A method and system for communication and sensing signal processing for low-altitude environments.

Abstract

This invention discloses a signal processing method and system for low-altitude communication and sensing, relating to the field of integrated low-altitude communication and sensing signal processing technology. Specifically, this invention combines a frame detection algorithm based on PES symbol autocorrelation characteristics with a residual carrier frequency offset estimation method in the fractional Fourier transform domain to achieve high-precision frame synchronization and frequency offset compensation for mixed signals. Furthermore, by combining full-band channel estimation based on Wiener interpolation and an MMSE frequency domain equalizer, the impact of noise and channel distortion on communication signals is effectively eliminated. By combining ideal constellation point perturbation analysis with continuous phase unwrapping technology using the Goldstein branch cutting method, continuous reconstruction of weak phase perturbation information in communication signals is achieved. Finally, by combining continuous wavelet transform and instantaneous frequency redistribution algorithms, a high-resolution micro-Doppler spectrum is generated, improving the sensing added value of the communication signal.

Description

Technical Field

[0001] This invention relates to the field of low-altitude communication and sensing integrated signal processing technology, and in particular to a communication and sensing signal processing method and system for low-altitude applications. Background Technology

[0002] With the gradual opening of low-altitude airspace management policies, the concept of integrated sensing and communication has been widely proposed. By achieving synchronous sensing of communication signals under the same spectrum and waveform resources, active monitoring of aerial targets can be achieved while ensuring communication performance.

[0003] However, existing technologies still have shortcomings. First, traditional frame detection algorithms are not robust to noise and interference, which can easily cause frame boundary positioning shifts. Second, the accuracy of frequency domain channel estimation is limited by the interpolation model, making it difficult to maintain phase consistency in the presence of residual carrier frequency offset. Third, micro-Doppler feature extraction mostly relies on continuous time domain signals, and direct transformation can easily cause phase jumps and spectral broadening. Fourth, existing synesthetic fusion mostly adopts simple weighting or confidence threshold decision-making, and has not established a dynamic fusion strategy based on target type, task priority, and historical confidence. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method and system for low-altitude communication and sensing signal processing to address the following issues: First, traditional frame detection algorithms have poor robustness to noise and interference, easily causing frame boundary positioning offsets; second, the accuracy of frequency domain channel estimation is limited by the interpolation model, making it difficult to maintain phase consistency in the presence of residual carrier frequency offsets; third, micro-Doppler feature extraction mostly relies on continuous time-domain signals, and direct transformation easily causes phase jumps and spectral broadening; fourth, existing sensing fusion methods mostly use simple weighting or confidence threshold decisions, without establishing a dynamic fusion strategy based on target type, task priority, and historical confidence.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a communication and sensing signal processing method for low-altitude environments, comprising, Collect mixed signals to generate an ideal constellation point sequence; Based on the ideal constellation point sequence, the Goldstein branching method is used to segment the continuous phase sequence, and the continuous wavelet transform is used to generate the micro-Doppler spectrum. Based on the micro-Doppler spectrum, statistical features are extracted, feature vectors are generated, and a deep neural network model is constructed to predict probability vectors, thus obtaining passive micro-feature results. Significant periodic micro-motion features are detected, active sensing results of target echo are extracted, and the passive equivalent distance between active sensing results and passive micro-feature results is calculated using RSSI. The fusion confidence is then calculated to generate low-altitude communication decisions. The system constructs a visualization interface for each frame, displaying the fusion results and target location, and stores the generated mixed signal data.

[0007] As a preferred embodiment of the low-altitude communication and sensing signal processing method of the present invention, the step of collecting mixed signals and generating an ideal constellation point sequence includes: Collect mixed signals, including communication signals, target scattered echoes, ground clutter, interference, and noise broadband radio frequency signals; A local oscillator is used for quadrature downconversion to generate two baseband analog signals, I and Q, from the mixed signal. The I / Q baseband analog signals are digitally sampled to obtain discrete complex baseband signals. Frame detection and coarse timing synchronization are performed on the discrete complex baseband signals to generate frame start indexes. Based on the frame start index, the residual carrier frequency offset is estimated using the fractional Fourier transform domain peak search method for discrete complex baseband signals, and compensation is performed to obtain the compensated time-domain signal. The compensated time-domain signal is then subjected to Fourier transform to obtain the compensated frequency-domain signal. The initial channel response is calculated at the pilot subcarrier position in the frequency domain of the compensated signal, and the full-band channel estimate is obtained through frequency domain interpolation. Based on the compensated frequency domain signal and full-band channel estimation, the MMSE equalizer is used for equalization processing to obtain a complex soft symbol sequence. Based on the modulation scheme indicated by the current frame signaling, an ideal constellation mapping dictionary is established, and each complex soft symbol sequence is mapped to the corresponding ideal constellation point by looking up the table, thereby generating an ideal constellation point sequence.

[0008] As a preferred embodiment of the low-altitude communication and sensing signal processing method of the present invention, the step of segmenting a continuous phase sequence using the Goldstein branching method and generating a micro-Doppler spectrum using continuous wavelet transform includes: Calculate the perturbation residuals of the ideal constellation points, generate a complex perturbation sequence, and normalize it. The real and imaginary parts of the normalized complex perturbation sequence are filtered bidirectionally using a zero-phase high-pass filter to obtain the filtered complex perturbation sequence. The instantaneous phase of the filtered complex perturbation sequence is calculated, and the Goldstein branch-cut method is used to unwrap each adjacent instantaneous phase to obtain a continuous phase sequence. Perform continuous wavelet transform (CWT) on the continuous phase sequence to obtain CWT coefficients, and calculate instantaneous frequency redistribution on the CWT coefficients to obtain local instantaneous frequencies; The CWT coefficients are re-aggregated onto the frequency axis according to the local instantaneous frequency to obtain a compact time-frequency representation, generating a micro-Doppler spectrum.

[0009] As a preferred embodiment of the low-altitude communication and sensing signal processing method of the present invention, the step of extracting statistical features, generating feature vectors, constructing a deep neural network model to predict probability vectors, and obtaining passive micro-feature results includes: Calculate the average spectrum of the micro-Doppler spectrum, use Top-K peak detection to find the local maximum frequency, and take the minimum positive value as the main frequency; Calculate the presence of harmonics based on the dominant frequency; Calculate the spectral entropy and micro-Doppler signal-to-noise ratio of the average spectrum, respectively; Calculate the standard deviation of phase perturbation in a continuous phase sequence as an indicator of modulation depth; The dominant frequency, modulation depth index, harmonic presence degree, spectral entropy, and micro-Doppler signal-to-noise ratio are used as the feature vectors of the micro-Doppler spectrum. Build a deep neural network model, collect historical feature vectors for training, use the trained deep neural network model to classify the feature vectors, output a probability vector, and select the value with the highest probability in the probability vector as the passive classification confidence. The classification is implemented through the Softmax output layer, and the categories include quadcopter drones, fixed-wing drones, birds, and no targets; The probability vector refers to the probability of each classification result; The passive classification confidence and feature vectors are used to construct the passive micro-feature result.

[0010] As a preferred embodiment of the low-altitude communication and sensing signal processing method of the present invention, the steps of performing significant periodic micro-motion feature detection, extracting the active sensing results of the target echo, using RSSI to inversely calculate the passive equivalent distance between the active sensing results and the passive micro-feature results, calculating the fusion confidence score, and generating low-altitude communication decisions include: Setting a passive classification confidence threshold based on historical regression estimation; If a target echo with significant periodic micro-motion characteristics is detected in the current analysis window, the active sensing result of the target echo is extracted, and the passive equivalent distance between the active sensing result and the passive micro-feature result is calculated using RSSI. The radial distance error is calculated for the feature of each active sensing result, and it is determined whether the match is successful. If the match is successful, the fusion confidence is calculated. The term "significant periodic micro-motion characteristics" refers to quadcopter drones, fixed-wing drones, and birds, and requires that the passive classification confidence score be greater than the passive classification confidence score threshold. The active sensing results include four parameters: target slant range, radial velocity, orientation angle, and active trajectory confidence. The determination of whether a match is successful is made by comparing a radial distance error threshold. If the radial distance error of a feature of the active sensing result is less than the radial distance error threshold, the match is considered successful; otherwise, the match is considered unsuccessful. The results of active perception, passive micro-features, and fusion confidence are combined to construct a fusion result. If no target echo with significant periodic micro-motion characteristics is detected in the current analysis window, the fusion result is an empty set; If the fusion result is not an empty set and there is a detected target with a slant range lower than the standard safe distance, it is determined to be a low-altitude high-threat response mode. The high-threat response mode embeds high-density perception enhancement symbols in the next frame of communication signal, points the beamforming main lobe toward the target direction angle, so as to improve the perception resolution of the low-altitude target, and triggers air traffic control reporting when the task priority is high. If the fusion result is an empty set, a low-overhead synergistic coordination strategy is implemented according to the priority order of UAV tasks. The low-overhead sensory collaboration strategy based on UAV mission priority includes: When the task priority is high, enable the active sensing symbols with medium density and start the initial inspection function. When the task priority is medium, enable low-density active sensing symbols and retain the initial detection function; When the task priority is low, all dedicated active sensing symbols are turned off, and passive micro-feature listening is performed solely by relying on communication pilots.

[0011] As a preferred embodiment of the low-altitude communication and sensing signal processing method of the present invention, wherein: React.js is used to construct the fusion result and target position displayed in the visualization interface of each frame; The target location refers to a location with significant periodic micro-motion characteristics superimposed on the background of the electronic fence map based on a geographic coordinate system. Different icons are used to distinguish quadcopter drones, fixed-wing drones, and birds, and the color intensity indicates the level of fusion confidence. When the target enters the safe distance, the alarm area is automatically highlighted and an audio-visual prompt is triggered. Users who have passed real-name verification are allowed to view this information.

[0012] As a preferred embodiment of the low-altitude communication and sensing signal processing method of the present invention, the step of storing the analyzed mixed signal data includes: The generated mixed signal data is stored in a database and secure access measures are set. The database backs up the stored data to the cloud and performs integrity checks on the stored data and backup data regularly. After the checks are completed, integrity check records are generated and stored synchronously in the database.

[0013] Secondly, the present invention provides a communication and sensing signal processing system for low-altitude environments, comprising, The collection and generation module is used to collect mixed signals and generate an ideal constellation point sequence; The segmentation and transformation module is used to segment a continuous phase sequence based on an ideal constellation point sequence using the Goldstein branching method, and to generate a micro-Doppler spectrum using continuous wavelet transform. The extraction module is used to extract statistical features based on micro-Doppler spectra, generate feature vectors, construct a deep neural network model to predict probability vectors, and obtain passive micro-feature results; The detection and calculation module is used to detect significant periodic micro-motion features, extract the active sensing results of the target echo, use RSSI to back-calculate the passive equivalent distance between the active sensing results and the passive micro-feature results, calculate the fusion confidence, and generate low-altitude communication decisions. The visualization and storage module is used to construct the fusion results and target locations displayed in each frame of the visualization interface, and to store the mixed signal data generated by the analysis.

[0014] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein the computer program, when executed by the processor, implements any step of the low-altitude communication and sensing signal processing method as described in the first aspect of the present invention.

[0015] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the low-altitude communication and sensing signal processing method as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: This invention combines a frame detection algorithm based on the autocorrelation characteristics of PES symbols with a residual carrier frequency offset estimation method in the fractional Fourier transform domain, thereby achieving high-precision frame synchronization and frequency offset compensation for mixed signals. Furthermore, by combining full-band channel estimation based on Wiener interpolation and MMSE frequency domain equalizer, the influence of noise and channel distortion on communication signals is effectively eliminated. By combining ideal constellation point perturbation analysis with continuous phase unwrapping technology of Goldstein branch cutting method, continuous reconstruction of weak phase perturbation information in communication signals is achieved. In addition, by combining continuous wavelet transform and instantaneous frequency redistribution algorithm, high-resolution micro-Doppler spectra are generated, thereby improving the perceived added value of communication signals. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0018] Figure 1 This is a flowchart of a low-altitude communication and sensing signal processing method in Example 1.

[0019] Figure 2 This is a schematic diagram of a low-altitude communication and sensing signal processing system in Example 1.

[0020] Figure 3 This is a flowchart of the signal processing in Example 1.

[0021] Figure 4 This is a diagram of the feature extraction and fusion decision architecture in Example 1. Detailed Implementation

[0022] 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.

[0023] 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.

[0024] Secondly, the term "an 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 throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0025] Example 1, referring to Figures 1 to 4 This is the first embodiment of the present invention, which provides a method for low-altitude communication and sensing signal processing, including the following steps: S1. Collect the mixed signal and generate the ideal constellation point sequence; Preferably, the collection includes mixed signals, including communication signals, target scattered echoes, ground clutter, interference, and noise broadband radio frequency signals; A local oscillator is used for quadrature downconversion to generate two baseband analog signals (in-phase and quadrature branches) from the mixed signal; The I / Q baseband analog signals are digitally sampled to obtain discrete complex baseband signals. Frame detection and coarse timing synchronization are performed on the discrete complex baseband signals to generate frame start indexes. The frame detection and coarse timing synchronization of the discrete complex baseband signal, and the generation of the frame start index, refer to the use of a frame detection algorithm based on the symbol autocorrelation characteristics of PES to calculate the cross-correlation value between the signal and its delayed replica within the sliding window. When the correlation peak exceeds a preset threshold and satisfies the known time-frequency structure of PES, it is determined to be the frame start position and the frame start index is generated. Based on the frame start index, the residual carrier frequency offset is estimated using the fractional Fourier transform domain peak search method for discrete complex baseband signals, and then compensated to obtain the compensated time-domain signal. The compensated time-domain signal is then subjected to Fourier transform to obtain the compensated frequency-domain signal. The initial channel response is calculated at the pilot subcarrier positions in the frequency domain of the compensated signal (which are predefined by the communication protocol or frame structure and are known quantities). The full-band channel estimate is then obtained through frequency domain interpolation (using the Wiener interpolation method based on the channel time-frequency correlation function). Based on the compensated frequency domain signal and full-band channel estimation, equalization processing is performed using an MMSE equalizer (which estimates the noise variance using the received signal power on the empty subcarrier and performs frequency domain equalization in conjunction with the full-band channel estimation) to obtain a complex soft symbol sequence. Based on the modulation scheme indicated by the current frame signaling (such as 64-QAM), an ideal constellation mapping dictionary is established. Each complex soft symbol sequence is mapped to the corresponding ideal constellation point by looking up the table (for each equalized complex symbol, the nearest ideal constellation point in the Euclidean distance sense is found as a reference signal, rather than for hard decision), and an ideal constellation point sequence is generated.

[0026] By using a local oscillator for quadrature downconversion, the mixed signal is converted into two baseband analog signals (I and Q), completing the conversion from radio frequency signal to baseband signal. The two signals carrying amplitude and phase information are separated. The residual carrier frequency offset is estimated and compensated by the peak search method in the fractional Fourier transform domain. Peak detection is performed in the fractional Fourier domain to improve the resolution and accuracy of frequency offset estimation. Through ideal constellation mapping and ideal constellation point sequence generation, the non-hard decision Euclidean distance matching method retains the soft information of the signal while reducing constellation distortion caused by misjudgment.

[0027] S2. Based on the ideal constellation point sequence, the Goldstein branching method is used to segment the continuous phase sequence, and the continuous wavelet transform is used to generate the micro-Doppler spectrum. Preferably, the perturbation residuals of the ideal constellation points are calculated, a complex perturbation sequence is generated, and the perturbation is normalized (symbol energy normalization). The real and imaginary parts of the normalized complex perturbation sequence are bidirectionally filtered using a zero-phase high-pass filter (to avoid phase distortion) to obtain the filtered complex perturbation sequence. The instantaneous phase of the filtered complex perturbation sequence is calculated, and the Goldstein branch-cut method is used to unwrap each adjacent instantaneous phase to obtain a continuous phase sequence. Perform a continuous wavelet transform (CWT) on the continuous phase sequence to obtain CWT coefficients. Calculate the instantaneous frequency redistribution from the CWT coefficients to obtain the local instantaneous frequency. The formula is: , in, For local instantaneous angular frequency, and These are the scale parameter and the time translation parameter, obtained through sampling. The imaginary unit, CWT coefficient The CWT coefficients are re-aggregated onto the frequency axis according to the local instantaneous frequency to obtain a compact time-frequency representation, generating a micro-Doppler spectrum.

[0028] By using zero-phase high-pass filtering to perform bidirectional filtering on the real and imaginary parts respectively, the phase delay or distortion problems caused by unidirectional filtering are avoided, so that the filtered complex perturbation sequence still maintains the original phase structure. Micro-Doppler spectra are generated by continuous wavelet transform (CWT), which has multi-scale resolution in time and frequency and can simultaneously reflect local time-domain changes and spectral structure. By calculating CWT coefficients and redistributing them according to instantaneous frequency, time-frequency joint localization is achieved, improving the time-frequency focusing of the micro-Doppler spectrum.

[0029] S3. Based on the micro-Doppler spectrum, extract statistical features, generate feature vectors, construct a deep neural network model to predict probability vectors, and obtain passive micro-feature results; Preferably, the average spectrum of the micro-Doppler spectrum is calculated, and the average spectrum is detected using Top-K peak detection. The local maximum frequency is found, and the minimum positive value is taken as the main frequency. Based on the dominant frequency, a harmonic existence function is defined as follows: , , in, For the degree of harmonic presence, It is an exponential function. The harmonic order is... Let i be the frequency of the i-th local maximum detected in the average spectrum. Main frequency, For frequency tolerance, For observation time window; Calculate the spectral entropy and micro-Doppler signal-to-noise ratio of the average spectrum, respectively; Calculate the standard deviation of phase perturbation in a continuous phase sequence as an indicator of modulation depth; The dominant frequency, modulation depth index, harmonic presence degree, spectral entropy, and micro-Doppler signal-to-noise ratio are used as the feature vectors of the micro-Doppler spectrum. Build a deep neural network model, collect historical feature vectors for training, use the trained deep neural network model to classify the feature vectors, output a probability vector, and select the value with the highest probability in the probability vector as the passive classification confidence. The dominant frequency in the feature vector reflects the micro-motion period (such as rotor speed), the modulation depth index reflects the micro-motion amplitude, the harmonic presence degree is used to distinguish between single-frequency vibration (fixed wing) and multi-harmonic vibration (multi-rotor), the spectral entropy is used to reflect the motion complexity (birds > drones), and the micro-Doppler signal-to-noise ratio is used to determine reliability. The classification is implemented through the Softmax output layer, and the categories include quadcopter drones, fixed-wing drones, birds, and no targets; The probability vector refers to the probability of each classification result; The passive classification confidence and feature vectors are used to construct the passive micro-feature result.

[0030] By extracting the average spectrum based on the micro-Doppler spectrum and obtaining the dominant frequency through Top-K peak detection, the influence of time-varying noise can be eliminated while ensuring feature stability. Spectral entropy reflects the dispersion of the signal spectrum, while signal-to-noise ratio reflects the concentration of spectral energy. Combining the two can simultaneously evaluate the complexity of the signal. The above features are combined into a multi-dimensional feature vector, which is then classified after training with a deep neural network. The probability vector is output through Softmax, which improves the accuracy of target recognition.

[0031] S4. Perform significant periodic micro-motion feature detection, extract the active sensing results of the target echo, use RSSI to back-calculate the passive equivalent distance between the active sensing results and the passive micro-feature results, calculate the fusion confidence, and generate low-altitude communication decisions. Preferably, a passive classification confidence threshold is set based on historical regression estimation (the passive classification confidence threshold is dynamically adjusted according to the statistical mean of the confidence of historically correctly identified samples, wherein historically correctly identified samples refer to classification results verified by the consistency of multiple frames of trajectories). If a target echo with significant periodic micro-motion characteristics is detected in the current analysis window, the active sensing result of the target echo is extracted, and the passive equivalent distance between the active sensing result and the passive micro-feature result is calculated using RSSI (based on the free space path loss model). The radial distance error is calculated for the feature of each active sensing result, and it is determined whether the match is successful. If the match is successful, the fusion confidence is calculated. The term "significant periodic micro-motion characteristics" refers to quadcopter drones, fixed-wing drones, and birds, and requires that the passive classification confidence score be greater than the passive classification confidence score threshold. The active sensing results include four parameters: target slant range (obtained by matching and filtering the echoes of actively transmitted sensing enhancement symbols, finding the peak position in the range cell, and converting it into physical distance by combining the system sampling rate and signal bandwidth), radial velocity (obtained by searching for the velocity peak in the Doppler frequency domain based on the phase change of the Doppler effect between multiple sensing symbols in the same frame, and then converting it into velocity), azimuth angle (obtained by beamforming estimation of the received echoes, locating the main lobe peak in the angle spectrum), and active trajectory confidence (from the weighted entropy output by the particle filter, after normalization). The determination of whether a match is successful is made by comparing a radial distance error threshold (e.g., 20 meters). If the radial distance error of a feature of the active sensing result is less than the radial distance error threshold, the match is considered successful; otherwise, the match is considered unsuccessful. The fusion confidence score is a weighted average of the active trajectory confidence score and the passive classification confidence score, such as an active trajectory confidence score of 60% and a passive classification confidence score of 40%. The results of active perception, passive micro-features, and fusion confidence are combined to construct a fusion result. If no target echo with significant periodic micro-motion characteristics is detected in the current analysis window, the fusion result is an empty set; If the fusion result is not an empty set and there is a detected target, and the target slant range is lower than the standard set safety distance (e.g., 200 meters, the specific value is set according to industry standards), then it is judged as a low-altitude high-threat response mode. The high-threat response mode embeds high-density perception enhancement symbols (such as 8 chirp symbols / frame) into the next frame of communication signal, points the beamforming main lobe toward the target direction angle, so as to improve the perception resolution of the low-altitude target, and triggers air traffic control reporting when the task priority is high. If the fusion result is an empty set, a low-overhead synergistic coordination strategy is implemented according to the priority order of UAV tasks. The low-overhead sensory collaboration strategy based on UAV mission priority includes: When the task priority is high, enable medium-density active sensing symbols (such as 4 chirp symbols / frame) and start the initial detection function; When the task priority is medium, enable low-density active sensing symbols (e.g., 2 chirp symbols / frame) and retain the initial detection function; When the task priority is low, all dedicated active sensing symbols are turned off, and passive micro-feature listening is carried out solely by communication pilots. This ensures continuous low-altitude surveillance capabilities while minimizing additional power consumption and communication interference.

[0032] By using a free-space path loss model, the passive equivalent distance between the active sensing results and the passive micro-feature results is inferred from the received signal strength indication, and the radial distance error between the two is calculated. This achieves spatial correlation and quantitative matching between the active and passive detection modes. By constructing a fusion result from the active sensing results, the passive micro-feature results, and the fusion confidence, and triggering a high-threat response mode under the condition that the target slant range is lower than the safe distance, the adaptive communication strategy scheduling of the integrated sensing system is realized.

[0033] S5. Construct the fusion result and target position displayed in the visualization interface for each frame, and store the generated mixed signal data; Preferably, the target location refers to a location with significant periodic micro-motion characteristics superimposed on the background of the electronic fence map based on a geographic coordinate system, with different icons distinguishing quadcopter drones, fixed-wing drones and birds, and the color depth indicating the level of fusion confidence; when the target enters the safe distance, the alarm area is automatically highlighted and an audio-visual prompt is triggered. Users who have passed real-name verification are allowed to view this information.

[0034] By visually displaying the fusion results and target locations frame by frame and storing the mixed signal data, the workflow, which emphasizes both real-time human-computer interaction and recording, allows operators to see the current fusion conclusions immediately. At the same time, the system retains the original or fused mixed signal of each frame for subsequent analysis. By overlaying the detected hotspots exhibiting periodic micro-movements onto the map background layer using standard geographic coordinates, maintenance personnel or regulators can intuitively determine whether abnormal activities are concentrated in certain geographic hotspots.

[0035] Furthermore, the generated mixed signal data is stored in a database, and secure access measures are set up. The database backs up the stored data to the cloud and performs integrity checks on the stored data and backup data regularly. After the checks are completed, integrity check records are generated and stored synchronously in the database.

[0036] By periodically or in real-time backing up the local database to cloud storage and performing periodic integrity checks on the master data and backups, a dual redundancy and verifiable data preservation chain are achieved: even if the master database is damaged or tampered with, the original data can be recovered or proven using backups and inspection records. The integrity records themselves are used as metadata for auditing.

[0037] This embodiment also provides a communication and sensing signal processing system for low-altitude environments, including: The collection and generation module is used to collect mixed signals and generate an ideal constellation point sequence; The segmentation and transformation module is used to segment a continuous phase sequence based on an ideal constellation point sequence using the Goldstein branching method, and to generate a micro-Doppler spectrum using continuous wavelet transform. The extraction module is used to extract statistical features based on micro-Doppler spectra, generate feature vectors, construct a deep neural network model to predict probability vectors, and obtain passive micro-feature results; The detection and calculation module is used to detect significant periodic micro-motion features, extract the active sensing results of the target echo, use RSSI to back-calculate the passive equivalent distance between the active sensing results and the passive micro-feature results, calculate the fusion confidence, and generate low-altitude communication decisions. The visualization and storage module is used to construct the fusion results and target locations displayed in each frame of the visualization interface, and to store the mixed signal data generated by the analysis.

[0038] This embodiment also provides a computer device applicable to the low-altitude communication and sensing signal processing method, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the low-altitude communication and sensing signal processing method proposed in the above embodiment.

[0039] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0040] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the low-altitude communication and sensing signal processing method proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0041] In summary, this invention achieves high-precision frame synchronization and frequency offset compensation for mixed signals by combining a frame detection algorithm based on PES symbol autocorrelation characteristics with a residual carrier frequency offset estimation method in the fractional Fourier transform domain. Furthermore, by combining full-band channel estimation based on Wiener interpolation with the MMSE frequency domain equalizer, the impact of noise and channel distortion on communication signals is effectively eliminated. By combining ideal constellation point perturbation analysis with continuous phase unwrapping technology using the Goldstein branch cutting method, continuous reconstruction of weak phase perturbation information in communication signals is achieved. Finally, by combining continuous wavelet transform and instantaneous frequency redistribution algorithms, high-resolution micro-Doppler spectra are generated, enhancing the perceived added value of communication signals.

[0042] 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 processing communication and sensing signals for low-altitude operations, characterized in that: include, Collect mixed signals to generate an ideal constellation point sequence; Based on the ideal constellation point sequence, the Goldstein branching method is used to segment the continuous phase sequence, and the continuous wavelet transform is used to generate the micro-Doppler spectrum. Based on the micro-Doppler spectrum, statistical features are extracted, feature vectors are generated, and a deep neural network model is constructed to predict probability vectors, thus obtaining passive micro-feature results. Significant periodic micro-motion features are detected, active sensing results of target echo are extracted, and the passive equivalent distance between active sensing results and passive micro-feature results is calculated using RSSI. The fusion confidence is then calculated to generate low-altitude communication decisions. The system constructs a visualization interface for each frame, displaying the fusion results and target location, and stores the generated mixed signal data.

2. The low-altitude communication and sensing signal processing method as described in claim 1, characterized in that: The process of collecting the mixed signals and generating an ideal constellation point sequence includes: Collect mixed signals, including communication signals, target scattered echoes, ground clutter, interference, and noise broadband radio frequency signals; A local oscillator is used for quadrature downconversion to generate two baseband analog signals, I and Q, from the mixed signal. The I / Q baseband analog signals are digitally sampled to obtain discrete complex baseband signals. Frame detection and coarse timing synchronization are performed on the discrete complex baseband signals to generate frame start indexes. Based on the frame start index, the residual carrier frequency offset is estimated using the fractional Fourier transform domain peak search method for discrete complex baseband signals, and compensation is performed to obtain the compensated time-domain signal. The compensated time-domain signal is then subjected to Fourier transform to obtain the compensated frequency-domain signal. The initial channel response is calculated at the pilot subcarrier position in the frequency domain of the compensated signal, and the full-band channel estimate is obtained through frequency domain interpolation. Based on the compensated frequency domain signal and full-band channel estimation, the MMSE equalizer is used for equalization processing to obtain a complex soft symbol sequence. Based on the modulation scheme indicated by the current frame signaling, an ideal constellation mapping dictionary is established, and each complex soft symbol sequence is mapped to the corresponding ideal constellation point by looking up the table, thereby generating an ideal constellation point sequence.

3. The low-altitude communication and sensing signal processing method as described in claim 2, characterized in that: The process of segmenting a continuous phase sequence using the Goldstein branching method and generating a micro-Doppler spectrum using continuous wavelet transform includes: Calculate the perturbation residuals of the ideal constellation points, generate a complex perturbation sequence, and normalize it. The real and imaginary parts of the normalized complex perturbation sequence are filtered bidirectionally using a zero-phase high-pass filter to obtain the filtered complex perturbation sequence. The instantaneous phase of the filtered complex perturbation sequence is calculated, and the Goldstein branch-cut method is used to unwrap each adjacent instantaneous phase to obtain a continuous phase sequence. Perform continuous wavelet transform (CWT) on the continuous phase sequence to obtain CWT coefficients, and calculate instantaneous frequency redistribution on the CWT coefficients to obtain local instantaneous frequencies; The CWT coefficients are re-aggregated onto the frequency axis according to the local instantaneous frequency to obtain a compact time-frequency representation, generating a micro-Doppler spectrum.

4. The low-altitude communication and sensing signal processing method as described in claim 3, characterized in that: The process of extracting statistical features, generating feature vectors, constructing a deep neural network model to predict probability vectors, and obtaining passive micro-feature results includes: Calculate the average spectrum of the micro-Doppler spectrum, use Top-K peak detection to find the local maximum frequency, and take the minimum positive value as the main frequency; Calculate the presence of harmonics based on the dominant frequency; Calculate the spectral entropy and micro-Doppler signal-to-noise ratio of the average spectrum, respectively; Calculate the standard deviation of phase perturbation in a continuous phase sequence as an indicator of modulation depth; The dominant frequency, modulation depth index, harmonic presence degree, spectral entropy, and micro-Doppler signal-to-noise ratio are used as the feature vectors of the micro-Doppler spectrum. Build a deep neural network model, collect historical feature vectors for training, use the trained deep neural network model to classify the feature vectors, output a probability vector, and select the value with the highest probability in the probability vector as the passive classification confidence. The classification is implemented through the Softmax output layer, and the categories include quadcopter drones, fixed-wing drones, birds, and no targets; The probability vector refers to the probability of each classification result; The passive classification confidence and feature vectors are used to construct the passive micro-feature result.

5. The low-altitude communication and sensing signal processing method as described in claim 4, characterized in that: The process of detecting significant periodic micro-motion features, extracting the active sensing results of the target echo, using RSSI to inversely calculate the passive equivalent distance between the active sensing results and the passive micro-feature results, calculating the fusion confidence score, and generating low-altitude communication decisions includes: Setting a passive classification confidence threshold based on historical regression estimation; If a target echo with significant periodic micro-motion characteristics is detected in the current analysis window, the active sensing result of the target echo is extracted, and the passive equivalent distance between the active sensing result and the passive micro-feature result is calculated using RSSI. The radial distance error is calculated for the feature of each active sensing result, and it is determined whether the match is successful. If the match is successful, the fusion confidence is calculated. The term "significant periodic micro-motion characteristics" refers to quadcopter drones, fixed-wing drones, and birds, and requires that the passive classification confidence score be greater than the passive classification confidence score threshold. The active sensing results include four parameters: target slant range, radial velocity, orientation angle, and active trajectory confidence. The determination of whether a match is successful is made by comparing a radial distance error threshold. If the radial distance error of a feature of the active sensing result is less than the radial distance error threshold, the match is considered successful; otherwise, the match is considered unsuccessful. The results of active perception, passive micro-features, and fusion confidence are combined to construct a fusion result. If no target echo with significant periodic micro-motion characteristics is detected in the current analysis window, the fusion result is an empty set; If the fusion result is not an empty set and there is a detected target with a slant range lower than the standard safe distance, it is determined to be a low-altitude high-threat response mode. The high-threat response mode embeds high-density perception enhancement symbols in the next frame of communication signal, points the beamforming main lobe toward the target direction angle, so as to improve the perception resolution of the low-altitude target, and triggers air traffic control reporting when the task priority is high. If the fusion result is an empty set, a low-overhead synergistic coordination strategy is implemented according to the priority order of UAV tasks. The low-overhead sensory collaboration strategy based on UAV mission priority includes: When the task priority is high, enable the active sensing symbols with medium density and start the initial inspection function. When the task priority is medium, enable low-density active sensing symbols and retain the initial detection function; When the task priority is low, all dedicated active sensing symbols are turned off, and passive micro-feature listening is performed solely by relying on communication pilots.

6. The low-altitude communication and sensing signal processing method as described in claim 5, characterized in that: The visualization of each frame's result and target location is built using React.js; The target location refers to a location with significant periodic micro-motion characteristics superimposed on the background of the electronic fence map based on the geographic coordinate system. Different icons are used to distinguish quadcopter drones, fixed-wing drones and birds, and the color depth indicates the level of fusion confidence. When the target enters the safe distance, the alarm area will be automatically highlighted and an audio-visual alert will be triggered. Users who have passed real-name verification are allowed to view this information.

7. The low-altitude communication and sensing signal processing method as described in claim 6, characterized in that: The process of storing the generated mixed signal data includes: The generated mixed signal data is stored in a database and secure access measures are set. The database backs up the stored data to the cloud and performs integrity checks on the stored data and backup data regularly. After the checks are completed, integrity check records are generated and stored synchronously in the database.

8. A low-altitude communication and sensing signal processing system, based on the low-altitude communication and sensing signal processing method according to any one of claims 1 to 7, characterized in that: include, The collection and generation module is used to collect mixed signals and generate an ideal constellation point sequence; The segmentation and transformation module is used to segment a continuous phase sequence based on an ideal constellation point sequence using the Goldstein branching method, and to generate a micro-Doppler spectrum using continuous wavelet transform. The extraction module is used to extract statistical features based on micro-Doppler spectra, generate feature vectors, construct a deep neural network model to predict probability vectors, and obtain passive micro-feature results; The detection and calculation module is used to detect significant periodic micro-motion features, extract the active sensing results of the target echo, use RSSI to back-calculate the passive equivalent distance between the active sensing results and the passive micro-feature results, calculate the fusion confidence, and generate low-altitude communication decisions. The visualization and storage module is used to construct the fusion results and target locations displayed in each frame of the visualization interface, and to store the mixed signal data generated by the analysis.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the low-altitude communication and sensing signal processing method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the low-altitude communication and sensing signal processing method according to any one of claims 1 to 7.

Images