission link performance; generating a multi-dimensional joint dynamic error compensation strategy, performing corresponding levels of error compensation in the baseband digital domain and RF analog domain; acquiring the compensated link data and performing real-time verification of the link transmission performance; and performing adaptive iterative optimization of the model and compensation strategy based on the verification results. This invention can accurately identify multi-dimensional error sources in millimeter-wave transmission links, construct a full-dimensional dynamic error coupling model, and effectively suppress time-varying errors in the link through multi-domain joint compensation.

Description

Technical Field

[0001] This invention relates to the field of radio communication technology, and in particular to a method and system for dynamic error compensation of millimeter-wave transmission links. Background Technology

[0002] Millimeter-wave communication, with its high bandwidth, high speed, and narrow beamwidth, has become a core technology supporting next-generation wireless communication systems, point-to-point high-speed data transmission, and satellite communication. However, the short carrier wavelength of millimeter-wave bands makes them susceptible to atmospheric absorption, rainfall attenuation, multipath reflection, and the Doppler effect during transmission. Furthermore, the nonlinear characteristics of radio frequency devices, local oscillator phase noise, and synchronization delay deviations at the transmitting and receiving ends all introduce multi-dimensional time-varying errors into the transmission link, directly causing amplitude distortion and phase distortion of the transmitted signal. This ultimately leads to increased bit error rate and decreased transmission reliability, becoming a core factor restricting the performance improvement of millimeter-wave communication systems. Traditional millimeter-wave transmission link error compensation technologies often design compensation schemes for single-type error sources, only achieving single-level error suppression such as channel equalization and nonlinear distortion pre-distortion. They do not consider the coupling relationships between different error sources and cannot fully characterize the comprehensive impact of all-dimensional errors in the millimeter-wave transmission link. In complex transmission scenarios with multiple error sources acting together, the compensation accuracy and error suppression effect are significantly limited. Meanwhile, traditional compensation schemes mostly use static error models with fixed parameters, which cannot adapt to the time-varying characteristics of millimeter-wave links in scenarios such as high-speed movement, changes in weather conditions, and adjustments in transmission distance. The matching degree between model parameters and actual link error characteristics will decrease rapidly with changes in the transmission environment, making it difficult to achieve continuous and effective error compensation.

[0003] Existing millimeter-wave error compensation technologies primarily focus on the baseband digital domain, correcting only quantifiable errors during baseband signal processing. They lack effective collaborative compensation methods for errors introduced by device characteristic fluctuations, transmission delay deviations, and gain fluctuations in the RF analog link, resulting in incomplete suppression of some errors and significant shortcomings in compensation effectiveness. Current compensation strategies often aim at single error suppression without considering multi-dimensional constraints based on core transmission performance factors such as system bit error rate and error vector magnitude. This can easily lead to over- or under-compensation, failing to ensure stable transmission link operation while suppressing errors. Furthermore, current error source identification methods often rely on single time-domain or frequency-domain analysis. For multiple error sources with significant differences in time- and frequency-domain characteristics, accurate separation and feature extraction are impossible, resulting in insufficient accuracy in subsequent error modeling and directly impacting the final compensation effect.

[0004] Existing millimeter-wave transmission link error compensation systems mostly employ open-loop processing architectures, capable of performing fixed compensation operations according to preset procedures. They lack the ability to perform reverse optimization based on the compensated link transmission performance and thus lack a complete closed-loop iterative optimization mechanism. Furthermore, the system's data acquisition stage lacks multi-channel synchronous processing capabilities, making it impossible to achieve precise time synchronization of link data from different sources and with different sampling rates. This results in timing deviations in error feature extraction and modeling, further reducing the adaptability of the compensation scheme. Existing systems also lack adaptability to different transmission scenarios, failing to adaptively adjust models and compensation strategies in various scenarios such as fixed terrestrial communication, high-speed mobile communication, and satellite-to-ground link communication. Consequently, they struggle to meet the application requirements of millimeter-wave communication systems in high-reliability, low-latency transmission scenarios. Summary of the Invention

[0005] This invention proposes a dynamic error compensation method and system for millimeter-wave transmission links to solve the problems mentioned in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a dynamic error compensation method for millimeter-wave transmission links, comprising the following steps: Collect real-time status data of the millimeter-wave transmission link and baseband signal data from the transceiver end, and complete the time synchronization alignment and standardization preprocessing of the data; Based on the preprocessed link data, multi-dimensional error sources of millimeter-wave transmission links are identified, and time-frequency domain feature extraction and characteristic analysis of various error sources are completed. Based on the identified error source characteristics, a dynamic error coupling model for millimeter-wave transmission links is constructed to quantify the impact weights of various error sources on transmission link performance. Based on the output of the dynamic error coupling model, a multi-dimensional joint dynamic error compensation strategy is generated, and corresponding levels of error compensation processing are performed in the baseband digital domain and the radio frequency analog domain. Collect compensated millimeter-wave transmission link data, perform real-time verification of link transmission performance, and calculate the link error suppression ratio and core transmission performance parameters before and after compensation. Based on the link performance verification results, adaptive iterative optimization is performed on the dynamic error coupling model and compensation strategy to complete the closed-loop dynamic error compensation of the millimeter-wave transmission link.

[0007] Furthermore, it also includes a full-dimensional time-varying error coupling modeling step for millimeter-wave transmission links, constructing a joint dynamic error coupling model in the time and frequency domains covering channel transmission characteristics, device nonlinearity, and transceiver synchronization deviation. The model is calculated as follows: ; for Time Frequency The comprehensive complex error value of the millimeter-wave transmission link at that location. for Time Frequency Atmospheric absorption attenuation coefficient at that location for Time Frequency Rainfall attenuation coefficient at the location, This represents the total number of multipath components in a millimeter-wave transmission link. for Time of the first Channel complex gain of multiple path components, for Time of the first The transmission delay of a multipath component. for Time of the first Doppler frequency shift of multipath components, for Time Frequency Phase noise introduced by the local oscillator at the transmitting and receiving ends. for Time Frequency Phase distortion introduced by the nonlinearity of the power amplifier. for The synchronization delay error at the transceiver end of the millimeter-wave transmission link is determined by integrating multiple error sources in the millimeter-wave transmission link, including time-varying channel attenuation, multipath propagation, Doppler effect, device nonlinearity, and synchronization delay.

[0008] Furthermore, it also includes a multi-dimensional error source time-frequency domain joint identification step, specifically performing short-time Fourier transform and wavelet transform on the preprocessed link data to complete the time-frequency domain joint decomposition of the link data, extracting the time-varying fluctuation characteristics of the link error in the time domain dimension, extracting the error amplitude and phase characteristics at different carrier frequency points in the frequency domain dimension, and combining the prior channel characteristics and device characteristics of the millimeter wave transmission link to simultaneously complete the time-domain correlation analysis and frequency-domain distribution characteristic analysis of various error sources.

[0009] Furthermore, it also includes a time-varying update step for the dynamic error coupling model. Specifically, a fixed-duration sliding time window is set. Within each sliding time window, real-time status data and error characteristic data of the millimeter-wave link are collected. The model is updated in real time based on the least squares fitting algorithm. At the same time, the weight distribution of different error sources in the model is dynamically adjusted based on the changing characteristics of the link scenario. When a switch in the link transmission scenario is detected, the rapid reconstruction process of the model is triggered.

[0010] Furthermore, it also includes an adaptive optimization solution step for the multi-dimensional joint error compensation coefficients, the solution method being as follows: ; The optimal multi-dimensional joint error compensation coefficient matrix. Let be the compensation coefficient matrix to be solved. For the compensated ideal baseband signal matrix, This is the original baseband signal matrix at the receiving end. For L2 norm operations, For L1 norm operations, For Frobenius norm operations, These are the sparse regularization weight coefficients. These are the total variation regularization weight coefficients. These are the time-varying smoothing regularization weight coefficients. For total variation operations, For time First-order partial differential operations The magnitude of the error vector corresponding to the compensation coefficient matrix. The system's preset error vector magnitude threshold, The system bit error rate corresponding to the compensation coefficient matrix, The preset bit error rate threshold for the system, To solve for the variable values ​​that minimize the objective function.

[0011] Furthermore, it also includes hierarchical error compensation execution steps. Specifically, based on the generated multi-dimensional joint dynamic error compensation strategy, channel equalization, phase noise compensation, Doppler frequency shift correction, and nonlinear distortion pre-distortion processing are performed in the baseband digital domain. Digital pre-distortion technology is used to pre-compensate the nonlinear distortion of the power amplifier. An adaptive equalization algorithm is used to compensate for the amplitude and phase errors caused by multipath channels and atmospheric attenuation. A phase-locked loop algorithm is used to complete carrier phase tracking and Doppler frequency shift correction. In the RF analog domain, adaptive adjustment of RF link gain, dynamic calibration of local oscillator frequency, and real-time compensation for time delay error are performed through hierarchical joint compensation in the baseband digital domain and the RF analog domain.

[0012] Furthermore, it also includes a closed-loop adaptive iterative optimization step, specifically, based on the results of real-time verification of link performance, calculating the error suppression effect and link performance improvement of the current compensation strategy. When the link error suppression ratio is lower than a preset threshold, the iterative optimization process of the model and compensation strategy is triggered. The weight parameters of the dynamic error coupling model are updated through an online gradient descent algorithm, and the optimal compensation coefficient matrix is ​​resolved based on the updated error model.

[0013] Furthermore, the millimeter-wave transmission link dynamic error compensation system includes the following modules: The link data acquisition and preprocessing module is used to acquire real-time status data and baseband signal data from the transceiver end of the millimeter-wave transmission link, and to complete the time synchronization alignment and standardization preprocessing of the data. The link error source identification and feature extraction module is used to identify multi-dimensional error sources of millimeter-wave transmission links based on preprocessed link data, and to complete time-frequency domain feature extraction and characteristic analysis of various error sources. The dynamic error coupling modeling module is used to construct a dynamic error coupling model of the millimeter-wave transmission link based on the identified error source characteristics, and to quantify the impact weight of various error sources on the transmission link performance. The multi-dimensional joint error compensation module is used to generate a multi-dimensional joint dynamic error compensation strategy based on the output of the dynamic error coupling model, and perform corresponding levels of error compensation processing in the baseband digital domain and the radio frequency analog domain respectively. The compensation effect verification module is used to collect the compensated millimeter-wave transmission link data, perform real-time verification of the link transmission performance, and calculate the link error suppression ratio and core transmission performance parameters before and after compensation. The adaptive iterative optimization module is used to perform adaptive iterative optimization of the dynamic error coupling model and compensation strategy based on the link performance verification results, so as to complete the closed-loop dynamic error compensation of the millimeter-wave transmission link.

[0014] Furthermore, the link data acquisition and preprocessing module incorporates a multi-channel synchronous acquisition unit, a time-frequency domain preprocessing unit, and a data synchronization unit. The multi-channel synchronous acquisition unit is used to complete the multi-channel synchronous acquisition of baseband IQ signals, RF link status parameters, channel environment parameters, and transceiver device operating parameters at the transceiver end. The time-frequency domain preprocessing unit is used to complete the denoising, normalization, and frequency offset correction preprocessing of the acquired data to eliminate noise interference and DC offset during the acquisition process. The data synchronization unit is used to complete the timestamp alignment and resampling processing of data from different sources and with different sampling rates based on the synchronous clock signal of the transceiver end.

[0015] Furthermore, the adaptive iterative optimization module incorporates an online learning optimization unit, a model parameter update unit, and a compensation strategy reconstruction unit. The online learning optimization unit is used to iteratively optimize and solve the model parameters and compensation coefficients based on the link performance verification results through online machine learning algorithms. The model parameter update unit is used to update the time-varying parameters of the dynamic error coupling model and adaptively reconstruct the scenario. The compensation strategy reconstruction unit is used to regenerate a multi-dimensional joint compensation strategy adapted to the current link state based on the optimized model output results, and simultaneously complete the smooth switching and execution of the compensation strategy.

[0016] Compared with existing technologies, the beneficial effects of this invention are: This invention, through joint time-frequency domain identification of multi-dimensional error sources, can accurately separate various core error sources in millimeter-wave transmission links, fully extract the time-frequency domain distribution characteristics and fluctuation characteristics of different errors, and provide accurate feature inputs for subsequent error modeling and compensation strategy generation, effectively improving the comprehensiveness and accuracy of link error identification.

[0017] This invention constructs a multi-dimensional time-frequency domain dynamic error coupling model, which can integrate multiple error sources in millimeter-wave transmission links, such as time-varying channel attenuation, multipath propagation, Doppler effect, device nonlinearity, and synchronization delay. It accurately characterizes the coupling relationship and time-varying characteristics between different errors, solves the problem that traditional fixed error models cannot adapt to the dynamic changes of millimeter-wave links, and greatly improves the accuracy of error modeling and the adaptability to different transmission scenarios.

[0018] This invention optimizes the multi-dimensional joint error compensation coefficient under multiple constraints, and can obtain the optimal compensation coefficient matrix under the dual constraints of system transmission performance, thereby achieving optimal compensation for time-varying coupling errors in millimeter-wave links. At the same time, through multiple regularization constraints, it ensures the convergence, robustness and dynamic smoothness of the compensation coefficient, taking into account both the accuracy of error compensation and the stability of link transmission.

[0019] This invention achieves coordinated suppression of all-dimensional errors in millimeter-wave transmission links through hierarchical joint compensation of the baseband digital domain and the radio frequency analog domain. It makes up for the coverage shortcomings of traditional single-domain compensation schemes, and can fully correct errors of different types and in different links in the link, effectively improving the overall suppression effect of link errors.

[0020] This invention, through a closed-loop adaptive iterative optimization mechanism, can update the parameters of the dynamic error coupling model and iteratively optimize the compensation strategy based on the real-time verification results of the compensated link performance. This enables the error compensation effect to continuously adapt to the time-varying characteristics of the millimeter-wave transmission link, maintaining a stable compensation effect in different transmission environments and application scenarios, thereby improving the system's scenario adaptability and operational robustness.

[0021] The system corresponding to this invention ensures the time synchronization and format consistency of multi-source link data through a multi-channel synchronous acquisition and preprocessing module, eliminating the impact of data timing deviation on error identification. Through an adaptive iterative optimization module, it realizes real-time optimization and smooth switching of the model and compensation strategy. While completing dynamic error compensation of the link, it ensures the continuity and reliability of the millimeter-wave transmission link, which can fully meet the high reliability and low latency transmission requirements of the new generation of millimeter-wave communication systems. Attached Figure Description

[0022] Figure 1This invention presents a schematic block diagram of a dynamic error compensation method and system for millimeter-wave transmission links. Figure 2 Flowchart for joint time-frequency domain identification and feature extraction of multi-dimensional error sources; Figure 3 Flowchart for the construction and time-varying update of a full-dimensional dynamic error coupling model; Figure 4 Execution logic diagram for hierarchical error compensation strategy; Figure 5 The flowchart shows the closed-loop adaptive iterative optimization and compensation coefficient solution. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0025] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly; for example, they may refer to a fixed connection, a detachable connection, or an integral connection; they may refer to a mechanical connection or an electrical connection; they may refer to a direct connection or an indirect connection through an intermediate medium; and they may refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The invention will now be described in further detail with reference to the accompanying drawings.

[0026] Reference Figures 1 to 5A method for dynamic error compensation in millimeter-wave transmission links includes the following steps: Collect real-time status data of the millimeter-wave transmission link and baseband signal data from the transceiver end, and complete the time synchronization alignment and standardization preprocessing of the data; Based on the preprocessed link data, multi-dimensional error sources of millimeter-wave transmission links are identified, and time-frequency domain feature extraction and characteristic analysis of various error sources are completed. Based on the identified error source characteristics, a dynamic error coupling model for millimeter-wave transmission links is constructed to quantify the impact weights of various error sources on transmission link performance. Based on the output of the dynamic error coupling model, a multi-dimensional joint dynamic error compensation strategy is generated, and corresponding levels of error compensation processing are performed in the baseband digital domain and the radio frequency analog domain. Collect compensated millimeter-wave transmission link data, perform real-time verification of link transmission performance, and calculate the link error suppression ratio and core transmission performance parameters before and after compensation. Based on the link performance verification results, adaptive iterative optimization is performed on the dynamic error coupling model and compensation strategy to complete the closed-loop dynamic error compensation of the millimeter-wave transmission link.

[0027] This invention also includes a full-dimensional time-varying error coupling modeling step for millimeter-wave transmission links. Specifically, based on the identified multi-dimensional error sources, a joint dynamic error coupling model in the time and frequency domains covering channel transmission characteristics, device nonlinearity, and transceiver synchronization deviation is constructed. The model is calculated as follows: ; for Time Frequency The comprehensive complex error value of the millimeter-wave transmission link at the location, with dimensions of 1. for Time Frequency The atmospheric absorption attenuation coefficient at that location, with dimensions in dB. for Time Frequency The rainfall attenuation coefficient at a given location, with dimensions in dB, The total number of multipath components in a millimeter-wave transmission link, dimensionless. for Time of the first The channel complex gain of the multipath components, with dimensions of 1. for Time of the first The propagation delay of a multipath component, in seconds. for Time of the first The Doppler frequency shift of the multipath components, in units of Hz. for Time Frequency The phase noise introduced by the local oscillator at the transmitting and receiving ends, with dimensions in rad. for Time Frequency The phase distortion introduced by the nonlinearity of the power amplifier is measured in rad. for The synchronization delay error between the transceiver ends of a millimeter-wave transmission link at any given time, measured in seconds (s). This refers to the millimeter-wave carrier frequency, with the dimension Hz. The variable is time, with the dimension in seconds. Using imaginary units and dimensionless, this model integrates multiple error sources in millimeter-wave transmission links, including time-varying channel attenuation, multipath propagation, Doppler effect, device nonlinearity, and synchronization delay, to construct a full-dimensional time-frequency domain dynamic error coupling model. This model accurately characterizes the time-varying characteristics and coupling relationships of millimeter-wave link errors under different scenarios, providing precise quantitative basis for the generation of subsequent dynamic compensation strategies and solving the problem that traditional fixed error models cannot adapt to the time-varying characteristics of millimeter-wave links.

[0028] This invention also includes a multi-dimensional error source joint time-frequency domain identification step. Specifically, it performs short-time Fourier transform and wavelet transform on the preprocessed link data to complete the joint time-frequency domain decomposition of the link data. In the time domain, it extracts the time-varying fluctuation characteristics of the link error, and in the frequency domain, it extracts the error amplitude and phase characteristics at different carrier frequencies. Combining the prior channel characteristics and device characteristics of the millimeter-wave transmission link, it accurately separates and identifies six core error sources: atmospheric attenuation error, multipath propagation error, Doppler frequency shift error, phase noise error, device nonlinear distortion error, and transmit / receive synchronization delay error. At the same time, it completes the time-domain correlation analysis and frequency-domain distribution characteristic analysis of various error sources, providing accurate feature inputs for subsequent error modeling.

[0029] This invention also includes a time-varying update step for the dynamic error coupling model. Specifically, a fixed-duration sliding time window is set. Within each sliding time window, real-time status data and error feature data of the millimeter-wave link are collected. The model is updated in real time based on the least squares fitting algorithm. At the same time, the weight allocation of different error sources in the model is dynamically adjusted based on the changing characteristics of the link scenario. When a switch in the link transmission scenario is detected, a rapid model reconstruction process is triggered. Based on the link features of the current scenario, the model parameters are rapidly converged and updated, ensuring the adaptability and modeling accuracy of the dynamic error coupling model under different time-varying scenarios.

[0030] This invention also includes an adaptive optimization solution step for multi-dimensional joint error compensation coefficients. Specifically, based on the link comprehensive error distribution characteristics output by the dynamic error coupling model, and with the goal of optimizing link transmission performance, an optimization solution model for compensation coefficients under multiple constraints is constructed. The solution method is as follows: ; This is the optimal multi-dimensional joint error compensation coefficient matrix, with a dimension of 1. Let be the compensation coefficient matrix to be solved, with a dimension of 1. The compensated ideal baseband signal matrix has a dimension of 1. This is the original baseband signal matrix at the receiving end, with a dimension of 1. For L2 norm operations, dimensionless. For L1 norm operations, dimensionless. For Frobenius norm operations, dimensionless. These are sparse regularization weight coefficients, dimensionless. These are the total variation regularization weight coefficients, which are dimensionless. These are time-varying smoothing regularization weight coefficients, dimensionless. For total variation operations, dimensionless. For time The first-order partial differential operation, with dimensions 1 / s. The magnitude of the error vector corresponding to the compensation coefficient matrix, with the dimension %. The preset error vector magnitude threshold for the system, with the dimension %. The system bit error rate corresponding to the compensation coefficient matrix is ​​dimensionless. The preset bit error rate threshold for the system is dimensionless. To solve the variable value operation that minimizes the objective function, a dimensionless method is used. Through weighted least squares optimization with multiple regularization constraints, the optimal multi-dimensional joint compensation coefficient matrix is ​​solved under the dual performance constraints of error vector magnitude and bit error rate. At the same time, through sparse regularization, total variational regularization and time-varying smoothing regularization, the convergence, robustness and dynamic smoothness of the compensation coefficients are guaranteed, so as to achieve the optimal joint compensation of time-varying coupling error of millimeter-wave link, taking into account both compensation accuracy and link transmission stability.

[0031] This invention also includes a hierarchical error compensation execution step, specifically based on the generated multi-dimensional joint dynamic error compensation strategy. In the baseband digital domain, channel equalization, phase noise compensation, Doppler frequency shift correction, and nonlinear distortion pre-distortion processing are performed. Digital pre-distortion technology is used to pre-compensate the nonlinear distortion of the power amplifier. An adaptive equalization algorithm is used to compensate for the amplitude and phase errors caused by multipath channels and atmospheric attenuation. A phase-locked loop algorithm is used to complete carrier phase tracking and Doppler frequency shift correction. In the radio frequency analog domain, adaptive adjustment of radio frequency link gain, dynamic calibration of local oscillator frequency, and real-time compensation for time delay error are performed. Through hierarchical joint compensation in the baseband digital domain and the radio frequency analog domain, the coordinated suppression of all-dimensional errors in the millimeter-wave transmission link is achieved.

[0032] This invention also includes a closed-loop adaptive iterative optimization step. Specifically, based on the results of real-time verification of link performance, the error suppression effect and link performance improvement of the current compensation strategy are calculated. When the link error suppression ratio is lower than a preset threshold, the iterative optimization process of the model and compensation strategy is triggered. The weight parameters of the dynamic error coupling model are updated through an online gradient descent algorithm. At the same time, the optimal compensation coefficient matrix is ​​resolved based on the updated error model to generate the optimized compensation strategy. During the iterative optimization process, the iterative convergence condition and the maximum number of iterations are set to ensure the real-time performance and convergence of the optimization process. Through continuous closed-loop iterative optimization, the error compensation effect is continuously adapted to the time-varying characteristics of the millimeter-wave transmission link.

[0033] In this invention, the millimeter-wave transmission link dynamic error compensation system includes the following modules: The link data acquisition and preprocessing module is used to acquire real-time status data and baseband signal data from the transceiver end of the millimeter-wave transmission link, and to complete the time synchronization alignment and standardization preprocessing of the data. The link error source identification and feature extraction module is used to identify multi-dimensional error sources of millimeter-wave transmission links based on preprocessed link data, and to complete time-frequency domain feature extraction and characteristic analysis of various error sources. The dynamic error coupling modeling module is used to construct a dynamic error coupling model of the millimeter-wave transmission link based on the identified error source characteristics, and to quantify the impact weight of various error sources on the transmission link performance. The multi-dimensional joint error compensation module is used to generate a multi-dimensional joint dynamic error compensation strategy based on the output of the dynamic error coupling model, and perform corresponding levels of error compensation processing in the baseband digital domain and the radio frequency analog domain respectively. The compensation effect verification module is used to collect the compensated millimeter-wave transmission link data, perform real-time verification of the link transmission performance, and calculate the link error suppression ratio and core transmission performance parameters before and after compensation. The adaptive iterative optimization module is used to perform adaptive iterative optimization of the dynamic error coupling model and compensation strategy based on the link performance verification results, so as to complete the closed-loop dynamic error compensation of the millimeter-wave transmission link.

[0034] In this invention, the link data acquisition and preprocessing module incorporates a multi-channel synchronous acquisition unit, a time-frequency domain preprocessing unit, and a data synchronization unit. The multi-channel synchronous acquisition unit is used to complete the multi-channel synchronous acquisition of baseband IQ signals, RF link status parameters, channel environment parameters, and transceiver device operating parameters at the transceiver end. The time-frequency domain preprocessing unit is used to complete the denoising, normalization, and frequency offset correction preprocessing of the acquired data to eliminate noise interference and DC offset during the acquisition process. The data synchronization unit is used to complete the timestamp alignment and resampling processing of data from different sources and with different sampling rates based on the synchronous clock signal of the transceiver end, ensuring the time synchronization and format consistency of the input data.

[0035] In this invention, the adaptive iterative optimization module incorporates an online learning optimization unit, a model parameter update unit, and a compensation strategy reconstruction unit. The online learning optimization unit is used to iteratively optimize and solve the model parameters and compensation coefficients based on the link performance verification results using an online machine learning algorithm. The model parameter update unit is used to update the time-varying parameters of the dynamic error coupling model and adaptively reconstruct the scenario. The compensation strategy reconstruction unit is used to regenerate a multi-dimensional joint compensation strategy adapted to the current link state based on the optimized model output results, and simultaneously complete the smooth switching and execution of the compensation strategy to ensure the real-time performance of the system's closed-loop optimization process and the continuity of link transmission.

[0036] The following two examples further illustrate the specific implementation of this system: Example 1

[0037] The application scenario of this embodiment is a point-to-point millimeter wave backbone transmission link in the 28GHz band of a metropolitan area communication network. The link transmission distance is 12km and the system transmission bandwidth is 500MHz. It is used for high-speed data backbone transmission between core urban areas. The link transmission environment includes open plains and some urban building clusters that block the view. The weather conditions change regularly with the seasons, which places strict requirements on the stability and reliability of the link transmission.

[0038] Before implementation, the deployment and joint debugging of millimeter-wave communication equipment at both the transmitting and receiving ends were completed. Both the transmitting and receiving ends were equipped with millimeter-wave radio frequency units, baseband signal processing units, high-precision synchronization clock units, link status acquisition units, and error compensation control units. The equipment at both ends achieved nanosecond-level time synchronization through a high-precision clock source. A link data processing server and performance monitoring platform were built in conjunction with the equipment. All functional units were connected to a unified control bus to achieve collaborative work.

[0039] During normal operation of the link, the link data acquisition and preprocessing module first completes multi-channel synchronous acquisition of all dimensions of millimeter-wave transmission link data. The acquired content includes baseband IQ signals at both ends, RF link gain and phase status parameters, transceiver device operating parameters, meteorological environmental parameters of the link transmission path, and channel status monitoring data. During the acquisition process, the timestamp of all data is completed based on the synchronous clocks at both ends. Then, the acquired data is subjected to standardized preprocessing to complete data resampling and time synchronization alignment. At the same time, noise reduction, normalization, and frequency offset correction are performed to eliminate noise interference, DC offset, and sampling rate deviation during the acquisition process, providing standardized data with unified format and time alignment for subsequent processing.

[0040] Based on the preprocessed link data, a multi-dimensional error source joint identification in the time and frequency domains is performed. Short-time Fourier transform and wavelet transform are applied to the link data to complete the joint time and frequency domain decomposition of the link data. In the time domain, the time-varying fluctuation characteristics of the link error are extracted, and in the frequency domain, the error amplitude and phase characteristics at different carrier frequencies are extracted. Combining the prior channel characteristics and device characteristics of the millimeter-wave transmission link, six core error sources are accurately separated and identified: atmospheric attenuation error, multipath propagation error, Doppler frequency shift error, phase noise error, device nonlinear distortion error, and transmit / receive synchronization delay error. At the same time, time-domain correlation analysis and frequency-domain distribution characteristic analysis of various error sources are completed, and the core feature parameters of various errors are extracted.

[0041] Based on the identified error source characteristics, a joint time-frequency domain dynamic error coupling model for millimeter-wave transmission links is constructed. The influence weights of various error sources on the transmission link performance are quantified. A fixed-duration sliding time window is set, and within each sliding time window, real-time status data and error feature data of the millimeter-wave link are collected. The model's various time-varying parameters are updated in real time based on the least squares fitting algorithm. At the same time, the weight allocation of different error sources in the model is dynamically adjusted based on the changing characteristics of the link scenario. When a switch in the link transmission scenario is detected, a rapid model reconstruction process is triggered, and the model parameters are rapidly converged and updated based on the link characteristics of the current scenario.

[0042] Based on the output of the dynamic error coupling model, a multi-dimensional joint dynamic error compensation strategy is generated, and hierarchical error compensation processing is performed. In the baseband digital domain, channel equalization, phase noise compensation, Doppler frequency shift correction, and nonlinear distortion pre-distortion processing are performed. Digital pre-distortion technology is used to pre-compensate the nonlinear distortion of the power amplifier. An adaptive equalization algorithm is used to compensate for the amplitude and phase errors caused by multipath channels and atmospheric attenuation. A phase-locked loop algorithm is used to complete carrier phase tracking and Doppler frequency shift correction. In the RF analog domain, adaptive adjustment of RF link gain, dynamic calibration of local oscillator frequency, and real-time compensation for delay error are performed. Through hierarchical joint compensation in the baseband digital domain and the RF analog domain, the coordinated suppression of all-dimensional errors in the millimeter-wave transmission link is achieved.

[0043] The system collects compensated millimeter-wave transmission link data, performs real-time verification of link transmission performance, and calculates the link error suppression ratio, bit error rate, and error vector amplitude—key transmission performance parameters—before and after compensation. Based on the link performance verification results, iterative optimization of the dynamic error coupling model and compensation strategy is performed. When the link error suppression ratio falls below a preset threshold, the iterative optimization process of the model and compensation strategy is triggered. The weight parameters of the dynamic error coupling model are updated using an online gradient descent algorithm. Simultaneously, the optimal compensation coefficient matrix is ​​resolved based on the updated error model to generate the optimized compensation strategy. During the iterative optimization process, iterative convergence conditions and the maximum number of iterations are set to ensure the real-time performance and convergence of the optimization process, thus completing the closed-loop dynamic error compensation of the millimeter-wave transmission link.

[0044] Table 1: Performance Comparison of This Method and Traditional Compensation Method for Fixed Point-to-Point Millimeter Wave Links

[0045] The data in Table 1 are derived from measured statistical data of the link's stable operation for 72 consecutive hours in this embodiment. Data for the traditional fixed compensation method is derived from comparative test results under the same link and configuration conditions. All indicators are set according to relevant industry standards for millimeter-wave communication system performance testing. The data shows that the method of this invention significantly outperforms the traditional fixed compensation method in core performance indicators such as link bit error rate, error vector amplitude, error suppression ratio, multipath error correction rate, and phase noise suppression ratio, while achieving stable link operation with zero interruption. This method achieves precise suppression of time-varying link errors through full-dimensional error coupling modeling and multi-domain joint compensation. Through closed-loop iterative optimization, it continuously adapts to changes in the link environment, effectively improving the transmission reliability and stability of fixed millimeter-wave backbone links.

[0046] Example 2

[0047] The application scenario of this embodiment is a vehicle-mounted millimeter-wave communication link in the 5G-Advanced vehicle-road cooperative scenario. It adopts the 39GHz millimeter-wave frequency band, with a system transmission bandwidth of 100MHz. The vehicle moving speed covers 0 to 120km / h. It is used for high-speed, low-latency data transmission between vehicles and roadside units, and between vehicles. The link transmission environment includes various scenarios such as urban roads, elevated roads, and tunnels. The fast time-varying channel characteristics, Doppler effect, and multipath reflection brought about by the high speed of vehicle movement place extremely high demands on the real-time performance and adaptability of the link dynamic error compensation.

[0048] Before implementation, the deployment and debugging of millimeter-wave communication equipment at the vehicle-mounted and roadside ends were completed. The vehicle-mounted end is equipped with a vehicle-mounted millimeter-wave radio frequency unit, an automotive-grade baseband signal processing unit, a high-precision vehicle-mounted synchronization clock unit, a vehicle motion status acquisition unit, a link status monitoring unit, and a real-time error compensation control unit. The roadside end is equipped with corresponding millimeter-wave communication equipment and a roadside data processing unit. The vehicle-mounted end and the roadside end achieve time synchronization through the Beidou high-precision positioning system and the synchronization clock. All vehicle-mounted equipment has completed automotive-grade environmental adaptation and anti-vibration and anti-interference debugging.

[0049] During link communication, the link data acquisition and preprocessing module completes the synchronous acquisition of multi-channel data. The acquired content includes baseband IQ signals from the vehicle-mounted end and the roadside end, RF link operating parameters, real-time vehicle speed and position data, roadside channel environment parameters, and transceiver device operating status data. Based on the high-precision synchronous clocks at both ends, the timestamp of all acquired data is completed, and then standardized preprocessing is performed to complete the resampling and time synchronization alignment of data at different sampling rates. At the same time, noise reduction, normalization, and frequency offset correction are performed to eliminate electromagnetic interference, DC offset, and frequency offset deviation caused by motion in the vehicle environment, ensuring the time synchronization and format consistency of the input data.

[0050] Based on the preprocessed link data, a multi-dimensional error source joint identification in the time and frequency domains is performed. Short-time Fourier transform and wavelet transform are performed on the link data to complete the joint time and frequency domain decomposition. In the time domain, the rapid time-varying fluctuation characteristics of link errors in high-speed mobile scenarios are extracted, and in the frequency domain, the error amplitude and phase characteristics at different carrier frequencies are extracted. Combining the prior channel characteristics and device characteristics of the vehicle-mounted millimeter-wave link, six types of core error sources are accurately separated and identified. The time-varying characteristics of Doppler frequency shift, fast fading channel, and multipath reflection errors caused by high-speed movement are analyzed in detail. The time-domain correlation analysis and frequency-domain distribution characteristic analysis of various error sources are completed, and the corresponding core feature parameters are extracted.

[0051] Based on the identified error source features, a dynamic error coupling model for millimeter-wave transmission links adapted to high-speed mobile scenarios is constructed. The influence weights of various error sources on the transmission link performance are quantified, and an adaptive sliding time window matching the vehicle's speed is set. Within each sliding time window, real-time link status data and error feature data are collected. The model's various time-varying parameters are updated in real time based on the recursive least squares algorithm. Simultaneously, the weight allocation of different error sources in the model is dynamically adjusted based on the vehicle's motion state and scene changes. When the vehicle is detected entering a tunnel, elevated road, or other scene transitions, a rapid model reconstruction process is triggered, and the model parameters are rapidly converged and updated based on the link features of the current scene.

[0052] Based on the output of the dynamic error coupling model, a multi-dimensional joint dynamic error compensation strategy adapted to fast time-varying links is generated. Hierarchical error compensation processing is performed. In the baseband digital domain, time-varying channel adaptive equalization, fast Doppler frequency shift tracking correction, real-time phase noise compensation, and digital pre-distortion processing of device nonlinear distortion are performed. Carrier phase tracking and fast Doppler frequency shift correction in high-speed mobile scenarios are completed through an improved phase-locked loop algorithm. Amplitude and phase error compensation of fast time-varying multipath channels are completed through an adaptive equalization algorithm. In the RF analog domain, fast automatic gain control of the RF link, dynamic local oscillator frequency calibration, and real-time transmission delay compensation are performed. Through hierarchical joint compensation in the baseband digital domain and the RF analog domain, the collaborative suppression of all-dimensional errors of the fast time-varying link is achieved.

[0053] The system collects compensated millimeter-wave transmission link data, performs real-time verification of link transmission performance, and calculates the link error suppression ratio, bit error rate, error vector magnitude, and transmission delay before and after compensation. Based on the link performance verification results, iterative optimization of the dynamic error coupling model and compensation strategy is performed. When the link error suppression ratio is lower than a preset threshold, the iterative optimization process of the model and compensation strategy is triggered. The weight parameters of the dynamic error coupling model are quickly updated using an online stochastic gradient descent algorithm. At the same time, the optimal compensation coefficient matrix is ​​resolved based on the updated error model to generate the optimized compensation strategy. During the iterative optimization process, the iterative convergence condition and the maximum number of iterations are set to ensure the real-time performance and convergence of the optimization process, thus completing the closed-loop dynamic error compensation of the vehicle-mounted millimeter-wave transmission link.

[0054] Table 2: Performance Comparison of This Method and Traditional Compensation Method for Vehicle-Mounted High-Speed ​​Mobile Millimeter-Wave Links

[0055] The data in Table 2 comes from real-vehicle test statistics of the vehicle-mounted link in this embodiment under full-scene urban road conditions, covering the entire speed range from 0 to 120 km / h. The data for the traditional fixed compensation method comes from comparative test results under the same equipment and scenario. All indicators meet the performance test specifications of automotive-grade millimeter-wave communication systems. The data shows that the method of this invention is significantly better than the traditional fixed compensation method in terms of link bit error rate, Doppler frequency shift correction accuracy, channel tracking response speed, and transmission delay stability in high-speed mobile scenarios, greatly reducing the link communication interruption rate. At the same time, the algorithm convergence time is much shorter than that of traditional methods. This method, through time-varying dynamic error coupling modeling and rapid iterative optimization, can accurately adapt to the fast time-varying characteristics of vehicle-mounted millimeter-wave links, realize real-time dynamic compensation of link errors in high-speed mobile scenarios, and effectively improve the reliability and real-time performance of vehicle-mounted millimeter-wave communication.

[0056] refer to Figure 1 This diagram illustrates the global operational logic of dynamic error compensation for millimeter-wave transmission links. The process begins with the synchronous acquisition of baseband signals and link status data from both the transceiver and receiver, establishing a data foundation through standardized preprocessing. Core components include time-frequency domain feature extraction from multi-dimensional error sources, construction of a dynamic error coupling model, and generation of a joint compensation strategy covering both the baseband digital domain and the radio frequency analog domain. Finally, the system verifies link performance (such as bit error rate and error vector amplitude) in real time and performs adaptive iterative optimization based on feedback results, forming a continuously evolving closed-loop management system to ensure high reliability of millimeter-wave transmission in time-varying environments.

[0057] Reference Figure 2 This figure details the technical approach for identifying error sources in the system. The system performs joint decomposition of the original signal using short-time Fourier transform and wavelet transform, analyzing it from both the time domain (wave characteristics) and frequency domain (amplitude and phase characteristics). The identification process covers six core errors unique to millimeter-wave links: atmospheric attenuation, multipath propagation, Doppler shift, phase noise, device nonlinear distortion, and transmit / receive synchronization delay. By analyzing the correlation and distribution characteristics of these error sources, high-quality feature inputs are provided for subsequently establishing an accurate coupling model.

[0058] Reference Figure 3This figure illustrates how an error-coupled model for a millimeter-wave link is constructed and updated in real time. This model comprehensively considers spatial transmission characteristics (atmosphere / rainfall), channel propagation characteristics (multipath / Doppler), and hardware characteristics (phase noise / synchronization error). To cope with rapid environmental changes, a sliding time window mechanism is introduced. Within each window, the model parameters are fitted and updated using a least-squares algorithm. Furthermore, when the system detects a change in transmission scenario (such as from clear weather to rain or from stationary to high-speed movement), a rapid reconstruction process is triggered to ensure that the model always accurately depicts the current link state.

[0059] Reference Figure 4 This figure illustrates the execution details of the millimeter-wave link compensation strategy at different physical levels. The compensation task is divided into two levels: the baseband digital domain and the RF analog domain, to achieve coordinated suppression. In the digital domain, the focus is on handling complex nonlinear distortion, channel equalization, Doppler shift correction, and phase noise compensation; while in the analog domain, the emphasis is on real-time hardware-level control, including adaptive adjustment of RF gain, local oscillator frequency calibration, and compensation for physical link delay. This hierarchical collaborative approach maximizes the advantages of each level, balancing computational complexity and real-time compensation performance.

[0060] refer to Figure 5 This diagram illustrates how the system drives strategy optimization through performance feedback. During the verification phase, the system calculates the error suppression ratio, error vector magnitude (EVM), and bit error rate (BER). If these metrics fail to meet the targets, the adaptive iterative optimization module is activated. This process uses an online gradient descent algorithm to update model parameters and employs multiple regularization constraints (including sparsity, total variation, and time-varying smoothness constraints) to solve for the optimal compensation coefficient matrix. This method, while ensuring compensation accuracy, enhances the robustness and smoothness of the compensation coefficients in dynamic environments through regularization, avoiding link instability caused by drastic changes in compensation parameters.

[0061] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A dynamic error compensation method for millimeter-wave transmission links, characterized in that, Includes the following steps: Collect real-time status data of the millimeter-wave transmission link and baseband signal data from the transceiver end, and complete the time synchronization alignment and standardization preprocessing of the data; Based on the preprocessed link data, multi-dimensional error sources of millimeter-wave transmission links are identified, and time-frequency domain feature extraction and characteristic analysis of various error sources are completed. Based on the identified error source characteristics, a dynamic error coupling model for millimeter-wave transmission links is constructed to quantify the impact weights of various error sources on transmission link performance. Based on the output of the dynamic error coupling model, a multi-dimensional joint dynamic error compensation strategy is generated, and corresponding levels of error compensation processing are performed in the baseband digital domain and the radio frequency analog domain. Collect compensated millimeter-wave transmission link data, perform real-time verification of link transmission performance, and calculate the link error suppression ratio and core transmission performance parameters before and after compensation. Based on the link performance verification results, adaptive iterative optimization is performed on the dynamic error coupling model and compensation strategy to complete the closed-loop dynamic error compensation of the millimeter-wave transmission link.

2. The method for dynamic error compensation of millimeter-wave transmission links according to claim 1, characterized in that, It also includes a full-dimensional time-varying error coupling modeling step for millimeter-wave transmission links, constructing a joint dynamic error coupling model in the time and frequency domains covering channel transmission characteristics, device nonlinearity, and transceiver synchronization deviation. The model is calculated as follows: ; for Time Frequency The comprehensive complex error value of the millimeter-wave transmission link at that location. for Time Frequency Atmospheric absorption attenuation coefficient at that location for Time Frequency Rainfall attenuation coefficient at the location, This represents the total number of multipath components in a millimeter-wave transmission link. for Time of the first Channel complex gain of multiple path components, for Time of the first The transmission delay of a multipath component. for Time of the first Doppler frequency shift of multipath components, for Time Frequency Phase noise introduced by the local oscillator at the transmitting and receiving ends. for Time Frequency Phase distortion introduced by the nonlinearity of the power amplifier. for The synchronization delay error at the transceiver end of the millimeter-wave transmission link is determined by integrating multiple error sources in the millimeter-wave transmission link, including time-varying channel attenuation, multipath propagation, Doppler effect, device nonlinearity, and synchronization delay.

3. The method for dynamic error compensation of millimeter-wave transmission links according to claim 1, characterized in that, It also includes a multi-dimensional error source time-frequency domain joint identification step, specifically performing short-time Fourier transform and wavelet transform on the preprocessed link data to complete the time-frequency domain joint decomposition of the link data, extracting the time-varying fluctuation characteristics of the link error in the time domain dimension, and extracting the error amplitude and phase characteristics at different carrier frequency points in the frequency domain dimension. Combining the prior channel characteristics and device characteristics of the millimeter wave transmission link, it simultaneously completes the time-domain correlation analysis and frequency-domain distribution characteristic analysis of various error sources.

4. The method for dynamic error compensation of millimeter-wave transmission links according to claim 1, characterized in that, It also includes a time-varying update step for the dynamic error coupling model. Specifically, a fixed-duration sliding time window is set. Within each sliding time window, real-time status data and error characteristic data of the millimeter-wave link are collected. The model is updated in real time based on the least squares fitting algorithm. At the same time, the weight distribution of different error sources in the model is dynamically adjusted based on the changing characteristics of the link scenario. When a switch in the link transmission scenario is detected, the model is triggered to perform a rapid reconstruction process.

5. The method for dynamic error compensation of millimeter-wave transmission links according to claim 1, characterized in that, It also includes an adaptive optimization solution step for the multi-dimensional joint error compensation coefficients, the solution method is as follows: ; The optimal multi-dimensional joint error compensation coefficient matrix. Let be the compensation coefficient matrix to be solved. For the compensated ideal baseband signal matrix, This is the original baseband signal matrix at the receiving end. For L2 norm operations, For L1 norm operations, For Frobenius norm operations, These are the sparse regularization weight coefficients. These are the total variation regularization weight coefficients. These are the time-varying smoothing regularization weight coefficients. For total variation operations, For time First-order partial differential operations The magnitude of the error vector corresponding to the compensation coefficient matrix. The system's preset error vector magnitude threshold, The system bit error rate corresponding to the compensation coefficient matrix, The preset bit error rate threshold for the system, To solve for the variable values ​​that minimize the objective function.

6. The method for dynamic error compensation of millimeter-wave transmission links according to claim 1, characterized in that, It also includes hierarchical error compensation execution steps, specifically based on the generated multi-dimensional joint dynamic error compensation strategy. In the baseband digital domain, channel equalization, phase noise compensation, Doppler frequency shift correction, and nonlinear distortion pre-distortion processing are performed. Digital pre-distortion technology is used to pre-compensate the nonlinear distortion of the power amplifier. An adaptive equalization algorithm is used to compensate for the amplitude and phase errors caused by multipath channels and atmospheric attenuation. A phase-locked loop algorithm is used to complete carrier phase tracking and Doppler frequency shift correction. In the RF analog domain, adaptive adjustment of RF link gain, dynamic calibration of local oscillator frequency, and real-time compensation for time delay error are performed through hierarchical joint compensation in the baseband digital domain and the RF analog domain.

7. The method for dynamic error compensation of millimeter-wave transmission links according to claim 1, characterized in that, It also includes a closed-loop adaptive iterative optimization step, which is to calculate the error suppression effect and link performance improvement of the current compensation strategy based on the results of real-time verification of link performance. When the link error suppression ratio is lower than the preset threshold, the iterative optimization process of the model and compensation strategy is triggered. The weight parameters of the dynamic error coupling model are updated through the online gradient descent algorithm, and the optimal compensation coefficient matrix is ​​resolved based on the updated error model.

8. A dynamic error compensation system for millimeter-wave transmission links, characterized in that, The system, employing the method described in any one of claims 1-7, comprises the following modules: The link data acquisition and preprocessing module is used to acquire real-time status data and baseband signal data from the transceiver end of the millimeter-wave transmission link, and to complete the time synchronization alignment and standardization preprocessing of the data. The link error source identification and feature extraction module is used to identify multi-dimensional error sources of millimeter-wave transmission links based on preprocessed link data, and to complete time-frequency domain feature extraction and characteristic analysis of various error sources. The dynamic error coupling modeling module is used to construct a dynamic error coupling model of the millimeter-wave transmission link based on the identified error source characteristics, and to quantify the impact weight of various error sources on the transmission link performance. The multi-dimensional joint error compensation module is used to generate a multi-dimensional joint dynamic error compensation strategy based on the output of the dynamic error coupling model, and perform corresponding levels of error compensation processing in the baseband digital domain and the radio frequency analog domain respectively. The compensation effect verification module is used to collect the compensated millimeter-wave transmission link data, perform real-time verification of the link transmission performance, and calculate the link error suppression ratio and core transmission performance parameters before and after compensation. The adaptive iterative optimization module is used to perform adaptive iterative optimization of the dynamic error coupling model and compensation strategy based on the link performance verification results, so as to complete the closed-loop dynamic error compensation of the millimeter-wave transmission link.

9. The millimeter-wave transmission link dynamic error compensation system according to claim 8, characterized in that, The link data acquisition and preprocessing module has a built-in multi-channel synchronous acquisition unit, a time-frequency domain preprocessing unit, and a data synchronization unit. The multi-channel synchronous acquisition unit is used to complete the multi-channel synchronous acquisition of baseband IQ signals, RF link status parameters, channel environment parameters, and transceiver device operating parameters at the transceiver end. The time-frequency domain preprocessing unit is used to complete the noise reduction, normalization, and frequency offset correction preprocessing of the acquired data to eliminate noise interference and DC offset during the acquisition process. The data synchronization unit is used to complete the timestamp alignment and resampling processing of data from different sources and with different sampling rates based on the synchronous clock signal of the transceiver end.

10. A dynamic error compensation system for millimeter-wave transmission links according to claim 8, characterized in that, The adaptive iterative optimization module incorporates an online learning optimization unit, a model parameter update unit, and a compensation strategy reconstruction unit. The online learning optimization unit is used to iteratively optimize and solve the model parameters and compensation coefficients based on the link performance verification results through online machine learning algorithms. The model parameter update unit is used to update the time-varying parameters of the dynamic error coupling model and adaptively reconstruct the scenario. The compensation strategy reconstruction unit is used to regenerate a multi-dimensional joint compensation strategy adapted to the current link state based on the optimized model output results, and simultaneously complete the smooth switching and execution of the compensation strategy.

Images