Unmanned aerial vehicle signal identification training platform

Abstract

The application relates to the technical field of unmanned aerial vehicle signal identification training, and discloses an unmanned aerial vehicle signal identification training platform, which comprises a main control unit, a signal simulation unit, an interference simulation unit, a signal receiving unit and a high-precision timing unit interconnected through a backplane bus, and a training terminal connected through an Ethernet. The high-precision timing unit provides a unified time-frequency reference for each unit, the signal simulation unit generates unmanned aerial vehicle simulation signals, the interference simulation unit generates electromagnetic interference signals and performs synchronous superposition, the signal receiving unit acquires signal measurement data, the main control unit generates evaluation results by comparing the identification results of the trainees with standard signal parameters and dynamically adjusts training parameters, and a closed-loop training platform is formed.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) signal recognition training technology, and more specifically, to a UAV signal recognition training platform. Background Technology

[0002] With the rapid development of drone technology, the demand for drone signal recognition capabilities in the military, security and other fields is becoming increasingly urgent. There is a pressing need for professional training platforms to improve trainees' ability to identify, distinguish and resist interference from different types of drone signals.

[0003] Current signal recognition training mainly utilizes general-purpose equipment in electronic measurement laboratories, namely a combination of signal generators and spectrum analyzers. This approach is essentially a testing environment rather than a training platform.

[0004] Existing technologies suffer from several problems: electromagnetic environment simulation distortion; the use of a single signal generator; lack of coherence simulation of multi-source signals in the time and frequency domains; inability to reproduce the real complex electromagnetic spectrum environment; limited training methods; trainees can only observe single-spectrum waveforms, lacking practical training in dynamically extracting and identifying weak signals under complex electromagnetic interference environments; poor platform integration and adaptability; independent and scattered functional modules connected through general interfaces, lacking a unified main control unit, resulting in complex system deployment and numerous connections, making it difficult to meet the rapid relocation and high reliability requirements of field mobile training; and a lack of evaluation and feedback mechanisms; existing platforms mainly rely on manual comparison for evaluation, which is highly subjective, has a single evaluation dimension, limited accuracy, and cannot adjust the training difficulty in real time based on the evaluation results to form an effective closed-loop training. Summary of the Invention

[0005] This invention provides a drone signal recognition training platform, which solves the technical problems in related technologies such as the lack of systematic training, difficulty in quantifying and evaluating training effects, and the inability to adaptively adjust training difficulty.

[0006] This invention discloses a training method for UAV signal recognition, comprising the following steps: a main control unit acquires training scenario data, generates signal configuration parameters and interference configuration parameters based on the training scenario data, and sends the signal configuration parameters to a signal simulation unit and the interference configuration parameters to an interference simulation unit via a backplane bus. The signal configuration parameters include the UAV signal type to be simulated, carrier frequency, modulation method, power level, and channel effect parameters; the interference configuration parameters include the interference signal type, interference frequency band, interference power, and interference timing parameters. The signal simulation unit generates a simulated UAV signal through digital signal processing based on the signal configuration parameters and outputs the simulated UAV signal through an RF port. The interference simulation unit generates an electromagnetic interference signal through digital signal processing based on the interference configuration parameters; the electromagnetic interference signal and the simulated UAV signal are combined via RF to form a mixed RF signal. The signal receiving unit receives the mixed RF signal and... The mixed radio frequency signal undergoes down-conversion, analog-to-digital conversion, and digital signal processing sequentially to obtain signal measurement data, including spectrum data, signal frequency, signal bandwidth, and signal level. The training terminal displays the spectrum data and receives signal recognition input from trainees based on the spectrum data, obtaining the trainees' recognition results, which include the signal type, signal frequency, signal bandwidth, and signal level identified by the trainees. The main control unit acquires the standard signal parameters corresponding to the current training subject, compares the trainees' recognition results with the standard signal parameters, calculates the deviation values ​​for each evaluation dimension, and generates evaluation results based on the deviation values. The main control unit sends the evaluation results to the training terminal for display and, based on the evaluation results and the preset training strategy, determines whether the training parameters need to be adjusted. If adjustment is required, the main control unit generates updated signal configuration parameters and interference configuration parameters and returns to the parameter distribution step.

[0007] Furthermore, before issuing configuration parameters, the main control unit also sends a synchronization trigger command to the high-precision timing unit via the backplane bus. The high-precision timing unit responds to the command and distributes a unified reference clock signal to the signal simulation unit, interference simulation unit, and signal receiving unit through the clock distributor, so that each unit is locked to the same time and frequency reference.

[0008] Furthermore, the steps for the signal simulation unit to generate the UAV simulated signal include: the digital upconversion module inside the FPGA generates a digital baseband signal and performs upconversion processing according to the carrier frequency and modulation method in the signal configuration parameters to obtain a digital intermediate frequency signal; the digital intermediate frequency signal is converted into an analog intermediate frequency signal by a digital-to-analog converter; the analog intermediate frequency signal is converted to the target radio frequency band by a radio frequency upconversion module to obtain the UAV simulated signal.

[0009] Furthermore, after generating the digital baseband signal, the signal simulation unit also performs channel effect superposition processing on the digital baseband signal inside the FPGA. The processing is as follows: according to the channel effect parameters in the signal configuration parameters, Rayleigh fading or shadow fading is superimposed in the digital domain, and the multipath delay spread characteristics are simulated by dynamically adjusting the coefficients of the FIR filter to obtain the digital baseband signal after superposition of channel effects.

[0010] Furthermore, the signal simulation unit is equipped with a digitally controlled attenuator at the RF output end. According to the power level requirements in the signal configuration parameters, the attenuation of the digitally controlled attenuator is adjusted to control the output power of the UAV simulated signal.

[0011] Furthermore, the types of electromagnetic interference signals include one or more combinations of white noise interference, narrowband interference, and frequency hopping interference.

[0012] Furthermore, the training terminal records the first timestamp of the signal occurrence and the second timestamp of the trainee completing the recognition operation, and calculates the response delay based on the first and second timestamps.

[0013] Furthermore, the main control unit compares the trainee's identification results with standard signal parameters by: matching the signal type identified by the trainee with the standard signal type; if the match is successful, the score for that dimension is full; if the match fails, the score for that dimension is zero; calculating the difference between the frequency of the signal identified by the trainee and the frequency of the standard signal to obtain the frequency deviation; calculating the difference between the bandwidth of the signal identified by the trainee and the bandwidth of the standard signal to obtain the bandwidth deviation; calculating the difference between the signal level identified by the trainee and the signal level of the standard signal to obtain the level deviation; mapping the frequency deviation, bandwidth deviation, level deviation, and response delay to the corresponding quantitative scores of each dimension according to the preset scoring rules; and weighted summing of the quantitative scores of each evaluation dimension to obtain a comprehensive score.

[0014] Furthermore, the preset training strategies include: if the comprehensive score is higher than the preset upper threshold, the training difficulty is increased; if the comprehensive score is lower than the preset lower threshold, the training difficulty is decreased; if the comprehensive score is between the upper and lower thresholds, the current training difficulty is maintained. Among them, the ways to increase the training difficulty include increasing the power of the electromagnetic interference signal, increasing the number of types of interference signals, reducing the output power of the UAV simulated signal, or increasing the number of UAV signals that appear simultaneously.

[0015] This invention discloses a UAV signal recognition training platform, comprising: a main control unit, used to acquire training scenario data, generate signal configuration parameters and interference configuration parameters based on the training scenario data, distribute configuration parameters via a backplane bus, acquire standard signal parameters and trainee recognition results for comparison to generate evaluation results, and adjust training parameters based on the evaluation results; a signal simulation unit, connected to the main control unit via a backplane bus, used to generate UAV simulated signals based on the signal configuration parameters; an interference simulation unit, connected to the main control unit via a backplane bus, used to generate electromagnetic interference signals based on interference configuration parameters; a signal receiving unit, connected to the main control unit via a backplane bus, used to receive mixed radio frequency signals and perform signal processing to obtain signal measurement data; a high-precision time synchronization unit, connected to the main control unit via a backplane bus, used to provide a unified reference clock signal to the signal simulation unit, interference simulation unit, and signal receiving unit; and a training terminal, connected to the main control unit via Ethernet, used to display spectrum data and receive signal recognition operations from trainees.

[0016] This invention employs a backplane integrated chassis architecture, tightly coupling the main control unit, signal simulation unit, interference simulation unit, signal receiving unit, and high-precision timing unit via a backplane bus. This solves the technical problems of poor platform integration and adaptability. The system has a compact structure and reliable connections, meeting the rapid relocation and high reliability requirements of field mobile training. By introducing a high-precision timing unit to provide a unified time and frequency reference for all units, the UAV simulated signal and electromagnetic interference signal are strictly synchronized in both the time and frequency domains. This solves the technical problem of electromagnetic environment simulation distortion and accurately simulates the coherence characteristics of interference signals and target signals in real environments. The system achieves several improvements. First, by configuring an independent interference simulation unit, it generates various types of electromagnetic interference signals, such as white noise interference, narrowband interference, and frequency hopping interference, and synchronously superimposes them with the UAV's simulated signals. This solves the technical problem of limited training methods, enabling trainees to dynamically extract and identify weak signals in complex electromagnetic interference environments. Second, by deploying an evaluation algorithm in the main control unit, it automatically quantifies and scores multiple dimensions, including signal type identification accuracy, frequency deviation, bandwidth deviation, level deviation, and response delay. Based on the evaluation results, it dynamically adjusts training parameters, solving the technical problem of a lack of evaluation feedback mechanisms and forming a closed loop of training, evaluation, and optimization. Attached Figure Description

[0017] Figure 1 This is a flowchart of the UAV signal recognition training method provided in an embodiment of the present invention; Figure 2 This is a sub-flowchart of generating simulated UAV signals in the UAV signal recognition training method provided in this embodiment of the invention; Figure 3This is a sub-flowchart of the UAV signal recognition training method provided in this embodiment of the invention, which compares the trainee's recognition results with standard signal parameters to generate evaluation results; Figure 4 This is a sub-flowchart of the training platform workflow in the UAV signal recognition training method provided in this embodiment of the invention. Detailed Implementation

[0018] With the rapid development of drone technology, various types of drones are widely used in military reconnaissance, civilian operations, and other fields. However, this also brings security risks such as illegal intrusion and signal interference. The military, security, and other fields have an increasingly urgent need for drone signal recognition capabilities, and there is a pressing need for professional training platforms to improve trainees' ability to identify, differentiate, and resist interference from different types of drone signals.

[0019] Current signal recognition training primarily utilizes general-purpose equipment from electronic measurement laboratories, namely a combination of a signal generator and a spectrum analyzer. This approach is essentially a testing environment rather than a training platform, and it presents the following technical problems: First, there is the problem of electromagnetic environment simulation distortion. Existing technologies use a single signal generator and lack coherence simulation of multi-source signals in the time, frequency, and spatial domains, which fails to reproduce the real complex electromagnetic spectrum environment, resulting in overly idealized training scenarios.

[0020] Secondly, there is the problem of limited training methods. Existing training methods mostly rely on the static mode of signal generators and spectrum analyzers, allowing trainees to observe only a single spectrum waveform. There is a lack of practical training in dynamically extracting and identifying weak signals in complex electromagnetic interference environments.

[0021] Third, there is the problem of poor platform integration and adaptability. The functional modules are independent and scattered, connected through a common interface, and lack a unified main control unit, resulting in complex system deployment and numerous connections, making it difficult to meet the requirements of rapid relocation and high reliability needed for field mobile training.

[0022] Fourth, there is a lack of evaluation and feedback mechanisms. Existing platforms mainly rely on manual comparison for evaluation, which is highly subjective, has a single evaluation dimension, limited accuracy, and cannot adjust the training difficulty in real time based on the evaluation results to form an effective closed-loop training.

[0023] According to an embodiment of this invention, a method for training unmanned aerial vehicle (UAV) signal recognition is provided, which is based on an integrated hardware closed-loop system. It should be understood that this hardware closed-loop system adopts a backplane integrated chassis architecture, including a main control unit, a signal simulation unit, an interference simulation unit, a signal receiving unit, and a high-precision timing unit interconnected via a backplane bus, and also includes a training terminal connected to the main control unit via Ethernet. The backplane bus refers to a high-speed internal communication channel that integrates all hardware modules such as the main control unit, signal simulation unit, interference simulation unit, and signal receiving unit onto the same chassis base plate; the high-precision timing unit refers to a dedicated module that provides a unified, high-precision time and frequency reference for all units within the platform that need to work collaboratively.

[0024] This embodiment provides a drone signal recognition training method including the following steps: Step 1: Configure training parameters and distribute them to each hardware unit The main control unit acquires training scenario data, generates signal configuration parameters and interference configuration parameters based on the training scenario data, and sends the signal configuration parameters to the signal simulation unit and the interference configuration parameters to the interference simulation unit via the backplane bus. The signal configuration parameters include the type of UAV signal to be simulated, carrier frequency, modulation method, power level, and channel effect parameters; the interference configuration parameters include the type of interference signal, interference frequency band, interference power, and interference timing parameters.

[0025] The aforementioned training scenario data is a training subject configuration file pre-stored in the main control unit. This configuration file defines the signal environment characteristics required for the training scenario. The training scenario data may include a single UAV signal scenario, a mixed signal scenario of multiple UAV models, or a weak signal scenario under strong interference.

[0026] The aforementioned signal types may include UAV control link signals and image transmission link signals, with different signal types corresponding to different frequency band ranges and modulation characteristics.

[0027] In this embodiment of the application, in order to ensure that the signals output by the signal simulation unit and the interference simulation unit are strictly synchronized in the time domain and frequency domain, the main control unit sends a synchronization trigger command to the high-precision timing unit through the backplane bus before issuing the configuration parameters. The high-precision timing unit responds to the command and distributes the unified reference clock signal to the signal simulation unit, the interference simulation unit and the signal receiving unit through the clock distributor, so that each unit is locked to the same time reference and frequency reference.

[0028] Step 2: Generate drone simulation signals The signal simulation unit generates a drone simulation signal through digital signal processing based on the signal configuration parameters, and outputs the drone simulation signal through the radio frequency port.

[0029] The aforementioned signal simulation unit adopts a FPGA-based direct digital synthesis architecture, and its signal generation process includes the following sub-steps: Step 201: The digital upconversion module inside the FPGA generates a digital baseband signal and performs upconversion processing based on the carrier frequency and modulation method in the signal configuration parameters to obtain a digital intermediate frequency signal; Step 202: The digital intermediate frequency signal is converted into an analog intermediate frequency signal by a digital-to-analog converter; Step 203: The simulated intermediate frequency signal is converted to the target radio frequency band by the radio frequency upconversion module to obtain the simulated UAV signal.

[0030] In this embodiment, to simulate the multipath effect in a real wireless propagation environment, the signal simulation unit performs channel effect superposition processing on the digital baseband signal within the FPGA after generating the digital baseband signal. This process involves superimposing Rayleigh fading or shadowing fading in the digital domain according to the channel effect parameters in the signal configuration parameters, and simulating multipath delay spread characteristics by dynamically adjusting the coefficients of the FIR filter to obtain the digital baseband signal after superimposing the channel effect.

[0031] In this embodiment of the application, in order to simulate the signal strength changes of the UAV under different distance conditions, the signal simulation unit is equipped with a digitally controlled attenuator at the radio frequency output end. According to the power level requirements in the signal configuration parameters, the attenuation of the digitally controlled attenuator is adjusted to precisely control the output power of the UAV simulated signal.

[0032] Step 3: Generate an electromagnetic interference signal and synchronously superimpose it with the simulated signal from the UAV. The interference simulation unit generates electromagnetic interference signals through digital signal processing based on interference configuration parameters. The electromagnetic interference signals and the UAV simulated signals are combined by radio frequency under the same clock reference to form a hybrid radio frequency signal.

[0033] The aforementioned types of electromagnetic interference signals may include one or more combinations of white noise interference, narrowband interference, and frequency hopping interference. White noise interference is used to simulate broadband background noise environments, narrowband interference is used to simulate co-channel interference at specific frequencies, and frequency hopping interference is used to simulate dynamically changing interference sources.

[0034] The aforementioned interference simulation unit can be a separate FPGA-based radio frequency board, or it can share hardware resources with the signal simulation unit and run as an independent algorithm module within the FPGA.

[0035] In this embodiment, since both the signal simulation unit and the interference simulation unit are connected to a unified reference clock provided by the high-precision timing unit, the electromagnetic interference signal and the UAV simulation signal are strictly synchronized in frequency and phase, avoiding false frequency deviations caused by clock drift, thereby ensuring that the interference signal can accurately align with the frequency domain position of the target signal and simulate real co-channel or adjacent-channel interference scenarios.

[0036] Step 4: Receive the mixed radio frequency signals and perform signal processing. The signal receiving unit receives the mixed radio frequency signal and sequentially performs down-conversion, analog-to-digital conversion, and digital signal processing on the mixed radio frequency signal to obtain signal measurement data. This signal measurement data includes spectrum data, signal frequency, signal bandwidth, and signal level.

[0037] The aforementioned signal receiving unit uses a broadband receiving and processing board, whose receiving frequency band covers the frequency range used by common UAV signals.

[0038] In this embodiment, since the signal receiving unit is connected to the unified reference clock provided by the high-precision timing unit, its local oscillator frequency has the same frequency reference as the carrier frequency of the signal simulation unit. Therefore, the frequency measurement value and bandwidth measurement value in the signal measurement data have high measurement accuracy, which can provide an accurate data basis for subsequent automatic evaluation.

[0039] Step 5: Obtain the signal recognition results from the trainees. The training terminal displays the spectrum data output by the signal receiving unit, receives signal recognition input from the trainee based on the spectrum data, and obtains the trainee's recognition results. These results include the signal type, frequency, bandwidth, and level identified by the trainee.

[0040] The training terminal records the first timestamp of the signal occurrence and the second timestamp of the trainee completing the recognition operation, and calculates the response delay based on the first and second timestamps.

[0041] The aforementioned training terminal is a ruggedized touch screen all-in-one machine, which is connected to the main control unit via Ethernet and is used to provide a human-computer interaction interface.

[0042] Step 6: Compare the trainee's recognition results with standard signal parameters to generate evaluation results. The main control unit acquires the standard signal parameters corresponding to the current training subject, compares the trainee's identification results with the standard signal parameters, calculates the deviation value of each evaluation dimension, and generates the evaluation result based on the deviation value.

[0043] The aforementioned standard signal parameters are derived from the signal configuration parameters sent by the main control unit to the signal simulation unit in step 1, including standard signal type, standard signal frequency, standard signal bandwidth, and standard signal level.

[0044] The aforementioned evaluation dimensions include signal type identification accuracy, frequency deviation, bandwidth deviation, level deviation, and response delay. The evaluation results include quantitative scores for each evaluation dimension and a comprehensive score.

[0045] The aforementioned comparison process includes the following sub-steps: Step 601: Match the signal type identified by the trainee with the standard signal type. If the match is successful, the score for that dimension is full; if the match fails, the score for that dimension is zero. Step 602: Calculate the difference between the signal frequency identified by the trainee and the standard signal frequency to obtain the frequency deviation; calculate the difference between the signal bandwidth identified by the trainee and the standard signal bandwidth to obtain the bandwidth deviation; calculate the difference between the signal level identified by the trainee and the standard signal level to obtain the level deviation. Step 603: According to the preset scoring rules, the frequency deviation, bandwidth deviation, level deviation and response delay are mapped to the corresponding quantitative scores of the dimensions. Step 604: Weight the quantitative scores of each evaluation dimension and sum them to obtain a comprehensive score.

[0046] Step 7: Adjust training parameters based on evaluation results and form a closed-loop feedback. The main control unit sends the evaluation results to the training terminal for display, and based on the evaluation results and the preset training strategy, determines whether the training parameters need to be adjusted. If adjustment is required, the main control unit generates updated signal configuration parameters and interference configuration parameters, and returns to step 1.

[0047] The aforementioned training strategies may include: increasing the training difficulty if the overall score is higher than a preset upper threshold; decreasing the training difficulty if the overall score is lower than a preset lower threshold; and maintaining the current training difficulty if the overall score is between the upper and lower thresholds.

[0048] The aforementioned methods for increasing training difficulty include: increasing the power of electromagnetic interference signals, increasing the types and number of interference signals, reducing the output power of the UAV simulated signals, or increasing the number of simultaneous UAV signals. Methods for reducing training difficulty are the opposite of those for increasing it.

[0049] In this embodiment of the application, in order to achieve a smooth transition in training difficulty, the main control unit adopts a gradual adjustment strategy when adjusting training parameters, that is, the adjustment range does not exceed the preset adjustment step size each time, so as to avoid the drastic change in training difficulty from affecting the training effect.

[0050] This implementation adopts a backplane integrated chassis architecture, which tightly couples the main control unit, signal simulation unit, interference simulation unit, signal receiving unit, and high-precision timing unit through a backplane bus, replacing the traditional approach where each module is independently connected by external cables. Therefore, the system has a compact structure and reliable connection, which can meet the requirements of rapid relocation and high reliability for field mobile training, and solves the technical problems of poor platform integration and adaptability.

[0051] This implementation introduces a high-precision timing unit to provide a unified time and frequency reference for the signal simulation unit, interference simulation unit, and signal receiving unit, enabling the UAV simulated signal and electromagnetic interference signal to be strictly synchronized in the time and frequency domains. Therefore, it can accurately simulate the coherence characteristics of interference signals and target signals in the real environment, solving the technical problem of electromagnetic environment simulation distortion.

[0052] This implementation method, by configuring an independent interference simulation unit, can generate various types of electromagnetic interference signals such as white noise interference, narrowband interference, and frequency hopping interference, and synchronously superimpose them with the UAV simulated signals. This enables trainees to perform dynamic extraction and identification training of weak signals in complex electromagnetic interference environments, solving the technical problem of limited training methods.

[0053] This implementation method deploys an evaluation algorithm in the main control unit, which can automatically acquire standard signal parameters and trainee identification results, and quantitatively score multiple dimensions such as signal type identification accuracy, frequency deviation, bandwidth deviation, level deviation and response delay. Based on the evaluation results, the training parameters are dynamically adjusted to form a closed loop of training, evaluation and optimization, thus solving the technical problem of the lack of evaluation feedback mechanism.

[0054] The embodiments of the present invention have been described above. However, the embodiments are not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make more equivalent embodiments under the guidance of the present embodiments, and all of them are within the protection scope of the present embodiments.

Claims

1. A method for training unmanned aerial vehicle (UAV) signal recognition, characterized in that, Includes the following steps: The main control unit acquires training scenario data, generates signal configuration parameters and interference configuration parameters based on the training scenario data, and sends the signal configuration parameters to the signal simulation unit and the interference configuration parameters to the interference simulation unit through the backplane bus. The signal configuration parameters include the type of UAV signal to be simulated, carrier frequency, modulation method, power level and channel effect parameters, and the interference configuration parameters include the interference signal type, interference frequency band, interference power and interference timing parameters. The signal simulation unit generates simulated UAV signals through digital signal processing based on signal configuration parameters, and outputs the simulated UAV signals through the radio frequency port. The interference simulation unit generates electromagnetic interference signals through digital signal processing based on interference configuration parameters. The electromagnetic interference signals and the UAV simulation signals are combined by radio frequency to form a mixed radio frequency signal. The signal receiving unit receives the mixed radio frequency signal and performs down-conversion, analog-to-digital conversion and digital signal processing on the mixed radio frequency signal in sequence to obtain signal measurement data, which includes spectrum data, signal frequency, signal bandwidth and signal level. The training terminal displays spectrum data, receives signal recognition operations input by trainees based on the spectrum data, and obtains the trainees' recognition results. The trainees' recognition results include the signal type, signal frequency, signal bandwidth, and signal level identified by the trainees. The main control unit acquires the standard signal parameters corresponding to the current training subject, compares the trainee's identification results with the standard signal parameters, calculates the deviation value of each evaluation dimension, and generates the evaluation result based on the deviation value; The main control unit sends the evaluation results to the training terminal for display. Based on the evaluation results and the preset training strategy, it determines whether the training parameters need to be adjusted. If adjustment is required, the main control unit generates updated signal configuration parameters and interference configuration parameters and returns to the parameter distribution step.

2. The UAV signal recognition training method according to claim 1, characterized in that, Before issuing configuration parameters, the main control unit also sends a synchronization trigger command to the high-precision timing unit via the backplane bus. The high-precision timing unit responds to the command and distributes the unified reference clock signal to the signal simulation unit, interference simulation unit and signal receiving unit through the clock distributor, so that each unit is locked to the same time reference and frequency reference.

3. The UAV signal recognition training method according to claim 1, characterized in that, The steps for the signal simulation unit to generate simulated UAV signals include: The digital upconversion module inside the FPGA generates a digital baseband signal and performs upconversion processing based on the carrier frequency and modulation method in the signal configuration parameters to obtain a digital intermediate frequency signal. The digital intermediate frequency signal is converted into an analog intermediate frequency signal by a digital-to-analog converter; The analog intermediate frequency signal is converted to the target radio frequency band by the radio frequency upconversion module to obtain the drone's analog signal.

4. The UAV signal recognition training method according to claim 3, characterized in that, After generating the digital baseband signal, the signal simulation unit also performs channel effect superposition processing on the digital baseband signal inside the FPGA. The process is as follows: according to the channel effect parameters in the signal configuration parameters, Rayleigh fading or shadow fading is superimposed in the digital domain, and the multipath delay spread characteristics are simulated by dynamically adjusting the coefficients of the FIR filter to obtain the digital baseband signal after superposition of channel effects.

5. The UAV signal recognition training method according to claim 1, characterized in that, The signal simulation unit is equipped with a digitally controlled attenuator at the RF output end. According to the power level requirements in the signal configuration parameters, the attenuation of the digitally controlled attenuator is adjusted to control the output power of the UAV simulated signal.

6. The UAV signal recognition training method according to claim 1, characterized in that, Electromagnetic interference signals include one or more combinations of white noise interference, narrowband interference, and frequency hopping interference.

7. The UAV signal recognition training method according to claim 1, characterized in that, The training terminal records the first timestamp of the signal occurrence and the second timestamp of the trainee completing the recognition operation, and calculates the response delay based on the first and second timestamps.

8. The UAV signal recognition training method according to claim 1 or 7, characterized in that, The steps by which the main control unit compares the trainee's identification results with standard signal parameters include: The trainee identifies the signal type and matches it with the standard signal type. If the match is successful, the dimension is scored as full marks; if the match fails, the dimension is scored as zero marks. The frequency deviation is obtained by calculating the difference between the signal frequency identified by the trainee and the standard signal frequency; the bandwidth deviation is obtained by calculating the difference between the signal bandwidth identified by the trainee and the standard signal bandwidth; and the level deviation is obtained by calculating the difference between the signal level identified by the trainee and the standard signal level. According to the preset scoring rules, frequency deviation, bandwidth deviation, level deviation and response delay are mapped to the corresponding quantitative scores of the dimensions. The quantitative scores of each evaluation dimension are weighted and summed to obtain a comprehensive score.

9. The UAV signal recognition training method according to claim 8, characterized in that, The preset training strategies include: if the overall score is higher than the preset upper threshold, the training difficulty is increased; if the overall score is lower than the preset lower threshold, the training difficulty is decreased; if the overall score is between the upper and lower thresholds, the current training difficulty is maintained. Among them, the ways to increase the training difficulty include increasing the power of the electromagnetic interference signal, increasing the number of types of interference signals, reducing the output power of the drone simulation signal, or increasing the number of drone signals that appear simultaneously.

10. A drone signal recognition training platform, used to execute the drone signal recognition training method according to any one of claims 1 to 9, characterized in that, include: The main control unit is used to acquire training scenario data, generate signal configuration parameters and interference configuration parameters based on the training scenario data, send configuration parameters through the backplane bus, acquire standard signal parameters and trainee identification results for comparison to generate evaluation results, and adjust training parameters based on the evaluation results. The signal simulation unit is connected to the main control unit via a backplane bus and is used to generate simulated UAV signals based on signal configuration parameters. The interference simulation unit is connected to the main control unit via the backplane bus and is used to generate electromagnetic interference signals based on interference configuration parameters. The signal receiving unit is connected to the main control unit via the backplane bus. It is used to receive mixed radio frequency signals and perform signal processing to obtain signal measurement data. The high-precision timing unit is connected to the main control unit via the backplane bus and is used to provide a unified reference clock signal to the signal simulation unit, interference simulation unit and signal receiving unit. The training terminal is connected to the main control unit via Ethernet and is used to display spectrum data and receive signal recognition operations from trainees.

Images