Intelligent control method and system for electromagnetic shielding room

Abstract

The application relates to the technical field of computer and industrial control systems, and discloses an intelligent control method and system for an electromagnetic shielding room. The method collects omnidirectional wideband electromagnetic data in real time through a distributed electromagnetic sensor array, constructs a three-dimensional dynamic electromagnetic situation model after time-space alignment, realizes abnormal interference detection in combination with multi-scale time-frequency analysis and a convolutional autoencoder, and completes sub-meter interference source positioning by using direction of arrival estimation and reverse ray tracing. According to the spectrum, position and threat level of the interference source, the reconfigurable metamaterial shielding units in the adjacent area are dynamically activated to form a local high-attenuation barrier. The system comprises a sensor array, a main control processing unit and an addressable shielding unit array. The system realizes millisecond-level response, accurate isolation and on-demand shielding, and improves shielding efficiency, energy efficiency ratio and adaptability to complex electromagnetic environments.

Description

Technical Field

[0001] This invention belongs to the field of computer and industrial control system technology, specifically relating to an intelligent control method and system for an electromagnetic shielding room. Background Technology

[0002] With the rapid development of technologies such as 5G communication, industrial IoT, and high-precision electronic measurement, electromagnetically sensitive equipment is facing higher requirements for the electromagnetic cleanliness of its working environment. Electromagnetic shielding rooms, as key infrastructure for isolating external electromagnetic interference and ensuring the stable operation of internal equipment, have been widely used in defense, medical imaging, chip testing, and scientific research. Their core function is to construct a sealed cavity using conductive or magnetic materials to attenuate the intrusion of external electromagnetic waves and suppress the leakage of internal signals, thereby maintaining a controllable electromagnetic silence environment.

[0003] Rapid response and precise suppression of sudden or unknown interference sources in complex dynamic electromagnetic environments have become an important development direction for current electromagnetic shielding technology. Traditional shielding rooms mostly adopt fixed structures and passive filtering designs, relying on the shielding effectiveness indicators of preset frequency bands, and lack the ability to sense and adaptively adjust the real-time electromagnetic situation. When unexpected narrowband interference, broadband impulse noise, or multi-source composite interference occurs in the environment, existing systems struggle to simultaneously complete interference feature extraction, spatial localization, and directional suppression in the time, frequency, and spatial domains, leading to a sharp drop in shielding effectiveness or even failure.

[0004] In existing technologies, some solutions attempt to introduce a single type of sensor for interference monitoring, but these are limited by the single sensing dimension, insufficient spatial resolution, and weak data fusion capabilities, making it impossible to accurately reconstruct the spatial location and radiation characteristics of the interference source. Other solutions, while integrating actuator arrays, lack a closed-loop collaborative mechanism with the sensing layer, resulting in lagging and coarse control strategies. Especially in scenarios with multiple concurrent, mobile, or intermittent interference sources, existing systems generally suffer from low positioning accuracy, large response delays, and poor suppression efficiency, failing to meet the urgent needs for high-dynamic, high-reliability electromagnetic protection. Therefore, there is an urgent need for an intelligent control method and system for electromagnetic shielding rooms that integrates multimodal sensing, intelligent identification, and multi-actuator collaborative control to achieve rapid location, precise isolation, and dynamic suppression of unknown interference sources. Summary of the Invention

[0005] This invention provides an intelligent control method and system for electromagnetic shielding rooms, aiming to solve the technical problem of difficulty in quickly locating and isolating sudden or unknown interference sources in complex electromagnetic environments. In existing technologies, electromagnetic shielding rooms mostly adopt static shielding structures, relying on fixed-band filters and preset shielding strategies, which cannot provide real-time responses to dynamically changing electromagnetic environments.

[0006] This invention provides an intelligent control method for an electromagnetic shielding room, comprising: The raw electromagnetic field data inside and at the boundary of the shielded room are collected in real time by a distributed electromagnetic sensor array. The raw electromagnetic field data includes electric field strength, magnetic field strength, signal frequency, phase information, polarization state and arrival time. The original electromagnetic field data is spatiotemporally aligned to generate an electromagnetic sensing dataset with a unified time reference and spatial coordinates. Based on the electromagnetic sensing dataset, a three-dimensional dynamic electromagnetic situation model covering the entire shielded room is constructed. Multi-scale time-frequency analysis and abnormal pattern recognition are performed on the three-dimensional dynamic electromagnetic situation model to detect sudden or unknown interference signals that deviate from the normal electromagnetic background. For the detected interference signal, a direction of arrival estimation algorithm and multipath reverse ray tracing are performed to determine the three-dimensional spatial coordinates of the interference source in the shielded room; The electromagnetic threat level of the interference source is assessed based on its three-dimensional spatial coordinates, spectral bandwidth, power intensity, and duration. Based on the electromagnetic threat level, a preset shielding strategy library is invoked to generate a local shielding command targeting the interference source; The local shielding command is sent to the reconfigurable electromagnetic shielding unit located in the vicinity of the interference source, and its electromagnetic parameters are controlled to switch to a high attenuation state in the corresponding frequency band to form a directional shielding barrier.

[0007] Preferably, the distributed electromagnetic sensing array is composed of a grid of planar electromagnetic sensors disposed on the six inner surfaces of the shielded room and a three-dimensional electromagnetic probe array suspended in the central region of the room. The planar electromagnetic sensor grid is arranged at equal intervals, with the distance between adjacent sensors not exceeding 1 / 4 of the wavelength corresponding to the highest frequency in the operating frequency band; the three-dimensional electromagnetic probe array consists of three orthogonally arranged dipole antennas, used to capture electromagnetic vector information in any direction in space.

[0008] Preferably, the spatiotemporal alignment process includes: synchronizing the sampling times of all sensor nodes at the nanosecond level based on the master control clock; Using the pre-calibrated spatial position parameters of the sensors, the electromagnetic data collected by each node is mapped to a unified Cartesian coordinate system; for sensor data with different sampling rates, the cubic spline interpolation method is used to resample to a unified frequency.

[0009] Preferably, the construction of the three-dimensional dynamic electromagnetic situation model specifically includes: slicing the spatiotemporally aligned electromagnetic sensing data according to a time window; For the data within each time window, the shielded room space is divided using a voxelization method; For each voxel unit, the amplitude, frequency, and phase information of all electromagnetic signals falling within it are aggregated to form a multidimensional electromagnetic feature vector containing spectral density, polarization ellipse parameters, and coherence indices; the voxel feature vector sequences of continuous time windows are stacked to form a four-dimensional electromagnetic state tensor.

[0010] Preferably, the multi-scale time-frequency analysis and anomaly pattern recognition include: The four-dimensional electromagnetic situation tensor is subjected to a short-time Fourier transform along the time dimension to extract the time spectrum of each voxel unit; the time spectrum is input into a pre-trained convolutional autoencoder network to calculate the reconstruction error; when the reconstruction error of a voxel unit is greater than a preset threshold in three consecutive time windows, it is determined that there is abnormal electromagnetic activity in the voxel region.

[0011] Preferably, the direction-of-arrival estimation algorithm employs a multi-signal classification algorithm, utilizing the phase difference data of the three-dimensional electromagnetic probe array to calculate the incident elevation angle and azimuth angle of the interference signal; The multipath reverse ray tracing is based on the electromagnetic reflection coefficient database of the inner wall of the shielded room. Combined with the principle of geometric optics, it reverse-engineers the possible location of the signal source from multiple receiving points and solves the optimal three-dimensional coordinates of the interference source through least squares optimization.

[0012] Preferably, the electromagnetic threat level assessment is based on the following rules: If the interference signal frequency band covers the operating frequency band of critical equipment, the threat level is increased by one level; if the signal power is greater than the safety threshold, the threat level is increased by one level. If the signal exhibits frequency hopping or spread spectrum characteristics, the threat level is increased by one level; taking into account the above factors, the threat level is divided into three levels: low, medium, and high.

[0013] Preferably, the reconfigurable electromagnetic shielding unit is composed of a flexible substrate, a tunable metamaterial layer, and a driving circuit; The tunable metamaterial layer comprises periodically arranged metal resonant rings and varactor diodes; After receiving the partial shielding command, the driving circuit outputs a corresponding bias voltage to the varactor diode to change its capacitance value, thereby adjusting the equivalent dielectric constant and permeability of the metamaterial layer.

[0014] Preferably, the local shielding command includes the target frequency band range, the required attenuation value, and the effective spatial area; the effective spatial area is a spherical area with a radius of 1 meter centered on the three-dimensional coordinates of the interference source; the reconfigurable electromagnetic shielding unit only activates the unit modules located in the spherical area that are projected onto the inner wall of the shielding room, while the remaining units remain in a normal shielding state.

[0015] The present invention also provides an intelligent control system for an electromagnetic shielding room, comprising: A distributed electromagnetic sensing array is used to collect raw electromagnetic field data inside and at the boundary of a shielded room in real time. The main control processing unit integrates the following: The spatiotemporal alignment module is used to synchronize the original electromagnetic field data in time and unify its spatial coordinates. The electromagnetic situation modeling module is used to construct a three-dimensional dynamic electromagnetic situation model based on aligned data. Anomaly detection module is used to perform multi-scale time-frequency analysis on the electromagnetic situation model and identify interference signals; The interference source localization module is used to perform direction-of-arrival estimation and reverse ray tracing to determine the location of the interference source; The threat assessment module is used to determine the electromagnetic threat level based on the characteristics of interference signals. The blocking policy generation module is used to generate local blocking instructions by calling the blocking policy library based on the threat level; A reconfigurable electromagnetic shielding unit array is deployed on the inner wall of the shielding room to receive the local shielding command and dynamically adjust the electromagnetic shielding performance of the local area. An instruction distribution bus connects the main control processing unit and the reconfigurable electromagnetic shielding unit array, and is used to transmit local shielding instructions.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention constructs a closed-loop intelligent control system that integrates distributed sensing, dynamic modeling, and reconfigurable execution, thereby achieving millisecond-level response, sub-meter-level positioning, and precise isolation of sudden or unknown interference sources in complex electromagnetic environments.

[0017] 2. Compared with traditional static shielding schemes, this invention abandons the one-size-fits-all shielding mode of "full frequency band, full space" and instead adopts a dynamic strategy of "activation on demand and local enhancement", which reduces system energy consumption and heat load.

[0018] 3. By introducing a deep learning-based anomaly detection mechanism and a high-precision ray tracing positioning algorithm, the shortcomings of existing technologies in terms of weak recognition ability and low positioning accuracy for non-cooperative and non-steady-state interference signals are overcome.

[0019] 4. The reconfigurable electromagnetic shielding unit is based on metamaterial design and has the characteristics of wide bandwidth, fast response and high attenuation. It can achieve directional suppression of interference in specific frequency bands without damaging the overall structure of the shielding room.

[0020] 5. The overall system improves the environmental adaptability, anti-interference robustness, and operational reliability of the electromagnetic shielding room in highly sensitive application scenarios such as military communications, precision measurement, and medical imaging. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall technical solution architecture of the present invention; Figure 2 This is a schematic diagram of the core principle framework of the collaborative mechanism of multi-source heterogeneous electromagnetic sensing and adaptive shielding control in this invention; Figure 3 This is a logical flowchart of the distributed electromagnetic sensing array and three-dimensional dynamic electromagnetic situation modeling in this invention. Figure 4 This is a logical framework diagram of the joint processing of interference signal detection, localization, and threat level assessment in this invention; Figure 5 This is a control logic framework diagram of the dynamic activation of the reconfigurable electromagnetic shielding unit and the formation of the local shielding barrier in this invention; Figure 6 This is a schematic diagram of the multi-level interaction relationship and data flow between the main control processing unit and the reconfigurable electromagnetic shielding unit array in this invention. Detailed Implementation

[0022] refer to Figures 1 to 6 This invention provides an intelligent control method and system for electromagnetic shielding rooms, aiming to solve the technical problem of difficulty in quickly locating and isolating sudden or unknown interference sources in complex electromagnetic environments. In existing technologies, electromagnetic shielding rooms mostly adopt static shielding structures, relying on fixed-band filters and preset shielding strategies, which cannot provide real-time responses to dynamically changing electromagnetic environments.

[0023] When unexpected high-frequency, broadband, or frequency-hopping interference signals occur, traditional systems lack the ability to jointly perceive the spatial location, spectral characteristics, and temporal evolution of the interference source. This leads to delayed or ineffective shielding measures, or even secondary interference, severely restricting the reliable operation of highly sensitive electronic equipment in complex electromagnetic environments.

[0024] To overcome the aforementioned shortcomings, this invention proposes an intelligent control method and system based on a collaborative mechanism of multi-source heterogeneous electromagnetic sensing, dynamic spatial modeling, and adaptive shielding regulation. This system utilizes a distributed electromagnetic sensing array deployed on the inner walls and interior space of a shielded room to collect real-time omnidirectional, wide-bandwidth electromagnetic field strength, phase, polarization direction, and time-frequency distribution data; combined with a high-precision spatiotemporal synchronization mechanism, it constructs a three-dimensional dynamic electromagnetic situation map. Based on this, multimodal signal separation and anomaly detection algorithms are used to identify potential interference sources, and sub-meter-level spatial positioning of interference sources is achieved through direction of arrival estimation and reverse ray tracing techniques. Ultimately, based on the spectral characteristics, spatial location, and threat level of the interference source, the reconfigurable electromagnetic shielding units in the corresponding area are dynamically activated to form a local high-attenuation shielding barrier, thereby achieving precise isolation and suppression of the interference source.

[0025] The intelligent control method for the electromagnetic shielding room includes the following steps: S1, real-time acquisition of raw electromagnetic field data inside and at the boundary of the shielded room through a distributed electromagnetic sensor array. The raw electromagnetic field data includes electric field strength, magnetic field strength, signal frequency, phase information, polarization state and arrival time. S2, perform spatiotemporal alignment processing on the original electromagnetic field data to generate an electromagnetic sensing dataset with a unified time reference and spatial coordinates; S3. Based on the electromagnetic sensing dataset, construct a three-dimensional dynamic electromagnetic situation model covering the entire shielded room. S4, perform multi-scale time-frequency analysis and abnormal pattern recognition on the three-dimensional dynamic electromagnetic situation model to detect sudden or unknown interference signals that deviate from the normal electromagnetic background. S5, for the detected interference signal, execute the direction of arrival estimation algorithm and multi-path reverse ray tracing to determine the three-dimensional spatial coordinates of the interference source in the shielded room; S6. Assess the electromagnetic threat level of the interference source based on its three-dimensional spatial coordinates, spectral bandwidth, power intensity, and duration. S7. Based on the electromagnetic threat level, call the preset shielding strategy library to generate a local shielding command for the interference source; S8, the local shielding command is sent to the reconfigurable electromagnetic shielding unit located in the vicinity of the interference source, and its electromagnetic parameters are switched to the high attenuation state of the corresponding frequency band to form a directional shielding barrier.

[0026] In step S1, raw electromagnetic field data inside and at the boundaries of the shielded room are collected in real time using a distributed electromagnetic sensor array. The distributed electromagnetic sensor array consists of a grid of planar electromagnetic sensors disposed on the six inner surfaces of the shielded room and a three-dimensional electromagnetic probe array suspended in the central region of the room.

[0027] The planar electromagnetic sensor grid is arranged at equal intervals, with the distance between adjacent sensors not exceeding 1 / 4 of the wavelength corresponding to the highest frequency in the operating band. This satisfies the Nyquist sampling theorem's sampling requirements for spatial frequencies and avoids spatial aliasing. The three-dimensional electromagnetic probe array consists of three orthogonally arranged dipole antennas, positioned along the X, Y, and Z axes respectively, used to capture electromagnetic vector information in any spatial direction, including the complete polarization state of the electric and magnetic fields.

[0028] Each sensor node incorporates a high dynamic range analog-to-digital converter with a sampling rate greater than 500 MHz and an effective bandwidth covering 0.1 MHz to 40 GHz, ensuring complete capture of wideband interference signals. All sensor nodes are equipped with temperature compensation circuitry and a self-calibration module, automatically performing internal calibration upon each startup or every 30 minutes to eliminate measurement biases caused by temperature drift or device aging. Acquired data is stored in raw complex form, including both real and imaginary parts, preserving complete phase information to provide necessary input for subsequent direction-of-arrival estimation.

[0029] In step S2, the raw electromagnetic field data undergoes spatiotemporal alignment processing. This process includes two sub-processes: time synchronization and spatial coordinate unification. Time synchronization uses the atomic clock built into the main control processing unit as a reference, aligning the sampling times of all sensor nodes to nanosecond-level precision through a precise time protocol. Each sensor node adds a high-precision timestamp with each sampling. Upon receiving this timestamp, the main control processing unit aligns the data streams of all channels accordingly. For time offsets caused by transmission delays, a sliding window cross-correlation algorithm is used for fine-tuning to ensure strict alignment of the signals in each channel on the time axis.

[0030] Regarding spatial coordinate unification, the electromagnetic data collected by each node is mapped to a unified Cartesian coordinate system using the pre-calibrated three-dimensional spatial position parameters of the sensors at the factory. The origin of this coordinate system is located at the geometric center of the shielded room, and the X, Y, and Z axes are parallel to the length, width, and height of the shielded room, respectively. For sensor data with different sampling rates, cubic spline interpolation is used to resample to a unified frequency, which is set to 500 MHz to balance computational efficiency and signal fidelity. The data after spatiotemporal alignment processing forms a structured electromagnetic sensing dataset. Each frame of data contains electromagnetic field vectors of N sensor nodes at M time points, where N is the total number of sensors and M is the number of sampling points within the time window.

[0031] In step S3, a three-dimensional dynamic electromagnetic situation model covering the entire shielded room is constructed based on the electromagnetic sensing dataset. Specifically, the spatiotemporally aligned electromagnetic sensing data is sliced ​​into time windows, each 10 milliseconds long, corresponding to 500 sampling points. For the data within each time window, the internal space of the shielded room is divided into cubic units with a side length of 10 centimeters, i.e., voxels, using a voxelization method.

[0032] Assuming the shielded room has internal dimensions of 5m × 4m × 3m, the entire space is divided into 5000 voxels. For each voxel, the electromagnetic signals collected by all sensor nodes falling within its spatial range within the given time window are aggregated. The comprehensive electromagnetic characteristics of the voxel within the current time window are calculated using a weighted average or maximum likelihood estimation method. These comprehensive electromagnetic characteristics include spectral density, polarization ellipse parameters (major axis, minor axis, tilt angle), coherence indices (phase consistency between signals), and instantaneous power. Each voxel's characteristics are represented as a seven-dimensional vector.

[0033] The sequence of voxel feature vectors from 100 consecutive time windows (i.e., 1 second) is stacked to form a four-dimensional electromagnetic situation tensor, whose dimensions are (number of time steps, number of X voxels, number of Y voxels, number of Z voxels, and feature dimension). This four-dimensional tensor fully describes the dynamic evolution of the electromagnetic field inside the shielded room in the spatiotemporal domain, providing a high-dimensional input for subsequent anomaly detection.

[0034] In step S4, multi-scale time-frequency analysis and anomaly pattern identification are performed on the three-dimensional dynamic electromagnetic situation model. First, a short-time Fourier transform is applied to the four-dimensional electromagnetic situation tensor along the time dimension, with a window length of 10 milliseconds and an overlap rate of 50%, to extract the time-frequency spectrum of each voxel unit in the frequency band from 0.1 MHz to 40 GHz. Each voxel outputs a two-dimensional time-frequency matrix, with time on the horizontal axis and frequency on the vertical axis, and the element value being the energy density at the corresponding time-frequency point.

[0035] Subsequently, the temporal spectra of all voxels are concatenated into a high-dimensional input tensor and fed into a pre-trained convolutional autoencoder network. This convolutional autoencoder consists of an encoder and a decoder. The encoder contains four convolutional layers, each followed by batch normalization and a ReLU activation function to progressively compress the input features. The decoder contains four transposed convolutional layers, reconstructing the original temporal spectra layer by layer. The network is trained unsupervised on a large amount of historical data collected under normal electromagnetic conditions, with the goal of minimizing reconstruction error.

[0036] During the inference phase, the reconstruction error of the time-spectrum for each voxel is calculated and defined as the mean square error between the original input and the network output. A dynamic threshold is set, determined based on the 99th quantile of the historical error distribution. When the reconstruction error of a voxel unit exceeds this threshold for three consecutive time windows, it is determined that there is abnormal electromagnetic activity in that voxel region and it is marked as a candidate interference region. This mechanism can effectively identify sudden, unknown, or non-steady-state interference signals, which fail to be learned in normal mode, leading to reconstruction failure.

[0037] In step S5, for the detected interference signal, a direction of arrival estimation algorithm and multipath reverse ray tracing are performed to determine the three-dimensional spatial coordinates of the interference source in the shielded room.

[0038] First, the corresponding raw electromagnetic field data is extracted from the candidate interference regions, focusing on the three-channel complex signals from a stereo electromagnetic probe array. A multiple signal classification algorithm is used for direction-of-arrival (DOA) estimation. Based on the orthogonality principle of the signal and noise subspaces, this algorithm constructs the received signal covariance matrix, performs eigenvalue decomposition, and separates the signal and noise subspaces. By searching all possible incident directions, the projected energy of the steering vector and the noise subspace is calculated; the direction corresponding to the minimum energy is the incident direction of the interference signal. This yields the elevation and azimuth angles of the interference signal.

[0039] Subsequently, multi-path reverse ray tracing is performed. This process is based on an electromagnetic reflection coefficient database of the shielded room's inner walls, which was pre-obtained through full-wave simulation or field measurements, recording the reflection amplitude and phase of each wall material at different frequencies and incident angles. Combining geometric optics principles, rays are extended backward along the estimated incident direction from the receiving point of the stereo electromagnetic probe array, considering first and second reflection paths. For each possible path, its theoretical arrival time and phase are calculated and compared with measured data. Using a least-squares optimization method, the three-dimensional coordinates of the interference source that minimize the sum of squared prediction errors for all paths are determined. The positioning results are output in meters, with an accuracy better than 30 centimeters.

[0040] In step S6, the electromagnetic threat level of the interference source is assessed based on its three-dimensional spatial coordinates, spectral bandwidth, power intensity, and duration.

[0041] Threat level assessment is based on the following rules: if the interference signal frequency band covers the operating frequency band of critical equipment, the threat level is increased by one level; if the signal power exceeds the safety threshold of 20 dB / µV / m, the threat level is increased by one level; if the signal exhibits frequency hopping or spread spectrum characteristics, the threat level is increased by one level. The operating frequency band information of critical equipment is pre-stored in the equipment configuration table of the main control processing unit, including the center frequency and bandwidth. Signal power is calculated from the maximum electric field strength within a voxel.

[0042] Frequency hopping or spread spectrum characteristics are determined by analyzing the rate of change of the center frequency of the interference signal within a continuous time window. If the frequency hopping interval is less than 10 milliseconds and the hopping range is greater than 10 MHz, it is identified as a frequency hopping signal. Considering the above factors, the initial threat level is set to low, and it is increased by one level for each condition met, up to a maximum of high. The threat level is divided into three levels: low, medium, and high, each corresponding to a different shielding response strategy.

[0043] In step S7, based on the electromagnetic threat level, a preset shielding strategy library is invoked to generate a local shielding command targeting the interference source. The shielding strategy library is stored in the non-volatile memory of the main control processing unit and contains shielding parameter templates corresponding to different threat levels. For low threat levels, only basic shielding in the vicinity of the interference source is activated; for medium threat levels, the shielding frequency band width and attenuation value are increased; for high threat levels, omnidirectional enhanced shielding is activated and the duration of action is extended.

[0044] The local shielding command includes the target frequency band range, the required attenuation value, and the effective spatial area. The effective spatial area is a spherical region with a radius of 1 meter, centered on the three-dimensional coordinates of the interference source. This spherical region intersects with the six inner surfaces of the shielding room, forming several projection surfaces. The system calculates a list of reconfigurable electromagnetic shielding unit numbers covered by these projection surfaces.

[0045] In step S8, the local shielding command is sent to the reconfigurable electromagnetic shielding unit located in the vicinity of the interference source, controlling its electromagnetic parameters to switch to a high attenuation state in the corresponding frequency band, forming a directional shielding barrier. The reconfigurable electromagnetic shielding unit consists of a flexible substrate, a tunable metamaterial layer, and a driving circuit. The tunable metamaterial layer includes periodically arranged metal resonant rings and varactor diodes. After receiving the local shielding command, the driving circuit analyzes the target frequency band and attenuation value, looks up the corresponding bias voltage value in a table, and outputs the voltage to the varactor diode.

[0046] The varactor diode capacitance continuously varies with the bias voltage, thereby adjusting the equivalent dielectric constant and permeability of the metamaterial layer to generate an insertion loss greater than 60 dB in the target interference frequency band. Each reconfigurable electromagnetic shielding unit has an independent address code, and the command distribution bus adopts the Controller Area Network protocol to ensure that local shielding commands are accurately delivered to the designated unit. The shielding action is completed within 5 milliseconds after the command is issued, forming a dynamic, localized, and highly attenuated electromagnetic barrier, effectively isolating interference sources.

[0047] The intelligent control system of the electromagnetic shielding room includes a distributed electromagnetic sensor array, a main control processing unit, a reconfigurable electromagnetic shielding unit array, and an instruction distribution bus.

[0048] As mentioned earlier, the distributed electromagnetic sensor array is responsible for acquiring raw data.

[0049] The main control processing unit adopts a heterogeneous computing architecture, which includes a central processing unit, a field-programmable gate array, and a neural network accelerator.

[0050] The central processing unit (CPU) runs a real-time operating system, responsible for task scheduling, logic control, and communication management. A field-programmable gate array (FPGA) enables high-speed signal preprocessing, spatiotemporal alignment, and time-frequency transformation, with a processing latency of less than 1 millisecond. A neural network accelerator is dedicated to running a convolutional autoencoder network for anomaly detection, capable of processing 1000 voxels of time-frequency data per second. A reconfigurable electromagnetic shielding unit array is deployed on the six inner surfaces of the shielded room, each unit measuring 20 cm × 20 cm, seamlessly spliced ​​together to cover the entire inner wall. The command distribution bus is a dual-redundant controller area network (CLAN) bus with a baud rate set to 1 megabit per second, ensuring the reliability and timeliness of command transmission. The entire system forms a closed-loop intelligent control circuit, with the entire process from perception to decision-making to execution taking less than 50 milliseconds, achieving millisecond-level response to sudden interference.

[0051] The drive circuit of the reconfigurable electromagnetic shielding unit includes a digital-to-analog converter (DAC), a voltage amplifier, and a protection diode. The DAC receives digital control words from the main control processing unit and converts them into analog voltage signals; the voltage amplifier amplifies this signal to a range of 0 to 30 volts to drive the varactor diode; and the protection diode prevents reverse voltage damage to the device. The resonant frequency of the metamaterial layer and the capacitance value of the varactor diode satisfy the following relationship: ; The resonant frequency, This is the equivalent inductance of the metal resonant ring. For varactor diodes under bias voltage The capacitance value is adjusted. Continuously tuned It covers the range from 0.5 GHz to 18 GHz.

[0052] Another formula is used to calculate the coherence index of electromagnetic signals within a voxel. : ; The cross power spectral density of the two sensor signals. and For their respective power spectral densities, The value ranges from 0 to 1, and the closer the value is to 1, the stronger the signal coherence.

[0053] In anomaly detection, reconstruction error Defined as:

[0054] This is the original time spectrum. The result of autoencoder reconstruction. For time points, This represents the number of frequency points. It is an index of a specific time point. It is an index of the frequency point.

[0055] In interference source localization, the least squares optimization objective function is: ; The coordinates of the interference source to be determined are: This represents the number of ray paths. For the first Measured arrival parameters (time, phase) for each path. Based on Predicted parameters This is the path weight, which is usually proportional to the signal strength.

[0056] During system operation, the main control processing unit continuously monitors the status of all modules. If a sensor node fails, the system automatically resets its data to zero and compensates through interpolation with neighboring nodes; if a reconfigurable shielding unit responds abnormally, the system records the fault address and switches to a backup unit. All operation logs are stored in real time, supporting post-event analysis and strategy optimization. Through the above methods and systems, this invention achieves intelligent sensing, localization, and dynamic suppression of complex electromagnetic interference, improving the environmental adaptability and anti-interference capability of the electromagnetic shielding room.

[0057] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0058] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent control method for an electromagnetic shielding room, characterized in that, include: The raw electromagnetic field data inside and at the boundary of the shielded room are collected in real time by a distributed electromagnetic sensor array. The raw electromagnetic field data includes electric field strength, magnetic field strength, signal frequency, phase information, polarization state and arrival time. The original electromagnetic field data is spatiotemporally aligned to generate an electromagnetic sensing dataset with a unified time reference and spatial coordinates. Based on the electromagnetic sensing dataset, a three-dimensional dynamic electromagnetic situation model covering the entire shielded room is constructed. Multi-scale time-frequency analysis and abnormal pattern recognition are performed on the three-dimensional dynamic electromagnetic situation model to detect sudden or unknown interference signals that deviate from the normal electromagnetic background. For the detected interference signal, a direction of arrival estimation algorithm and multipath reverse ray tracing are performed to determine the three-dimensional spatial coordinates of the interference source in the shielded room; The electromagnetic threat level of the interference source is assessed based on its three-dimensional spatial coordinates, spectral bandwidth, power intensity, and duration. Based on the electromagnetic threat level, a preset shielding strategy library is invoked to generate a local shielding command targeting the interference source; The local shielding command is sent to the reconfigurable electromagnetic shielding unit located in the vicinity of the interference source, and its electromagnetic parameters are controlled to switch to a high attenuation state in the corresponding frequency band to form a directional shielding barrier.

2. The intelligent control method for an electromagnetic shielding room according to claim 1, characterized in that, The distributed electromagnetic sensing array is composed of a grid of planar electromagnetic sensors set on the six inner surfaces of the shielded room and a three-dimensional electromagnetic probe array suspended in the central area of ​​the room. The planar electromagnetic sensor grid is arranged at equal intervals, with the distance between adjacent sensors not exceeding 1 / 4 of the wavelength corresponding to the highest frequency in the operating frequency band; the three-dimensional electromagnetic probe array consists of three orthogonally arranged dipole antennas, used to capture electromagnetic vector information in any direction in space.

3. The intelligent control method for an electromagnetic shielding room according to claim 2, characterized in that, The spatiotemporal alignment process includes: synchronizing the sampling times of all sensor nodes at the nanosecond level based on the master control clock; Using the pre-calibrated spatial position parameters of the sensors, the electromagnetic data collected by each node is mapped to a unified Cartesian coordinate system; for sensor data with different sampling rates, the cubic spline interpolation method is used to resample to a unified frequency.

4. The intelligent control method for an electromagnetic shielding room according to claim 3, characterized in that, The construction of the three-dimensional dynamic electromagnetic situation model specifically includes: slicing the spatiotemporally aligned electromagnetic sensing data according to a time window. For the data within each time window, the shielded room space is divided using a voxelization method; For each voxel unit, the amplitude, frequency, and phase information of all electromagnetic signals falling within it are aggregated to form a multidimensional electromagnetic feature vector containing spectral density, polarization ellipse parameters, and coherence indices; the voxel feature vector sequences of continuous time windows are stacked to form a four-dimensional electromagnetic state tensor.

5. The intelligent control method for an electromagnetic shielding room according to claim 4, characterized in that, The multi-scale time-frequency analysis and anomaly pattern recognition include: The four-dimensional electromagnetic situation tensor is subjected to a short-time Fourier transform along the time dimension to extract the time spectrum of each voxel unit; the time spectrum is input into a pre-trained convolutional autoencoder network to calculate the reconstruction error; when the reconstruction error of a voxel unit is greater than a preset threshold in three consecutive time windows, it is determined that there is abnormal electromagnetic activity in the voxel region.

6. The intelligent control method for an electromagnetic shielding room according to claim 5, characterized in that, The direction-of-arrival estimation algorithm employs a multi-signal classification algorithm, utilizing the phase difference data of a three-dimensional electromagnetic probe array to calculate the incident elevation and azimuth angles of the interference signal. The multipath reverse ray tracing is based on the electromagnetic reflection coefficient database of the inner wall of the shielded room. Combined with the principle of geometric optics, it reverse-engineers the possible location of the signal source from multiple receiving points and solves the optimal three-dimensional coordinates of the interference source through least squares optimization.

7. The intelligent control method for an electromagnetic shielding room according to claim 6, characterized in that, The electromagnetic threat level assessment is based on the following rules: If the interference signal frequency band covers the operating frequency band of critical equipment, the threat level is increased by one level; if the signal power is greater than the safety threshold, the threat level is increased by one level. If the signal exhibits frequency hopping or spread spectrum characteristics, the threat level is increased by one level; taking into account the above factors, the threat level is divided into three levels: low, medium, and high.

8. The intelligent control method for an electromagnetic shielding room according to claim 7, characterized in that, The reconfigurable electromagnetic shielding unit consists of a flexible substrate, a tunable metamaterial layer, and a driving circuit. The tunable metamaterial layer comprises periodically arranged metal resonant rings and varactor diodes; After receiving the partial shielding command, the driving circuit outputs a corresponding bias voltage to the varactor diode to change its capacitance value, thereby adjusting the equivalent dielectric constant and permeability of the metamaterial layer.

9. The intelligent control method for an electromagnetic shielding room according to claim 8, characterized in that, The local shielding command includes the target frequency band range, the required attenuation value, and the effective spatial area; the effective spatial area is a spherical area with a radius of 1 meter centered on the three-dimensional coordinates of the interference source; the reconfigurable electromagnetic shielding unit only activates the unit modules located in the spherical area that are projected onto the inner wall of the shielding room, while the remaining units remain in a normal shielding state.

10. An intelligent control system for an electromagnetic shielding room, characterized in that, include: A distributed electromagnetic sensing array is used to collect raw electromagnetic field data inside and at the boundary of a shielded room in real time. The main control processing unit integrates the following: The spatiotemporal alignment module is used to synchronize the original electromagnetic field data in time and unify its spatial coordinates. The electromagnetic situation modeling module is used to construct a three-dimensional dynamic electromagnetic situation model based on aligned data. Anomaly detection module is used to perform multi-scale time-frequency analysis on the electromagnetic situation model and identify interference signals; The interference source localization module is used to perform direction-of-arrival estimation and reverse ray tracing to determine the location of the interference source; The threat assessment module is used to determine the electromagnetic threat level based on the characteristics of interference signals. The blocking policy generation module is used to generate local blocking instructions by calling the blocking policy library based on the threat level; A reconfigurable electromagnetic shielding unit array is deployed on the inner wall of the shielding room to receive the local shielding command and dynamically adjust the electromagnetic shielding performance of the local area. An instruction distribution bus connects the main control processing unit and the reconfigurable electromagnetic shielding unit array, and is used to transmit local shielding instructions.

Images