Method and system for monitoring low-altitude aircraft at an airport

Abstract

The present invention discloses a method and system for monitoring low-altitude aircraft at an airport, which relates to the field of aerospace monitoring technology, including arranging multiple software-defined radio receivers in a preset area around the airport; scanning and capturing radio signals in the low-altitude area to generate data packets to be transmitted; transmitting the data packets to be transmitted to a central processing server through an edge network, parsing and extracting features of the data packets to be transmitted in the central processing server, and generating signal feature information; based on the signal feature information, using a multi-source fusion positioning algorithm to locate low-altitude aircraft and generate comprehensive positioning information; combining the comprehensive positioning information and signal feature information, using a graph neural network to build a dynamic flight behavior model, and evaluating potential threats in real time. The present invention realizes the accurate capture, processing and classification of radio signals in low-altitude areas by utilizing high-precision software-defined radio receivers, combined with multi-level wavelet transform noise reduction and advanced signal classification algorithms.

Description

Technical Field

[0001] The invention relates to the field of aviation and aerospace monitoring technology, and in particular to a method and system for monitoring low-altitude aircraft at an airport. Background Art

[0002] In modern airport operations, the monitoring and management of low-altitude aircraft has become an important safety issue. With the widespread use of drones and other low-altitude aircraft, traditional radar and communication systems have limited capabilities in monitoring low-altitude aircraft and cannot meet the needs of high-precision, high-real-time monitoring of aircraft. Existing technologies usually rely on limited monitoring methods, such as ground radar, fixed sensors, and manual monitoring. These methods are difficult to effectively handle dense low-altitude aircraft traffic, resulting in monitoring blind spots and response delays, thus affecting the safety of airports and their surrounding airspace.

[0003] In addition, traditional radio signal processing technology is also unable to cope with complex signal environments and multiple signal overlaps. When processing radio signals emitted by low-altitude aircraft, existing systems have problems such as signal interference, noise influence, and insufficient real-time analysis capabilities. These technical limitations make it difficult to accurately and timely classify, locate, and predict the behavior of aircraft. Therefore, a more advanced monitoring method and system is urgently needed to improve the overall effect and reliability of low-altitude aircraft monitoring. Summary of the invention

[0004] In view of the fact that the existing low-altitude aircraft monitoring system has deficiencies in accuracy, real-time performance and signal processing capability, resulting in ineffective monitoring and management of low-altitude aircraft, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to efficiently and accurately monitor and locate low-altitude aircraft, thereby effectively ensuring the safety of the airport and its surrounding airspace.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, an embodiment of the present invention provides a method for monitoring low-altitude aircraft at an airport, which includes arranging multiple software-defined radio receivers in a preset area around the airport; using the software-defined radio receivers to scan and capture radio signals in the low-altitude area to generate data packets to be transmitted; transmitting the data packets to be transmitted to a central processing server through an edge network, parsing and extracting features of the data packets to be transmitted in the central processing server to generate signal feature information; based on the signal feature information, using a multi-source fusion positioning algorithm, combining the signal time difference and frequency difference of multiple software-defined radio receivers, locating the low-altitude aircraft and generating comprehensive positioning information; combining the comprehensive positioning information and signal feature information, using a graph neural network to build a dynamic flight behavior model, and evaluating potential threats in real time; when a threat is detected, starting a directional antenna array to transmit a simulated control signal with the same frequency as the low-altitude aircraft to the target area, and safely guiding or taking over control of the low-altitude aircraft.

[0008] As a preferred solution of the method for monitoring low-altitude aircraft at an airport described in the present invention, generating a data packet to be transmitted includes the following steps: setting the scanning parameters of a software-defined radio receiver to adapt to the signal characteristics of the low-altitude area; using a programmable RF front end and a high-speed analog-to-digital converter to receive and digitize the radio signal in the low-altitude area to obtain a digitized radio signal; using a digital signal processor to perform multi-level wavelet transform noise reduction on the digitized radio signal; using a signal analysis software module to perform time-frequency analysis on the digitized radio signal after noise reduction to generate a time-frequency spectrum; starting a signal classification module to classify the radio signal based on the time-frequency spectrum; controlling a multi-element antenna array to perform beamforming to generate an adjustable beam signal for each type of signal; performing a fast Fourier transform on the adjustable beam signal to obtain a signal spectrum feature; combining the signal spectrum feature with the corresponding signal type label to generate a data packet to be transmitted.

[0009] As a preferred solution of the method for monitoring low-altitude aircraft at an airport according to the present invention, a multi-level wavelet transform is performed to reduce noise, and a noise-reduced signal x'(n) is obtained. The specific formula is as follows:

[0010]

[0011] Where x'(n) is the digitized radio signal after noise reduction, a J is the approximate coefficient of the Jth layer, J is the number of layers of wavelet decomposition, φ J,k (n) is the scaling function, ψ J,k (n) is the wavelet function, d' j,k is the detail coefficient of the jth layer after processing, and the specific formula is as follows:

[0012] d' j,k =sign(d j,k )(|dj,k |-λ j ) +

[0013] Among them, d' j,k is the detail coefficient of the jth layer after processing, sign(·) is the sign function, (·) + is the positive part in soft threshold processing, λ j The threshold of the jth layer is as follows:

[0014]

[0015] Among them, σ j is the estimated value of the standard deviation of the noise in the jth layer, and N is the signal length.

[0016] As a preferred solution of the method for monitoring low-altitude aircraft at an airport described in the present invention, the classification of radio signals based on the time-frequency spectrum includes the following steps: image preprocessing of the time-frequency spectrum; extracting feature vectors from the preprocessed time-frequency spectrum, wherein the feature vectors include frequency bandwidth, time continuity and energy distribution; using a pre-trained convolutional neural network model to perform initial classification on the feature vectors; based on the initial classification results, applying an integrated learning method to perform refined classification; applying a timing consistency check to the refined classification results to eliminate instantaneous misclassification; using a spectrum occupancy analysis method to identify multiple signal overlapping areas in the time-frequency spectrum; applying a blind source separation algorithm to the multiple signal overlapping areas to separate independent signal components; re-inputting the separated signal components into the classification process for secondary classification; combining the initial classification results and the secondary classification results to generate a signal type label; associating the signal type label with the original time-frequency spectrum to form a labeled time-frequency spectrum.

[0017] As a preferred solution of the airport low-altitude aircraft monitoring method described in the present invention, the following steps are included in the generation of steerable beam signals for various types of signals: obtaining the geometric layout parameters of the multi-element antenna array and the phase center position of each antenna unit; calculating the wavelength corresponding to each signal based on the frequency characteristics of each type of signal; determining the desired main beam direction and sidelobe suppression level for each identified signal type; using an adaptive beamforming algorithm to calculate the amplitude and phase weights of each antenna unit required for the main beam direction and sidelobe suppression level; applying the calculated amplitude and phase weights to each antenna unit of the multi-element antenna array; performing phase control by adjusting the digital phase shifter of the antenna unit; adjusting the signal amplitude of each antenna unit using a variable gain amplifier to cooperate with the phase control; performing real-time signal synthesis to superimpose the signals received by each antenna unit according to the set amplitude and phase; digitally filtering the synthesized signal to remove interference from non-target frequency bands; generating an steerable beam signal and storing it in association with the corresponding signal type identifier.

[0018] As a preferred solution of the airport low-altitude aircraft monitoring method described in the present invention, positioning the low-altitude aircraft and generating comprehensive positioning information includes the following steps: dynamically selecting a positioning algorithm combination based on signal type labels and key parameters to utilize time correlation and frequency correlation to calculate the signal time difference and frequency difference between multiple software-defined radio receivers; inputting the signal time difference and frequency difference into the selected positioning algorithm combination to calculate the three-dimensional coordinates and velocity vector of the low-altitude aircraft; based on the changing trend of time correlation and frequency correlation, constructing a low-altitude aircraft motion model and calculating the predicted trajectory; according to the time-varying characteristics of the key parameters, adjusting the weight coefficients of each algorithm in the positioning algorithm combination in real time; utilizing the spatial distribution information of multiple software-defined radio receivers to construct a collaborative positioning network to improve positioning accuracy and reduce the risk of single-point failure; based on the positioning results and predicted trajectories, dynamically adjusting the parameters of the software-defined radio receiver, including beam direction and gain; calculating the credibility index of the positioning results to evaluate the positioning accuracy; generating comprehensive positioning information, associating the comprehensive positioning information with the corresponding signal feature information, and updating the records in the database.

[0019] As a preferred solution of the method for monitoring low-altitude aircraft at airports described in the present invention, the dynamic selection of a positioning algorithm combination includes the following steps: constructing a positioning algorithm library; preliminarily screening a suitable positioning algorithm subset from the positioning algorithm library based on a signal type label; calculating the time-frequency resolution index of the signal using the center frequency and bandwidth in the key parameters, and prioritizing the algorithms in the positioning algorithm subset according to the time-frequency resolution index; estimating the signal-to-noise ratio of the signal based on the peak power in the key parameters, and selecting the top N algorithms from the sorted positioning algorithm subset according to the signal-to-noise ratio; constructing a positioning algorithm combination scoring model, the scoring model taking into account the algorithm complexity, computing resource requirements and historical positioning accuracy; and using the positioning algorithm combination scoring model to perform combined optimization on the selected N algorithms to obtain the optimal positioning algorithm combination.

[0020] In a second aspect, an embodiment of the present invention provides an airport low-altitude aircraft monitoring system, which includes a receiver deployment module for deploying multiple software-defined radio receivers in a preset area around the airport; a data packet generation module for using a software-defined radio receiver to scan and capture radio signals in the low-altitude area to generate a data packet to be transmitted; a feature extraction module for transmitting the data packet to be transmitted to a central processing server through an edge network, parsing and extracting features of the data packet to be transmitted in the central processing server, and generating signal feature information; a positioning module for locating the low-altitude aircraft and generating comprehensive positioning information based on the signal feature information, using a multi-source fusion positioning algorithm, combining the signal time difference and frequency difference of multiple software-defined radio receivers; a threat assessment module for combining the comprehensive positioning information and signal feature information, using a graph neural network to build a dynamic flight behavior model, and evaluating potential threats in real time; a control module for starting a directional antenna array when a threat is detected, transmitting a simulated control signal with the same frequency as the low-altitude aircraft to the target area, and safely guiding or taking over control of the low-altitude aircraft.

[0021] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the method for monitoring low-altitude aircraft at an airport as described in the first aspect of the present invention are implemented.

[0022] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the method for monitoring low-altitude aircraft at an airport as described in the first aspect of the present invention are implemented.

[0023] The beneficial effects of the present invention are as follows: the present invention realizes the accurate capture, processing and classification of radio signals in low-altitude areas by utilizing a high-precision software-defined radio receiver, combined with multi-level wavelet transform noise reduction and advanced signal classification algorithms. The multi-source fusion positioning algorithm is adopted, combined with the signal time difference and frequency difference, to efficiently realize the real-time positioning of low-altitude aircraft, and dynamically adjust the positioning algorithm and parameters to improve the positioning accuracy. The flight behavior is evaluated in real time through the graph neural network, and potential threats are warned early. When a threat is detected, a simulated control signal is transmitted using a directional antenna array to achieve safe guidance or takeover control of the target aircraft, effectively ensuring the safety of the airspace around the airport. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 The framework flow chart of the low-altitude aircraft monitoring method at the airport.

[0026] Figure 2 A method for monitoring low-altitude aircraft at airports.

[0027] Figure 3 A method for monitoring low-altitude aircraft at airports.

[0028] Figure 4 Computer equipment diagram for low-flying aircraft surveillance methods at airports. DETAILED DESCRIPTION

[0029] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0030] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0031] 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 term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0032] Example 1, reference Figure 1 to Figure 4 , which is the first embodiment of the present invention, provides a method for monitoring low-altitude aircraft at an airport, comprising:

[0033] S1: Multiple software defined radio receivers are deployed in a pre-set area around the airport.

[0034] It should be noted that the software-defined radio receiver includes a programmable RF front end, a high-speed analog-to-digital converter, a digital signal processor, a multi-element antenna array, a signal classification module, and a signal analysis software module.

[0035] It should be noted that the preset area around the airport includes but is not limited to the area around the airport runway and taxiway, the airport approach and departure route area, the low-altitude airspace around the airport, the key infrastructure near the airport, and the buffer zone around the airport. The specific scope of this preset area will vary depending on the size, geographical location, surrounding environment and local regulations of the airport.

[0036] S2: Use a software defined radio receiver to scan and capture radio signals in the low-altitude area and generate data packets to be transmitted.

[0037] Specifically, the following steps are included:

[0038] S2.1: Set the scanning parameters of the software defined radio receiver to adapt to the signal characteristics of the low altitude area.

[0039] Preferably, the scanning parameters include frequency range (eg, 400 MHz-6 GHz), sampling rate (eg, 20 MHz), and adjustable gain (eg, 0-60 dB).

[0040] S2.2: Receive and digitize radio signals in low-altitude areas using a programmable RF front end and a high-speed analog-to-digital converter to obtain a digitized radio signal.

[0041] S2.3: Using a digital signal processor, perform multi-level wavelet transform noise reduction on the digitized radio signal.

[0042] Specifically, the digitized radio signal x(n) is decomposed by J-layer wavelet to obtain the approximate coefficient a J and detail factor d j (j=1,2,3,...,J), the wavelet decomposition can be expressed as:

[0043]

[0044] Among them, φ J,k (n) is the scaling function, ψ J,k (n) is the wavelet function, x(n) is the original digitized radio signal, J is the number of layers of wavelet decomposition, and a J is the approximation coefficient of the Jth layer, d j is the detail coefficient of the jth layer, j = 1, 2, 3, ..., J, k is the index of the approximate coefficient and the detail coefficient.

[0045] It should be noted that φ J,k (n) is the scaling function, which is used to calculate the approximate coefficient a of the Jth layer. J , which is a low-pass filter that retains the low-frequency part of the signal, ψ J,k (n) is the wavelet function used to calculate the detail coefficient d of the jth layer j, which is a bandpass filter that extracts the high-frequency components of the signal.

[0046] Furthermore, in order to reduce the influence of noise, the detail coefficient d j Perform threshold processing to obtain the processed detail coefficient d' j , the threshold processing formula is as follows:

[0047] d' j,k =sign(d j,k )(|d j,k |-λ j ) +

[0048] Among them, d' j,k is the detail coefficient of the jth layer after processing, sign(·) is the sign function, (·) + is the positive part in soft threshold processing ((x) + =max(x,0), i.e. negative values ​​are set to zero), λ j The threshold of the jth layer is as follows:

[0049]

[0050] Among them, σ j is the estimated value of the standard deviation of the noise in the jth layer, and N is the signal length.

[0051] Furthermore, using the processed detail coefficient d' j and approximation coefficient a J Perform wavelet reconstruction to obtain the denoised signal x'(n). The specific formula is as follows:

[0052]

[0053] Where x'(n) is the digitized radio signal after noise reduction, d' j,k is the detail coefficient of the jth layer after processing.

[0054] Preferably, the method separates the signal into approximate coefficients and detail coefficients of different frequency scales by wavelet decomposition, effectively distinguishing the main structure and noise components of the signal. By thresholding the detail coefficients, noise can be selectively suppressed while retaining the important features of the signal. Multi-level decomposition allows noise removal at different scales to adapt to the noise characteristics of different frequency ranges. The soft threshold processing method can smooth the processing results and avoid artificial discontinuities. By adaptively calculating the threshold using the estimation of the noise standard deviation and the signal length, the noise reduction process can be adjusted according to the specific characteristics of the signal. Finally, the processed coefficients are used for wavelet reconstruction to obtain the signal after noise reduction, which effectively improves the signal-to-noise ratio and enhances the clarity and recognizability of the useful signal.

[0055] S2.4: Use the signal analysis software module to perform time-frequency analysis on the digitized radio signal after noise reduction to generate a time-frequency spectrum.

[0056] Specifically, the denoised digitized radio signal x'(n) is segmented into multiple overlapping time windows, each window length is L, and the overlapping rate is R; the Hanning window function w(n) is applied to the signal data of each time window, and the specific formula is as follows:

[0057]

[0058] Where L is the window length, w(n) is the Hanning window function, and n is each time point in the window.

[0059] Furthermore, a fast Fourier transform is performed on each time window signal after applying the Hanning window function. The specific formula is as follows:

[0060]

[0061] Among them, X(k,m) is the spectral component of the mth time window at frequency index k, m is the time index, k is the frequency index (the position of the frequency component in the Fourier transform), H is the step size between adjacent time windows, H = L×(1-R), L is the window length, R is the overlap rate, and n is each time point in the window.

[0062] Furthermore, the square of the amplitude of the fast Fourier transform result is calculated to obtain the power spectrum of each time window. The specific formula is as follows:

[0063] P(k,m)=|X(k,m)| 2

[0064] Among them, X(k,m) is the spectral component of the mth time window at the frequency index k, and P(k,m) is the energy of the mth time window at the frequency index k.

[0065] Furthermore, the power spectrum is arranged in chronological order to construct a time-frequency matrix S(k,m); bilinear interpolation is performed on the time-frequency matrix to obtain the interpolated time-frequency matrix S i (k',m'); normalize the interpolated time-frequency matrix. The specific formula is as follows:

[0066]

[0067] Among them, S i (k',m') is the time-frequency matrix after interpolation, S n (k',m') is the normalized time-frequency matrix, max(S i ) and min(Si ) represent the interpolated time-frequency matrix S i The maximum and minimum values ​​of (k',m').

[0068] Furthermore, a time-frequency spectrum is generated based on the normalized time-frequency matrix.

[0069] S2.5: Start the signal classification module to classify the radio signals based on the time-frequency spectrum, including remote control signals, data transmission signals, navigation signals, image transmission signals and flight control signals.

[0070] Specifically, the time-frequency spectrum is subjected to image preprocessing, including denoising and contrast enhancement; feature vectors are extracted from the preprocessed time-frequency spectrum, including frequency bandwidth, time continuity and energy distribution; the feature vectors are initially classified using a pre-trained convolutional neural network model; based on the initial classification results, an ensemble learning method (such as random forest or gradient boosting tree) is applied for refined classification; a temporal consistency check is applied to the refined classification results to eliminate instantaneous misclassification; a spectrum occupancy analysis method is used to identify the overlapping areas of multiple signals in the time-frequency spectrum; a blind source separation algorithm is applied to the overlapping areas of multiple signals to separate independent signal components; the separated signal components are re-input into the classification process for secondary classification; the initial classification results and the secondary classification results are combined to generate a final signal type label; the signal type label is associated with the original time-frequency spectrum to form a labeled time-frequency spectrum.

[0071] S2.6: Control the multi-element antenna array to perform beamforming, and generate steerable beam signals for various types of signals.

[0072] Specifically, the geometric layout parameters of the multi-element antenna array and the phase center position of each antenna unit are obtained; based on the frequency characteristics of various types of signals, the wavelength corresponding to each signal is calculated; for each identified signal type, the desired main beam direction and sidelobe suppression level are determined; using an adaptive beamforming algorithm, the amplitude and phase weights of each antenna unit required for the main beam direction and sidelobe suppression level are calculated; the calculated amplitude and phase weights are applied to each antenna unit of the multi-element antenna array; phase control is performed by adjusting the digital phase shifter of the antenna unit; the signal amplitude of each antenna unit is adjusted by a variable gain amplifier, and coordinated with the phase control; real-time signal synthesis is performed, and the signals received by each antenna unit are superimposed according to the set amplitude and phase; the synthesized signal is digitally filtered to remove interference from non-target frequency bands; an adjustable beam signal is generated, and it is associated with the corresponding signal type identifier and stored.

[0073] S2.7: Perform fast Fourier transform on the steerable beam signal to obtain the signal spectrum characteristics.

[0074] S2.8: Combine the signal spectrum characteristics with the corresponding signal type label to generate a data packet to be transmitted.

[0075] Preferably, metadata such as timestamp and geographic location information (if any) are added to the data packet to facilitate subsequent processing and analysis.

[0076] S3: The data packets to be transmitted are transmitted to the central processing server through the edge network, and the data packets to be transmitted are analyzed and feature extracted in the central processing server to generate signal feature information.

[0077] Specifically, the data packets to be transmitted are transmitted to the central processing server through the edge network; in the central processing server, the data packets to be transmitted are parsed to extract the signal spectrum characteristics and the corresponding signal type labels; the signal spectrum characteristics are normalized to eliminate the gain differences between different receivers; based on the signal type label, the corresponding feature extraction algorithm is selected; key parameters are extracted from the normalized signal spectrum characteristics, including the center frequency, bandwidth and peak power; the time correlation and frequency correlation of the signal spectrum characteristics are calculated; the signal characteristic information is generated by combining the key parameters, time correlation and frequency correlation; the signal characteristic information is associated with the original signal type label, and the database record is updated.

[0078] S4: Based on signal feature information, a multi-source fusion positioning algorithm is used to combine the signal time difference and frequency difference of multiple software-defined radio receivers to locate low-altitude aircraft and generate comprehensive positioning information.

[0079] Specifically, the following steps are included:

[0080] S4.1: Extract signal type labels, key parameters, time correlation, and frequency correlation from signal feature information.

[0081] S4.2: Dynamically select a positioning algorithm combination based on the signal type label and key parameters.

[0082] Preferably, a positioning algorithm library is constructed, which includes relevant positioning algorithms such as time difference of arrival, frequency difference of arrival, angle of arrival, Doppler frequency shift, etc.; based on the signal type label, a suitable positioning algorithm subset is preliminarily screened from the positioning algorithm library; the time-frequency resolution index of the signal is calculated using the center frequency and bandwidth in the key parameters; the algorithms in the positioning algorithm subset are prioritized according to the time-frequency resolution index; based on the peak power in the key parameters, the signal-to-noise ratio of the signal is estimated; according to the signal-to-noise ratio, the top N algorithms are selected from the sorted positioning algorithm subset, where N is a preset threshold; a positioning algorithm combination scoring model is constructed, and the scoring model takes into account the algorithm complexity, computing resource requirements and historical positioning accuracy; the positioning algorithm combination scoring model is used to perform combination optimization on the selected N algorithms to obtain the optimal positioning algorithm combination.

[0083] S4.3: Calculate the signal time difference and frequency difference between multiple software defined radio receivers using the time correlation and frequency correlation.

[0084] S4.4: Input the signal time difference and frequency difference into the selected positioning algorithm combination to calculate the three-dimensional coordinates and velocity vector of the low-altitude aircraft.

[0085] S4.5: Based on the changing trends of time correlation and frequency correlation, construct a low-altitude aircraft motion model and calculate the predicted trajectory.

[0086] Preferably, a historical data sequence of time correlation and frequency correlation is obtained, and the historical data sequence is segmented into time windows to obtain data subsequences of multiple time windows; fast Fourier transform is performed on the data subsequences of multiple time windows respectively to extract the main frequency components and their amplitudes in each time window; based on the main frequency components and their amplitudes, a time-varying autoregressive moving average model is constructed, and the parameters of the time-varying autoregressive moving average model are dynamically adjusted with the time window; the short-term prediction values ​​of time correlation and frequency correlation are calculated using the time-varying autoregressive moving average model; the short-term prediction values ​​are input into a pre-trained long short-term memory neural network, and the long short-term memory neural network is used to capture the long-term change characteristics of time correlation and frequency correlation; based on the output of the long short-term memory neural network, a nonlinear state space model of the low-altitude aircraft is constructed, and the nonlinear state space model includes state variables such as position, velocity and acceleration; an unscented Kalman filter algorithm is applied to the nonlinear state space model to obtain an optimal estimate of the state of the low-altitude aircraft; based on the state of the optimal estimate, the nonlinear state space model is numerically integrated using the Runge-Kutta method to obtain a set of predicted trajectory points of the low-altitude aircraft in the future.

[0087] S4.6: According to the time-varying characteristics of the key parameters, the weight coefficients of each algorithm in the positioning algorithm combination are adjusted in real time.

[0088] S4.7: Leverage the spatial distribution information of multiple software-defined radio receivers to build a collaborative positioning network to improve positioning accuracy and reduce the risk of single point failures.

[0089] S4.8: Dynamically adjust software-defined radio receiver parameters, including beam direction and gain, based on positioning results and predicted trajectory.

[0090] S4.9: Calculate the confidence index of the positioning result, considering key parameters including peak power, number of receivers involved in positioning, and geometric distribution, to evaluate the positioning accuracy.

[0091] S4.10: Generate comprehensive positioning information, associate the comprehensive positioning information with corresponding signal characteristic information, and update the records in the database.

[0092] It should be noted that the comprehensive positioning information includes the three-dimensional coordinates of the low-altitude aircraft, the three-dimensional coordinates of the low-altitude aircraft, the predicted trajectory point set, the positioning accuracy estimate and the credibility index.

[0093] Preferably, this step achieves high-precision, real-time positioning and trajectory prediction of low-altitude aircraft by adopting a multi-source fusion positioning algorithm. By dynamically selecting a combination of positioning algorithms, the system can flexibly adjust the positioning strategy according to different signal characteristics, thereby improving adaptability and accuracy. By utilizing time correlation and frequency correlation and combining the information of multiple receivers, not only the positioning accuracy is improved, but also the anti-interference ability of the system is enhanced. The constructed low-altitude aircraft motion model and predicted trajectory function provide key support for the timely identification of potential threats. Real-time adjustment of positioning algorithm weights and receiver parameters further optimizes system performance. The construction of a collaborative positioning network improves system reliability, and the calculation of the credibility index provides quality assurance for the positioning results.

[0094] S5: Combine comprehensive positioning information and signal feature information, use graph neural network to build a dynamic flight behavior model, and evaluate potential threats in real time.

[0095] Specifically, the three-dimensional coordinates, velocity vector and predicted trajectory point set of the low-altitude aircraft are extracted from the comprehensive positioning information, and the signal type label, key parameters, time correlation and frequency correlation are extracted from the signal feature information; a spatiotemporal graph structure is constructed, in which the graph nodes represent the state of the low-altitude aircraft at different time points, and the graph edges represent the state transition between adjacent time points; feature vectors are assigned to the graph nodes, and the feature vectors contain three-dimensional coordinates, velocity vectors, signal type labels and key parameters; weights are assigned to the graph edges, and the weights are calculated based on time correlation, frequency correlation and state changes between adjacent time points; a temporal convolutional network is used to extract local features from the feature vectors to obtain the local temporal features of the nodes; the local temporal features are input into the graph attention network, and global spatial features are aggregated to obtain the global spatiotemporal features of the nodes; a gated recurrent unit network is used to perform sequence modeling on the global spatiotemporal features, capture long-term dependencies, and output dynamic flight behavior features; the dynamic flight behavior features are calculated for similarity with the predefined threat pattern library to obtain a threat similarity score; based on the threat similarity score and the preset threshold, it is determined whether the low-altitude aircraft poses a potential threat.

[0096] S6: When a threat is detected, the directional antenna array is activated to transmit a simulated control signal with the same frequency as the low-altitude aircraft to the target area to safely guide the low-altitude aircraft or take over control.

[0097] It should be noted that the target area refers to the area where low-altitude aircraft are located.

[0098] Furthermore, the present embodiment also provides an airport low-altitude aircraft monitoring system, including a receiver deployment module, which is used to deploy multiple software-defined radio receivers in a preset area around the airport; a data packet generation module, which is used to use the software-defined radio receiver to scan and capture radio signals in the low-altitude area to generate a data packet to be transmitted; a feature extraction module, which is used to transmit the data packet to be transmitted to the central processing server through the edge network, parse and extract features of the data packet to be transmitted in the central processing server, and generate signal feature information; a positioning module, which is used to locate the low-altitude aircraft and generate comprehensive positioning information based on the signal feature information, using a multi-source fusion positioning algorithm, combining the signal time difference and frequency difference of multiple software-defined radio receivers; a threat assessment module, which is used to combine the comprehensive positioning information and signal feature information, use a graph neural network to build a dynamic flight behavior model, and evaluate potential threats in real time; a control module, which is used to start the directional antenna array when a threat is detected, and transmit a simulated control signal with the same frequency as the low-altitude aircraft to the target area, so as to safely guide or take over the control of the low-altitude aircraft.

[0099] This embodiment also provides a computer device, which is suitable for the case of a method for monitoring low-altitude aircraft at an airport, and includes a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the method for monitoring low-altitude aircraft at an airport as proposed in the above embodiment.

[0100] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0101] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for monitoring low-altitude aircraft at an airport as proposed in the above embodiment is implemented.

[0102] In summary, the present invention realizes the accurate capture, processing and classification of radio signals in low-altitude areas by utilizing a high-precision software-defined radio receiver, combined with multi-level wavelet transform noise reduction and advanced signal classification algorithms. A multi-source fusion positioning algorithm is adopted, combined with signal time difference and frequency difference, to efficiently realize the real-time positioning of low-altitude aircraft, and dynamically adjust the positioning algorithm and parameters to improve positioning accuracy. The flight behavior is evaluated in real time through the graph neural network, potential threats are warned early, and when a threat is detected, a simulated control signal is emitted using a directional antenna array to achieve safe guidance or takeover control of the target aircraft, effectively ensuring the safety of the airspace around the airport.

[0103] Example 2, reference Figure 1 to Figure 4 , which is the second embodiment of the present invention, provides a method for monitoring low-altitude aircraft at an airport. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0104] Eight software-defined radio receivers were deployed within a 5km radius around the airport to form an evenly distributed monitoring network. Each receiver is equipped with a 4-element antenna array, and the programmable RF front-end has a frequency range of 400MHz-6GHz, a sampling rate of 20MHz, and an adjustable gain range of 0-60dB.

[0105] During the experiment, three different types of low-altitude aircraft were simulated: a civilian quadcopter drone, a fixed-wing drone and a micro-airship. These aircraft flew at different trajectories and speeds in the airspace of 1-3 kilometers around the airport, with altitudes ranging from 50-300 meters. The flight duration was 30 minutes.

[0106] First, the software-defined radio receiver successfully captured various radio signals emitted by these aircraft. Through multi-level wavelet transform noise reduction and time-frequency analysis, a clear time-frequency spectrum was obtained. The signal classification module correctly identified the remote control signal (2.4GHz), digital transmission signal (915MHz) and image transmission signal (5.8GHz).

[0107] Subsequently, the multi-source fusion positioning algorithm successfully calculated the three-dimensional coordinates of the aircraft based on the time difference and frequency difference of the signals captured by different receivers. The positioning accuracy is better than 5 meters in the horizontal direction and better than 10 meters in the vertical direction. The average deviation between the predicted trajectory and the actual flight trajectory does not exceed 20 meters.

[0108] Next, the dynamic flight behavior model constructed by the graph neural network evaluated the behavior of the aircraft in real time. During the experiment, a potential threat scenario was simulated: the fixed-wing drone suddenly changed its flight direction and approached the runway quickly. The system successfully identified this abnormal behavior within 10 seconds, and the threat similarity score reached 0.85 (the threshold was 0.7).

[0109] Finally, the system activated the directional antenna array and transmitted a simulated control signal with the same frequency as the drone to the target area. The signal power was gradually adjusted, and finally it successfully took over the control of the drone 300 meters away from the target and guided it safely to the designated area for landing.

[0110] During the entire experiment, the system's average signal processing delay was 50 milliseconds, the positioning update frequency was 10Hz, and the threat assessment cycle was 1 second. The experimental results show that the airport low-altitude aircraft monitoring method and system can effectively identify, locate and respond to potential low-altitude aircraft threats, providing strong support for airport safety management.

[0111] Preferably, the comparison indicators of the present invention and the traditional method are shown in Table 1.

[0112] Table 1 Comparison indexes of the present invention and the traditional method

[0113] index The present invention Traditional methods Signal capture frequency range 400MHz-6GHz 800MHz-2.4GHz Signal classification accuracy 98％ 85％ Positioning accuracy (horizontal) 5 meters 15 meters Positioning accuracy (vertical) 10 meters 25 meters Threat Identification Time 10 seconds 30 seconds Signal processing delay 50 ms 200 ms Positioning update frequency 10Hz 2Hz Threat Assessment Cycle 1 second 5 seconds

[0114] As can be seen from Table 1, the present invention is significantly superior to traditional methods in all key indicators. The signal capture frequency range has been expanded by nearly 3 times, covering 400MHz-6GHz; the signal classification accuracy has been increased by 13 percentage points to 98%; the positioning accuracy has been greatly improved, and the horizontal and vertical directions have been increased to 5 meters and 10 meters respectively; the threat identification time has been shortened to one-third of the original, only 10 seconds; the system response performance has been comprehensively improved, the signal processing delay has been reduced by 75% to 50 milliseconds, the positioning update frequency has been increased by 5 times to 10Hz, and the threat assessment cycle has been shortened by 80% to 1 second. These improvements fully reflect the technical advantages of the present invention in low-altitude aircraft monitoring, positioning and threat assessment, and provide a more accurate, fast and comprehensive solution for airport safety management.

[0115] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for monitoring low-altitude aircraft at an airport, characterized in that: include, Deploy multiple software-defined radio receivers in pre-set areas around the airport; Using a software defined radio receiver to scan and capture radio signals in low altitude areas, generating data packets to be transmitted; Transmitting the data packets to be transmitted to the central processing server through the edge network, parsing and extracting features of the data packets to be transmitted in the central processing server, and generating signal feature information; Based on signal feature information, a multi-source fusion positioning algorithm is used to combine the signal time difference and frequency difference of multiple software-defined radio receivers to locate low-altitude aircraft and generate comprehensive positioning information; Combining comprehensive positioning information and signal feature information, a graph neural network is used to build a dynamic flight behavior model to assess potential threats in real time; When a threat is detected, the directional antenna array is activated to transmit a simulated control signal with the same frequency as the low-altitude aircraft to the target area, thereby safely guiding the low-altitude aircraft or taking over control; The generating of the data packet to be transmitted comprises the following steps: Set the scanning parameters of the software-defined radio receiver to adapt to the signal characteristics of the low-altitude area; Using a programmable RF front end and a high-speed analog-to-digital converter, the radio signal in the low-altitude area is received and digitized to obtain a digitized radio signal; Using a digital signal processor, performing multi-level wavelet transform noise reduction on the digitized radio signal; Using the signal analysis software module, the time-frequency analysis of the digitized radio signal after noise reduction is performed to generate a time-frequency spectrum; Start the signal classification module to classify the radio signals based on the time-frequency spectrum; Control the multi-element antenna array to perform beamforming and generate steerable beam signals for various types of signals; Performing fast Fourier transform on the steerable beam signal to obtain the signal spectrum characteristics; Combining the signal spectrum characteristics with the corresponding signal type label to generate a data packet to be transmitted; The classification of radio signals based on the time-frequency spectrum comprises the following steps: Perform image preprocessing on the time-frequency spectrum; Extracting feature vectors from the preprocessed time-frequency spectrum, wherein the feature vectors include frequency bandwidth, time persistence, and energy distribution; The pre-trained convolutional neural network model is used to perform initial classification of feature vectors. Based on the initial classification results, an ensemble learning method is used to perform refined classification. Apply temporal consistency checks to the refined classification results to eliminate instantaneous misclassifications; The spectrum occupancy analysis method is used to identify the overlapping areas of multiple signals in the time-frequency spectrum, and the blind source separation algorithm is applied to the overlapping areas of multiple signals to separate independent signal components. The separated signal components are re-input into the classification process for secondary classification; Combining the initial classification results and the secondary classification results, a signal type label is generated; The signal type label is associated with the original time-frequency spectrum to form a labeled time-frequency spectrum.

2. The method for monitoring low-altitude aircraft at an airport as claimed in claim 1, characterized in that: The multi-level wavelet transform is performed to reduce noise, and the noise-reduced signal x'(n) is obtained. The specific formula is as follows: Where x'(n) is the digitized radio signal after noise reduction, a J is the approximate coefficient of the Jth layer, J is the number of layers of wavelet decomposition, φ J,k (n) is the scaling function, ψ J,k (n) is the wavelet function, d' j,k is the detail coefficient of the jth layer after processing, and the specific formula is as follows: d' j,k =sign(d j,k )(|d j,k |-λ j ) + Among them, d' j,k is the detail coefficient of the jth layer after processing, sign(·) is the sign function, (·) + is the positive part in soft threshold processing, λ j The threshold of the jth layer is as follows: Among them, σ j is the estimated value of the standard deviation of the noise in the jth layer, and N is the signal length.

3. The method for monitoring low-altitude aircraft at an airport as claimed in claim 1, characterized in that: The generating of steerable beam signals for various types of signals comprises the following steps: Obtaining the geometrical layout parameters of the multi-element antenna array and the phase center position of each antenna element; Based on the frequency characteristics of various signals, calculate the wavelength corresponding to each signal; For each identified signal type, determine the desired main beam direction and sidelobe suppression level; Using an adaptive beamforming algorithm, the amplitude and phase weights for each antenna element are calculated to determine the main beam direction and sidelobe suppression level. applying the calculated amplitude and phase weights to each antenna element of the multi-element antenna array; Phase control is performed by adjusting the digital phase shifter of the antenna unit; The signal amplitude of each antenna element is adjusted using a variable gain amplifier, in conjunction with phase control; Perform real-time signal synthesis to superimpose the signals received by each antenna unit according to the set amplitude and phase; Digitally filter the synthesized signal to remove interference from non-target frequency bands; A steerable beam signal is generated and stored in association with a corresponding signal type identifier.

4. The method for monitoring low-altitude aircraft at an airport as claimed in claim 1, characterized in that: The method of locating the low-altitude aircraft and generating comprehensive positioning information comprises the following steps: Dynamically select positioning algorithm combinations based on signal type labels and key parameters; Utilizing time correlation and frequency correlation, the signal time difference and frequency difference between multiple software defined radio receivers are calculated; Input the signal time difference and frequency difference into the selected positioning algorithm combination to calculate the three-dimensional coordinates and velocity vector of the low-altitude aircraft; Based on the changing trends of time correlation and frequency correlation, a low-altitude aircraft motion model is constructed to calculate the predicted trajectory; According to the time-varying characteristics of key parameters, the weight coefficients of each algorithm in the positioning algorithm combination are adjusted in real time; Utilize the spatial distribution information of multiple software-defined radio receivers to build a collaborative positioning network to improve positioning accuracy and reduce the risk of single point failure; Dynamically adjust the parameters of the software-defined radio receiver, including beam direction and gain, based on the positioning results and predicted trajectory; Calculate the credibility index of the positioning result and evaluate the positioning accuracy; Generate comprehensive positioning information, associate the comprehensive positioning information with corresponding signal characteristic information, and update the records in the database.

5. The method for monitoring low-altitude aircraft at an airport as claimed in claim 4, characterized in that: The dynamic selection positioning algorithm combination comprises the following steps: Build a positioning algorithm library; Based on the signal type label, preliminarily select a subset of applicable positioning algorithms from the positioning algorithm library; The time-frequency resolution index of the signal is calculated using the center frequency and bandwidth among the key parameters, and the algorithms in the positioning algorithm subset are prioritized according to the time-frequency resolution index; Based on the peak power among the key parameters, the signal-to-noise ratio of the signal is estimated, and the top N algorithms are selected from the sorted positioning algorithm subset according to the signal-to-noise ratio; Construct a positioning algorithm combination scoring model that takes into account algorithm complexity, computing resource requirements, and historical positioning accuracy; The positioning algorithm combination scoring model is used to perform combination optimization on the selected N algorithms to obtain the optimal positioning algorithm combination.

6. An airport low-altitude aircraft monitoring system, based on the airport low-altitude aircraft monitoring method according to any one of claims 1 to 5, characterized in that: Also includes, A receiver placement module, for placing a plurality of software defined radio receivers in a preset area around the airport; A data packet generation module, used to use a software defined radio receiver to scan and capture radio signals in a low-altitude area and generate data packets to be transmitted; A feature extraction module is used to transmit the data packet to be transmitted to the central processing server through the edge network, and parse and extract features of the data packet to be transmitted in the central processing server to generate signal feature information; A positioning module is used to locate low-altitude aircraft and generate comprehensive positioning information based on signal feature information, using a multi-source fusion positioning algorithm and combining the signal time difference and frequency difference of multiple software-defined radio receivers; The threat assessment module is used to combine comprehensive positioning information and signal feature information, use graph neural networks to build a dynamic flight behavior model, and evaluate potential threats in real time; The control module is used to activate the directional antenna array when a threat is detected, and transmit a simulated control signal with the same frequency as the low-altitude aircraft to the target area, so as to safely guide the low-altitude aircraft or take over control.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for monitoring low-altitude aircraft at an airport described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for monitoring low-altitude aircraft at an airport described in any one of claims 1 to 5 are implemented.

Images