Recording shielding method and device based on ultrasonic interference

Abstract

The present invention discloses a recording shielding method and device based on ultrasonic interference, wherein the method comprises: detecting and identifying recording devices in a target area and obtaining type information of the recording devices; configuring a first ultrasonic interference signal and a second ultrasonic interference signal based on the type information; and performing recording interference on the target area according to the first ultrasonic interference signal and the second ultrasonic interference signal. This application solves the technical problem in the prior art of lacking a precise shielding method for different types of recording devices, resulting in poor recording interference effects, and achieves the technical effect of customizing ultrasonic interference signals according to the type of recording device, thereby achieving efficient and precise shielding of recording content.

Description

Technical Field

[0001] The present invention relates to the field of recording shielding, and in particular to a recording shielding method and device based on ultrasonic interference. Background Art

[0002] With the continuous advancement of information technology, recording equipment has become widely used in various fields. However, in some situations where confidentiality is crucial, such as business negotiations and confidential meetings, unauthorized recording can lead to information leaks, resulting in serious security risks and financial losses. Therefore, recording shielding is necessary in these situations. Currently, common recording shielding techniques on the market include white noise interference, acoustic wave cancellation, and electromagnetic interference. However, most of these methods employ a unified interference strategy and fail to provide targeted shielding based on the characteristics of different types of recording equipment. Furthermore, due to significant differences in sound pickup principles, frequency response ranges, and signal processing mechanisms among various recording devices, universal interference methods often struggle to effectively shield all recording devices. For example, interference signals targeting traditional analog recording devices may have limited effectiveness against digital recording devices; interference methods optimized for a specific frequency range may be ineffective against broadband recording devices. Consequently, existing technologies lack precise shielding methods tailored to different types of recording devices, resulting in poor recording shielding effectiveness. Summary of the Invention

[0003] This application provides a recording shielding method and device based on ultrasonic interference, aiming to solve the technical problem in the prior art of lacking precise shielding methods for different types of recording equipment, resulting in poor recording interference effects.

[0004] The first aspect disclosed in the present application provides a recording shielding method based on ultrasonic interference, the method comprising: detecting and identifying the recording device in the target area, and obtaining type information of the recording device; configuring a first ultrasonic interference signal and a second ultrasonic interference signal based on the type information, wherein the first ultrasonic interference signal is a signal that is inaudible to human ears but can be captured by the recording device, and the second ultrasonic interference signal is a signal for generating a nonlinear mixing effect inside the recording device; and performing recording interference on the target area according to the first ultrasonic interference signal and the second ultrasonic interference signal.

[0005] Optionally, the detecting and identifying of the recording device in the target area and obtaining the type information of the recording device include: scanning the electromagnetic signal in the target area through a radio frequency scanner; performing spectrum analysis on the obtained electromagnetic signal to extract multiple signal features, wherein the multiple signal features include frequency features, intensity features and modulation features; performing device matching in a preset recording device feature database based on the frequency features, the intensity features and the modulation features, and determining the type information of the recording device based on the matching results.

[0006] Optionally, performing device matching in a preset recording device feature database based on the frequency feature, the intensity feature and the modulation feature, and determining the type information of the recording device based on the matching result, includes: obtaining a device feature matching formula; determining a first sample device in the preset recording device feature database, and extracting a first frequency feature, a first intensity feature and a first modulation feature of the first sample device; based on the device feature matching formula, obtaining a first device matching degree between the recording device and the first sample device according to the frequency feature, the intensity feature and the modulation feature of the recording device, and the first frequency feature, the first intensity feature and the first modulation feature of the first sample device; according to the method of obtaining the first device matching degree, continuing to obtain the device matching degrees of other sample devices in the preset recording device feature database with the recording device to obtain multiple device matching degrees; selecting the maximum device matching degree among the multiple device matching degrees as the matching sample device; and extracting the type information of the matching sample device in the preset recording device feature database as the type information of the recording device.

[0007] Optionally, the device feature matching formula is:

[0008]

[0009] in, Indicates the device matching degree, represents the frequency characteristics, represents the first frequency characteristic, Indicates the similarity between the frequency feature and the first frequency feature, Indicates the intensity characteristics, represents the first intensity feature, Indicates the similarity between the intensity feature and the first intensity feature, represents the modulation characteristics, represents the first modulation characteristic, represents the similarity between the modulation feature and the first modulation feature, is the weight coefficient, and .

[0010] Optionally, configuring the first ultrasonic interference signal and the second ultrasonic interference signal based on the type information includes: constructing an interference signal configuration model, the interference signal configuration model including a first signal configuration channel and a second signal configuration channel; inputting the type information into the first signal configuration channel to obtain the first ultrasonic interference signal; inputting the type information into the second signal configuration channel to obtain the second ultrasonic interference signal.

[0011] Optionally, an interference signal configuration model is constructed, wherein the interference signal configuration model includes a first signal configuration channel and a second signal configuration channel, including: obtaining experimental interference data of multiple sample recording devices, wherein the experimental interference data includes recording interference data and device interference data; establishing a first correspondence library between recording device types and first ultrasonic interference parameters based on the recording interference data of multiple sample recording devices; establishing a second correspondence library between recording device types and second ultrasonic interference parameters based on the device interference data of multiple sample recording devices; training a first deep learning network as the first signal configuration channel according to the first correspondence library; training a second deep learning network as the second signal configuration channel according to the second correspondence library; and combining the first deep learning network with the second deep learning network to form the interference signal configuration model.

[0012] Optionally, the recording interference of the target area according to the first ultrasonic interference signal and the second ultrasonic interference signal includes: configuring a first interference duration interval of the first ultrasonic interference signal and a second interference duration interval of the second ultrasonic interference signal; randomly selecting a first execution duration from the first interference duration interval; randomly selecting a second execution duration from the second interference duration interval; after executing the first ultrasonic interference signal according to the first execution duration, executing the second ultrasonic interference signal according to the second execution duration; repeating the random selection of durations and execution of interference signals to continuously perform recording interference on the target area.

[0013] Another aspect disclosed in the present application provides a recording shielding device based on ultrasonic interference, which includes: a type information acquisition unit, used to detect and identify the recording equipment in the target area, and obtain the type information of the recording equipment; an interference signal configuration unit, used to configure a first ultrasonic interference signal and a second ultrasonic interference signal based on the type information, wherein the first ultrasonic interference signal is a signal that is inaudible to human ears but can be captured by the recording equipment, and the second ultrasonic interference signal is a signal used to generate a nonlinear mixing effect inside the recording equipment; a recording interference execution unit, used to perform recording interference on the target area according to the first ultrasonic interference signal and the second ultrasonic interference signal.

[0014] The present invention proposes a recording shielding method and device based on ultrasonic interference. First, the recording device in the target area is detected and identified, the type information of the recording device is obtained, the specific object to be interfered with and its characteristics are determined, and basic data is provided for the subsequent targeted configuration of the interference signal; secondly, based on the obtained recording device type information, a first ultrasonic interference signal and a second ultrasonic interference signal are configured, wherein the first ultrasonic interference signal is inaudible to the human ear but can be captured by the recording device, thereby ensuring that the interference process does not affect normal interpersonal communication; the second ultrasonic interference signal is a signal used to generate a nonlinear mixing effect inside the recording device, which interacts with the first signal inside the recording device to produce a more effective interference effect; then, the target area is interfered with by the configured first ultrasonic interference signal and the second ultrasonic interference signal, so that the recording device cannot normally capture effective sound information, thereby achieving recording shielding of the target area and improving the effect of recording interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flow chart of a recording shielding method based on ultrasonic interference is provided for an embodiment of the present application;

[0016] Figure 2 A structural schematic diagram of a recording shielding device based on ultrasonic interference is provided for an embodiment of the present application.

[0017] Description of reference numerals:

[0018] 11. Type information acquisition unit; 12. Interference signal configuration unit; 13. Recording interference execution unit. DETAILED DESCRIPTION

[0019] The overall idea of ​​the technical solution provided by this application is as follows:

[0020] The embodiments of the present application provide a recording shielding method and device based on ultrasonic interference, which achieves efficient shielding of recording content by accurately identifying the type of recording equipment and configuring a customized ultrasonic interference signal according to its characteristics.

[0021] Specifically, first, by detecting and identifying the recording devices in the target area, their type information is obtained, laying the foundation for subsequent precise interference; second, based on the obtained recording device type information, two ultrasonic interference signals are configured in a targeted manner, namely, a first ultrasonic interference signal that is inaudible to the human ear but can be captured by the recording device, and a second ultrasonic interference signal used to produce a nonlinear mixing effect inside the recording device; then, the configured ultrasonic interference signal is used to perform precise recording interference on the target area, making it impossible for the recording device to record valid information normally.

[0022] The above technical solution solves the technical problem in the prior art of lacking precise shielding methods for different types of recording equipment, resulting in poor recording interference effects, and provides a more effective and precise recording shielding solution for various occasions requiring confidentiality.

[0023] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced in conjunction with the drawings in the specification.

[0024] Example 1, as Figure 1 As shown, an embodiment of the present application provides a recording shielding method based on ultrasonic interference, the method comprising:

[0025] S100: Detect and identify recording devices in a target area, and obtain type information of the recording devices.

[0026] Specifically, first, determine the recording devices present in the target area and their specific type information. In actual application scenarios, different types of recording devices have different hardware structures, signal processing characteristics, and frequency response characteristics. Therefore, accurately identifying the type of recording device is crucial for the subsequent configuration of targeted interference signals. Specifically, the target area is scanned by using specific detection methods to detect possible recording devices. Among them, the detection methods may include but are not limited to electromagnetic signal detection, radio spectrum analysis, nonlinear node detection and other technologies. When a recording device is detected in the target area, the acquired signal characteristics are further analyzed and compared with the preset recording device feature database to identify the specific type of recording device, such as the recording function of a smartphone, a professional voice recorder, a concealed recording device, etc.

[0027] By detecting and identifying the recording equipment in the target area, the necessary parameter basis is provided for the subsequent configuration of targeted ultrasonic interference signals, ensuring the accuracy and effectiveness of the interference effect.

[0028] S200: Based on the type information, configure a first ultrasonic interference signal and a second ultrasonic interference signal, wherein the first ultrasonic interference signal is a signal that is inaudible to human ears but can be captured by the recording device, and the second ultrasonic interference signal is a signal used to generate a nonlinear mixing effect inside the recording device.

[0029] Specifically, after obtaining the type information of the recording device, a targeted ultrasonic interference signal is customized and configured, including a first ultrasonic interference signal and a second ultrasonic interference signal. By configuring two ultrasonic interference signals with different functions, a dual interference effect on the recording device is achieved.

[0030] The frequency range of the first ultrasonic interference signal is set to a frequency band that is imperceptible to the human ear but can be captured by the recording device. Although this signal has no effect on the human auditory system, it can be captured and recorded by the recording device's sensors, introducing a large amount of noise interference into the recorded content. The second ultrasonic interference signal exploits the nonlinear characteristics of the recording device's internal microphone elements and amplifier circuits and is designed to produce a nonlinear mixing effect within the recording device. When two or more ultrasonic signals of different frequencies act simultaneously on the recording device, due to the device's inherent nonlinear characteristics, these signals mix internally, generating new difference and sum frequency components, further interfering with the recorded content.

[0031] For example, a table lookup configuration method is used to achieve precise configuration of the interference signal. Specifically, a mapping database of recording device types and optimal interference signal parameters is pre-established. The database stores interference parameter sets corresponding to various common recording devices, including but not limited to the frequency, amplitude, and waveform parameters of the first ultrasonic interference signal, and the frequency combination, phase relationship, and modulation method of the second ultrasonic interference signal. After the type of recording device in the target area is identified, the mapping database is directly queried to quickly extract the optimal interference parameter set corresponding to the type of device, and the first and second ultrasonic interference signals are configured accordingly. This table lookup configuration method has the characteristics of fast response speed and high efficiency, and is particularly suitable for real-time interference scenarios of known types of recording devices.

[0032] By configuring the first ultrasonic interference signal and the second ultrasonic interference signal, the most effective ultrasonic interference signal combination can be generated for a specific type of recording equipment, laying the foundation for subsequent interference implementation, thereby achieving precise shielding for different recording equipment.

[0033] S300: Performing recording interference on the target area according to the first ultrasonic interference signal and the second ultrasonic interference signal.

[0034] Specifically, after configuring the first ultrasonic interference signal and the second ultrasonic interference signal, the two configured ultrasonic interference signals are actually transmitted to the target area, thereby effectively interfering with the recording equipment in the target area. During the specific implementation process, the configured first ultrasonic interference signal and the second ultrasonic interference signal are first converted into sound wave energy through an ultrasonic transducer array. The ultrasonic transducer array is usually composed of multiple precisely calibrated ultrasonic transmitting units, which can be independently controlled and work together to ensure that the interference signal can accurately cover the target area.

[0035] To improve the jamming effect and reduce the likelihood of circumvention, intelligent scheduling strategies can be used to control the transmission order, duration, and intensity of the two jamming signals. For example, the two jamming signals can be transmitted alternately or simultaneously depending on the characteristics of the recording equipment, creating a composite jamming effect. Furthermore, the directionality and energy distribution of the jamming signal can be adjusted in real time based on factors such as the spatial characteristics of the target area and the ambient noise level to maximize the jamming effect. Due to the strong directionality and attenuation characteristics of ultrasound, the formation and projection of the ultrasonic beam are precisely controlled during jamming, concentrating the jamming signal on the target area while minimizing the impact on non-target areas, achieving precise jamming.

[0036] By performing recording interference based on the first ultrasonic interference signal and the second ultrasonic interference signal, it is possible to effectively interfere with the recording equipment in the target area, making it impossible for it to normally record clear and discernible sound content, thereby achieving safe shielding of the recording content and protecting sensitive information from being obtained through illegal recording.

[0037] Furthermore, the detecting and identifying the recording device in the target area and obtaining the type information of the recording device includes:

[0038] S110: Scanning electromagnetic signals in the target area using a radio frequency scanner;

[0039] S120: performing spectrum analysis on the acquired electromagnetic signal to extract multiple signal features, where the multiple signal features include frequency features, intensity features, and modulation features;

[0040] S130: performing device matching in a preset recording device feature database according to the frequency feature, the intensity feature, and the modulation feature, and determining type information of the recording device according to a matching result.

[0041] Specifically, when obtaining the type information of the recording device, first, a dedicated RF scanner is used to scan the target area for full-band electromagnetic signals. The RF scanner can be configured to cover the frequency range of 300MHz to 6GHz, which includes the operating frequency bands of most recording devices. During the scanning process, the RF scanner continuously samples the electromagnetic environment in the target area at a preset frequency step size to obtain raw data such as electromagnetic signal strength and phase at different frequency points. To improve scanning efficiency and detection sensitivity, the RF scanner can use a directional antenna array to perform fine grid detection of the target area through beam scanning technology to ensure that no possible electromagnetic signal sources are missed.

[0042] The acquired electromagnetic signal is then subjected to in-depth signal processing and feature extraction to extract multiple signal features. The time-domain signal is converted to a frequency-domain representation using a fast Fourier transform (FFT), yielding a map of the electromagnetic spectrum within the target area. This map is then used to extract frequency, intensity, and modulation features. Frequency features include parameters such as the signal's center frequency, bandwidth, frequency offset, harmonic components, and their proportions. These parameters reflect the characteristics of the recording device's RF oscillator and modulation circuit. Intensity features include parameters such as the signal's power spectral density, peak intensity, average intensity, and its time-varying characteristics. These parameters reflect the device's transmission power and distance. Modulation features include parameters such as the signal's modulation method, modulation depth, symbol rate, and pulse characteristics. These parameters reflect the device's data transmission protocol and encoding scheme.

[0043] Subsequently, based on the extracted frequency, intensity, and modulation features, matching and identification are performed against a pre-set recording device feature database. This database contains electromagnetic signal feature templates for various common recording devices. Each template contains the frequency, intensity, and modulation characteristics of that type of device. The matching process utilizes a multi-feature fusion similarity calculation method, comprehensively considering the degree of matching across all dimensions and calculating the degree of match between the target signal and each template in the database. Ultimately, the device type with the highest matching degree is selected as the identification result, which serves as the recording device type information and provides an accurate basis for subsequent interference signal configuration.

[0044] Furthermore, performing device matching in a preset recording device feature database based on the frequency feature, the intensity feature, and the modulation feature, and determining the type information of the recording device based on the matching result, includes:

[0045] S131: Obtaining a device feature matching formula;

[0046] S132: Determine a first sample device in the preset recording device feature database, and extract a first frequency feature, a first intensity feature, and a first modulation feature of the first sample device;

[0047] S133: Obtaining a first device matching degree between the recording device and the first sample device based on the device feature matching formula and the frequency feature, the intensity feature, and the modulation feature of the recording device, and the first frequency feature, the first intensity feature, and the first modulation feature of the first sample device;

[0048] S134: Continuing to obtain device matching degrees between other sample devices in a preset recording device feature database and the recording device according to the method for obtaining the first device matching degree, to obtain multiple device matching degrees;

[0049] S135: Selecting the device with the maximum matching degree among the multiple device matching degrees as the matching sample device;

[0050] S136: Extracting the type information of the matching sample device from a preset recording device feature database as the type information of the recording device.

[0051] As an optional implementation, a device feature matching formula is first obtained. This formula is used to calculate the similarity between the recording device and the sample device. This formula comprehensively considers the varying importance of three different features in device identification: frequency, intensity, and modulation, and rationally regulates the contribution of each feature using weight coefficients. Next, a sample device, designated as the first sample device, is selected from a preset recording device feature database as the comparison target. This selection process can be based on a preset search order or a likelihood ranking estimated based on prior knowledge. After determining the first sample device, its features are extracted from the preset recording device feature database, including its first frequency feature, first intensity feature, and first modulation feature. These features are reference features obtained through testing and analysis of known devices during the database establishment phase and serve as baseline data for subsequent matching calculations. Next, the obtained device feature matching formula is applied to the extracted device features of the recording device within the target area and the extracted features of the first sample device to calculate the similarity between them, namely the first device matching degree. During the calculation process, the similarity between the frequency, intensity, and modulation features and the first frequency, intensity, and modulation features is calculated. These similarities are then weighted and combined using the weight coefficients in the device feature matching formula to produce a comprehensive matching score, which serves as the first device matching degree. This matching score quantitatively represents the degree of similarity between the recording device and the first sample device, providing a basis for subsequent matching determinations.

[0052] Then, using the same method as that used to obtain the matching degree of the first device, the matching degrees of the recording device and the remaining sample devices in the database are calculated one by one. By traversing all the sample devices in the database, multiple device matching degrees are finally obtained, and then the maximum value is found. The sample device with the maximum matching degree is considered to be the device type most similar to the recording device, and is therefore selected as the matching sample device. Subsequently, based on the determined matching sample device, the type information of the matching sample device is extracted from the preset recording device feature database, including key attributes such as the device's brand, model, functional parameters, and hardware architecture. This information can fully describe the characteristics of the recording device. The extracted type information serves as the identification result of the recording device, providing an accurate reference basis for the subsequent interference signal configuration.

[0053] Through precise matching and type identification of recording devices based on multi-dimensional features, the foundation is laid for subsequent interference signal configuration.

[0054] Furthermore, the device feature matching formula is:

[0055]

[0056] in, Indicates the device matching degree, represents the frequency characteristics, represents the first frequency characteristic, Indicates the similarity between the frequency feature and the first frequency feature, Indicates the intensity characteristics, represents the first intensity feature, Indicates the similarity between the intensity feature and the first intensity feature, represents the modulation characteristics, represents the first modulation characteristic, represents the similarity between the modulation feature and the first modulation feature, is the weight coefficient, and .

[0057] The above device feature matching formula uses a weighted linear combination to comprehensively evaluate the similarity between the recording device and the sample device in terms of frequency characteristics, intensity characteristics, and modulation characteristics. Quantified the frequency characteristics of recording equipment First frequency characteristics of the sample device In actual calculations, mathematical methods such as cosine similarity and the normalized inverse of Euclidean distance can be used to calculate the similarity of two frequency feature vectors. Frequency feature similarity reflects the degree of matching between two devices in terms of operating frequency, frequency band characteristics, etc. Intensity feature similarity Measured the intensity characteristics of the recording device First intensity characteristics of the sample device The intensity feature is usually expressed as the statistical characteristics of the signal energy distribution. Its similarity can be calculated by using methods such as histogram comparison or statistical moment difference comparison. The intensity feature similarity can reflect the device's transmission power characteristics and near-field and far-field performance. Modulation feature similarity Evaluated the modulation characteristics of the recording equipment The first modulation characteristics of the sample device The modulation feature usually contains information such as signal modulation mode and symbol features. Its similarity calculation can be based on feature comparison in the modulation domain or comparison of the results of the modulation recognition algorithm. These weights represent the importance of frequency, intensity, and modulation features in the final match determination, respectively. The sum of these three factors is 1, ensuring a normalized match score. These weights can be dynamically adjusted to optimize matching performance based on different application scenarios and device type distributions.

[0058] The device matching degree calculated by the device feature matching formula It is a scalar value between 0 and 1. The closer its value is to 1, the more similar the recording device is to the sample device, and the more likely they are of the same type.

[0059] Furthermore, configuring the first ultrasonic interference signal and the second ultrasonic interference signal based on the type information includes:

[0060] S210: Construct an interference signal configuration model, where the interference signal configuration model includes a first signal configuration channel and a second signal configuration channel;

[0061] S220: Inputting the type information into the first signal configuration channel to obtain a first ultrasonic interference signal;

[0062] S230: Input the type information into the second signal configuration channel to obtain a second ultrasonic interference signal.

[0063] In one feasible approach, first, a dual-channel interference signal configuration model is constructed as a processing unit that converts the recording device type information into specific interference signal parameters. The model adopts a parallel processing architecture and contains two independent but collaborative signal configuration channels. The first signal configuration channel is specifically responsible for generating the parameter configuration of the first ultrasonic interference signal, which is an interference signal that is inaudible to the human ear but can be captured by the recording device; the second signal configuration channel is specifically responsible for generating the parameter configuration of the second ultrasonic interference signal, which is used to produce a nonlinear mixing effect inside the recording device. Although the two configuration channels have different functional positioning, they maintain consistency in the model architecture and both use a mapping conversion mechanism to effectively convert the type information of the recording device into the corresponding interference signal parameter set.

[0064] When configuring the first and second ultrasonic interference signals, the acquired recording device type information is input into the first signal configuration channel for processing. The first signal configuration channel implements a mapping relationship between device type and signal parameters. Based on the input type information, the first ultrasonic interference signal parameters, including key parameters such as frequency, amplitude, waveform, and duration, are generated that are suitable for the type of recording device. Because the first ultrasonic interference signal primarily targets frequencies that are captureable by the recording device but inaudible to the human ear, this channel specifically focuses on the upper frequency response limit of the recording device, optimizing the frequency setting of the interference signal to ensure that the generated interference signal can be effectively captured by the target recording device while remaining imperceptible to the human ear. This configuration channel ultimately produces a highly targeted and optimized first ultrasonic interference signal. The recording device type information is also input into the second signal configuration channel. The second signal configuration channel focuses on generating ultrasonic signal parameters that can induce nonlinear mixing effects within the recording device. This channel implements a specialized mapping relationship between device type and nonlinear mixing interference signal parameters. Based on the varying nonlinear characteristics of different recording device types, it calculates the ultrasonic frequency combination, phase relationship, and modulation parameters that produce the optimal mixing effect. The second signal configuration channel specifically considers the nonlinear transfer function characteristics of the recording device and optimizes the frequency combination of the interference signal, ensuring that the generated difference frequency components effectively cover the core frequency band of human speech, thereby minimizing interference with the recorded speech content. This configuration channel generates a second ultrasonic interference signal optimized for nonlinear mixing interference.

[0065] By converting the configuration from recording device type information to dual-channel ultrasonic interference signals, a parameter basis is provided for the subsequent recording interference implementation, ensuring the targetedness and effectiveness of interference against different types of recording devices.

[0066] Furthermore, an interference signal configuration model is constructed, wherein the interference signal configuration model includes a first signal configuration channel and a second signal configuration channel, including:

[0067] S211: Acquire test interference data of a plurality of sample recording devices, wherein the test interference data includes recording interference data and device interference data;

[0068] S212: establishing a first correspondence relationship library between recording device types and first ultrasonic interference parameters based on recording interference data of multiple sample recording devices;

[0069] S213: establishing a second correspondence database between recording device types and second ultrasonic interference parameters based on device interference data of a plurality of sample recording devices;

[0070] S214: Training a first deep learning network as the first signal configuration channel according to the first correspondence library;

[0071] S215: Training a second deep learning network as the second signal configuration channel according to the second correspondence library;

[0072] S216: Combine the first deep learning network and the second deep learning network to form the interference signal configuration model.

[0073] Specifically, when constructing the interference signal configuration model, first, systematic interference tests are conducted on various types of sample recording devices to obtain comprehensive experimental interference data. The test process adopts a strictly controlled experimental environment, applies ultrasonic interference signals with different parameters to each type of recording device, and records the interference effects in detail. The experimental interference data obtained are divided into two categories: recording interference data and device interference data. The recording interference data mainly reflects the interference effect of ultrasonic signals with different parameters on the quality of the recorded content, including quantitative indicators such as the degree of reduction in signal-to-noise ratio and the level of reduction in speech intelligibility; the device interference data mainly reflects the situation in which ultrasonic signals cause nonlinear mixing effects inside the recording device, including characteristic parameters such as the spectral distribution and relative intensity of the mixing products. These experimental data constitute the basic data set for subsequent model training, and their quality and comprehensiveness directly affect the performance of the final interference model.

[0074] The acquired recording interference data was then subjected to in-depth analysis and data mining, extracting mappings between recording device types and optimal first ultrasonic interference parameters from a large amount of experimental data. The analysis process employed statistical methods such as multivariate correlation analysis and parameter sensitivity analysis to determine the optimal interference frequency, amplitude, waveform, and other parameter combinations for different recording device types. Based on the analysis results, a structured first correspondence database was established, indexed by recording device type and storing the corresponding optimal first ultrasonic interference parameter sets. This database not only contains discrete parameter correspondence records but also incorporates the inherent correlations between parameters, providing a high-quality structured knowledge source for subsequent deep learning model training. Simultaneously, a similar approach was employed, but with the device interference data analyzed and processed, to establish a mapping database between recording device type and optimal second ultrasonic interference parameters. Because the second ultrasonic interference signal primarily generates nonlinear mixing effects, the analysis focused specifically on the nonlinear characteristics of different recording device types and their sensitivity to mixing parameters. Through specialized methods such as nonlinear system identification and harmonic analysis, the ultrasonic signal parameter combinations that produce the strongest nonlinear mixing effects for each type of recording device were determined. Based on the analysis results, a second correspondence library was established, which accurately records the optimal second ultrasonic interference parameters corresponding to various types of recording equipment, providing a professional knowledge foundation for nonlinear mixing effects for subsequent model training.

[0075] Subsequently, based on the established first correspondence library, deep learning techniques are used to train the first signal configuration channel. This deep learning network utilizes a multi-layer neural network architecture, consisting of an input layer, multiple hidden layers, and an output layer. The network input is the recording device type feature vector, and the output is the parameter vector of the first ultrasonic interference signal. During training, data from the first correspondence library is used as training samples, and the network parameters are optimized using a backpropagation algorithm, enabling the network to accurately map the relationship between device type and optimal interference parameters. To improve the model's generalization, regularization techniques and early stopping strategies are incorporated into the training process to avoid overfitting. After training, the first deep learning network can quickly output optimized parameters for the first ultrasonic interference signal based on the input device type information, achieving intelligent conversion from device type to interference parameters. Simultaneously, a second deep learning network for nonlinear mixing interference is trained using a similar deep learning approach, but based on the established second correspondence library. This network also utilizes a multi-layer neural network architecture, but its design is specifically optimized to learn the nonlinear characteristics of the recording device. The network's input layer receives the recording device type features, and the output layer generates the parameter vector of the second ultrasonic interference signal. During the training process, data from the second correspondence database is used as training samples. Network weights are adjusted through optimization algorithms such as gradient descent, enabling the network to accurately predict the optimal nonlinear mixing interference parameters. To enhance model robustness, cross-validation and data augmentation techniques are employed during training to ensure robust performance even with unknown device types. After training, the second deep learning network can output the optimal second ultrasonic interference signal parameters for a given device type, providing parameter support for effective nonlinear mixing interference. The trained first and second deep learning networks are then integrated to form a complete interference signal configuration model. This integration utilizes a parallel architecture, with the two deep learning networks independently processing the same input (recording device type information) and outputting parameters for the first and second ultrasonic interference signals, respectively. The integrated interference signal configuration model boasts dual-channel parallel processing capabilities, capable of simultaneously generating two optimized interference signal parameters tailored to the recording content and device nonlinear characteristics, providing intelligent configuration support for comprehensive and efficient recording interference.

[0076] Through the complete process from experimental data acquisition to deep learning model training to system integration, an interference signal configuration model with intelligent learning and precise configuration capabilities was constructed, providing technical support for configuring interference signals.

[0077] Furthermore, the recording interference of the target area according to the first ultrasonic interference signal and the second ultrasonic interference signal includes:

[0078] S310: configuring a first interference duration interval for the first ultrasonic interference signal and a second interference duration interval for the second ultrasonic interference signal;

[0079] S320: Randomly select a first execution duration from the first interference duration interval;

[0080] S330: Randomly select a second execution duration from the second interference duration interval;

[0081] S340: After executing the first ultrasonic interference signal according to the first execution duration, executing the second ultrasonic interference signal according to the second execution duration;

[0082] S350: Repeat the random selection of duration and the execution of the interference signal to continuously interfere with the recording of the target area.

[0083] In an optional embodiment, first, a first interference duration interval of the first ultrasonic interference signal and a second interference duration interval of the second ultrasonic interference signal are configured. The first interference duration interval is for the first ultrasonic interference signal and is usually set to [ , ],in and are the minimum and maximum execution durations of the first ultrasonic interference signal respectively; similarly, the second interference duration interval is set to [ , The specific duration interval setting must take into account the sampling characteristics of the recording device, the signal processing mechanism, and the requirements for the persistence of the interference effect. A reasonable duration interval configuration can ensure the effectiveness of each interference, while providing sufficient variation in subsequent randomization strategies and enhancing the unpredictability of the interference.

[0084] Then, a random number generation algorithm is used to generate the first interference duration interval [ , ] Randomly select a duration value as the first execution duration The random selection process can adopt uniform distribution or other suitable probability distribution to ensure the randomness and uniformity of the duration value within the set interval. This random selection mechanism makes the execution duration of each interference different, increases the complexity and unpredictability of the interference pattern, and effectively prevents the recording equipment or recording software from filtering out or compensating the interference signal through pattern recognition and adaptive algorithms. At the same time, a similar random selection method is used, but for the second interference duration interval [ , ], randomly select a duration value as the second execution duration The random selection of the second execution duration is independent of the first execution duration, further increasing the randomness and complexity of the interference sequence. By doubly randomizing the execution durations of the two interference signals, the entire interference process exhibits highly irregular temporal characteristics, significantly improving the interference's anti-identification and anti-avoidance capabilities.

[0085] Then, the configured first ultrasonic interference signal is executed according to the selected first execution time. Emitted to the target area, the first stage of interference is implemented; after the first stage of interference is completed, the configured second ultrasonic interference signal is immediately sent according to the selected second execution time. The second stage of interference is then transmitted to the target area. This sequential arrangement creates an alternating interference pattern between the two interference signals, leveraging their respective interference strengths while also creating a sequential, composite interference effect. The first ultrasonic interference signal primarily interferes directly with the recorded content, while the second focuses on inducing nonlinear mixing effects within the recording device. The alternating execution of these two signals forms a multi-layered, comprehensive interference mechanism. The process from S320 to S340 is considered a jamming cycle, and multiple cycles are repeated to achieve continuous jamming of the target area. Within each cycle, a randomized duration is selected, resulting in the overall jamming process appearing as a time-varying, unpredictable jamming sequence. This continuous, randomly changing jamming pattern effectively prevents the recording device or post-processing software from identifying and filtering out the jamming signal through time or frequency domain analysis. The duration or number of continuous jamming cycles can be set based on actual needs, and the continuation of jamming can also be dynamically determined by an external control signal.

[0086] By realizing an ultrasonic interference sequence with time-varying and random characteristics, the effectiveness and difficulty of avoiding recording interference are improved, providing a more reliable technical means for protecting sensitive sound information.

[0087] The second embodiment is based on the same inventive concept as the recording shielding method based on ultrasonic interference in the above embodiment. Figure 2 As shown, an embodiment of the present application provides a recording shielding device based on ultrasonic interference, the device comprising:

[0088] A type information acquisition unit 11 is configured to detect and identify recording devices in a target area and acquire type information of the recording devices;

[0089] an interference signal configuration unit 12, configured to configure a first ultrasonic interference signal and a second ultrasonic interference signal based on the type information, wherein the first ultrasonic interference signal is a signal that is inaudible to human ears but can be captured by the recording device, and the second ultrasonic interference signal is a signal for generating a nonlinear mixing effect within the recording device;

[0090] The recording interference execution unit 13 is configured to perform recording interference on the target area according to the first ultrasonic interference signal and the second ultrasonic interference signal.

[0091] Furthermore, the execution steps of the type information acquisition unit 11 include:

[0092] scanning electromagnetic signals in a target area using a radio frequency scanner;

[0093] Performing spectrum analysis on the acquired electromagnetic signal to extract multiple signal features, wherein the multiple signal features include frequency features, intensity features, and modulation features;

[0094] According to the frequency feature, the intensity feature, and the modulation feature, device matching is performed in a preset recording device feature database, and type information of the recording device is determined according to the matching result.

[0095] Furthermore, the execution steps of the type information acquisition unit 11 also include:

[0096] Get the device feature matching formula;

[0097] Determine a first sample device in the preset recording device feature database, and extract a first frequency feature, a first intensity feature, and a first modulation feature of the first sample device;

[0098] Obtaining a first device matching degree between the recording device and the first sample device based on the device feature matching formula and the frequency feature, the intensity feature, and the modulation feature of the recording device, and the first frequency feature, the first intensity feature, and the first modulation feature of the first sample device;

[0099] According to the method for obtaining the first device matching degree, continue to obtain device matching degrees between other sample devices in the preset recording device feature database and the recording device, and obtain multiple device matching degrees;

[0100] Select the device with the maximum matching degree among multiple device matching degrees as the matching sample device;

[0101] The type information of the matching sample device is extracted from a preset recording device feature database as the type information of the recording device.

[0102] Furthermore, the device feature matching formula is:

[0103]

[0104] in, Indicates the device matching degree, represents the frequency characteristics, represents the first frequency characteristic, Indicates the similarity between the frequency feature and the first frequency feature, Indicates the intensity characteristics, represents the first intensity feature, Indicates the similarity between the intensity feature and the first intensity feature, represents the modulation characteristics, represents the first modulation characteristic, represents the similarity between the modulation feature and the first modulation feature, is the weight coefficient, and .

[0105] Furthermore, the execution steps of the interference signal configuration unit 12 include:

[0106] Constructing an interference signal configuration model, wherein the interference signal configuration model includes a first signal configuration channel and a second signal configuration channel;

[0107] Inputting the type information into the first signal configuration channel to obtain a first ultrasonic interference signal;

[0108] The type information is input into the second signal configuration channel to obtain a second ultrasonic interference signal.

[0109] Furthermore, the execution steps of the interference signal configuration unit 12 also include:

[0110] Acquire test interference data of a plurality of sample recording devices, wherein the test interference data includes recording interference data and device interference data;

[0111] Establishing a first correspondence library between recording device types and first ultrasonic interference parameters based on recording interference data of a plurality of sample recording devices;

[0112] Establishing a second correspondence library between recording device types and second ultrasonic interference parameters based on device interference data of a plurality of sample recording devices;

[0113] According to the first correspondence library, training a first deep learning network as the first signal configuration channel;

[0114] According to the second correspondence library, training a second deep learning network as the second signal configuration channel;

[0115] The first deep learning network and the second deep learning network are combined to form the interference signal configuration model.

[0116] Furthermore, the recording interference execution unit 13 executes the following steps:

[0117] Configure a first interference duration interval for the first ultrasonic interference signal and a second interference duration interval for the second ultrasonic interference signal;

[0118] Randomly selecting a first execution duration from the first interference duration interval;

[0119] Randomly selecting a second execution duration from the second interference duration interval;

[0120] After executing the first ultrasonic interference signal according to the first execution duration, executing the second ultrasonic interference signal according to the second execution duration;

[0121] The random selection of duration and execution of interference signals are repeated to continuously interfere with the recording of the target area.

[0122] The above are only preferred embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A recording shielding method based on ultrasonic interference, characterized in that: The method comprises: Detecting and identifying recording devices in a target area, and obtaining type information of the recording devices; Based on the type information, configuring a first ultrasonic interference signal and a second ultrasonic interference signal, wherein the first ultrasonic interference signal is a signal that is inaudible to human ears but can be captured by the recording device, and the second ultrasonic interference signal is a signal used to generate a nonlinear mixing effect inside the recording device; Performing recording interference on the target area according to the first ultrasonic interference signal and the second ultrasonic interference signal; The detecting and identifying the recording device in the target area and obtaining the type information of the recording device includes: scanning electromagnetic signals in a target area using a radio frequency scanner; Performing spectrum analysis on the acquired electromagnetic signal to extract multiple signal features, wherein the multiple signal features include frequency features, intensity features, and modulation features; performing device matching in a preset recording device feature database according to the frequency feature, the intensity feature, and the modulation feature, and determining type information of the recording device according to a matching result; The performing recording interference on the target area according to the first ultrasonic interference signal and the second ultrasonic interference signal includes: Configure a first interference duration interval for the first ultrasonic interference signal and a second interference duration interval for the second ultrasonic interference signal; Randomly selecting a first execution duration from the first interference duration interval; Randomly selecting a second execution duration from the second interference duration interval; After executing the first ultrasonic interference signal according to the first execution duration, executing the second ultrasonic interference signal according to the second execution duration; The random selection of duration and execution of interference signals are repeated to continuously interfere with the recording of the target area.

2. The method according to claim 1, characterized in that The performing device matching in a preset recording device feature database according to the frequency feature, the intensity feature, and the modulation feature, and determining the type information of the recording device according to the matching result, includes: Get the device feature matching formula; Determine a first sample device in the preset recording device feature database, and extract a first frequency feature, a first intensity feature, and a first modulation feature of the first sample device; Obtaining a first device matching degree between the recording device and the first sample device based on the device feature matching formula and the frequency feature, the intensity feature, and the modulation feature of the recording device, and the first frequency feature, the first intensity feature, and the first modulation feature of the first sample device; According to the method for obtaining the first device matching degree, continue to obtain device matching degrees between other sample devices in the preset recording device feature database and the recording device, and obtain multiple device matching degrees; Select the device with the maximum matching degree among multiple device matching degrees as the matching sample device; The type information of the matching sample device is extracted from a preset recording device feature database as the type information of the recording device.

3. The method according to claim 2, characterized in that The device feature matching formula is: M=α·sim(F,F ′ )+β·sim(A,A ′ )+γ·sim(T,T ′ ) Among them, M represents the device matching degree, F represents the frequency characteristics, and F ′ Represents the first frequency characteristic, sim(F, F ′ ) represents the similarity between the frequency feature and the first frequency feature, A represents the intensity feature, A ′ represents the first intensity feature, sim(A, A ′ ) represents the similarity between the intensity feature and the first intensity feature, T represents the modulation feature, T ′ represents the first modulation characteristic, sim(T, T ′ ) represents the similarity between the modulation feature and the first modulation feature, α, β, γ are weight coefficients, and α+β+γ=1.

4. The method according to claim 1, wherein The configuring the first ultrasonic interference signal and the second ultrasonic interference signal based on the type information includes: Constructing an interference signal configuration model, wherein the interference signal configuration model includes a first signal configuration channel and a second signal configuration channel; Inputting the type information into the first signal configuration channel to obtain a first ultrasonic interference signal; The type information is input into the second signal configuration channel to obtain a second ultrasonic interference signal.

5. The method according to claim 1, characterized in that Constructing an interference signal configuration model, wherein the interference signal configuration model includes a first signal configuration channel and a second signal configuration channel, including: Acquire test interference data of a plurality of sample recording devices, wherein the test interference data includes recording interference data and device interference data; Establishing a first correspondence library between recording device types and first ultrasonic interference parameters based on recording interference data of a plurality of sample recording devices; Establishing a second correspondence library between recording device types and second ultrasonic interference parameters based on device interference data of a plurality of sample recording devices; According to the first correspondence library, training a first deep learning network as the first signal configuration channel; According to the second correspondence library, training a second deep learning network as the second signal configuration channel; The first deep learning network and the second deep learning network are combined to form the interference signal configuration model.

6. The recording shielding device based on ultrasonic interference is characterized in that: The device is used to perform the method according to any one of claims 1 to 5, and the device comprises: a type information acquiring unit, configured to detect and identify a recording device in a target area and acquire type information of the recording device; an interference signal configuration unit, configured to configure a first ultrasonic interference signal and a second ultrasonic interference signal based on the type information, wherein the first ultrasonic interference signal is a signal that is inaudible to human ears but can be captured by the recording device, and the first ultrasonic interference signal is a signal used to generate a nonlinear mixing effect within the recording device; A recording interference execution unit is used to perform recording interference on the target area according to the first ultrasonic interference signal and the second ultrasonic interference signal.

Images