Wireless communication signal analysis method and system

Abstract

The invention discloses a wireless communication signal analysis method and system, relates to the technical field of wireless communication signal analysis, and aims at remarkably improving the dynamic anti-interference capability in a millimeter wave communication scene. A time-frequency double-domain compressed sampling mechanism is introduced, so that the signal integrity is guaranteed, the front-end processing load is reduced, and a high-density spectrum environment is adapted; the combination of the multipath feature library and the real-time phase compensation effectively deals with the time-varying phase distortion caused by the metal reflector, and the robustness of a complex industrial scene is enhanced; knowledge sharing among edge nodes is realized by adopting a lightweight model under a federated learning architecture, and the dependence of traditional centralized training on data timeliness is overcome; in addition, the dual-mode resource scheduling strategy gives consideration to both steady-state optimization and burst response requirements, and ensures the service quality of key services.

Description

Technical Field

[0001] This invention relates to the field of wireless communication signal analysis technology, and in particular to a method and system for analyzing wireless communication signals. Background Technology

[0002] Currently, the millimeter wave band 24-100GHz has become the core carrier for 5G communication, industrial IoT and vehicle networking. Its large bandwidth characteristics can support emerging applications such as ultra-high-definition video transmission and millisecond-level control of industrial equipment.

[0003] In scenarios such as smart manufacturing and smart ports, facilities such as metal frame factory buildings and automated production lines create strong reflection environments, causing phase distortion and inter-symbol interference in signal propagation through multiple paths. These complex propagation characteristics place stringent requirements on existing signal analysis systems, which need to complete multipath separation and signal reconstruction within microsecond time windows.

[0004] In recent years, mainstream solutions have adopted large-scale MIMO beamforming combined with deep learning methods. For example, the intelligent reflector RIS technology uses a programmable surface to control the phase of electromagnetic waves, and some solutions use time delay estimation models based on convolutional neural networks. However, in real industrial scenarios, dynamic reflectors such as mobile robotic arms and metal conveyor belts can cause time-varying multipath effects, making it difficult for fixed-configuration RIS units to match environmental changes in real time. At the same time, deep learning models rely on a large amount of labeled data for training, making it difficult to cope with the domain offset problem caused by frequent adjustments to the layout of workshop equipment. Some cutting-edge research has attempted to introduce federated learning frameworks, but the computing power limitations of edge nodes cause model updates to lag behind the speed of environmental changes. Summary of the Invention

[0005] In view of the aforementioned existing problems, the present invention is proposed.

[0006] This invention provides a method and system for analyzing wireless communication signals, addressing the core pain points of existing millimeter-wave communication systems in dynamic industrial environments, such as severe multipath interference, insufficient real-time compensation capabilities, and limited edge computing power, which restrict the realization of high-reliability communication.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide a method for analyzing wireless communication signals, comprising, Step S1: Receive millimeter-wave band signals through a multiple-input multiple-output (MIMO) antenna array, and perform time-frequency dual-domain compressed sampling on the signals at the signal processing front end; Step S2: Based on the preset multipath feature library, match the material parameters of the reflective surface in the current environment to generate a dynamic phase compensation template; Step S3: The phase offset of the signal propagation path is corrected in real time using a phase fine-tuning array. The adjustment parameters of the phase fine-tuning array are dynamically adjusted based on the real-time spatial reflector distribution information obtained by the environmental perception module. Step S4: Perform joint blind source separation and symbol demodulation on the corrected signal to extract valid communication data; Step S5: The local demodulation features are collaboratively trained with neighboring nodes through a federated learning network to update the global interference suppression model; Step S6: Output the signal analysis results after multipath interference compensation and trigger the dynamic allocation command for spectrum resources.

[0008] As a preferred embodiment of the wireless communication signal analysis method of the present invention, the time-frequency dual-domain compressed sampling includes: in the time domain, a non-uniform sampling window allocation strategy is adopted to dynamically adjust the sampling density according to the signal burst probability; in the frequency domain, a sub-band energy detection algorithm is used to select high signal-to-noise ratio frequency bands for key sampling, and the remaining frequency bands are sampled using dimension reduction projection.

[0009] In a preferred embodiment of the wireless communication signal analysis method of the present invention, the phase fine-tuning array comprises: The piezoelectric ceramic sheet, physically integrated with the MIMO antenna unit, achieves micron-level length adjustment of the electromagnetic wave propagation path through mechanical deformation; The driving voltage of the piezoelectric ceramic sheet is dynamically adjusted according to the real-time channel impulse response calculation results, and the adjustment period is less than 1 / 5 of the current signal symbol interval.

[0010] In a preferred embodiment of the wireless communication signal analysis method of the present invention, during the dynamic adjustment process based on the real-time channel impulse response calculation results in step S3, the following operations are performed on the i-th antenna element within the n-th symbol period to perform real-time correction of the propagation path phase offset. The specific process includes: Phase shift is estimated based on weighted summation of multipath impulse responses: , in, This represents the phase shift of the i-th antenna element in the n-th symbol period. This indicates that the i-th antenna element has a time delay of in the nth symbol period. The impulse response sample value at the location, This represents the time delay of the l-th multipath component. This represents the weighting coefficient of the l-th multipath component. Indicates the total number of multipath components. For complex number argument operations; The phase offset is converted into the desired fine-tuning path length using the following formula: , in, The micrometer-level adjustment representing the path length. The wavelength of a millimeter-wave signal. Pi is a constant. Calculate the voltage driving the piezoelectric ceramic sheet: , in, This represents the driving voltage applied to the i-th unit. This represents the voltage-strain conversion factor of piezoelectric ceramics, in units of... ; set up: , in, Indicates the symbol interval of the current signal. This indicates the update period for phase fine-tuning. This represents the mechanical response time constant of the piezoelectric ceramic sheet; Further, from the mechanical bandwidth requirements, we can obtain: , in, This indicates the -3dB bandwidth of the piezoelectric ceramic sheet, in Hz.

[0011] Secondly, the present invention provides a wireless communication signal analysis system, comprising, The signal acquisition module includes a MIMO antenna array supporting the millimeter-wave frequency band and a time-frequency dual-domain compressed sampling unit. The compressed sampling unit uses sub-band segmentation technology to isolate interference signals outside the target frequency band. The dynamic compensation module integrates a phase fine-tuning array and a multipath feature matching unit. The multipath feature matching unit stores electromagnetic parameters of at least 16 types of reflective surface materials in industrial scenarios and generates compensation strategies in conjunction with real-time environmental scanning data. The collaborative processing module, deployed on edge computing nodes, includes federated learning agent units and lightweight signal demodulation models. The federated learning agent units achieve incremental model synchronization among multiple nodes through low-power wide-area networks. The resource scheduling module dynamically adjusts the channel allocation scheme based on real-time spectrum analysis results and sends power control commands to the communication equipment. The resource scheduling module includes a dual-mode decision-making unit: In a steady-state environment, a long-term optimization strategy based on reinforcement learning is adopted; In the event of sudden interference, the rapid response mode is activated, shortening the decision-making cycle to 1-5 milliseconds.

[0012] In a preferred embodiment of the wireless communication signal analysis system of the present invention, the dynamic compensation module further includes: The environmental perception submodule integrates millimeter-wave radar and optical sensors to construct a three-dimensional spatial reflector distribution map; The phase calibration submodule adopts a two-level compensation mechanism. The first level performs coarse compensation based on a preset feature library, and the second level performs microsecond-level fine adjustment based on real-time environmental data.

[0013] As a preferred embodiment of the wireless communication signal analysis system of the present invention, the two-level compensation mechanism includes a first-level coarse compensation and a second-level fine adjustment. The first-level coarse compensation generates an initial phase correction matrix by matching the preset material reflection coefficients in the multipath feature library. The second-level fine adjustment updates the compensation parameters based on the real-time motion trajectory of the reflector obtained by the millimeter-wave radar using a sliding window prediction algorithm.

[0014] In a preferred embodiment of the wireless communication signal analysis system described in this invention, the lightweight signal demodulation model in the collaborative processing module is generated using knowledge distillation technology, and includes: The teacher model deployed at the central node uses a deep residual network structure to process the full spectrum data; The student model deployed at the edge node adopts a pruned capsule network structure and is updated synchronously with the teacher model through a time-sensitive distillation strategy. The time-sensitive distillation strategy includes: during working hours, the teacher model supervises the output layer features of the student model online; during idle periods, the intermediate layer features of the teacher model are encoded into knowledge vectors and transmitted to edge nodes for cached training through a differential privacy mechanism.

[0015] In a preferred embodiment of the wireless communication signal analysis system of the present invention, the process of activating the fast response mode in the resource scheduling module under sudden interference scenarios includes: When the system detects that the instantaneous interference power exceeds a preset threshold, the resource scheduling module automatically switches to fast response mode. The trigger condition is defined as follows: , in, Indicates the first Instantaneous interference power index within each decision cycle, This indicates the preset interference power threshold; The decision-making cycle in fast response mode consists of sampling delay and computation delay: , in, Indicates the decision-making cycle. This indicates the number of sampling points processed in each decision. This indicates the hardware sampling rate, measured in samples per second. This represents the algorithm's complexity coefficient, in units of... , Indicates the complexity index of the algorithm; satisfy: , in, This indicates the maximum allowable decision-making period range, in milliseconds (ms). If a linear time algorithm is chosen ,but , in, For the sampling delay at each point, This represents the processing delay coefficient for each point.

[0016] As a preferred embodiment of the wireless communication signal analysis system of the present invention, it further includes a hardware multiplexing unit: The power management chip is shared between the driving circuit of the phase fine-tuning array and the RF components of the local LoRaWAN communication module.

[0017] The beneficial effects of this invention are as follows: It significantly improves the dynamic anti-interference capability in millimeter-wave communication scenarios; the introduction of a time-frequency dual-domain compressed sampling mechanism reduces the front-end processing load while ensuring signal integrity, adapting to high-density spectrum environments; the combination of a multipath feature library and real-time phase compensation effectively addresses time-varying phase distortion caused by metal reflectors, enhancing robustness in complex industrial scenarios; the use of a lightweight model under a federated learning architecture enables knowledge sharing among edge nodes, overcoming the dependence of traditional centralized training on data timeliness; furthermore, the dual-mode resource scheduling strategy balances steady-state optimization and burst response requirements, ensuring the service quality of critical businesses; and the hardware reuse design reduces deployment costs through functional module integration, solving a core obstacle to the commercialization of advanced compensation technologies. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a wireless communication signal analysis method in Example 1.

[0020] Figure 2This is a schematic diagram of the framework of a wireless communication signal analysis system in Example 2. Detailed Implementation

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0023] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0024] Example 1, referring to Figure 1 This embodiment provides a method for analyzing wireless communication signals, including the following steps: Step S1: Receive millimeter-wave band signals through a multiple-input multiple-output (MIMO) antenna array, and perform time-frequency dual-domain compressed sampling on the signals at the signal processing front end; Time-frequency dual-domain compressed sampling includes: in the time domain, a non-uniform sampling window allocation strategy is adopted to dynamically adjust the sampling density according to the signal burst probability; in the frequency domain, a sub-band energy detection algorithm is used to select high signal-to-noise ratio frequency bands for key sampling, while other frequency bands are sampled using dimension reduction projection. Step S2: Based on the preset multipath feature library, match the material parameters of the reflective surface in the current environment to generate a dynamic phase compensation template; Step S3: The phase offset of the signal propagation path is corrected in real time using a phase fine-tuning array. The control parameters of the phase fine-tuning array are dynamically adjusted based on the real-time spatial reflector distribution information obtained by the environmental perception module. The phase fine-tuning array includes: The piezoelectric ceramic sheet, physically integrated with the MIMO antenna unit, achieves micron-level length adjustment of the electromagnetic wave propagation path through mechanical deformation; The driving voltage of the piezoelectric ceramic sheet is dynamically adjusted based on the real-time channel impulse response calculation results, and the adjustment period is less than 1 / 5 of the current signal symbol interval; In step S3, during the dynamic adjustment process based on the real-time channel impulse response calculation results, the following operations are performed on the i-th antenna element within the n-th symbol period to perform real-time correction of the propagation path phase offset. The specific process includes: Phase shift is estimated based on weighted summation of multipath impulse responses: , in, This represents the phase shift of the i-th antenna element in the n-th symbol period. This indicates that the i-th antenna element has a time delay of in the nth symbol period. The impulse response sample value at the location, This represents the time delay of the l-th multipath component. This represents the weighting coefficient of the l-th multipath component. Indicates the total number of multipath components. For complex number argument operations; The phase offset is converted into the desired fine-tuning path length using the following formula: , in, The micrometer-level adjustment representing the path length. The wavelength of a millimeter-wave signal. Pi is a constant. Calculate the voltage driving the piezoelectric ceramic sheet: , in, This represents the driving voltage applied to the i-th unit. This represents the voltage-strain conversion factor of piezoelectric ceramics, in units of... ; To ensure that the adjustment period is less than one-fifth of the symbol interval, let: , in, Indicates the symbol interval of the current signal. This indicates the update period for phase fine-tuning. This represents the mechanical response time constant of the piezoelectric ceramic sheet; Further, from the mechanical bandwidth requirements, we can obtain: , in, This indicates the -3dB bandwidth of the piezoelectric ceramic sheet, in Hz. Specifically, the phase shift is calculated by real-time sampling of the multipath impulse response and mapped to a micrometer-level mechanical displacement to form a closed-loop correction. The weighted argument calculation takes into account the signal-to-noise ratio and time delay characteristics of each multipath, improving the phase estimation accuracy. The length conversion formula and voltage mapping formula are connected to reduce the calculation delay between steps. The update cycle is strictly limited to within one-fifth of the symbol interval. Combined with the mechanical time constant and bandwidth index of piezoelectric ceramics, fine-tuning will not cause lag or additional phase error. While ensuring real-time performance, it also takes into account the hardware response limit and signal processing requirements, enabling the system to have a fast and stable phase compensation capability for environmental changes, thereby improving the stability of the communication link and the efficiency of spectrum utilization. Step S4: Perform joint blind source separation and symbol demodulation on the corrected signal to extract valid communication data; Step S5: The local demodulation features are collaboratively trained with neighboring nodes through a federated learning network to update the global interference suppression model; Step S6: Output the signal analysis results after multipath interference compensation and trigger the dynamic allocation command for spectrum resources.

[0025] Example 2, refer to Figure 2 This embodiment provides a wireless communication signal analysis system, including: The signal acquisition module includes a MIMO antenna array supporting the millimeter-wave frequency band and a time-frequency dual-domain compressed sampling unit. The compressed sampling unit uses sub-band segmentation technology to isolate interference signals outside the target frequency band. The dynamic compensation module integrates a phase fine-tuning array and a multipath feature matching unit. The multipath feature matching unit stores electromagnetic parameters of at least 16 types of reflective surface materials in industrial scenarios and generates compensation strategies in conjunction with real-time environmental scanning data. The dynamic compensation module further includes: The environmental perception submodule integrates millimeter-wave radar and optical sensors to construct a three-dimensional spatial reflector distribution map; The phase calibration submodule adopts a two-level compensation mechanism. The first level performs coarse compensation based on a preset feature library, and the second level performs microsecond-level fine adjustment based on real-time environmental data. The two-level compensation mechanism includes a first-level coarse compensation and a second-level fine adjustment. The first-level coarse compensation generates an initial phase correction matrix by matching the preset material reflection coefficients in the multipath feature library. The second-level fine adjustment updates the compensation parameters based on the real-time motion trajectory of the reflector obtained by the millimeter-wave radar using a sliding window prediction algorithm. The collaborative processing module, deployed on edge computing nodes, includes federated learning agent units and lightweight signal demodulation models. The federated learning agent units achieve incremental model synchronization among multiple nodes through low-power wide-area networks. The lightweight signal demodulation model in the collaborative processing module is generated using knowledge distillation techniques, including: The teacher model deployed at the central node uses a deep residual network structure to process the full spectrum data; The student model deployed at the edge nodes adopts a pruned capsule network structure and is updated synchronously with the teacher model through a time-sensitive distillation strategy. The time-sensitive distillation strategy includes: during working hours, the teacher model supervises the output layer features of the student model online; during idle periods, the intermediate layer features of the teacher model are encoded into knowledge vectors and transmitted to edge nodes for cached training through a differential privacy mechanism. The resource scheduling module dynamically adjusts the channel allocation scheme based on real-time spectrum analysis results and sends power control commands to the communication equipment. The resource scheduling module includes a dual-mode decision-making unit: In a steady-state environment, a long-term optimization strategy based on reinforcement learning is adopted, with a decision cycle of 10-60 seconds; In the event of sudden interference, a rapid response mode is activated, shortening the decision-making cycle to 1-5 milliseconds, and prioritizing the communication quality of critical equipment. In the resource scheduling module, the process for activating the rapid response mode in the event of sudden interference includes: When the system detects that the instantaneous interference power exceeds a preset threshold, the resource scheduling module automatically switches to fast response mode. The trigger condition is defined as follows: , in, Indicates the first Instantaneous interference power index within each decision cycle, This indicates the preset interference power threshold; The decision-making cycle in fast response mode consists of sampling delay and computation delay: , in, Indicates the decision-making cycle. This indicates the number of sampling points processed in each decision. This indicates the hardware sampling rate, measured in samples per second. This represents the algorithm's complexity coefficient, in units of... , Indicates the complexity index of the algorithm; satisfy: , in, This indicates the maximum allowable decision-making period range, in milliseconds (ms). If a linear time algorithm is chosen ,but , in, For the sampling delay at each point, This represents the processing delay coefficient for each point. Specifically, a fast switching strategy based on a preset power threshold enables the system to immediately enter a low-latency mode when sudden interference occurs. The decision cycle is determined by both the hardware sampling rate and the algorithm complexity, and the general expression is as follows: The system performance can be quantitatively evaluated under different numbers of sampling points and algorithm configurations. In the most commonly used linear algorithm scenarios, closed-form solutions can be used to evaluate the system performance. The maximum processable sampling volume is directly determined, simplifying online calculations and ensuring the strict latency requirement of 1-5 milliseconds. The latency time can be adjusted as needed. The wireless communication signal analysis system further includes a hardware multiplexing unit: The power management chip is shared between the driving circuit of the phase fine-tuning array and the RF components of the local LoRaWAN communication module. In smart street light deployment scenarios, some intermediate spectrum analysis data is transmitted via power line carrier communication links.

[0026] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for analyzing wireless communication signals, characterized in that, include, Step S1: Receive millimeter-wave band signals through a multiple-input multiple-output (MIMO) antenna array, and perform time-frequency dual-domain compressed sampling on the signals at the signal processing front end; Step S2: Based on the preset multipath feature library, match the material parameters of the reflective surface in the current environment to generate a dynamic phase compensation template; Step S3: The phase offset of the signal propagation path is corrected in real time using a phase fine-tuning array. The control parameters of the phase fine-tuning array are dynamically adjusted based on the real-time spatial reflector distribution information obtained by the environmental perception module. Step S4: Perform joint blind source separation and symbol demodulation on the corrected signal to extract valid communication data; Step S5: The local demodulation features are collaboratively trained with neighboring nodes through a federated learning network to update the global interference suppression model; Step S6: Output the signal analysis results after multipath interference compensation and trigger the dynamic allocation command for spectrum resources.

2. The method for analyzing wireless communication signals as described in claim 1, characterized in that, The time-frequency dual-domain compressed sampling includes: in the time domain, a non-uniform sampling window allocation strategy is adopted to dynamically adjust the sampling density according to the signal burst probability; in the frequency domain, a sub-band energy detection algorithm is used to select high signal-to-noise ratio frequency bands for key sampling, and the remaining frequency bands are sampled using dimensionality reduction projection.

3. The method for analyzing wireless communication signals as described in claim 1, characterized in that, The phase fine-tuning array includes: The piezoelectric ceramic sheet, physically integrated with the MIMO antenna unit, achieves micron-level length adjustment of the electromagnetic wave propagation path through mechanical deformation; The driving voltage of the piezoelectric ceramic sheet is dynamically adjusted according to the real-time channel impulse response calculation results, and the adjustment period is less than 1 / 5 of the current signal symbol interval.

4. The method for analyzing wireless communication signals as described in claim 3, characterized in that, In step S3, during the dynamic adjustment process based on the real-time channel impulse response calculation results, the following operations are performed on the i-th antenna element within the n-th symbol period to perform real-time correction of the propagation path phase offset. The specific process includes: Phase shift is estimated based on weighted summation of multipath impulse responses: ， in, This represents the phase shift of the i-th antenna element in the n-th symbol period. This indicates that the i-th antenna element has a time delay of in the nth symbol period. The impulse response sample value at the location, This represents the time delay of the l-th multipath component. This represents the weighting coefficient of the l-th multipath component. Indicates the total number of multipath components. For complex number argument operations; The phase offset is converted into the desired fine-tuning path length using the following formula: ， in, The micrometer-level adjustment representing the path length. The wavelength of a millimeter-wave signal. Pi is a constant. Calculate the voltage driving the piezoelectric ceramic sheet: ， in, This represents the driving voltage applied to the i-th unit. This represents the voltage-strain conversion factor of piezoelectric ceramics, in units of... ; set up: ， in, Indicates the symbol interval of the current signal. This indicates the update period for phase fine-tuning. This represents the mechanical response time constant of the piezoelectric ceramic sheet; Further, from the mechanical bandwidth requirements, we can obtain: ， in, This indicates the -3dB bandwidth of the piezoelectric ceramic sheet, in Hz.

5. A wireless communication signal analysis system, based on the wireless communication signal analysis method according to any one of claims 1 to 4, characterized in that, include: The signal acquisition module includes a MIMO antenna array supporting the millimeter-wave frequency band and a time-frequency dual-domain compressed sampling unit. The compressed sampling unit uses sub-band segmentation technology to isolate interference signals outside the target frequency band. The dynamic compensation module integrates a phase fine-tuning array and a multipath feature matching unit. The multipath feature matching unit stores electromagnetic parameters of at least 16 types of reflective surface materials in industrial scenarios and generates compensation strategies in conjunction with real-time environmental scanning data. The collaborative processing module, deployed on edge computing nodes, includes federated learning agent units and lightweight signal demodulation models. The federated learning agent units achieve incremental model synchronization among multiple nodes through low-power wide-area networks. The resource scheduling module dynamically adjusts the channel allocation scheme based on real-time spectrum analysis results and sends power control commands to the communication equipment. The resource scheduling module includes a dual-mode decision-making unit: In a steady-state environment, a long-term optimization strategy based on reinforcement learning is adopted; In the event of sudden interference, the rapid response mode is activated, shortening the decision-making cycle to 1-5 milliseconds.

6. The wireless communication signal analysis system as described in claim 5, characterized in that, The dynamic compensation module further includes: The environmental perception submodule integrates millimeter-wave radar and optical sensors to construct a three-dimensional spatial reflector distribution map; The phase calibration submodule adopts a two-level compensation mechanism. The first level performs coarse compensation based on a preset feature library, and the second level performs microsecond-level fine adjustment based on real-time environmental data.

7. The wireless communication signal analysis system as described in claim 6, characterized in that, The two-level compensation mechanism includes a first-level coarse compensation and a second-level fine adjustment. The first-level coarse compensation generates an initial phase correction matrix by matching the preset material reflection coefficients in the multipath feature library. The second-level fine adjustment updates the compensation parameters based on the real-time motion trajectory of the reflector obtained by the millimeter-wave radar using a sliding window prediction algorithm.

8. The wireless communication signal analysis system as described in claim 5, characterized in that, The lightweight signal demodulation model in the collaborative processing module is generated using knowledge distillation technology, including: The teacher model deployed at the central node uses a deep residual network structure to process the full spectrum data; The student model deployed at the edge nodes adopts a pruned capsule network structure and is updated synchronously with the teacher model through a time-sensitive distillation strategy. The time-sensitive distillation strategy includes: during working hours, the teacher model supervises the output layer features of the student model online; during idle periods, the intermediate layer features of the teacher model are encoded into knowledge vectors and transmitted to edge nodes for cached training through a differential privacy mechanism.

9. The wireless communication signal analysis system as described in claim 5, characterized in that, The process for activating the rapid response mode in the resource scheduling module under sudden interference scenarios includes: When the system detects that the instantaneous interference power exceeds a preset threshold, the resource scheduling module automatically switches to fast response mode. The trigger condition is defined as follows: ， in, Indicates the first Instantaneous interference power index within each decision cycle, This indicates the preset interference power threshold; The decision-making cycle in fast response mode consists of sampling delay and computation delay: ， in, Indicates the decision-making cycle. This indicates the number of sampling points processed in each decision. This indicates the hardware sampling rate, measured in samples per second. This represents the algorithm's complexity coefficient, in units of... , Indicates the complexity index of the algorithm; satisfy: ， in, This indicates the maximum allowable decision-making period range, in milliseconds (ms). If a linear time algorithm is chosen ,but ， in, For the sampling delay at each point, This represents the processing delay coefficient for each point.

10. The wireless communication signal analysis system as described in claim 9, characterized in that, It also includes a hardware multiplexing unit: The power management chip is shared between the driving circuit of the phase fine-tuning array and the RF components of the local LoRaWAN communication module.

Images