An ADS-B signal flight identification method and device

Abstract

The application provides an ADS-B signal flight identification method and device, wherein the method comprises the following steps: acquiring information pulses of ADS-B signals frame by frame to generate an information pulse set; calculating the instantaneous phase of each information pulse in the information pulse set, and calculating the instantaneous phase difference between two adjacent information pulses according to the instantaneous phase; performing smoothing processing on the instantaneous phase difference to generate smoothed phase difference data; performing multi-row fusion processing on the smoothed phase difference data, and generating a feature image set according to the fused data; and inputting the feature image set into a neural network model for identification. The ADS-B signal flight identification method and device provided by the application can improve the identification rate of ADS-B signals and meet the actual engineering application requirements.

Description

Technical Field

[0001] This invention belongs to the field of signal processing technology, and in particular relates to an ADS-B signal flight identification method and device. Background Technology

[0002] With ground stations receiving tens of thousands of different types of aircraft flights, flight identification and classification statistics become crucial. Current flight identification methods primarily extract ICAO numbers through ADS-B signal demodulation and protocol parsing. These methods are not only labor-intensive but also prone to labeling errors during long processing times. To address this issue, those skilled in the art utilize AI technology to automate flight identification, thereby increasing accuracy and reducing the statistical workload for ground station personnel.

[0003] However, since ADS-B signals all use the same modulation scheme, the actual automatic flight identification rate remains low. Although those skilled in the art can optimize the process using methods such as normalized power spectrum and instantaneous frequency phase, the complex background of actual signal reception means that even increasing the depth of the neural network does not fundamentally improve the flight identification rate and cannot meet practical application requirements. Summary of the Invention

[0004] In view of this, the present invention aims to provide an ADS-B signal flight identification method and apparatus to improve the automatic identification rate of ADS-B signals.

[0005] To achieve the above objectives, the technical solution created by this invention is implemented as follows:

[0006] In one aspect, embodiments of the present invention provide an ADS-B signal flight identification method, comprising:

[0007] Acquire information pulses of the ADS-B signal frame by frame and generate an information pulse set;

[0008] Calculate the instantaneous phase of each information pulse in the information pulse set, and calculate the instantaneous phase difference between two adjacent frames of information pulses based on the instantaneous phase;

[0009] The instantaneous phase difference is smoothed to generate smoothed phase difference data;

[0010] Multi-row fusion processing is performed on the smoothed phase difference data, and a feature image set is generated based on the fused data;

[0011] The feature image set is input into a neural network model for recognition.

[0012] Furthermore, the step of smoothing the instantaneous phase difference to generate smoothed phase difference data includes:

[0013] A window function is set based on the frame length and instantaneous phase difference of the information pulse, and the instantaneous phase difference is smoothed through the window function to generate smooth phase difference data.

[0014] Furthermore, the step of setting the window function based on the frame length and instantaneous phase difference of the information pulse includes:

[0015] The window length of the window function is set according to the frame length of the information pulse, and the window length of the window function is less than or equal to one-twentieth of the frame length of the information pulse.

[0016] The smoothing step size of the window function is set according to the length of the instantaneous phase difference, and the smoothing step size of the window function is between one-hundredth and one-twentieth of the length of the instantaneous phase difference.

[0017] Furthermore, before performing multi-line fusion processing on the smoothed phase difference data, the ADS-B signal flight identification method further includes:

[0018] The number of fused rows for multi-line fusion processing is calculated based on the preset number of feature patterns and the number of frames of the ADS-B signal.

[0019] Furthermore, the step of generating a feature image set based on the fused data includes:

[0020] The imagesc function is used to process the fused data to generate a feature image set.

[0021] On the other hand, embodiments of the present invention also provide an ADS-B signal flight identification device, comprising:

[0022] The acquisition module is used to acquire information pulses of the ADS-B signal frame by frame and generate information pulse sets;

[0023] The calculation module is used to calculate the instantaneous phase of each information pulse in the information pulse set, and to calculate the instantaneous phase difference between two adjacent frames of information pulses based on the instantaneous phase.

[0024] The smoothing module is used to smooth the instantaneous phase difference and generate smoothed phase difference data;

[0025] The generation module is used to perform multi-row fusion processing on smooth phase difference data and generate a feature image set based on the fused data;

[0026] The recognition module is used to input the feature image set into the neural network model for recognition.

[0027] Compared with existing technologies, the ADS-B signal flight identification method and device described in this invention have the following advantages:

[0028] The present invention provides an ADS-B signal flight identification method and device, which can significantly improve the identification rate under different ADS-B signal reception backgrounds by extracting the instantaneous smooth phase difference pattern features of intra-frame pulses. It is applicable to neural networks for various identification and classification, reduces the workload of protocol parsing and manual identification, and improves the intelligent identification level of ground stations for different flights, thus having high engineering application value. Attached Figure Description

[0029] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0030] Figure 1 A flowchart of the ADS-B signal flight identification method described in Embodiment 1 of this invention is provided.

[0031] Figure 2 This is a schematic diagram of the structure of the ADS-B signal flight identification device described in Embodiment 2 of the present invention. Detailed Implementation

[0032] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0033] Example 1

[0034] Figure 1 The flowchart of the ADS-B signal flight identification method provided in Embodiment 1 of the present invention is shown in the figure. In this embodiment, the ADS-B signal flight identification method specifically includes the following steps:

[0035] Step 110: Acquire information pulses of the ADS-B signal frame by frame and generate an information pulse set.

[0036] When the ground receiving station receives the ADS-B signal, in order to improve the recognition rate of subsequent signals, the information pulses of the ADS-B signal can be extracted frame by frame and an information pulse set p(t) can be generated for subsequent signal processing.

[0037] When there are m ADS-B signals to be identified, and each ADS-B signal includes n frames of information pulses, the generated information pulse set p(t) is as follows:

[0038]

[0039] Therefore, in subsequent signal processing, p can be used. ijp(t) represents the j-th frame information pulse of the i-th ADS-B signal in the information pulse set p(t), where i = 1, 2, 3, ..., m; j = 1, 2, 3, ..., n.

[0040] Correspondingly, data fusion can be performed on information pulses belonging to the same ADS-B signal to generate a fused information pulse set R(t):

[0041]

[0042] In subsequent signal processing, R can be used i (t) represents all information pulses (i = 1, 2, 3, ..., m) of the i-th ADS-B signal, in order to reduce the complexity of subsequent signal processing.

[0043] Step 120: Calculate the instantaneous phase of each information pulse in the information pulse set, and calculate the instantaneous phase difference between two adjacent frames of information pulses based on the instantaneous phase.

[0044] In each frame of the ADS-B signal, information pulse p ij Once (t) is determined, the instantaneous phase ξ of each frame of information pulse can be calculated using the angle function. ij (t), that is:

[0045] ξ ij (t)=angle(p ij (t))

[0046] Accordingly, the instantaneous phase ξ(t) of each frame of each ADS-B signal is as follows:

[0047]

[0048] The instantaneous phase difference Φ between two adjacent frames of information pulses in the same ADS-B signal can then be calculated using the following formula. ia (t):

[0049] Φ ia (t)=ξ iv (t)-ξ 1,v-1 (t)

[0050] Where a = 1, 2, 3, ..., n-1; v = 2, 3, 4, ..., n.

[0051] Correspondingly, the instantaneous phase difference Φ(t) of each ADS-B signal is as follows:

[0052]

[0053] Step 130: Smooth the instantaneous phase difference to generate smoothed phase difference data.

[0054] Once the instantaneous phase difference Φ(t) has been calculated, a common window function can be used to smooth it. Accordingly, the selected window function should be set according to the frame length of the information pulse and the instantaneous phase difference to improve the smoothing effect.

[0055] For example, this embodiment uses the Hanning window as an example to illustrate the smoothing process of the instantaneous phase difference Φ(t). The window function expression of the Hanning window is as follows:

[0056] win(t)=0.5*[1-cos(2*π*t / N)]

[0057] In the formula, N is the window length of the window function. In this embodiment, the window length of the window function should be set according to the frame length of the information pulse, and the window length of the window function should be less than or equal to one-twentieth of the frame length of the information pulse. In addition, the smoothing step size of the window function should be set according to the length of the instantaneous phase difference, and the smoothing step size of the window function should be between one-hundredth and one-twentieth of the length of the instantaneous phase difference.

[0058] After smoothing the instantaneous phase difference Φ(t) using a window function, the smoothed phase difference data θ(t) can be obtained as shown below:

[0059]

[0060] Therefore, θ can be used in subsequent processing. ia (t) represents the smoothed phase difference data between any two adjacent information pulses in the i-th ADS-B signal, where i = 1, 2, 3, ..., m; a = 1, 2, 3, ..., n-1.

[0061] Step 140: Perform multi-row fusion processing on the smoothed phase difference data, and generate a feature image set based on the fused data.

[0062] Before performing multi-line fusion processing on the smoothed phase difference data, the operator should determine the number of fusion rows based on the actual required number of feature patterns and the number of frames in the ADS-B signal. Specifically, the relationship between the number of feature patterns and the number of fusion rows is shown in the following formula:

[0063] num = floor(n / k)

[0064] Where floor is the floor function, num is the number of feature patterns that can be generated, n is the number of frames of the ADS-B signal, and k is the number of fused rows for multi-line fusion processing.

[0065] The fused data is then processed using the imagesc function to generate the feature image set λ shown below:

[0066]

[0067] Step 150: Input the feature image set into the neural network model for recognition.

[0068] This embodiment provides an ADS-B signal flight identification method that can significantly improve the identification rate under different ADS-B signal reception backgrounds by extracting the instantaneous smooth phase difference pattern features of intra-frame pulses. This not only reduces the workload of protocol parsing and manual identification, but also adapts to various neural networks for identification and classification. Therefore, it can improve the intelligent identification level of ground stations for different flights and meet the needs of practical engineering applications.

[0069] Example 2

[0070] Figure 2 This is a schematic diagram of the structure of the ADS-B signal flight identification device provided in Embodiment 2 of the present invention, as shown below. Figure 2 As shown, the ADS-B signal flight identification device includes:

[0071] The acquisition module 210 is used to acquire information pulses of the ADS-B signal frame by frame and generate an information pulse set;

[0072] The calculation module 220 is used to calculate the instantaneous phase of each information pulse in the information pulse set, and to calculate the instantaneous phase difference between two adjacent frames of information pulses based on the instantaneous phase.

[0073] The smoothing module 230 is used to smooth the instantaneous phase difference and generate smoothed phase difference data;

[0074] The generation module 240 is used to perform multi-row fusion processing on the smoothed phase difference data and generate a feature image set based on the fused data;

[0075] The recognition module 250 is used to input the feature image set into the neural network model for recognition.

[0076] Based on the above embodiments, the smoothing module may include:

[0077] The first setting unit is used to set the window length of the window function according to the frame length of the information pulse, and the window length of the window function is less than or equal to one-twentieth of the frame length of the information pulse.

[0078] The second setting unit is used to set the smoothing step size of the window function according to the length of the instantaneous phase difference, and the smoothing step size of the window function is between one-hundredth and one-twentieth of the length of the instantaneous phase difference.

[0079] Based on the above embodiments, the ADS-B signal flight identification device further includes:

[0080] The fusion row count calculation module is used to calculate the fusion row count for multi-line fusion processing based on the preset number of feature patterns and the number of frames of the ADS-B signal.

[0081] The ADS-B signal flight identification device provided in this embodiment of the invention can execute the ADS-B signal flight identification method provided in this embodiment of the invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0082] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A method for identifying ADS-B signals, characterized in that... include: Acquire information pulses of the ADS-B signal frame by frame and generate an information pulse set; Calculate the instantaneous phase of each information pulse in the information pulse set, and calculate the instantaneous phase difference between two adjacent frames of information pulses based on the instantaneous phase; The instantaneous phase difference is smoothed to generate smoothed phase difference data; Multi-row fusion processing is performed on the smoothed phase difference data, and a feature image set is generated based on the fused data; The feature image set is input into a neural network model for recognition.

2. The ADS-B signal flight identification method according to claim 1, characterized in that: The process of smoothing the instantaneous phase difference to generate smoothed phase difference data includes: A window function is set based on the frame length and instantaneous phase difference of the information pulse, and the instantaneous phase difference is smoothed through the window function to generate smooth phase difference data.

3. The ADS-B signal flight identification method according to claim 2, characterized in that: The step of setting the window function based on the frame length and instantaneous phase difference of the information pulse includes: The window length of the window function is set according to the frame length of the information pulse, and the window length of the window function is less than or equal to one-twentieth of the frame length of the information pulse. The smoothing step size of the window function is set according to the length of the instantaneous phase difference, and the smoothing step size of the window function is between one-hundredth and one-twentieth of the length of the instantaneous phase difference.

4. The ADS-B signal flight identification method according to claim 1, characterized in that: Prior to performing multi-line fusion processing on the smoothed phase difference data, the ADS-B signal flight identification method further includes: The number of fused rows for multi-line fusion processing is calculated based on the preset number of feature patterns and the number of frames of the ADS-B signal.

5. The ADS-B signal flight identification method according to claim 1, characterized in that: The step of generating a feature image set based on the fused data includes: The imagesc function is used to process the fused data to generate a feature image set.

6. An ADS-B signal flight identification device, characterized in that, include: The acquisition module is used to acquire information pulses of the ADS-B signal frame by frame and generate information pulse sets; The calculation module is used to calculate the instantaneous phase of each information pulse in the information pulse set, and to calculate the instantaneous phase difference between two adjacent frames of information pulses based on the instantaneous phase. The smoothing module is used to smooth the instantaneous phase difference and generate smoothed phase difference data; The generation module is used to perform multi-row fusion processing on smooth phase difference data and generate a feature image set based on the fused data; The recognition module is used to input the feature image set into the neural network model for recognition.

Images