Portable secret-related environment anti-eavesdropping and photo-stealing equipment detection device

Abstract

The invention relates to the field of safety detection, and discloses a portable secret-related environment anti-eavesdropping and anti-photo stealing equipment detection device, which comprises a lower shell, an upper shell is mounted on one side of the lower shell, a thermal imaging scanning module is mounted in the middle of the upper shell, wireless detection modules are symmetrically mounted on one side of the upper shell, and the wireless detection modules are mounted on the other side of the upper shell. An endoscopic detection module, a nonlinear node detection module and a vibration spectrum analysis module are sequentially installed on the other side of the upper shell, a lens detection module is installed on the outer wall of the upper shell, handles are installed on the two sides of the lower shell, and a power module is installed at the bottom of the lower shell. The detection data of each sensor is collected in real time through the display control module, the weight of each mode is dynamically adjusted based on the environment change, and when it is detected that the false alarm rate of the wireless detection module is increased due to strong electromagnetic noise, the weight ratio is automatically reduced; and the weight of the nonlinear node detection module or the thermal imaging scanning module with higher anti-interference capability is synchronously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of security detection, and in particular to a portable anti-eavesdropping and anti-photography equipment detection device in a confidential environment. Background Art

[0002] As the demand for information security in confidential government and business scenarios continues to escalate, anti-eavesdropping and anti-photography device detection technology, a core branch of security detection, must address increasingly covert and diverse espionage methods (such as miniature cameras and low-power wireless transmitters). Currently, multimodal fusion detection (integrating electromagnetic, thermal, and vibration detection) has become a technological development trend.

[0003] Traditional detection equipment mostly integrates modules such as wireless signal receivers (to detect signal strength in the 2.4GHz / 5.8GHz frequency bands), uncooled infrared thermal imagers (to capture device temperature rise), and MEMS vibration sensors (to analyze mechanical vibration characteristics). However, each module operates independently or in a simple cascade manner, lacking a unified policy scheduling center. When strong electromagnetic noise occurs in confidential scenarios (such as interference from multiple Wi-Fi devices), the false alarm rate of the wireless detection module increases. However, due to the continuous consumption of battery power and computing resources due to the fixed high weight, the device's battery life drops sharply, and the detection accuracy decreases instead of increases, resulting in the risk of missed detection. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides a portable anti-eavesdropping and anti-photography equipment detection device for confidential environments, which solves the problem that traditional detection equipment lacks a unified policy scheduling center. When strong electromagnetic noise occurs in confidential scenes, the false alarm rate of the wireless detection module increases, but due to the fixed high weight, the battery power and computing resources are continuously consumed, resulting in a sudden drop in the device's battery life and a decrease in detection accuracy, which creates the risk of missed detection.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a portable anti-eavesdropping and anti-photography equipment detection device for confidential environments, comprising a lower shell, an upper shell is installed on one side of the lower shell, a thermal imaging scanning module is installed in the middle of the upper shell, a wireless detection module is symmetrically installed on one side of the upper shell, an endoscopy detection module, a nonlinear node detection module, and a vibration spectrum analysis module are installed in sequence on the other side of the upper shell, a lens detection module is installed on the outer wall of the upper shell, handles are installed on both sides of the lower shell, a power module is installed on the bottom of the lower shell, a charging port is provided on the other side of the lower shell, heat dissipation vents are evenly opened on one side of the bottom of the lower shell, and a display control module is installed on the top of the lower shell, and the display control module is used to dynamically adjust the modal weight and visualize the detection strategy based on multimodal detection data.

[0006] By adopting the above technical solution, the detection data of each sensor can be collected in real time through the display control module, and the weights of each modality can be dynamically adjusted based on environmental changes. When strong electromagnetic noise is detected, causing the false alarm rate of the wireless detection module to increase, its weight ratio is automatically reduced, and the weight of the nonlinear node detection module or thermal imaging scanning module with stronger anti-interference ability is simultaneously increased. While reducing invalid computing power and power consumption, the detection reliability is enhanced through the complementarity of multimodal data, thereby constructing a multimodal collaborative detection system with the display control module as the core, realizing unified policy scheduling and dynamic resource allocation for modules such as thermal imaging scanning, wireless detection, endoscopic detection, nonlinear node detection, vibration spectrum analysis and lens detection, etc., solving the problem of traditional detection equipment lacking a unified policy scheduling center. When strong electromagnetic noise occurs in confidential scenes, the false alarm rate of the wireless detection module increases, but due to the continuous consumption of battery power and computing resources due to the fixed high weight, the device battery life drops sharply and the detection accuracy decreases instead of increases, resulting in the risk of missed detection.

[0007] Preferably, the display control module includes: Solving unit: used to solve the problem of meeting power consumption constraints The modal weight vector ,in and , and obtain the three-dimensional weight-accuracy surface; Computational unit: used to calculate the pairwise mutual information of nonlinear node detection signals, thermal imaging data, and vibration voiceprint data ; Display unit: used for real-time rendering of three-dimensional weight-accuracy surface and mutual information heat map, the three-dimensional weight-accuracy surface is generated by radial basis function interpolation, the cell color depth of the mutual information heat map maps the mutual information size; Interaction unit: used to adjust the modal weight and perform dynamic power consumption mode switching, and the adjustment of the modal weight maintains the weight and constraint ,in They are the fusion coefficients of nonlinear node detection, thermal imaging scanning, and vibration spectrum analysis respectively.

[0008] Preferably, the solution is obtained by an interior point method, and its objective function is , the constraints include power consumption constraints ,in, is the unit weight power consumption of the corresponding mode.

[0009] Preferably, the iterative termination condition of the interior point method is the gradient norm , where the Lagrangian function , is the power consumption constraint multiplier, are weight and constraint multipliers.

[0010] Preferably, the calculation is performed using Gaussian kernel density estimation and is based on the mutual information formula Quantify the modal correlation, where is the modal data sample, including the nonlinear node harmonic amplitude and thermal imaging temperature value, is the marginal probability distribution of the modal data.

[0011] Preferably, when adjusting the modal weight, the adjusted detection accuracy is calculated in real time. and power consumption , and draw the power consumption-accuracy feasible region boundary, satisfying .

[0012] Preferably, the three-dimensional positioning area of ​​the display unit is used to calculate the coordinates of the abnormal signal source by triangulation positioning method. ,satisfy in, is the sensor space coordinate, is the detection distance between the sensor and the signal source, positioning error .

[0013] Preferably, when the dynamic power consumption mode is switched, When the non-essential modes are turned off and the weights are re-solved, the power consumption after switching satisfies ,in This is the upper limit of low power mode.

[0014] Preferably, the display control module and the thermal imaging scanning module, the nonlinear node detection module, and the vibration spectrum analysis module are synchronized via the LVDS differential bus, and the synchronization error satisfy and is quantified by the following formula ,in They are the data acquisition timestamps of the thermal imaging scanning module, nonlinear node detection module, and vibration spectrum analysis module respectively.

[0015] Preferably, the display control module also includes an encryption unit, which is used to encrypt and tamper-proof various data, including the harmonic signal of the nonlinear node detection module, the temperature matrix of the thermal imaging scanning module, and the time domain waveform of the vibration spectrum analysis module. The encryption is performed using the AES-256 algorithm, and the encryption key is stored in the ASIC chip hardware encryption area of ​​the display control module. The tamper-proof processing is to trigger the modal weight reset when it is detected that the hash value of the decrypted data is inconsistent with the original hash value, switch to the preset security policy, and start the full-band scanning mode to forcibly cover the suspected tampering scenario.

[0016] Working principle: A protective structure is formed by the lower shell and the upper shell. The thermal imaging scanning module uses a microlens array to collect infrared radiation and convert it into an electrical signal. The temperature distribution map is generated through digital signal processing. The wireless detection module adopts a superheterodyne receiver architecture and uses fast Fourier transform to analyze the signal strength and spectrum characteristics of common frequency bands. The endoscopic detection module uses a flexible detection rod to penetrate into the gap space to collect optical images. The nonlinear node detection module identifies the nonlinear response unique to electronic equipment. The vibration spectrum analysis module uses window function noise and fast Fourier transform to extract the vibration characteristics of the 10-500Hz frequency band. The lens detection module identifies the speckle pattern of reflected light by analyzing it. The optical surface of the lens is hidden, and the device is easy to carry through the handle. The power module dynamically distributes power to each module through the power management chip. After receiving multimodal data, the display control module constructs a modal weight optimization model with power consumption constraints based on convex optimization theory. It combines mutual information to quantify the correlation between each modal data to assist feature fusion, and dynamically solves the modal weight vector that satisfies the maximum detection accuracy and power consumption constraints. At the same time, the optimization strategy is visualized in the form of three-dimensional weight-accuracy surfaces, mutual information heat maps, etc. Through theory-driven dynamic adjustment of strategies and visual interaction, the coordinated optimization of detection accuracy and device power consumption in complex scenarios is achieved, while taking into account the power consumption control and ease of use of portable devices.

[0017] The present invention provides a portable device for detecting devices that can prevent eavesdropping or stealing photos in confidential environments. It has the following beneficial effects: 1. The present invention collects detection data from various sensors in real time through a display control module, and dynamically adjusts the weights of various modalities based on environmental changes. When strong electromagnetic noise is detected, causing the false alarm rate of the wireless detection module to increase, its weight ratio is automatically reduced, and the weight of the nonlinear node detection module or thermal imaging scanning module with stronger anti-interference capabilities is simultaneously increased. While reducing invalid computing power and power consumption, the detection reliability is enhanced through the complementarity of multimodal data, thereby constructing a multimodal collaborative detection system with the display control module as the core, and realizing unified strategic scheduling and dynamic resource allocation for modules such as thermal imaging scanning, wireless detection, endoscopic detection, nonlinear node detection, vibration spectrum analysis, and lens detection.

[0018] 2. The present invention integrates multiple types of sensors such as thermal imaging scanning modules, wireless detection modules, and nonlinear node detection modules to build a full-dimensional detection system covering electromagnetic radiation, temperature anomalies, mechanical vibrations, and optical characteristics. When detecting low-power eavesdropping devices with strong concealment, the synergistic effect of thermal imaging and nonlinear node detection modules can improve the recognition rate, effectively solving the problem of missed detection caused by insufficient feature coverage in traditional single-modal detection. It is especially suitable for complex and confidential environments with dense equipment such as conference rooms and vehicles.

[0019] 3. The present invention constructs a weight allocation model with power consumption constraints through a convex optimization algorithm, and dynamically adjusts the detection strategy in real time according to the correlation of multimodal data. When the false alarm rate of the wireless detection module increases due to strong electromagnetic interference, its weight is reduced, and the weight of the nonlinear node detection module is simultaneously increased. While reducing the power consumption of invalid signal processing, the detection accuracy is maintained through harmonic detection with stronger anti-interference ability, avoiding the problem of high-power consumption modules continuously occupying resources but having low efficiency in traditional fixed weight strategies, and achieving the resource scheduling goals of on-demand allocation and efficiency priority in complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the three-dimensional structure of the present invention; Figure 2 It is a schematic diagram of the local structure of the lower shell of the present invention; Figure 3 It is a schematic diagram of the partial structure of the upper shell of the present invention; Figure 4 FIG. 4 is a schematic diagram of the module architecture of the display control module of the present invention.

[0021] Among them, 1. Lower shell; 2. Upper shell; 3. Thermal imaging scanning module; 4. Lens detection module; 5. Endoscopic detection module; 6. Wireless detection module; 7. Nonlinear node detection module; 8. Vibration spectrum analysis module; 9. Power module; 10. Heat dissipation port; 11. Handle; 12. Charging port; 13. Display control module. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solution of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0023] Please see the attached Figure 1 -Attached Figure 4An embodiment of the present invention provides a portable anti-eavesdropping and anti-photography equipment detection device for confidential environments, comprising a lower shell 1, an upper shell 2 installed on one side of the lower shell 1, a thermal imaging scanning module 3 installed in the middle of the upper shell 2, a wireless detection module 6 symmetrically installed on one side of the upper shell 2, an endoscopy detection module 5, a nonlinear node detection module 7, and a vibration spectrum analysis module 8 installed in sequence on the other side of the upper shell 2, a lens detection module 4 installed on the outer wall of the upper shell 2, handles 11 installed on both sides of the lower shell 1, a power module 9 installed on the bottom of the lower shell 1, a charging port 12 provided on the other side of the lower shell 1, heat dissipation vents 10 evenly opened on one side of the bottom of the lower shell 1, and a display control module 13 installed on the top of the lower shell 1. The display control module 13 is used to dynamically adjust the modal weight based on multimodal detection data and visualize the detection strategy.

[0024] Specifically, a compact protective structure is formed by the lower shell 1 and the upper shell 2. The thermal imaging scanning module 3 in the middle of the upper shell 2 adopts an uncooled infrared focal plane array, which collects infrared radiation through a microlens array and converts it into an electrical signal, and generates a temperature distribution map through digital signal processing. In some embodiments, the thermal imaging scanning module 3 can adopt a FLIRRepton series infrared thermal imager module; the symmetrically arranged wireless detection module 6 adopts a superheterodyne receiver architecture, and analyzes the signal strength and spectrum characteristics of common frequency bands such as 2.4GHz and 5.8GHz through fast Fourier transform; the endoscopy detection module 5 integrates a micro camera and an LED lighting unit, and collects optical images through a flexible detection rod into the gap space; the nonlinear node detection module 7 is based on the Schottky diode mixing principle, transmits the fundamental signal and receives the second and third harmonics, and identifies the nonlinear response unique to electronic devices; the vibration spectrum analysis module 8 adopts a three-axis MEMS accelerometer, and extracts the vibration characteristics of the 10-500Hz frequency band through window function noise and fast Fourier transform; the lens detection module 4 adopts laser speckle imaging technology to identify hidden mirrors by analyzing the speckle pattern of reflected light. The optical surface of the head is smooth; the handles 11 on both sides of the lower housing 1 are ergonomically designed for easy carrying; the power module 9 at the bottom integrates a lithium battery pack and DC-DC conversion circuit, and dynamically distributes power to each module through the power management chip; the charging port 12 uses a waterproof USB-C interface and supports the PD fast charging protocol; the heat dissipation vent 10 uses a honeycomb structure combined with a thermally conductive silicone sheet to effectively reduce the operating temperature of the ASIC chip; after receiving multimodal data, the display control module 13 at the top of the lower housing 1 constructs a modal weight optimization model with power consumption constraints based on convex optimization theory. It combines mutual information to quantify the correlation between each modal data to assist in feature fusion, and dynamically solves the modal weight vector that satisfies the maximum detection accuracy and power consumption constraints. The optimization strategy is also visualized in the form of a three-dimensional weight-accuracy surface and mutual information heat map. Through theory-driven dynamic strategy adjustment and visual interaction, the coordinated optimization of detection accuracy and device power consumption in complex scenarios is achieved, improving the device's recognition efficiency and environmental adaptability against eavesdropping and stealing devices. The adjusted detection accuracy and power consumption changes are calculated in real time and feedback is displayed, while taking into account power consumption control and ease of use of portable devices. Thus, a multimodal collaborative detection system with the display control module as the core was constructed, which realized unified strategy scheduling and dynamic resource allocation for modules such as thermal imaging scanning, wireless detection, endoscopic detection, nonlinear node detection, vibration spectrum analysis and lens detection. It solved the problems of sudden drop in battery life and risk of missed detection caused by excessive resource consumption of wireless detection modules in strong electromagnetic noise scenarios, achieved an intelligent balance between detection accuracy and equipment power consumption in complex confidential environments, and improved the device's adaptability to dynamic interference environments and safety protection effectiveness.

[0025] The display control module 13 includes: Solving unit: used to solve the problem of meeting power consumption constraints The modal weight vector ,in and , and obtain the three-dimensional weight-accuracy surface; the solution is solved by the interior point method, and its objective function is , the constraints include power consumption constraints ,in, is the unit weight power consumption of the corresponding mode. The iterative termination condition of the interior point method is the gradient norm , where the Lagrangian function , is the power consumption constraint multiplier, are weight and constraint multipliers.

[0026] Specifically, the solving unit is used to construct and solve the convex optimization model of multimodal detection to determine the optimal weight distribution of each mode under the power consumption constraint. It first obtains the unit weight power consumption of the nonlinear node detection module, thermal imaging scanning module, and vibration spectrum analysis module. This parameter is obtained by monitoring the module working current in real time through the current sensor of the power module and converting it into working time. At the same time, based on the multimodal data collected by each module, the detection accuracy under the current weight is calculated through the classification algorithm. Then, the solving unit constructs the optimization goal: guided by maximizing the detection accuracy, the objective function is transformed into , and introduce constraints: one is the power consumption constraint, that is, the weighted sum of each modal weight and the corresponding unit weight power consumption does not exceed the maximum allowable power consumption , the expression is ; The second is the weight and constraint and non-negative constraint, that is, the sum of the modal weights is 1 and the single weight is non-negative, that is and In order to solve the constrained convex optimization problem, the interior point method is used to construct the Lagrangian function Fusion objectives and constraints, where is the power consumption constraint multiplier, used to penalize solutions that exceed the power consumption limit. is the weight and constraint multiplier to ensure that the weight and meet the normalization requirements. During the iteration process, the solution unit continuously calculates the gradient of the Lagrangian function. Convergence is determined when , and the weight vector obtained at this time This is a modal weight allocation scheme that satisfies power consumption constraints and optimizes detection accuracy. This balances multimodal detection accuracy with device power consumption, provides core computing support for the display control module 13 to output a dynamically adjustable optimal detection strategy, and ensures the device's detection performance in complex environments.

[0027] In this embodiment, when the solving unit is working, the unit weight power consumption of the nonlinear node detection, thermal imaging scanning, and vibration spectrum analysis modules is first obtained. , which is generated by the correlation between the power module monitoring module current and weight ratio, as well as the detection accuracy after multimodal data fusion The unit power consumption is compared with the modal weight calculated in real time. Substitute into the power consumption constraint formula , during the weight allocation process, the total power consumption of the device does not exceed the preset upper limit ; At the same time, construct the Lagrangian function ,Integrate the goal of maximizing detection accuracy with power consumption, weights and constraints, and iteratively optimize through the interior point method , when the gradient norm of the Lagrangian function is less than When , the output modal weight vector that meets the power consumption constraint and has the best detection accuracy is This vector guides the display control module 13 to dynamically adjust the working intensity of each mode, achieve a coordinated balance between detection accuracy and device power consumption, and improve the device's adaptability to identifying eavesdropping and stealing devices.

[0028] Computational unit: used to calculate the pairwise mutual information of nonlinear node detection signals, thermal imaging data, and vibration voiceprint data ; The calculation is done using Gaussian kernel density estimation and based on the mutual information formula Quantify the modal correlation, where is the modal data sample, including the nonlinear node harmonic amplitude and thermal imaging temperature value, is the marginal probability distribution of the modal data.

[0029] Specifically, the calculation unit receives the harmonic amplitude of the nonlinear node detection module, the temperature value of the thermal imaging scanning module, and the vibration signal characteristics of the vibration spectrum analysis module as input. For each two modal data, such as nonlinear node and thermal imaging, nonlinear node and vibration, thermal imaging and vibration, the calculation unit first normalizes the data to eliminate dimensional differences. Subsequently, Gaussian kernel density estimation is used to model the joint distribution of the two modal data with a preset Gaussian kernel function to obtain the joint probability density. ; At the same time, model the single modal data separately to obtain the marginal probability density and Substituting these probability densities into the mutual information formula , through numerical integration, such as Monte Carlo sampling, the integral is calculated to obtain the mutual information value between the two modes. This value quantifies the degree of modal correlation. If the mutual information between two groups of modalities is high, it means that the two are highly complementary when detecting eavesdropping and stealing equipment. The calculation unit outputs all pairwise mutual information results to the display control module for constructing a mutual information heat map, and assists the solution unit in optimizing the modal weights, so that multimodal fusion is more in line with data association, and improves the rationality and performance of the detection strategy. Among them, the calculation unit first obtains the harmonic amplitude sequence output by the nonlinear node detection module, the temperature matrix output by the thermal imaging scanning module, and the vibration soundprint waveform output by the vibration spectrum analysis module, and uses these data as modal samples. (such as nonlinear node harmonic amplitudes) and (such as thermal imaging temperature value). Then, the calculation unit uses the Gaussian kernel density estimation method to fit the joint probability distribution based on the collected multiple sets of sample data. , and the marginal probability distribution of unimodal data .Bundle 、 Substitute into the mutual information formula ,Through numerical integration operations, Monte Carlo sampling can be used to approximate the integral calculation and output the mutual information value between two modal data, such as nonlinear node detection signals and thermal imaging data.

[0030] Display unit: used for real-time rendering of 3D weight-accuracy surface and mutual information heat map. 3D weight-accuracy surface is generated by radial basis function interpolation. Cell color depth of mutual information heat map maps mutual information. The three-dimensional positioning area of ​​the display unit calculates the coordinates of the abnormal signal source through the triangulation positioning method ,satisfy in, is the sensor space coordinate, is the detection distance between the sensor and the signal source, positioning error .

[0031] Specifically, the display unit serves as a carrier for interaction between the device and the user. When working, it first receives multiple groups of modal weight combinations and corresponding detection accuracy data output by the solution unit, and fits the discrete weight-accuracy data based on the radial basis function interpolation algorithm to generate a continuous three-dimensional weight-accuracy surface, which intuitively presents the changing trend of detection performance under different modal weight distributions. At the same time, it receives the pairwise modal mutual information results calculated by the calculation unit, assigns color depth to the heat map cells according to the mutual information value, constructs a mutual information heat map, and clearly shows the degree of correlation between the modes. When the device detects an abnormal signal, the display unit obtains the spatial coordinate information of multiple sensors (such as nonlinear node detection modules, thermal imaging scanning modules, etc.) and the distance data of the signal sources detected by each, and uses the triangulation positioning method to calculate the spatial coordinates of the sensors. , detection distance Substitute into the positioning formula , solve to get the coordinates of the abnormal signal source , and combined with the positioning error formula , rendering the location of the abnormal signal source and the error range in the 3D positioning area. Using the 3D weight-accuracy surface, users can quickly understand the impact of modal weight adjustments on detection results. Mutual information heat maps help grasp modal correlation characteristics and facilitate the optimization of fusion strategies. The 3D positioning function accurately locates the abnormal source, and the multi-dimensional visualization output allows users to efficiently obtain detection information, assist in decision-making, and enhance the device's practicality and ease of use in confidential environment detection tasks.

[0032] Interaction unit: used to adjust modal weights and perform dynamic power consumption mode switching, maintaining weights and constraints when adjusting modal weights ,in They are the fusion coefficients of nonlinear node detection, thermal imaging scanning, and vibration spectrum analysis. When adjusting the modal weight, the adjusted detection accuracy is calculated in real time. and power consumption , and draw the power consumption-accuracy feasible region boundary, satisfying When dynamic power mode switching is performed, when it is detected When the non-essential modes are turned off and the weights are re-solved, the power consumption after switching satisfies ,in This is the upper limit of low power mode.

[0033] Specifically, the interactive unit responds to the operation initiated by the user through the touch interface of the display control module, and adjusts the fusion coefficient of the nonlinear node detection, thermal imaging scanning, and vibration spectrum analysis modules. When maintaining weights and constraints During the adjustment process, the interactive unit calls the calculation logic of the solution unit in real time, and calculates the detection accuracy based on the current weight combination and multi-modal detection data. , and calculate the total power consumption based on the unit weight power consumption of each module and the current weight At the same time, the display unit is driven to draw the power consumption-accuracy feasible domain boundary on the interface, which follows The constraints of the weight adjustment are intuitively presented to show the impact of the weight adjustment on the detection performance and power consumption. When the interaction unit refers to the mutual information matrix output by the calculation unit, it first determines the modes with low mutual information and weak detection contribution as unnecessary modes (such as vibration spectrum analysis module), shuts down some of its functions or reduces its working intensity, and then triggers the solution unit to re-solve the mode that meets the upper limit of low power consumption mode. In this way, the interactive unit not only supports users to flexibly adjust detection strategies, but also automatically responds to power consumption exceeding the limit, achieving a dynamic balance between detection performance and device battery life, and improving the device's adaptability in complex and confidential environments.

[0034] The display control module 13 and the thermal imaging scanning module 3, the nonlinear node detection module 7, and the vibration spectrum analysis module 8 are synchronized through the LVDS differential bus. satisfy and is quantified by the following formula ,in They are the data acquisition timestamps of the thermal imaging scanning module 3, the nonlinear node detection module 7, and the vibration spectrum analysis module 8 respectively.

[0035] Specifically, the display control module 13 transmits data with the thermal imaging scanning module 3, the nonlinear node detection module 7, and the vibration spectrum analysis module 8 via the LVDS differential bus. By leveraging the anti-interference characteristics of low voltage differential signals, high-speed synchronous transmission of multimodal data is achieved. When each module collects detection data such as harmonic amplitude, temperature matrix, and vibration waveform, it synchronously generates a data acquisition timestamp. Nonlinear node detection module, Thermal imaging scanning module, Vibration spectrum analysis module. After the display control module receives the data, it substitutes these three timestamps into the formula , calculate the time offset between multimodal data .like , then the data synchronization is determined to meet the requirements, and the aligned multimodal data is directly input into the solution unit and the calculation unit; if If the threshold is exceeded, the display control module triggers the data resynchronization mechanism and recalibrates the acquisition timing of each module. This time synchronization check ensures that the solver unit optimizes modal weights based on detection data from the same time dimension, avoiding distortions in detection accuracy and power consumption calculations due to time offsets. It also ensures the accuracy of the joint probability distribution constructed by the calculation unit, providing a reliable timing foundation for mutual information quantization and multimodal feature fusion, and improving the accuracy and stability of multimodal detection strategies.

[0036] The display control module 13 also includes an encryption unit, which is used to encrypt and tamper-proof various data. The various data include the harmonic signal of the nonlinear node detection module 7, the temperature matrix of the thermal imaging scanning module 3, and the time domain waveform of the vibration spectrum analysis module 8. The encryption is performed using the AES-256 algorithm, and the encryption key is stored in the ASIC chip hardware encryption area of ​​the display control module 13. The anti-tampering processing is that when the hash value of the decrypted data is detected to be inconsistent with the original hash value, the modal weight reset is triggered, the preset security policy is switched to, and the full-band scanning mode is started to forcibly cover the suspected tampering scenario.

[0037] Specifically, the encryption unit in the display control module first captures the harmonic signal from the nonlinear node detection module, the temperature matrix from the thermal imaging scanning module, and the time-domain waveform from the vibration spectrum analysis module as encryption targets. During the encryption phase, the data is encrypted segment by segment using the AES-256 algorithm. The encryption key is generated by the ASIC's built-in true random number generator and stored only in the chip's hardware encryption area, which is physically isolated and logically locked to prevent unauthorized external access. Simultaneously, the encryption unit calculates a SHA-256 hash value for the original data and stores it. During the decryption phase, if the calculated hash value of the decrypted data does not match the stored value, the data is deemed tampered with. In this case, the encryption unit immediately triggers a chain reaction: it sends a signal to the solver unit, invoking the preset security weights (such as the modal weight configuration that prioritizes nonlinear node detection), rapidly resetting the current modal weights and restoring basic detection capabilities. It also simultaneously controls the nonlinear node detection module to activate full-scan mode in both the 2.4GHz and 3.6GHz bands, switches the thermal imaging scanning module to high-sensitivity acquisition, and increases the sampling frequency of the vibration spectrum analysis module to ensure detection covers suspected tampering scenarios. Through the closed-loop design of encryption protection and tampering response, the strategic recovery capability of the solution unit and the mandatory detection function of the multimodal module are connected to resist the risk of data tampering and ensure the stability of confidential environment detection.

[0038] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A portable anti-eavesdropping and anti-photography device detection device for confidential environments, comprising a lower housing (1), characterized in that: An upper shell (2) is installed on one side of the lower shell (1), a thermal imaging scanning module (3) is installed in the middle of the upper shell (2), a wireless detection module (6) is symmetrically installed on one side of the upper shell (2), an endoscopy detection module (5), a nonlinear node detection module (7), and a vibration spectrum analysis module (8) are installed in sequence on the other side of the upper shell (2), a lens detection module (4) is installed on the outer wall of the upper shell (2), handles (11) are installed on both sides of the lower shell (1), a power module (9) is installed on the bottom of the lower shell (1), a charging port (12) is provided on the other side of the lower shell (1), heat dissipation ports (10) are uniformly opened on one side of the bottom of the lower shell (1), and a display control module (13) is installed on the top of the lower shell (1), and the display control module (13) is used to dynamically adjust the modal weight based on multimodal detection data and visualize the detection strategy.

2. The portable anti-eavesdropping and anti-photography device detection device for confidential environments according to claim 1, characterized in that: The display control module (13) comprises: Solving unit: used to solve the problem of meeting power consumption constraints The modal weight vector ,in and , and obtain the three-dimensional weight-accuracy surface; Computational unit: used to calculate the pairwise mutual information of nonlinear node detection signals, thermal imaging data, and vibration voiceprint data ; Display unit: used for real-time rendering of three-dimensional weight-accuracy surface and mutual information heat map, the three-dimensional weight-accuracy surface is generated by radial basis function interpolation, the cell color depth of the mutual information heat map maps the mutual information size; Interaction unit: used to adjust the modal weight and perform dynamic power consumption mode switching, and the adjustment of the modal weight maintains the weight and constraint ,in They are the fusion coefficients of nonlinear node detection, thermal imaging scanning, and vibration spectrum analysis respectively.

3. The portable anti-eavesdropping and anti-photography device detection device for confidential environments according to claim 2, characterized in that: The solution is obtained by the interior point method, and its objective function is , the constraints include power consumption constraints ,in, is the unit weight power consumption of the corresponding mode.

4. The portable anti-eavesdropping and anti-photography device detection device for confidential environments according to claim 3, characterized in that: The iterative termination condition of the interior point method is the gradient norm , where the Lagrangian function , is the power consumption constraint multiplier, are weight and constraint multipliers.

5. The portable anti-eavesdropping and anti-photography device detection device for confidential environments according to claim 2, characterized in that: The calculation is performed using Gaussian kernel density estimation and is based on the mutual information formula Quantify the modal correlation, where is the modal data sample, including the nonlinear node harmonic amplitude and thermal imaging temperature value, is the marginal probability distribution of the modal data.

6. The portable anti-eavesdropping and anti-photography device detection device for confidential environments according to claim 2, characterized in that: When the modal weight is adjusted, the adjusted detection accuracy is calculated in real time and power consumption , and draw the power consumption-accuracy feasible region boundary, satisfying .

7. The portable anti-eavesdropping and anti-photography device detection device for confidential environments according to claim 2, characterized in that: The three-dimensional positioning area of ​​the display unit calculates the coordinates of the abnormal signal source by triangulation positioning method ,satisfy in, is the sensor space coordinate, is the detection distance between the sensor and the signal source, positioning error .

8. The portable anti-eavesdropping and anti-photography device detection device for confidential environments according to claim 2, characterized in that: When the dynamic power consumption mode is switched, When the non-essential modes are turned off and the weights are re-solved, the power consumption after switching satisfies ,in This is the upper limit of low power mode.

9. The portable anti-eavesdropping and anti-photography device detection device for confidential environments according to claim 1, characterized in that: The display control module (13) is synchronized with the thermal imaging scanning module (3), the nonlinear node detection module (7), and the vibration spectrum analysis module (8) via the LVDS differential bus. satisfy and is quantified by the following formula ,in They are the data acquisition timestamps of the thermal imaging scanning module (3), the nonlinear node detection module (7), and the vibration spectrum analysis module (8).

10. The portable anti-eavesdropping and anti-photography device detection device for confidential environments according to claim 1, characterized in that: The display control module (13) further includes an encryption unit, which is used to encrypt and perform tamper-proof processing on various data, including the harmonic signal of the nonlinear node detection module (7), the temperature matrix of the thermal imaging scanning module (3), and the time domain waveform of the vibration spectrum analysis module (8). The encryption is performed using the AES-256 algorithm, and the encryption key is stored in the ASIC chip hardware encryption area of ​​the display control module (13). The tamper-proof processing is to trigger a modal weight reset when it is detected that the hash value of the decrypted data is inconsistent with the original hash value, switch to a preset security policy, and start a full-band scanning mode to forcibly cover the suspected tampering scene.

Images