A protection system for a timekeeping isolation device

Abstract

The application discloses a protection system for a time service isolation device, which is characterized by the following steps: firstly, a feature extraction unit is used to perform feature engineering on the preprocessed signal to extract the Doppler frequency offset, pseudo-code phase of the satellite signal, the time sequence fluctuation feature of the device state signal and the spectrum peak key feature of the electromagnetic signal; then, a multi-modal fusion unit is used to fuse the satellite signal feature, the device state feature and the electromagnetic environment feature according to a deep learning model of an attention mechanism; finally, an interference / deception recognition unit is used to identify the specific threat type by using a pre-training model. Based on the above, the application realizes the accurate recognition of complex interference and deception signals by fusing the multi-dimensional features of the satellite signal, the device state and the electromagnetic environment through a multi-modal AI recognition module and combining the filtering, noise reduction and weak signal enhancement processing of the signal preprocessing and enhancement module.

Description

Technical Field

[0001] This invention relates to the field of time synchronization isolation equipment technology, specifically to a protection system for time synchronization isolation equipment. Background Technology

[0002] With the widespread application of satellite navigation technology in fields such as power, finance, and mobile communications, the security and stability of time synchronization have become crucial for ensuring the operation of core business processes. While time synchronization isolation equipment, as a core device protecting satellite time synchronization systems from interference and spoofing, can achieve basic signal isolation, compatibility, and monitoring alarms, it still has many shortcomings in practical applications: 1. Insufficient interference identification accuracy: For example, in power dispatching systems, existing time synchronization isolation equipment has a low recognition rate for new composite interference signals (which simultaneously contain electromagnetic interference and deception components that simulate satellite signals), resulting in time synchronization deviations exceeding millisecond levels, which directly affects the accuracy of power grid load distribution and fault location.

[0003] 2. High mode switching latency: In high-frequency financial trading scenarios, when the device detects a spoofed signal targeting the timestamp, the mode switching response time can be as long as hundreds of microseconds, causing transaction time synchronization errors and posing potential transaction and compliance risks to financial institutions.

[0004] 3. Equipment compatibility limitations: Existing time synchronization isolation devices are mostly designed for specific manufacturers and models of time synchronization devices. When upgrading the system or replacing the equipment, large-scale modifications to the original architecture are required (such as changing interfaces and re-debugging), which increases deployment costs and time (for example, a provincial data center spent 2 weeks modifying the time synchronization isolation system due to changing the time synchronization device manufacturer, during which the time synchronization reliability of some services decreased).

[0005] Based on the above, a protection system for timing isolation devices is invented. Summary of the Invention

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A protection system for a time synchronization isolation device, comprising: The signal acquisition module is used to acquire satellite navigation signals, timing equipment operating status signals, and environmental electromagnetic signals in real time. The signal preprocessing and enhancement module is used to filter, reduce noise, and enhance weak signals of the acquired satellite navigation signals and equipment status signals. The multimodal AI recognition module is used to perform multi-dimensional analysis of the collected signals based on the pre-trained deep learning model, including signal characteristics, timing patterns, and electromagnetic environment, to identify interference signals, deceptive signals, and potential abnormal patterns. The adaptive switching module is used to automatically switch the working mode based on the results of the built-in high-speed switching circuit and the multimodal AI recognition module. The autonomous timing backup module is used to automatically switch to local autonomous timing mode when the satellite signal is completely interfered with / spoofed and cannot be recovered, based on the built-in high-precision local clock source, to ensure the continuity of time synchronization in extreme scenarios. The deep compatibility module is used to automatically identify the type and communication protocol of the access timing device, enabling seamless integration with timing devices from different manufacturers and models; and it supports hot-swappable deployment without interrupting the original timing link. The remote intelligent decision-making module is used to receive alarm information and equipment operation data from the multimodal AI recognition module, and perform trend analysis and threat prediction through cloud AI algorithms; it also supports automatic updating of the recognition model, remote adjustment of equipment parameters, and provides a manual remote intervention interface. The security audit and traceability module records all critical operations and generates traceable security audit logs; at the same time, it uses blockchain technology to store evidence of critical logs. The energy efficiency management module is used to monitor the energy consumption and load of each module in real time, dynamically adjust the computing power allocation and power supply strategy, and achieve the optimal energy efficiency of the system.

[0007] As a preferred embodiment of the protection system for a time synchronization isolation device according to the present invention, the signal acquisition module includes: The satellite signal receiving unit is used to acquire satellite navigation radio frequency signals in real time through a multi-band antenna. The equipment status signal acquisition unit is used to collect the operating parameters of the timing device through the sensor interface and monitor the health status of the device in real time. The environmental electromagnetic signal monitoring unit is used to integrate broadband electromagnetic sensors to collect electromagnetic interference signals from the environment in which the equipment is deployed.

[0008] As a preferred embodiment of the protection system for timing isolation equipment described in this invention, the signal preprocessing and enhancement module includes: The adaptive filtering unit is used to dynamically filter periodic clutter in satellite signals according to the minimum mean square error algorithm, and filter out frequency band signals that are not related to time synchronization. The noise reduction unit is used to process random noise in the device status signal using wavelet transform algorithm to improve the signal-to-noise ratio. The weak signal enhancement unit is used to enhance the signal strength of weak satellite signals in obstructed scenarios through coherent accumulation and signal reconstruction algorithms, ensuring that effective time information can still be extracted in extreme environments.

[0009] As a preferred embodiment of the protection system for timing isolation equipment described in this invention, the multimodal AI recognition module includes: The feature extraction unit is used to perform feature engineering on the preprocessed signal, extracting Doppler frequency offset and pseudocode phase of satellite signals, time-series fluctuation characteristics of equipment status signals, and key spectral peak characteristics of electromagnetic signals. The multimodal fusion unit is used to fuse satellite signal features, equipment status features, and electromagnetic environment features based on a deep learning model with an attention mechanism. The interference / deception identification unit is used to identify specific threat types through a pre-trained model; The threat level assessment unit is used to classify threats into three levels: low, medium, and high, based on the intensity of interference, duration, and impact on timing accuracy.

[0010] As a preferred embodiment of the protection system for a time synchronization isolation device according to the present invention, the adaptive switching module includes: The mode determination unit is used to automatically determine the target's working mode based on the threat level assessment results, and supports manual preset switching thresholds. The high-speed switching circuit unit is used to design switching logic based on FPGA hardware acceleration technology to achieve uninterrupted switching of signal links. The switching verification unit is used to verify the synchronization accuracy of the signal after the switching is completed by comparing the timestamp. If the deviation exceeds the threshold, a second switching or alarm is triggered to avoid synchronization abnormalities caused by switching failure.

[0011] As a preferred embodiment of the protection system for timing isolation equipment described in this invention, the autonomous timing backup module includes: A local high-precision clock source unit is used to house a miniaturized rubidium atomic clock or a high-stability crystal oscillator, serving as a core backup source when satellite signals fail, ensuring time accuracy during independent operation; The clock calibration unit is used to calibrate the local clock in real time using the satellite timestamp when the satellite signal is normal, eliminating accumulated errors; and after the satellite fails, it predicts clock drift based on historical calibration data to maintain accuracy. The switching trigger unit is used to monitor satellite signal availability, automatically trigger the switch from satellite time synchronization to local autonomous time synchronization, and supports manual forced activation of backup mode.

[0012] As a preferred embodiment of the protection system for timing isolation equipment described in this invention, the deep compatibility module includes: The interface identification unit is used to automatically identify the type and interface parameters of the access timing device through hardware interface scanning and protocol handshake testing. The protocol conversion unit is used to integrate the conversion logic of mainstream time synchronization protocols to achieve seamless adaptation between different protocols; The parameter adaptive unit is used to automatically adjust the parameters of this system according to the performance parameters of the access device to avoid synchronization errors caused by device incompatibility. The hot-swap control unit supports hot-swap operations of timing devices. Through circuit protection design, it ensures that the original timing link is not interrupted during the insertion and removal process, thereby reducing system downtime.

[0013] As a preferred embodiment of the protection system for a time synchronization isolation device described in this invention, the remote intelligent decision-making module includes: The data transmission unit is used to upload the operating data of the local device to the cloud platform through an encrypted transmission channel, and supports edge computing preprocessing. The cloud-based AI analysis unit is used to perform trend analysis on historical data of devices in multiple regions based on the federated learning framework to predict potential threats; The model update unit is used to automatically push incremental update packages for the multimodal AI recognition model based on cloud analysis results, and supports offline update mode. The remote control unit provides an encrypted remote operation interface, enabling administrators to remotely adjust device parameters, manually trigger mode switching, and view real-time signal waveforms.

[0014] As a preferred embodiment of the protection system for timing isolation equipment described in this invention, the security audit and tracing module includes: The full log recording unit is used to record all critical operations; The blockchain evidence storage unit is used to store key logs on the blockchain for evidence storage. The source tracing analysis unit is used to reverse analyze the location and type of interference source by analyzing the spectral characteristics, occurrence time, and geographical distribution of interference signals, thus assisting in locating the source of the threat. The compliance reporting unit is used to automatically generate audit reports that comply with the standards of the power and financial industries.

[0015] As a preferred embodiment of the protection system for timing isolation equipment described in this invention, the energy efficiency management module includes: The energy consumption monitoring unit is used to monitor the instantaneous power consumption of each module in real time through current sensors, calculate the energy consumption per unit time, and generate an energy consumption curve. The computing power scheduling unit is used to automatically reduce the computing power allocation of the multimodal AI recognition module in low-load scenarios to reduce ineffective energy consumption; and to dynamically increase computing power to ensure real-time performance under high load. The power management unit is designed with a wide power adaptive range, supports sleep mode and fast wake-up, and adapts to the power supply requirements of different scenarios. The energy efficiency analysis unit is used to analyze the relationship between energy consumption and business load, and output energy efficiency optimization suggestions to help users reduce long-term operating costs.

[0016] Compared with existing technologies: 1. By integrating satellite signals, equipment status, and electromagnetic environment features through a multi-modal AI recognition module, and combining filtering, noise reduction, and weak signal enhancement processing of the signal preprocessing and enhancement module, it can accurately identify complex interference and deceptive signals.

[0017] 2. Through the coordinated operation of the high-speed switching circuit of the adaptive switching module and the mode determination unit, it can achieve rapid mode switching and ensure the continuity of time synchronization of the timing device in interference scenarios.

[0018] 3. Through the cooperation of interfaces, protocols, and parameters, the deep compatibility module enables seamless compatibility with multiple manufacturers and models of timing devices, without requiring large-scale modifications to the original timing system architecture. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall framework of the present invention; Figure 2 This is a schematic diagram of the signal acquisition module framework of the present invention; Figure 3 This is a schematic diagram of the signal preprocessing and enhancement module framework of the present invention; Figure 4 This is a schematic diagram of the multimodal AI recognition module framework of the present invention; Figure 5 This is a schematic diagram of the adaptive switching module framework of the present invention; Figure 6 This is a schematic diagram of the autonomous time synchronization backup module framework of the present invention; Figure 7 This is a schematic diagram of the deep compatibility module framework of the present invention; Figure 8 This is a schematic diagram of the remote intelligent decision-making module framework of the present invention; Figure 9 This is a schematic diagram of the security audit and traceability module framework of the present invention; Figure 10 This is a schematic diagram of the energy efficiency management module framework of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0021] This invention provides a protection system for time synchronization isolation devices. Please refer to [link / reference]. Figures 1-10 ,include: The signal acquisition module is used to acquire satellite navigation signals, timing equipment operating status signals, and environmental electromagnetic signals in real time. The signal preprocessing and enhancement module is used to filter, reduce noise, and enhance weak signals of the acquired satellite navigation signals and equipment status signals. The multimodal AI recognition module is used to perform multi-dimensional analysis of the collected signals based on the pre-trained deep learning model, including signal characteristics, timing patterns, and electromagnetic environment, to identify interference signals, deceptive signals, and potential abnormal patterns. The adaptive switching module is used to automatically switch the working mode based on the results of the built-in high-speed switching circuit and the multimodal AI recognition module. The autonomous timing backup module is used to automatically switch to local autonomous timing mode when the satellite signal is completely interfered with / spoofed and cannot be recovered, based on the built-in high-precision local clock source, to ensure the continuity of time synchronization in extreme scenarios. The deep compatibility module is used to automatically identify the type and communication protocol of the access timing device, enabling seamless integration with timing devices from different manufacturers and models; and it supports hot-swappable deployment without interrupting the original timing link. The remote intelligent decision-making module is used to receive alarm information and equipment operation data from the multimodal AI recognition module, and perform trend analysis and threat prediction through cloud AI algorithms; it also supports automatic updating of the recognition model, remote adjustment of equipment parameters, and provides a manual remote intervention interface. The security audit and traceability module records all critical operations and generates traceable security audit logs; at the same time, it uses blockchain technology to store evidence of critical logs. The energy efficiency management module is used to monitor the energy consumption and load of each module in real time, dynamically adjust the computing power allocation and power supply strategy, and achieve the optimal energy efficiency of the system.

[0022] The signal acquisition module includes: The satellite signal receiving unit is used to acquire satellite navigation radio frequency signals in real time through a multi-band antenna (supporting GPS, BeiDou, GLONASS, etc.); The equipment status signal acquisition unit is used to acquire the operating parameters of the timing device through sensor interfaces (such as RS485, Ethernet), including time synchronization deviation (nanosecond level), equipment temperature, power supply voltage, interface communication rate, etc., and monitor the health status of the device in real time. The environmental electromagnetic signal monitoring unit is used to integrate a wideband electromagnetic sensor to collect electromagnetic interference signals (frequency range 10kHz-6GHz) from the environment in which the equipment is deployed.

[0023] The signal preprocessing and enhancement module includes: The adaptive filtering unit is used to dynamically filter periodic clutter (such as 50Hz harmonic interference from power transformers) in satellite signals according to the minimum mean square error (LMS) algorithm, and filter out frequency band signals that are not related to time synchronization. The noise reduction unit is used to perform noise reduction processing on random noise (such as interface transmission jitter) in the device status signal using wavelet transform algorithm, thereby improving the signal-to-noise ratio (SNR), for example, reducing the noise of financial transaction timestamp signals by more than 30%. The weak signal enhancement unit is used to enhance the signal strength of weak satellite signals in obstructed scenarios (such as urban canyons and shielded environments in computer rooms) through coherent accumulation and signal reconstruction algorithms, ensuring that effective time information can still be extracted in extreme environments.

[0024] The multimodal AI recognition module includes: The feature extraction unit is used to perform feature engineering on the preprocessed signal, extracting key features such as Doppler frequency offset and pseudocode phase of satellite signals, time-series fluctuation characteristics of equipment status signals, and spectral peak values ​​of electromagnetic signals. The multimodal fusion unit is used to fuse satellite signal features, device status features, and electromagnetic environment features based on deep learning models with attention mechanisms (such as CNN+Transformer architecture) to solve the limitations of single-signal dimension recognition (such as the difficulty in distinguishing between natural attenuation and human deception based solely on satellite signals). The interference / spoofing identification unit is used to identify specific threat types, such as forwarding spoofing, generative false signals, and electromagnetic pulse interference, through a pre-trained model (training samples include 200+ types of interference / spoofing scenarios), with an identification accuracy of ≥99.9%. The threat level assessment unit is used to classify threats into three levels—low (≤10ns deviation), medium (10ns-1μs deviation), and high (≥1μs deviation)—based on interference intensity, duration, and impact on timing accuracy, providing a basis for subsequent handover strategies.

[0025] The adaptive switching module includes: The mode determination unit is used to automatically determine the target's operating mode (normal mode: relying on satellite signals; emergency mode: switching to backup satellite source; autonomous mode: enabling local clock) based on the threat level assessment results, and supports manual preset switching thresholds (such as "switching at 100ns deviation" in financial scenarios). The high-speed switching circuit unit is used to design switching logic based on FPGA hardware acceleration technology to achieve uninterrupted switching of signal links and control the switching delay to ≤50 microseconds (the traditional solution is hundreds of microseconds), meeting the real-time requirements of high-frequency trading and power relay protection. The switching verification unit is used to verify the synchronization accuracy of the signal after the switching is completed by comparing the timestamp (calibrating with the backup clock source). If the deviation exceeds the threshold (e.g., 5ns), a second switching or alarm is triggered to avoid synchronization abnormalities caused by switching failure.

[0026] The autonomous time synchronization backup module includes: A local high-precision clock source unit is used to house a miniaturized rubidium atomic clock (daily drift ≤1e-12) or a high-stability crystal oscillator (daily drift ≤5e-9) as the core backup source when the satellite signal fails, ensuring time accuracy when operating independently; The clock calibration unit is used to calibrate the local clock in real time using the satellite timestamp (calibration period ≤ 1 minute) when the satellite signal is normal, eliminating accumulated errors; and after the satellite fails, it predicts clock drift based on historical calibration data to maintain accuracy. The switching trigger unit is used to monitor satellite signal availability (such as no effective signal for 3 consecutive seconds or identification as a high-level spoofing), automatically trigger the switch from satellite time synchronization to local autonomous time synchronization, and supports manual forced activation of backup mode.

[0027] The deep compatibility module includes: The interface identification unit is used to automatically identify the type of access timing device (such as Beidou receiver, NTP server, PTP master clock) and interface parameters (such as baud rate and level standard) through hardware interface scanning (supporting BNC, SFP, RJ45, etc.) and protocol handshake testing. The protocol conversion unit is used to integrate the conversion logic of mainstream time synchronization protocols such as NTPv4, PTPv2, IRIG-B, and 1588, so as to achieve seamless adaptation between different protocols (such as converting PTP signals into IRIG-B codes for output to traditional power equipment). The parameter adaptive unit is used to automatically adjust the signal output frequency, verification period and other parameters of this system according to the performance parameters of the access device (such as maximum synchronization deviation and communication delay) to avoid synchronization errors caused by device mismatch. The hot-swap control unit supports hot-swap operations of timing devices. Through circuit protection design (such as overcurrent protection and signal isolation), it ensures that the original timing link is not interrupted during the insertion and removal process, thereby reducing system downtime.

[0028] The remote intelligent decision-making module includes: The data transmission unit is used to upload the operating data (interference logs, switching records, energy consumption data) of the local device to the cloud platform through an encrypted transmission channel (such as TLS1.3), and supports edge computing preprocessing (reducing transmission bandwidth). The cloud-based AI analysis unit is used to perform trend analysis on historical data of devices in multiple regions based on the federated learning framework, and to predict potential threats (such as the frequent occurrence of interference signals of a specific spectrum in a certain region recently, and to push out protection strategies in advance). The model update unit is used to automatically push incremental update packages of multimodal AI recognition models based on cloud analysis results (avoiding the time-consuming full update), and supports offline update mode (storing update packages when there is no network and automatically installing them when connected to the network). The remote control unit provides an encrypted remote operation interface, allowing administrators to remotely adjust device parameters (such as switching thresholds and filter strength), manually trigger mode switching, and view real-time signal waveforms.

[0029] The security audit and tracing module includes: The full log recording unit is used to record all key operations, including signal acquisition data (one record every 10ms), AI recognition results, mode switching actions, parameter modification records, remote control commands, etc. The log retention period is ≥1 year (to meet financial regulatory requirements). The blockchain evidence storage unit is used to store key logs (such as the timing status corresponding to the timestamp of a financial transaction) on the blockchain to ensure that the logs are tamper-proof and traceable, thus solving the compliance risk of "timestamp forgery". The source tracing analysis unit is used to reverse analyze the location (accuracy ≤1km) and type (such as malicious devices, natural interference) of the interference source by analyzing the spectral characteristics, occurrence time, and geographical distribution (multi-device collaboration) of the interference signal, thus assisting in locating the source of the threat. The compliance reporting unit is used to automatically generate audit reports that comply with the power (DL / T1870) and financial (ISO8601) industry standards. It includes time synchronization accuracy statistics, abnormal event handling records, compliance verification results, etc., and supports one-click export.

[0030] The energy efficiency management module includes: The energy consumption monitoring unit is used to monitor the instantaneous power consumption of each module (such as the GPU of the AI ​​recognition module and the high-speed switching circuit) in real time through current sensors, calculate the energy consumption per unit time (such as kWh / day), and generate an energy consumption curve. The computing power scheduling unit is used to automatically reduce the computing power allocation of the multimodal AI recognition module (e.g., from 8-core GPU to 2 cores) in low-load scenarios (such as when there is no interference signal) to reduce unnecessary energy consumption; and to dynamically increase computing power to ensure real-time performance in high-load scenarios (such as when there is strong interference recognition). The power management unit is designed to adapt to a wide power supply range (110V-220VAC / 24VDC), supports sleep mode (50% reduction in power consumption when idle) and fast wake-up (wake-up time ≤100ms), and is adapted to the power supply requirements of different scenarios. The energy efficiency analysis unit is used to analyze the relationship between energy consumption and business load (such as the correlation between interference identification frequency and energy consumption), and output energy efficiency optimization suggestions (such as adjusting the sampling frequency) to help users reduce long-term operating costs.

[0031] In practical use, the specific steps are as follows: S1: Real-time acquisition of satellite navigation signals, timing equipment operation status signals, and environmental electromagnetic signals via the signal acquisition module; S2: The signal preprocessing and enhancement module performs filtering, noise reduction, and weak signal enhancement on the acquired satellite navigation signals and equipment status signals; S3: Through the multimodal AI recognition module, based on the pre-trained deep learning model, the collected signals are analyzed from multiple dimensions, including signal characteristics, time series patterns, and electromagnetic environment, to identify interference signals, deceptive signals, and potential abnormal patterns. S4: The adaptive switching module automatically switches the working mode based on the built-in high-speed switching circuit and the results of the multimodal AI recognition module. S5: Through the autonomous time synchronization backup module, based on the built-in high-precision local clock source, when the satellite signal is completely interfered with / spoofed and cannot be recovered, it automatically switches to the local autonomous time synchronization mode to ensure the continuity of time synchronization in extreme scenarios. S6: Through a deep compatibility module, it automatically identifies the type and communication protocol of the access timing device, achieving seamless integration with timing devices from different manufacturers and models; and supports hot-swappable deployment without interrupting the original timing link; S7: Receives alarm information and equipment operation data from the multimodal AI recognition module through the remote intelligent decision module, and performs trend analysis and threat prediction through cloud-based AI algorithms; it also supports automatic updating of the recognition model, remote adjustment of equipment parameters, and provides a manual remote intervention interface; S8: Record all critical operations through the security audit and traceability module to generate traceable security audit logs; at the same time, use blockchain technology to store the key logs as evidence. S9: The energy efficiency management module monitors the energy consumption and load of each module in real time, and dynamically adjusts the computing power allocation and power supply strategy to achieve the optimal energy efficiency of the system.

[0032] Although the present invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the disclosed embodiments can be combined with each other in any manner. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A protection system for a time synchronization isolation device, characterized in that, include: The signal acquisition module is used to acquire satellite navigation signals, timing equipment operating status signals, and environmental electromagnetic signals in real time. The signal preprocessing and enhancement module is used to filter, reduce noise, and enhance weak signals of the acquired satellite navigation signals and equipment status signals. The multimodal AI recognition module is used to perform multi-dimensional analysis of the collected signals based on the pre-trained deep learning model, including signal characteristics, timing patterns, and electromagnetic environment, to identify interference signals, deceptive signals, and potential abnormal patterns. The adaptive switching module is used to automatically switch the working mode based on the results of the built-in high-speed switching circuit and the multimodal AI recognition module. The autonomous timing backup module is used to automatically switch to local autonomous timing mode when the satellite signal is completely interfered with / spoofed and cannot be recovered, based on the built-in high-precision local clock source, to ensure the continuity of time synchronization in extreme scenarios. The deep compatibility module is used to automatically identify the type and communication protocol of the access timing device, enabling seamless integration with timing devices from different manufacturers and models; and it supports hot-swappable deployment without interrupting the original timing link. The remote intelligent decision-making module is used to receive alarm information and equipment operation data from the multimodal AI recognition module, and perform trend analysis and threat prediction through cloud AI algorithms; it also supports automatic updating of the recognition model, remote adjustment of equipment parameters, and provides a manual remote intervention interface. The security audit and traceability module records all critical operations and generates traceable security audit logs; at the same time, it uses blockchain technology to store evidence of critical logs. The energy efficiency management module is used to monitor the energy consumption and load of each module in real time, dynamically adjust the computing power allocation and power supply strategy, and achieve the optimal energy efficiency of the system.

2. The protection system for a time synchronization isolation device according to claim 1, characterized in that, The signal acquisition module includes: The satellite signal receiving unit is used to acquire satellite navigation radio frequency signals in real time through a multi-band antenna. The equipment status signal acquisition unit is used to collect the operating parameters of the timing device through the sensor interface and monitor the health status of the device in real time. The environmental electromagnetic signal monitoring unit is used to integrate broadband electromagnetic sensors to collect electromagnetic interference signals from the environment in which the equipment is deployed.

3. The protection system for a time synchronization isolation device according to claim 1, characterized in that, The signal preprocessing and enhancement module includes: The adaptive filtering unit is used to dynamically filter periodic clutter in satellite signals according to the minimum mean square error algorithm, and filter out frequency band signals that are not related to time synchronization. The noise reduction unit is used to process random noise in the device status signal using wavelet transform algorithm to improve the signal-to-noise ratio. The weak signal enhancement unit is used to enhance the signal strength of weak satellite signals in obstructed scenarios through coherent accumulation and signal reconstruction algorithms, ensuring that effective time information can still be extracted in extreme environments.

4. The protection system for a time synchronization isolation device according to claim 1, characterized in that, The multimodal AI recognition module includes: The feature extraction unit is used to perform feature engineering on the preprocessed signal, extracting Doppler frequency offset and pseudocode phase of satellite signals, time-series fluctuation characteristics of equipment status signals, and key spectral peak characteristics of electromagnetic signals. The multimodal fusion unit is used to fuse satellite signal features, equipment status features, and electromagnetic environment features based on a deep learning model with an attention mechanism. The interference / deception identification unit is used to identify specific threat types through a pre-trained model; The threat level assessment unit is used to classify threats into three levels: low, medium, and high, based on the intensity of interference, duration, and impact on timing accuracy.

5. A protection system for a time synchronization isolation device according to claim 1, characterized in that, The adaptive switching module includes: The mode determination unit is used to automatically determine the target's working mode based on the threat level assessment results, and supports manual preset switching thresholds. The high-speed switching circuit unit is used to design switching logic based on FPGA hardware acceleration technology to achieve uninterrupted switching of signal links. The switching verification unit is used to verify the synchronization accuracy of the signal after the switching is completed by comparing the timestamp. If the deviation exceeds the threshold, a second switching or alarm is triggered to avoid synchronization abnormalities caused by switching failure.

6. A protection system for a time synchronization isolation device according to claim 1, characterized in that, The autonomous time synchronization backup module includes: A local high-precision clock source unit is used to house a miniaturized rubidium atomic clock or a high-stability crystal oscillator, serving as a core backup source when satellite signals fail, ensuring time accuracy during independent operation; The clock calibration unit is used to calibrate the local clock in real time using the satellite timestamp when the satellite signal is normal, eliminating accumulated errors; and after the satellite fails, it predicts clock drift based on historical calibration data to maintain accuracy. The switching trigger unit is used to monitor satellite signal availability, automatically trigger the switch from satellite time synchronization to local autonomous time synchronization, and supports manual forced activation of backup mode.

7. A protection system for a time synchronization isolation device according to claim 1, characterized in that, The deep compatibility module includes: The interface identification unit is used to automatically identify the type and interface parameters of the access timing device through hardware interface scanning and protocol handshake testing. The protocol conversion unit is used to integrate the conversion logic of mainstream time synchronization protocols to achieve seamless adaptation between different protocols; The parameter adaptive unit is used to automatically adjust the parameters of this system according to the performance parameters of the access device to avoid synchronization errors caused by device incompatibility. The hot-swap control unit supports hot-swap operations of timing devices. Through circuit protection design, it ensures that the original timing link is not interrupted during the insertion and removal process, thereby reducing system downtime.

8. A protection system for a time synchronization isolation device according to claim 1, characterized in that, The remote intelligent decision-making module includes: The data transmission unit is used to upload the operating data of the local device to the cloud platform through an encrypted transmission channel, and supports edge computing preprocessing. The cloud-based AI analysis unit is used to perform trend analysis on historical data of devices in multiple regions based on the federated learning framework to predict potential threats; The model update unit is used to automatically push incremental update packages for the multimodal AI recognition model based on cloud analysis results, and supports offline update mode. The remote control unit provides an encrypted remote operation interface, enabling administrators to remotely adjust device parameters, manually trigger mode switching, and view real-time signal waveforms.

9. A protection system for a time synchronization isolation device according to claim 1, characterized in that, The security audit and tracing module includes: The full log recording unit is used to record all critical operations; The blockchain evidence storage unit is used to store key logs on the blockchain for evidence storage. The source tracing analysis unit is used to reverse analyze the location and type of interference source by analyzing the spectral characteristics, occurrence time, and geographical distribution of interference signals, thus assisting in locating the source of the threat. The compliance reporting unit is used to automatically generate audit reports that comply with the standards of the power and financial industries.

10. A protection system for a time synchronization isolation device according to claim 1, characterized in that, The energy efficiency management module includes: The energy consumption monitoring unit is used to monitor the instantaneous power consumption of each module in real time through current sensors, calculate the energy consumption per unit time, and generate an energy consumption curve. The computing power scheduling unit is used to automatically reduce the computing power allocation of the multimodal AI recognition module in low-load scenarios to reduce ineffective energy consumption; and to dynamically increase computing power to ensure real-time performance under high load. The power management unit is designed with a wide power adaptive range, supports sleep mode and fast wake-up, and adapts to the power supply requirements of different scenarios. The energy efficiency analysis unit is used to analyze the relationship between energy consumption and business load, and output energy efficiency optimization suggestions to help users reduce long-term operating costs.

Images