Unmanned aerial vehicle illegal intrusion countering system and method based on electromagnetic spectrum detection

Abstract

The invention discloses an unmanned aerial vehicle illegal intrusion countering system and method based on electromagnetic spectrum detection, and relates to the technical field of unmanned aerial vehicle countering, and the system comprises an electromagnetic spectrum detection module, a spectrum analysis processing module, an intelligent decision module and a countering execution module. By constructing a distributed electromagnetic spectrum detection network, the comprehensiveness and the accuracy of signal acquisition are improved, a spectrum analysis algorithm is combined, spectrum features are deeply mined, illegal intrusion signals of the unmanned aerial vehicle are accurately identified, high-precision detection and rapid early warning of illegal intrusion behaviors of the unmanned aerial vehicle are realized, and an intelligent self-adaptive countering strategy is utilized, so that the safety of the unmanned aerial vehicle is improved. According to the real-time state information of the unmanned aerial vehicle, the intensity, frequency and mode of countering measures are automatically adjusted, the countering effect is improved, the working frequency band of surrounding legal electronic equipment can be dynamically avoided in the countering process, interference to the electronic equipment is greatly reduced, stability of the electromagnetic environment in the area is guaranteed, and the safety of the unmanned aerial vehicle is improved. And the capability of preventing and coping with illegal invasion of the unmanned aerial vehicle is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of drone countermeasures technology, specifically to a drone illegal intrusion countermeasure system and method based on electromagnetic spectrum detection. Background Technology

[0002] Unmanned aerial vehicles (UAVs), or unmanned aerial vehicles, are unmanned aircraft controlled by radio remote control devices or their own preset programs. In recent years, with the rapid development of technology, UAVs have been widely used in many fields, such as aerial photography and mapping, agricultural plant protection, and logistics distribution. With their advantages of high flexibility and low cost, they have brought great convenience to various industries. However, the widespread use of UAVs has also caused a series of security problems. Cases of UAVs illegally intruding into specific areas are not uncommon. UAVs are used to illegally spy on sensitive areas, steal important information, and even carry dangerous items to threaten people and facilities, seriously endangering public safety, information security, and the normal order of important places. Therefore, it is urgent to effectively counter the illegal intrusion of UAVs.

[0003] However, existing drone countermeasure technologies still have certain shortcomings in use. In the detection stage, facing complex electromagnetic environments, the detection accuracy is low, making it difficult to accurately identify illegally intruding drone signals, and misjudgments and omissions are prone to occur. In terms of countermeasures, the methods are relatively simple and lack the ability to flexibly adjust according to the real-time status of the drone and the surrounding environment. Moreover, in the process of countermeasures, interference is often generated with surrounding legitimate electronic devices, affecting their normal operation. Therefore, it is of great significance to develop a drone illegal intrusion countermeasure system and method based on electromagnetic spectrum detection. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a countermeasure system and method for illegal drone intrusion based on electromagnetic spectrum detection. By constructing a distributed electromagnetic spectrum detection network and combining advanced spectrum analysis algorithms and intelligent decision-making mechanisms, it can achieve high-precision detection and rapid early warning of illegal drone intrusion. Utilizing an intelligent adaptive countermeasure strategy, it automatically adjusts the strength, frequency, and method of countermeasures based on the real-time status information of the drone. This not only improves the countermeasure effect but also significantly reduces interference with surrounding legitimate electronic devices, ensuring the stability of the electromagnetic environment in the area and effectively enhancing the ability to prevent and respond to illegal drone intrusion.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an anti-unmanned aerial vehicle (UAV) intrusion countermeasure system based on electromagnetic spectrum detection, the system comprising: an electromagnetic spectrum detection module, a spectrum analysis and processing module, an intelligent decision-making module, and a countermeasure execution module;

[0006] The electromagnetic spectrum detection module consists of multiple distributed spectrum sensors, used to collect and transmit electromagnetic spectrum signals within the monitoring area;

[0007] The spectrum analysis and processing module receives the signal from the electromagnetic spectrum detection module, performs preprocessing and feature extraction, and compares it with a pre-established normal spectrum activity model to determine whether there is an abnormal signal. If so, it identifies whether it is an illegal intrusion signal from a drone.

[0008] The intelligent decision-making module receives the detection results from the spectrum analysis and processing module, obtains the real-time location information of the UAV, and selects a countermeasure strategy according to the preset strategy library.

[0009] The countermeasure execution module executes countermeasure operations using jamming devices based on instructions issued by the intelligent decision-making module.

[0010] Furthermore, the distributed spectrum sensors in the electromagnetic spectrum detection module are evenly distributed in the monitoring area. Each sensor has an independent shielding and filtering device, covering the common UAV communication and control frequency bands of 2.4GHz-5.8GHz, and has the function of automatically adjusting the acquisition frequency and gain.

[0011] Furthermore, in the preprocessing stage, the spectrum analysis and processing module uses wavelet transform algorithm to remove Gaussian noise and impulse noise from the signal, and then optimizes the signal quality through adaptive filtering algorithm. During feature extraction, it combines time-frequency analysis technology to obtain the energy distribution characteristics of the signal at different times and frequencies. When constructing a normal spectrum activity model, it uses dynamic clustering analysis and principal component analysis algorithms to update the model parameters according to the real-time changes in the electromagnetic environment of the monitoring area. When judging abnormal signals, it calculates the difference ΔH between the information entropy H of the real-time spectrum signal and the information entropy H0 of the normal spectrum signal, as well as the correlation coefficient r between the real-time spectrum and the normal spectrum. The abnormality index D is calculated by the formula D=ω1·|ΔH|+ω2·(1-r), where ω1 and ω2 are weight coefficients determined by machine learning training on historical electromagnetic spectrum data. The training objective is to minimize the misjudgment rate of D value when distinguishing between normal and abnormal signals. When D is greater than a set threshold, it is judged as an abnormal signal.

[0012] Furthermore, the intelligent decision-making module selects countermeasures based on a deep reinforcement learning algorithm, constructing a state space that includes the UAV's position, speed, communication protocol, spectrum characteristics, and frequency band information of surrounding legitimate electronic devices. A reward function R is defined. During decision-making, the intelligent decision-making module selects the countermeasure that maximizes the long-term cumulative reward based on the current state, while continuously updating the policy network. When adjusting the interference power, the formula is used. Where P old P represents the current interference power. new$P$ is the adjusted interference power, $d$ is the real-time distance between the UAV and the sensitive area, $d_0$ is the preset critical distance, and $\beta$ is the power adjustment coefficient.

[0013] Further, the countermeasure execution module integrates an adaptive noise jammer, an intelligent co-frequency jammer, and a precise signal spoofing device. The adaptive noise jammer automatically adjusts the interference power and noise type according to the UAV signal strength. The intelligent co-frequency jammer identifies the UAV communication frequency and generates a co-frequency interference signal in real time. The precise signal spoofing device sends a simulated control signal based on the analysis of the UAV control protocol. Each interference device has a power adjustment function with an adjustment accuracy of 0.1 dBm. When selecting the interference method, the formula $E = \gamma\cdot S$ jamming / S signal -δ·I legal is used, where $S$ jamming is the interference signal strength, $S$ signal is the UAV communication signal strength, $I$ legal is the interference degree on surrounding legal electronic devices, and $\gamma$ and $\delta$ are the interference effect and legal device interference influence coefficients. The interference method with the largest $E$ value is selected.

[0014] Further, the intelligent decision-making module dynamically adjusts the countermeasure strategy according to the distance between the UAV and the sensitive area. When the distance between the UAV and the sensitive area is greater than the safety distance $d_1$, a low-power interference and early warning strategy is adopted. When the distance is between $d_2$ and $d_1$, where $d_2 < d_1$, the interference power is gradually increased and the interference method is switched. When the distance is less than $d_2$, the maximum power interference and multiple interference methods work together. The distance judgment formula is where $(x, y)$ is the UAV coordinate and $(x_0, y_0)$ is the center coordinate of the sensitive area.

[0015] Further, when the spectrum analysis and processing module extracts the features of the collected signal, the formula is used, where $f$ k is the $k$-th spectrum feature value, $w$ k is the importance weight of this feature, which is dynamically adjusted according to historical data by machine learning algorithms.

[0016] The method for countering illegal UAV intrusion based on electromagnetic spectrum detection is applicable to the above-mentioned system for countering illegal UAV intrusion based on electromagnetic spectrum detection. This method includes the following steps:

[0017] Collect electromagnetic spectrum signals in the monitoring area through multiple distributed spectrum sensors;

[0018] Preprocess and extract features from the collected signals, compare them with the pre-established normal spectrum activity model to determine whether there are abnormal signals, and if so, identify whether they are illegal UAV intrusion signals;

[0019] Acquire the real-time location information of the drone and select a countermeasure strategy based on a preset strategy library;

[0020] Countermeasures are carried out using jamming equipment.

[0021] Furthermore, when establishing a normal spectrum activity model, electromagnetic spectrum data are collected according to different seasons, different time periods of the day, and different weather conditions. Cluster analysis is performed on each type of data to determine the spectrum characteristic patterns of different categories. Principal component analysis algorithm is used to reduce the dimensionality of the data, extract key features to build the model, and when updating the model parameters, the update step size is adaptively adjusted according to the degree of difference between the newly collected data and the original model.

[0022] Compared with existing technologies, this anti-drone intrusion system and method based on electromagnetic spectrum detection has the following advantages:

[0023] This invention constructs a distributed electromagnetic spectrum detection network to achieve multi-node collaborative acquisition of electromagnetic spectrum signals, improving the comprehensiveness and accuracy of signal acquisition. Combined with spectrum analysis algorithms, it deeply mines spectrum features to accurately identify unauthorized drone intrusion signals, achieving high-precision detection and rapid early warning of unauthorized drone intrusion behavior. Utilizing an intelligent adaptive countermeasure strategy, it automatically adjusts the strength, frequency, and method of countermeasures based on the real-time status information of the drone. This not only improves the countermeasure effect but also dynamically avoids the operating frequency bands of nearby legitimate electronic devices during the countermeasure process, significantly reducing interference and ensuring the stability of the electromagnetic environment in the area. This effectively enhances the ability to prevent and respond to unauthorized drone intrusion.

[0024] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0026] Figure 1 This is a schematic diagram of a drone intrusion countermeasure system based on electromagnetic spectrum detection.

[0027] Figure 2 This is a flowchart illustrating a method for countering unauthorized drone intrusion based on electromagnetic spectrum detection. Detailed Implementation

[0028] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0029] Example 1

[0030] In a highly confidential research park setting, which stores a large amount of cutting-edge research results and key technical data, the security level is extremely high, and unauthorized drones are strictly prohibited from entering. To ensure the security of the park, the drone intrusion countermeasure system based on electromagnetic spectrum detection of this invention has been deployed.

[0031] After the system is started, the electromagnetic spectrum detection module begins to work. Its multiple distributed spectrum sensors are evenly distributed around and inside key locations of the research park. These sensors are equipped with independent shielding and filtering devices, which can effectively resist external interference and stably cover common UAV communication and control frequency bands such as 2.4GHz-5.8GHz. The sensors can automatically adjust the acquisition frequency and gain, and flexibly optimize the acquisition parameters according to the complexity of the surrounding electromagnetic environment to ensure that the acquired electromagnetic spectrum signals are complete and accurate. The acquired data is transmitted to the spectrum analysis and processing module in real time through a high-speed and stable network.

[0032] Upon receiving the signal, the spectrum analysis and processing module immediately begins preprocessing. Using wavelet transform algorithms, it selectively filters Gaussian and impulse noise. Subsequently, an adaptive filtering algorithm further optimizes the signal quality, enhancing its stability and reliability. In the feature extraction stage, time-frequency analysis techniques are combined to comprehensively analyze the signal from different time and frequency dimensions, obtaining its energy distribution characteristics. When constructing a normal spectrum activity model, dynamic clustering and principal component analysis algorithms are employed to continuously track and monitor real-time changes in the regional electromagnetic environment and update the model parameters accordingly, ensuring the model accurately reflects the normal electromagnetic spectrum activity state.

[0033] During the process of judging abnormal signals, the spectrum analysis and processing module calculates the difference ΔH between the information entropy H of the real-time spectrum signal and the information entropy H0 of the normal spectrum signal, as well as the correlation coefficient r between the real-time spectrum and the normal spectrum. The abnormality index D is calculated using the formula D=ω1·ΔH|+ω2·(1-r), where ω1 and ω2 are determined by machine learning training on a large amount of historical electromagnetic spectrum data. The training goal is to minimize the misjudgment rate of D value when distinguishing between normal and abnormal signals. Once D is greater than the set threshold, the system judges that an abnormal signal has occurred.

[0034] Next, by comparing with the pre-established UAV spectrum feature library, it is identified whether the abnormal signal is an illegal intrusion signal of a UAV. If it is confirmed to be an illegal intrusion signal of a UAV, the spectrum analysis and processing module quickly transmits relevant information, such as the approximate spectrum features of the UAV, possible intrusion directions, etc., to the intelligent decision-making module.

[0035] After receiving the information, the intelligent decision-making module combines the real-time position information of the UAV obtained from the positioning system, and based on the deep reinforcement learning algorithm, selects the most appropriate countermeasure strategy from the preset strategy library. The positioning system accurately obtains the coordinates (x, y) of the UAV through high-precision satellite positioning technology and positioning base stations inside the park. The intelligent decision-making module calculates the distance d between the UAV and the central coordinates (x0, y0) of the scientific research park (sensitive area) according to the formula When d is greater than the safety distance d1, the intelligent decision-making module issues an instruction to the countermeasure execution module, adopting a low-power interference and warning strategy. By sending a mild interference signal in a specific frequency band, the UAV is warned, and at the same time, the internal warning mechanism of the park is activated to notify the security personnel to be vigilant.

[0036] When d is between d2 and d1 (d2 < d1), the intelligent decision-making module gradually increases the interference power according to the formula

[0037] and switches the interference method according to the communication protocol and spectrum characteristics of the UAV. For example, if the UAV uses the WiFi communication protocol, the intelligent decision-making module controls the intelligent co-frequency jammer in the countermeasure execution module to send a co-frequency high-intensity interference signal to disrupt the communication link of the UAV; if the UAV has specific narrowband spectrum characteristics, an adaptive noise jammer is enabled to emit targeted narrowband noise interference signals.

[0038] For the selection of the interference method, the intelligent decision-making module evaluates through the formula E = γ·S jamming / S signal legal -δ·I legal where S jamming is the interference signal strength, S signal is the UAV communication signal strength, I legal is the interference degree on surrounding legal electronic devices, γ and δ are the interference effect and legal device interference influence coefficients. The intelligent decision-making module preferentially selects the interference method with the largest E value to ensure that while effectively countering the UAV, the interference to surrounding legal electronic devices is minimized. When d is less than d2, the intelligent decision-making module activates an emergency countermeasure strategy of maximum power interference and coordinated operation of multiple interference methods to ensure that the UAV cannot approach the scientific research park.

[0039] Upon receiving instructions from the intelligent decision-making module, the countermeasure execution module responds swiftly. Its integrated adaptive noise jammer automatically adjusts the jamming power and noise type based on the drone's signal strength, effectively interfering with the drone's communication and control signals. The intelligent co-frequency jammer accurately identifies the drone's communication frequency and generates co-frequency jamming signals in real time, blocking the communication link between the drone and its operator. The precise signal decoy, based on in-depth analysis of the drone's control protocol, sends simulated control signals to guide the drone away from its intrusion route, or even cause it to land in a safe area. Each jamming device has a power adjustment function with an adjustment accuracy of up to 0.1dBm, enabling precise adjustment of the jamming power according to the instructions of the intelligent decision-making module, achieving refined countermeasures.

[0040] In summary, this embodiment, through a complete and closely collaborative system process and technical solution, achieves comprehensive monitoring and efficient and precise countermeasures against illegal drone intrusion around the research park. High-precision electromagnetic spectrum detection and analysis technology ensures timely and accurate detection of illegally intruding drones. The intelligent and adaptive countermeasure strategy flexibly adjusts countermeasures based on the drone's real-time status and the surrounding environment, effectively preventing drone intrusions and minimizing the impact on legitimate electronic devices in the vicinity. This effectively safeguards the information security and normal research order of the research park and maintains the stability of the electromagnetic environment around the park.

[0041] Example 2

[0042] In a venue hosting a large international sporting event, where a large number of spectators, athletes, and media personnel gather, and where there are numerous communication devices and a complex electromagnetic environment, in order to ensure the smooth running of the event and prevent security risks and interference with live broadcasts caused by unauthorized drone intrusion, the unauthorized drone intrusion countermeasure system based on electromagnetic spectrum detection of this invention was deployed.

[0043] Once the system is powered on, the electromagnetic spectrum detection module enters its working state. Its distributed spectrum sensors are arranged around the stadium, evenly distributed at various key points, such as the four corners of the stadium and surrounding high points. With independent shielding and filtering devices, these sensors effectively filter out the noise interference generated by a large number of communication devices on site, stably covering the communication frequency bands of common UAVs. At the same time, they can automatically adjust the acquisition frequency and gain in real time according to the electromagnetic environment on site, ensuring comprehensive and accurate acquisition of electromagnetic spectrum signals. The acquired data is quickly transmitted to the spectrum analysis and processing module through a dedicated high-speed communication line.

[0044] After receiving the signal, the spectrum analysis and processing module immediately performs preprocessing. Using the wavelet transform algorithm, various types of noise in the signal are removed. Subsequently, the adaptive filtering algorithm is used to further optimize the signal quality and highlight the characteristics of the effective signal. In the feature extraction link, combining time-frequency analysis technology, detailed energy distribution characteristics of the signal are obtained from different time and frequency dimensions. When constructing the normal spectrum activity model, dynamic clustering analysis and principal component analysis algorithms are adopted, and the model parameters are updated in real time according to the changing electromagnetic environment at the event site to accurately reflect the spectrum activity under normal conditions.

[0045] When judging abnormal signals, the spectrum analysis and processing module calculates the difference ΔH between the information entropy H of the real-time spectrum signal and the information entropy H0 of the normal spectrum signal, as well as the correlation coefficient r between the real-time spectrum and the normal spectrum. The abnormal degree index D is calculated through the formula D = ω1·|ΔH| + ω2·(1 - r), where ω1 and ω2 are determined through machine learning training based on a large amount of historical spectrum data at the event site and similar complex electromagnetic environments, aiming to minimize the misjudgment rate when distinguishing normal and abnormal signals. When D is greater than the set threshold, it is determined that an abnormal signal has occurred.

[0046] After that, it is compared with the pre-constructed and continuously updated UAV spectrum feature library to determine whether the abnormal signal is an illegal UAV intrusion signal. Once it is confirmed that it is an illegal UAV intrusion signal, the spectrum analysis and processing module quickly transmits data such as signal characteristics and preliminary positioning information to the intelligent decision-making module.

[0047] After receiving the information, the intelligent decision-making module combines the real-time position information of the UAV obtained through the high-precision positioning system in the venue, and selects a suitable countermeasure strategy from the preset strategy library based on the deep reinforcement learning algorithm. The positioning system uses multiple positioning base stations in the venue, combines satellite positioning information, and accurately obtains the coordinates (x, y) of the UAV. The intelligent decision-making module is based on the formula to calculate the distance d between the coordinates (x0, y0) of the UAV and the center of the sports venue (center of the sensitive area).

[0048] When d is greater than the safety distance d1, the intelligent decision-making module instructs the countermeasure execution module to adopt low-power interference and warning strategies. On the one hand, low-intensity interference signals are sent through a specific frequency band to warn the UAV to stay away; on the other hand, a warning message is sent to the venue security command center to prompt security personnel to pay attention to the corresponding area.

[0049] When d is between d2 and d1 (d2 < d1), the intelligent decision-making module is based on the formula The interference power is gradually increased, and an appropriate interference method is selected according to the communication protocol and spectrum characteristics of the UAV. For example, if the UAV uses the Bluetooth communication protocol, the intelligent decision module controls the intelligent co-channel jammer in the countermeasure execution module to emit interference signals targeting the Bluetooth frequency band to disrupt the UAV's communication. If the UAV's spectrum shows broadband communication characteristics, the adaptive noise jammer is activated to emit broadband noise interference signals.

[0050] Regarding the selection of interference methods, the intelligent decision-making module uses the formula E = γ·S jamming / S signal -δ·I legal An evaluation was conducted, in which S jamming For the interference signal strength, S signal For the communication signal strength of the drone, I legal To determine the degree of interference to surrounding legitimate electronic devices, γ and δ represent the interference effect and the interference impact coefficient of legitimate devices. The intelligent decision-making module selects the interference method with the largest E value to minimize interference to legitimate electronic devices such as live broadcast equipment and audience communication equipment while ensuring effective countermeasures against drones. When d is less than d2, the intelligent decision-making module activates an emergency strategy that combines maximum power interference with multiple interference methods to ensure that drones cannot approach the event venue.

[0051] Upon receiving instructions from the intelligent decision-making module, the countermeasure execution module swiftly initiates countermeasures. Its integrated adaptive noise jammer automatically adjusts the jamming power and noise type based on the drone's signal strength, effectively interfering with the drone's communication and control signals. The intelligent co-channel jammer accurately identifies the drone's communication frequency and generates co-channel jamming signals in real time, severing the connection between the drone and its operator. The precise signal decoy sends simulated control signals based on the analysis of the drone's control protocol, guiding the drone to a safe area or causing it to land. Each jamming device possesses precise power adjustment capabilities, with an adjustment accuracy of up to 0.1 dBm, and can strictly adjust the jamming power according to the instructions of the intelligent decision-making module, achieving precise countermeasures.

[0052] In summary, in this embodiment, through the close cooperation of various modules of the system, the complete technical solution is effectively implemented. From precise electromagnetic spectrum detection to intelligent analysis and decision-making, and then to efficient countermeasure execution, the system achieves full-process monitoring and powerful countermeasures against illegal drone intrusion around sports event venues. The high-precision detection technology can quickly detect illegally intruding drones in complex electromagnetic environments, and the intelligent adaptive countermeasure strategy ensures that while maintaining the safety of the event, it minimizes interference with various legitimate electronic devices on site, thus guaranteeing the smooth progress of the event and the stability and order of the on-site electromagnetic environment.

[0053] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A drone intrusion countermeasure system based on electromagnetic spectrum detection, characterized in that, The system includes: an electromagnetic spectrum detection module, a spectrum analysis and processing module, an intelligent decision-making module, and a countermeasure execution module; The electromagnetic spectrum detection module consists of multiple distributed spectrum sensors, used to collect and transmit electromagnetic spectrum signals within the monitoring area; The spectrum analysis and processing module receives the signal from the electromagnetic spectrum detection module, performs preprocessing and feature extraction, and compares it with a pre-established normal spectrum activity model to determine whether there is an abnormal signal. If so, it identifies whether it is an illegal intrusion signal from a drone. The intelligent decision-making module receives the detection results from the spectrum analysis and processing module, obtains the real-time location information of the UAV, and selects a countermeasure strategy according to the preset strategy library. The countermeasure execution module executes countermeasure operations using jamming devices based on instructions issued by the intelligent decision-making module.

2. The anti-drone intrusion system based on electromagnetic spectrum detection according to claim 1, characterized in that, The distributed spectrum sensors in the electromagnetic spectrum detection module are evenly distributed in the monitoring area. Each sensor has an independent shielding and filtering device, covering the common UAV communication and control frequency bands of 2.4GHz-5.8GHz, and has the function of automatically adjusting the acquisition frequency and gain.

3. The anti-drone intrusion system based on electromagnetic spectrum detection according to claim 1, characterized in that, In the preprocessing stage, the spectrum analysis and processing module uses wavelet transform algorithm to remove Gaussian noise and impulse noise from the signal, and then optimizes the signal quality through adaptive filtering algorithm. During feature extraction, it combines time-frequency analysis technology to obtain the energy distribution characteristics of the signal at different times and frequencies. When constructing a normal spectrum activity model, it uses dynamic clustering analysis and principal component analysis algorithms to update the model parameters according to the real-time changes in the electromagnetic environment of the monitoring area. When judging abnormal signals, it calculates the difference ΔH between the information entropy H of the real-time spectrum signal and the information entropy H0 of the normal spectrum signal, as well as the correlation coefficient r between the real-time spectrum and the normal spectrum. The abnormality index D is calculated by the formula D=ω1·|ΔH|+ω2·(1-r), where ω1 and ω2 are weight coefficients determined by machine learning training on historical electromagnetic spectrum data. The training objective is to minimize the misjudgment rate of D value when distinguishing between normal and abnormal signals. When D is greater than a set threshold, it is judged as an abnormal signal.

4. The anti-drone intrusion system based on electromagnetic spectrum detection according to claim 1, characterized in that, The intelligent decision-making module selects countermeasures based on a deep reinforcement learning algorithm. It constructs a state space containing the UAV's position, speed, communication protocol, spectral characteristics, and frequency band information of surrounding legitimate electronic devices, and defines a reward function R. During decision-making, the intelligent decision-making module selects the countermeasure that maximizes the long-term cumulative reward based on the current state, while continuously updating the policy network. When adjusting the interference power, it uses the formula... Where P old P represents the current interference power. new d is the adjusted interference power, d is the real-time distance between the UAV and the sensitive area, d0 is the preset critical distance, and β is the power adjustment coefficient.

5. The anti-drone intrusion system based on electromagnetic spectrum detection according to claim 1, characterized in that, The countermeasure execution module integrates an adaptive noise jammer, an intelligent co-channel jammer, and a precise signal decoy. The adaptive noise jammer automatically adjusts the jamming power and noise type based on the UAV signal strength. The intelligent co-channel jammer identifies the UAV's communication frequency and generates a co-channel jamming signal in real time. The precise signal decoy sends simulated control signals based on the analysis of the UAV's control protocol. Each jamming device has a power adjustment function with an adjustment accuracy of 0.1 dBm. When selecting the jamming method, the formula E = γ·S is used. jamming / S signal -δ·I legal S jamming For the interference signal strength, S signal For the communication signal strength of the drone, I legal To determine the degree of interference to surrounding legitimate electronic devices, γ and δ are the interference effect and the interference impact coefficient of legitimate devices, respectively. The interference method with the largest E value is selected.

6. The anti-drone intrusion system based on electromagnetic spectrum detection according to claim 1, characterized in that, The intelligent decision-making module dynamically adjusts the countermeasure strategy according to the distance between the UAV and the sensitive area. When the distance between the UAV and the sensitive area is greater than the safety distance d1, a low-power interference and warning strategy is adopted. When the distance is between d2 and d1, where d2 < d1, the interference power is gradually increased and the interference mode is switched. When the distance is less than d2, the maximum-power interference and multiple interference modes work together. The distance judgment formula is where (x,y) are the coordinates of the UAV and (x0,y0) are the coordinates of the center of the sensitive area.

7. The anti-drone intrusion system based on electromagnetic spectrum detection according to claim 3, characterized in that, When the spectrum analysis and processing module extracts features from the acquired signal, it uses the formula... Where f k For the k-th spectral feature value, w k The importance weight of this feature is dynamically adjusted based on historical data using a machine learning algorithm.

8. A method for countering unmanned aerial vehicle (UAV) intrusion based on electromagnetic spectrum detection, applicable to the UAV intrusion countermeasure system based on electromagnetic spectrum detection as described in claims 1-7, characterized in that, The method includes the following steps: Electromagnetic spectrum signals within the monitoring area are collected using multiple distributed spectrum sensors; The collected signals are preprocessed and feature extracted, and compared with a pre-established normal spectrum activity model to determine whether there are abnormal signals. If so, it is identified whether they are signals of illegal intrusion by drones. Acquire the real-time location information of the drone and select a countermeasure strategy based on a preset strategy library; Countermeasures are carried out using jamming equipment.

9. The method for countering unauthorized intrusion by drones based on electromagnetic spectrum detection according to claim 8, characterized in that, When establishing a normal spectrum activity model, electromagnetic spectrum data are collected according to different seasons, different time periods of day, and different weather conditions. Cluster analysis is performed on each type of data to determine the spectrum characteristic patterns of different categories. Principal component analysis algorithm is used to reduce the dimensionality of the data, extract key features to build the model, and when updating the model parameters, the update step size is adaptively adjusted according to the degree of difference between the newly collected data and the original model.

Images