Multi-dimensional abnormal time screening time server security protection method and system

Abstract

The application discloses a time server security protection method and system for multi-dimensional discrimination of abnormal time, and belongs to the technical field of time synchronization security protection. The method comprises the following steps: acquiring an external reference time; performing wide-range discrimination, instantaneous discrimination and long-term cumulative deviation discrimination on the external reference time, the wide-range discrimination locks a reasonable time interval through preset time boundaries, the instantaneous discrimination adopts hierarchical threshold values to adapt to different scene precision requirements, and the long-term cumulative deviation discrimination dynamically sets threshold values based on a high-precision time-keeping unit; a three-level discrimination progressive verification logic is constructed, and time that continuously passes the three-level discrimination is determined as valid time; and the synchronization process of the time server is controlled according to the determination result. The system is correspondingly provided with a time acquisition module, three types of discrimination modules and a synchronization control module. The application realizes multi-level dead-angle-free interception of different types of time abnormalities, improves the anti-interference capability of the time server, and guarantees the security and reliability of time synchronization in a complex environment.

Description

Technical Field

[0001] This invention relates to the field of time synchronization security protection technology, specifically to a time server security protection method and system for multi-dimensional identification of abnormal time. Background Technology

[0002] Time synchronization is a core foundation for the stable operation of critical fields such as power dispatching, communication base stations, and industrial automation. As the core node of the time synchronization system, the time server needs to continuously track external reference time sources to ensure the accuracy of the output time. Currently, external reference time sources for time servers are mainly divided into three categories: wireless sources, network time sources, and cable / fiber optic transmission sources. Under ideal conditions, these sources can achieve nanosecond-level synchronization accuracy. However, in practical applications, they are susceptible to factors such as signal interference, upstream source failures, and transmission link anomalies, resulting in time drift, jumps, or even errors. If the time server directly synchronizes abnormal time, it will cause major safety accidents such as equipment control disorder and data recording failure. Therefore, effectively identifying external reference time has become a key requirement for the security protection of time servers.

[0003] Most time discrimination methods employ a single threshold comparison mechanism, identifying anomalies solely by determining whether the difference between the external time and the local time exceeds a fixed range. This approach struggles to address time spoofing and interference issues in complex scenarios. For instance, when faced with large-scale time jumps caused by wireless signal interference, a single threshold cannot effectively distinguish between normal synchronization and malicious tampering. Furthermore, traditional methods lack the dynamic adaptability to the decay of local timekeeping unit accuracy over long-term operation due to accumulated deviations in external time, resulting in a significant decrease in discrimination accuracy over time and failing to meet the application requirements of high-precision time synchronization scenarios.

[0004] Chinese Patent Publication No. (CN108880726A) discloses a time deviation measurement method and system. This technology acquires multi-source time signals and calculates the deviation values ​​between the signals to achieve preliminary detection of time anomalies. Compared with the traditional single threshold comparison method, it improves the accuracy of deviation measurement through cross-validation of multi-source signals and solves the problem of single-source time signals being susceptible to interference to a certain extent. However, this technology still has significant shortcomings. It only performs quantitative measurement of time deviation and does not construct a hierarchical anomaly identification logic. It cannot distinguish between different types of anomalies such as large-scale time jumps, instantaneous fluctuations, and long-term cumulative deviations. At the same time, this technology does not consider the accuracy decay characteristics of the local timekeeping unit throughout its entire life cycle, making it difficult to accurately identify cumulative deviations during long-term operation.

[0005] To address the technical shortcomings of the aforementioned patents, Chinese Patent Publication No. (CN121172970A) discloses a method and device for monitoring time deviation of power secondary equipment using multi-source time signals. This technology introduces a multi-dimensional deviation analysis model, combining the stability and continuity characteristics of time signals for comprehensive judgment, thus optimizing the ability to identify instantaneous time fluctuations. Furthermore, it adapts to the accuracy requirements of different application scenarios by configuring different monitoring thresholds. However, this technology still has limitations. It lacks a large-scale anomaly interception mechanism based on time axis intervals, making it unable to effectively identify time jumps and backtracking anomalies after wireless signals are interfered with. Moreover, in the long-term cumulative deviation identification stage, it fails to achieve dynamic correlation between threshold parameters and local timekeeping unit performance parameters, making it difficult to adapt to the accuracy characteristics of different types of timekeeping units. This results in the risk of missed or misjudged anomalies when facing complex interference scenarios.

[0006] Therefore, there is an urgent need for a time anomaly identification method that can cover a wide range, instantaneous, and long-term cumulative dimensions, to achieve accurate identification and blocking of different types of time anomalies. At the same time, it is necessary to dynamically adjust the identification threshold by combining the performance parameters of the local timekeeping unit, adapt to the accuracy decay characteristics of the timekeeping unit throughout its entire life cycle, and thus improve the time synchronization security and reliability of the time server in complex interference environments. Summary of the Invention

[0007] To address the aforementioned technical issues, this application discloses a method and system for protecting time servers by identifying anomalous times from multiple dimensions; the method for protecting time servers by identifying anomalous times from multiple dimensions includes: Obtain an external reference time, the source of which includes at least one of a wireless source, a network time source, or a cable or fiber optic transmission source; The external reference time is sequentially subjected to large-scale discrimination, instantaneous discrimination, and long-term cumulative deviation discrimination; Based on the results of the wide-range screening, instantaneous screening, and long-term cumulative deviation screening, it is determined whether the external reference time is a valid time. Based on the determination of the effective time, the synchronization process between the time server and the external reference time is controlled.

[0008] Preferably, the wide-ranging identification of external reference times includes: The time server is preset to accept a minimum time of the device's factory date. The maximum acceptable time is preset to be the equipment's manufacturing date and maximum service life. The sum, the formula is: ; Determine the external reference time Does it meet the requirements? ; If the conditions are met, the external reference time is determined to have passed a wide range of screening; otherwise, it is determined to be an invalid time.

[0009] Preferably, the wide-range identification uses a time-axis interval locking mechanism to identify two typical abnormal times that occur after the wireless source is interfered with, specifically: Time jump anomalies that occur after wireless signals are interfered with or spoofed, i.e., errors exceeding the maximum allowable time; Time rollback anomalies caused by wireless positioning device malfunctions, i.e., errors in time below the minimum allowable time; The time axis interval locking mechanism achieves targeted identification of abnormal scenarios of wireless time sources through preset time boundaries.

[0010] Preferably, the instantaneous discrimination of the external reference time includes: Get the current time from the time server. Preset time difference threshold ; Calculate the absolute value of the difference between the external reference time and the current time. ; judge Is it less than ; If the time is less than the specified time, the external reference time is determined to have passed instantaneous identification; otherwise, it is determined to be an invalid time.

[0011] Preferably, the time difference threshold A tiered configuration strategy is adopted, specifically as follows: At the basic configuration level, It meets the need for time continuity identification in typical scenarios; At the high-precision application level, the accuracy of the timekeeping unit based on the time server will... Scalable down to milliseconds and below, suitable for application scenarios with stringent time synchronization accuracy requirements; The hierarchical configuration strategy enables flexible matching of instantaneous fluctuation identification accuracy under different application scenarios.

[0012] Preferably, the long-term cumulative deviation identification of the external reference time includes: When the time server synchronizes with an external reference for the first time, a synchronization timer based on a high-precision timekeeping unit is started. The high-precision timekeeping unit includes an OCXO high-precision temperature-controlled crystal oscillator or a rubidium atomic clock. Obtain the timing time of the synchronization timer. Preset cumulative deviation threshold ; Calculate the absolute value of the difference between the external reference time and the timing time. ; judge Is it less than ; If the time is less than the specified value, the external reference time is determined to be identified through long-term cumulative deviation; otherwise, it is determined to be an invalid time.

[0013] Preferably, the cumulative deviation threshold The frequency stability parameters and lifecycle aging coefficient of the high-precision timekeeping unit are dynamically tuned, specifically as follows: For OCXO high-precision temperature-controlled crystal oscillators, first calculate the theoretical deviation value per unit time. ,in For the duration of the time, To ensure crystal oscillator frequency stability, a lifetime aging margin factor is then introduced. ,get ; For rubidium atomic clocks, the tuning is directly based on their inherent frequency stability characteristics. ; The dynamic tuning method is adapted to the accuracy decay characteristics of the timekeeping unit throughout its entire life cycle, enabling accurate identification of accumulated deviations.

[0014] Preferably, a three-level progressive verification logic is constructed, specifically as follows: S0. External reference times need to be thoroughly screened to filter out invalid times outside the time axis interval; S1. After a large-scale screening process, the system enters the instantaneous screening stage to eliminate small-scale time fluctuation anomalies. S2. After the time spent in the first two stages of screening, the system enters the long-term cumulative deviation screening stage to achieve full-cycle precision control. S3. Only external reference times that pass three levels of continuous screening are determined to be valid times. The progressive verification logic realizes multi-level, all-round interception of abnormal times.

[0015] The aforementioned multi-dimensional time server security protection system for identifying abnormal times includes: The time acquisition module acquires an external reference time, the source of which includes at least one of a wireless source, a network time source, or a cable or fiber optic transmission source; A wide-range discrimination module is connected to the time acquisition module and has a built-in time axis interval locking mechanism for performing wide-range discrimination on the external reference time. An instantaneous discrimination module, connected to the time acquisition module, is a threshold tuning unit employing a hierarchical configuration strategy, used to perform instantaneous discrimination on the external reference time; The long-term cumulative deviation discrimination module is connected to the time acquisition module and the high-precision timekeeping unit respectively, and is configured with a dynamically tuned cumulative deviation threshold calculation unit for performing long-term cumulative deviation discrimination on the external reference time. The synchronization control module is connected to the large-scale discrimination module, the instantaneous discrimination module, and the long-term cumulative deviation discrimination module, respectively. It has a built-in three-level discrimination progressive verification logic, which is used to control the synchronization process of the time server based on the results of each discrimination module.

[0016] Preferably, the long-term cumulative deviation discrimination module is a dual-reference source comparison unit, including a synchronization timer subunit and a deviation dynamic calculation subunit, specifically: The synchronization timer subunit starts when the time server first synchronizes with an external reference. It is based entirely on the inherent frequency of the high-precision timekeeping unit and keeps time independently without relying on an external time source, thus avoiding timing deviations introduced by external interference. The deviation dynamic calculation subunit calculates the absolute value of the difference between the external reference time and the timing time of the synchronization timer subunit in real time, and calls the cumulative deviation threshold dynamically tuned based on frequency stability and lifetime aging coefficient to output the discrimination result. The dual reference source comparison unit enables independent verification between the external time source and the local timekeeping unit, ensuring the anti-interference and accuracy of long-term accumulated deviation identification.

[0017] Compared with the prior art, the technical solution of this application has the following technical effects: This invention constructs a time axis interval locking mechanism through large-scale screening. Relying on the non-reversible nature of time, it presets time boundaries and can accurately identify time jumps and rollback anomalies after wireless signals are interfered with. It intercepts invalid time that exceeds the reasonable range from the source, avoids the disruptive impact of large-scale time errors on the time server synchronization process, and builds a solid first line of defense for time security.

[0018] The instantaneous discrimination of this invention sets a difference threshold based on the characteristics of time continuity, and combines a hierarchical configuration strategy to adapt to the accuracy requirements of different application scenarios. It can effectively eliminate small-scale time fluctuation anomalies, control instantaneous time errors within a reasonable range, and ensure that the time server maintains a stable time output state when faced with brief disturbances from the upper-level source, thereby improving the anti-interference capability of the synchronization process.

[0019] This invention utilizes a local high-precision timekeeping unit to construct an independent timing reference for long-term cumulative deviation identification. By dynamically adjusting the cumulative deviation threshold to adapt to the accuracy decay characteristics of the timekeeping unit throughout its entire life cycle, it achieves accurate identification of long-term drift of external time sources, solving the cumulative deviation problem that traditional methods struggle to address. It controls long-term synchronization errors to the microsecond level, meeting the stringent requirements of high-precision time synchronization scenarios.

[0020] This invention constructs a three-level progressive verification logic to achieve multi-level, blind-spot-free interception of large-scale, instantaneous, and long-term accumulated anomalies. Compared with traditional single-screening methods, it significantly improves the comprehensiveness and accuracy of anomaly time identification, while ensuring the synchronization security and reliability of the time server in complex interference environments.

[0021] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the preferred embodiments of this application are described in detail below with reference to the accompanying drawings.

[0022] The above and other objects, advantages and features of this application will become more apparent to those skilled in the art from the following detailed description of specific embodiments in conjunction with the accompanying drawings. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0024] Based on the description of the figures and their corresponding technical content in the document, the titles of the figures are as follows: Figure 1 Flowchart of core steps in a multi-dimensional method for identifying abnormal time periods in a server security protection system; Figure 2 : Schematic diagram of the execution logic of the time server security protection method for identifying abnormal times from multiple dimensions; Figure 3 : A schematic diagram illustrating the process and judgment logic for large-scale identification of external reference time from a time server; Figure 4 : Schematic diagram of the instantaneous identification process and threshold comparison logic of the external reference time of the time server; Figure 5 : Schematic diagram of the long-term cumulative deviation identification process and judgment for external reference time of time server; Figure 6 : A diagram illustrating the three-level progressive verification logic and anomaly interception steps of the time server; Figure 7 : A schematic diagram of a time server security protection system architecture that identifies abnormal times from multiple dimensions; Figure 8Statistical chart showing the trend of server Beidou / GPS wireless source time synchronization accuracy over 30 days; Figure 9 Comparison and smoothing trend statistics of server identification response time under three types of time anomaly conditions; Figure 10 : Schematic diagram of data processing volume and effective time determination results in the three-level screening process of the time server. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. In the following description, specific details such as specific configurations and components are provided merely to help fully understand the embodiments of this application. Therefore, those skilled in the art should understand that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. In addition, for clarity and brevity, descriptions of known functions and structures are omitted in the embodiments.

[0026] It should be understood that the phrase "an embodiment" or "this embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "an embodiment" or "this embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments.

[0027] Furthermore, reference numerals and / or letters may be repeated in different examples within this application. Such repetition is for the purpose of simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or settings discussed.

[0028] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" in this article describes another type of relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " in this article generally indicates that the related objects before and after it are in an "or" relationship.

[0029] In this article, the term "at least one" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, "at least one of A and B" can mean: A exists alone, A and B exist simultaneously, or B exists alone.

[0030] It should also 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.

[0031] Example 1 This embodiment mainly describes a time server security protection method for multi-dimensional identification of abnormal times, such as... Figures 1-2 As shown, it specifically includes: Obtain an external reference time, the source of which includes at least one of a wireless source, a network time source, or a cable or fiber optic transmission source; The external reference time is sequentially subjected to large-scale discrimination, instantaneous discrimination, and long-term cumulative deviation discrimination; Based on the results of the wide-range screening, instantaneous screening, and long-term cumulative deviation screening, it is determined whether the external reference time is a valid time. Based on the determination of the effective time, the synchronization process between the time server and the external reference time is controlled.

[0032] Furthermore, the external reference time acquisition process performs standardized data collection and format parsing operations for three types of time sources: wireless sources, network time sources, and cable or fiber optic transmission sources.

[0033] Wireless time sources include BeiDou, GPS, and long / shortwave signals. These sources capture time signals from satellites or shortwave transmitters using dedicated receiving modules, extracting time data containing year, month, day, hour, minute, second, and millisecond, with a time granularity accurate to 20 nanoseconds. Network time sources include time signals transmitted via NTP and PTP protocols. These sources receive time messages from upstream time servers through a 1000Mbps Ethernet communication interface, extracting timestamp information according to NTPv4 and PTPv2 protocol specifications, with a time granularity accurate to 100µs. Cable or fiber optic transmission sources include time signals in IRIG-B and TOD+PPS formats. These sources receive serial time codes and second pulse signals via RS485 cables or single-mode fiber optic links, combining the rising edge of the second pulse to achieve precise alignment of the time data, with a time granularity accurate to 20 nanoseconds.

[0034] This step requires converting time data from different sources into a unified UTC time format to create an external reference time with consistent time granularity. The time granularity is uniformly calibrated to 100 nanoseconds, providing a standardized and high-precision data foundation for subsequent full-process identification operations. Simultaneously, the signal strength and link bit error rate of each time source are monitored in real time. Signal strength below -120dBm or bit error rate above 1×10⁻⁻⁻⁻⁶ is considered acceptable. 6 When this happens, the time source is marked as unstable.

[0035] Furthermore, such as Figure 3 As shown, based on the inherent physical characteristic that time cannot be reversed, a large-scale identification mechanism is constructed by performing three core steps: time boundary presetting, time format verification, and interval judgment.

[0036] Time boundary preset: Presets the minimum time that the time server can accept to the device's factory default time. , The initial time is the time permanently stored in the local Flash memory module at the time of device manufacture, in the format YYYY-MM-DDHH:MM:SS, accurate to 1 second; the maximum acceptable time is preset to the sum of the device's manufacturing time and its maximum service life, calculated using the following formula: ,in The maximum design lifespan of the time server is set to 87,600 hours (i.e., 10 years). This parameter is preset based on the lifespan characteristics of the device's hardware components and is stored in the local storage module. It supports subsequent updates via dedicated maintenance software, with update increments of 100 hours.

[0037] Time format validation: Before performing interval checks, the external reference time needs to be validated. Perform format validity checks, including verifying the completeness of time data fields and the reasonableness of field value ranges. The year range is limited to 1970-2100, the month range to 1-12, the day range to 1-31, the hour range to 0-23, the minute range to 0-59, and the second range to 0-59. This ensures... To ensure that the time data conforms to the UTC time standard and avoids identification bias caused by format errors.

[0038] Interval determination: External reference time that has passed format validation. Substituting into the interval inequality for judgment, the inequality is: .

[0039] Result determination: If the external reference time satisfies the inequality, the process passes the large-scale screening and proceeds to the next stage, the instantaneous screening process; if the external reference time is greater than or equal to... or less than or equal to If the time is invalid, an abnormal alarm message is generated directly and the synchronization process of that time source is terminated. At the same time, the abnormal time data and the identification result are recorded to the local log module, and the log recording frequency is 1 time / second.

[0040] Furthermore, such as Figure 4 As shown, instantaneous identification relies on the inherent physical characteristics of time continuity and performs four core steps: threshold grading preset, current time extraction, absolute value of difference calculation, and threshold comparison judgment.

[0041] Threshold grading preset: preset time difference threshold This threshold uses a hierarchical configuration strategy, setting threshold parameters for two levels: a basic configuration level and a high-precision application level. Both levels' threshold parameters are stored in a local EEPROM configuration module, allowing for flexible switching based on actual application scenarios. The basic configuration level... The value is set to 10 seconds, which meets the requirements for time continuity identification in normal scenarios; high-precision application-level... Based on the precision parameters of the time server's timekeeping unit, a value of 10 milliseconds is set, extending the threshold down to the millisecond level, which is suitable for application scenarios with stringent time synchronization accuracy requirements.

[0042] Current Time Extraction: Retrieves the current time from the time server in real time. , The time is provided by the time server's built-in timekeeping unit. This time is the standard time currently maintained by the server, with a time granularity of 100 nanoseconds, and is consistent with external reference time. Maintaining a consistent time granularity ensures the accuracy of difference calculations.

[0043] Absolute difference calculation: Calculate the absolute value of the difference between the external reference time and the current time. The calculation formula is as follows: The absolute value calculation eliminates the influence of time sequence on the difference result, ensuring the consistency of the identification logic. The calculation cycle of the difference is 10 milliseconds.

[0044] Threshold comparison judgment: The absolute value of the calculated difference is used as the threshold comparison judgment. Preset thresholds corresponding to the current application scenario Perform a comparison, if If the instantaneous screening fails, the process proceeds to the next stage: the long-term cumulative deviation screening process. If the time is invalid, an abnormal alarm message is generated and the synchronization process of that time source is terminated. At the same time, the abnormal difference data and the identification results are recorded to the local log module, and the log storage capacity is 16GB.

[0045] Furthermore, such as Figure 5As shown, the long-term cumulative deviation identification relies on the high-precision timekeeping unit built into the time server to perform five core steps: synchronous timer initialization and startup, dynamic adjustment of cumulative deviation threshold, acquisition of timing time, calculation of absolute value of difference, and threshold comparison and judgment. The high-precision timekeeping unit includes two types: OCXO high-precision temperature-controlled crystal oscillator and rubidium atomic clock. Both types of timekeeping units have high frequency stability characteristics and can independently maintain a high-precision time reference.

[0046] Synchronization timer initialization and startup: When the time server synchronizes with the external reference for the first time, the synchronization timer initialization and startup process is triggered. The initial time of the synchronization timer is calibrated to the external reference time of the first synchronization, ensuring that the initial timing time is completely consistent with the external reference time. After the synchronization timer starts, it performs independent timing based entirely on the inherent frequency of the high-precision timekeeping unit. The frequency stability of the OCXO high-precision temperature-controlled crystal oscillator is 0.1 PPM, and the frequency stability of the rubidium atomic clock is ±1×10⁻¹¹. The timing process does not rely on any external time source, avoiding timing deviations introduced by external interference. The synchronization timer runs continuously and updates the timing time in real time. The timing granularity is 100 nanoseconds, compared with the external reference time. Maintain consistency.

[0047] Dynamic tuning of cumulative deviation threshold: The cumulative deviation threshold is dynamically tuned based on the frequency stability parameter and the aging coefficient of the high-precision timekeeping unit. The threshold tuning process is divided into two different logics for OCXO high-precision thermostatic crystal oscillators and rubidium atomic clocks.

[0048] For OCXO high-precision temperature-controlled crystal oscillators, first calculate the theoretical deviation value per unit time, using the following formula: ,in The set timeout duration is 3600 seconds (i.e., 1 hour). The frequency stability parameter of the crystal oscillator is set to 0.1 PPM, which is determined by the crystal oscillator's hardware characteristics; a lifetime aging margin factor is also introduced. The value is 3. This coefficient is set based on the accuracy decay characteristics of the crystal oscillator throughout its entire lifespan and is used to compensate for the accuracy decrease caused by crystal oscillator aging. The final threshold calculation formula is as follows: .

[0049] For rubidium atomic clocks, the cumulative deviation threshold is set directly based on their own frequency stability characteristics. The value is set to 1 microsecond / hour, eliminating the need for an additional aging margin coefficient and ensuring that the threshold parameter matches the high precision characteristics of the rubidium atomic clock.

[0050] Timing acquisition: each time an external reference time is received. At the same time, read the current timing time of the synchronization timer in real time. The reading latency is controlled within 1 microsecond to ensure that the acquired timing data is the local timing data at the same moment as the external reference time.

[0051] Calculation of absolute difference: Calculate the absolute value of the difference between the external reference time and the timing time. The formula is as follows: The influence of time sequence on the difference result is eliminated by absolute value calculation, and the calculation accuracy of the difference is 1 nanosecond.

[0052] Threshold comparison judgment: The absolute value of the difference is used for comparison. The cumulative deviation threshold of dynamic tuning Perform a comparison, if If so, it is determined that the identification is based on long-term cumulative bias; if If the time is invalid, an abnormal alarm message is generated and the synchronization process of that time source is terminated. At the same time, the abnormal cumulative deviation data and the identification results are recorded to the local log module.

[0053] Furthermore, such as Figure 6 As shown, a three-level progressive verification logic is constructed to filter external reference times layer by layer in a fixed order. The specific steps are as follows: S0. External reference times need to be thoroughly screened to filter out invalid times outside the time axis interval; S1. After a large-scale screening process, the system enters the instantaneous screening stage to eliminate small-scale time fluctuation anomalies. S2. After the time spent in the first two stages of screening, the system enters the long-term cumulative deviation screening stage to achieve full-cycle precision control. S3. Only external reference times that pass three levels of continuous screening are determined to be valid times. The progressive verification logic realizes multi-level, all-round interception of abnormal times.

[0054] In S0, the large-scale screening process requires external reference time to first pass through a large-scale screening. Using a time axis interval locking mechanism, invalid times that exceed the range from the device's manufacturing time to its maximum service life are filtered out. This step is the first line of defense for the entire verification logic, intercepting all large-scale time jumps and rollback anomalies.

[0055] In the instantaneous screening process in S1, after a large-scale screening period, the system enters the instantaneous screening stage. Based on the characteristics of time continuity, it eliminates small-scale time fluctuation anomalies whose difference from the current time of the server exceeds a preset threshold. This step is the second line of defense for the verification logic, intercepting instantaneous time anomalies caused by brief disturbances from the upper-level time source.

[0056] The long-term cumulative deviation screening in S2, after the time of the first two screening stages, finally enters the long-term cumulative deviation screening. Relying on the independent timing benchmark of the local high-precision timekeeping unit, it realizes accurate control of the cumulative deviation of the external time source throughout the entire cycle. This step is the third line of defense of the verification logic, intercepting the long-term accuracy anomaly caused by the slow drift of the external time source.

[0057] The S3 system comprehensively determines the validity of time. Only external reference times that pass the three levels of screening—wide-range screening, instantaneous screening, and long-term cumulative deviation screening—are considered valid. If any step fails, the time is deemed invalid, and subsequent screening processes for failed steps are no longer executed. The total time consumption of the entire progressive verification logic is controlled within 100 milliseconds, achieving multi-level, comprehensive interception of abnormal times.

[0058] Furthermore, for valid time synchronization control, the time server performs a time synchronization operation on external reference times that are determined to be valid, and maintains the current time of its own timekeeping unit. Calibrate to this external reference time The calibration process employs a smooth adjustment strategy with an adjustment rate of 1 millisecond per second to avoid abrupt time changes affecting the stability of the server's output time. After synchronization is complete, the initial time of the synchronization timer is updated to ensure the accuracy of the subsequent cumulative deviation discrimination benchmark.

[0059] Invalid Time Blocking Control: For external reference times determined to be invalid, the control time server blocks synchronization operations with that external reference time while maintaining its own time based on a high-precision timekeeping unit. The independent operating accuracy of the OCXO oven-controlled crystal oscillator timekeeping unit is ±1 millisecond / day, and the independent operating accuracy of the rubidium atomic clock timekeeping unit is ±1 microsecond / day, ensuring that the server continuously outputs a stable standard time.

[0060] Synchronization result traceability management: The execution results of synchronization control are fed back to the local log module in real time. The log records include the synchronization timestamp, the time difference before and after synchronization, and the judgment results of the screening process. The timestamp accuracy is 100 nanoseconds. The log data can be exported through the network interface to achieve full-process traceability management.

[0061] This implementation details how a three-level progressive screening logic accurately intercepts anomalies such as large-scale time jumps, instantaneous fluctuations, and long-term accumulated deviations. Combined with a local timekeeping unit, it achieves dynamic threshold tuning, significantly improving the time server's anti-interference capability, ensuring the security and reliability of time synchronization in complex scenarios, and providing stable and accurate time service support for key areas.

[0062] Example 2 describes in detail a time server security protection system for multi-dimensional anomaly detection, used to implement the aforementioned time server security protection method for multi-dimensional anomaly detection, such as... Figure 7 As shown, it includes: The time acquisition module acquires an external reference time, the source of which includes at least one of a wireless source, a network time source, or a cable or fiber optic transmission source; A wide-range discrimination module is connected to the time acquisition module and has a built-in time axis interval locking mechanism for performing wide-range discrimination on the external reference time. An instantaneous discrimination module, connected to the time acquisition module, is a threshold tuning unit employing a hierarchical configuration strategy, used to perform instantaneous discrimination on the external reference time; The long-term cumulative deviation discrimination module is connected to the time acquisition module and the high-precision timekeeping unit respectively, and is configured with a dynamically tuned cumulative deviation threshold calculation unit for performing long-term cumulative deviation discrimination on the external reference time. The synchronization control module is connected to the large-scale discrimination module, the instantaneous discrimination module, and the long-term cumulative deviation discrimination module, respectively. It has a built-in three-level discrimination progressive verification logic, which is used to control the synchronization process of the time server based on the results of each discrimination module.

[0063] Furthermore, the time acquisition module integrates multiple types of signal receiving and parsing units to adapt to the signal acquisition needs of three types of external reference time sources.

[0064] Wireless signal receiving unit: Built-in Beidou and GPS dual-mode receiver chip and long and short wave signal receiver, supports signal acquisition with a sensitivity of -120dBm, can resolve time data with a time granularity of 1 millisecond, and monitor signal strength and lock status in real time.

[0065] Network time receiving unit: Equipped with a 1000Mbps Ethernet communication interface, compatible with NTPv4 and PTPv2 protocols, capable of extracting high-precision timestamps with a time granularity of 100 nanoseconds, while simultaneously monitoring the bit error rate of the network link, with a bit error rate higher than 1×10⁻ 6 The time source of this path is marked as abnormal.

[0066] Cable / fiber optic signal receiving unit: Equipped with RS485 interface and single-mode fiber optic interface, it supports the reception of IRIG-B serial time code and TOD+PPS second pulse signal, and achieves time alignment with 1 microsecond accuracy by combining the rising edge of the second pulse.

[0067] The module has a built-in time format conversion unit that converts time data from different sources into UTC format and outputs a standardized external reference time. The data is simultaneously transmitted to the wide-range discrimination module, the instantaneous discrimination module, and the long-term cumulative deviation discrimination module.

[0068] Furthermore, the wide-range identification module and the time acquisition module are connected via a high-speed data bus, and have built-in time axis interval locking unit, time format verification unit and identification result output unit.

[0069] Time axis interval locking unit: Pre-stores the factory time of the time server and maximum service life Through formula The upper limit of the calculation time, where The default value is 87600 hours, and it can be updated via maintenance software; please enter the value... Substitute into the inequality Perform interval judgment.

[0070] Time format verification unit: for The system validates the integrity and value range of the fields, limiting the years to 1970-2100, months to 1-12, days to 1-31, hours to 0-23, minutes to 0-59, and seconds to 0-59, while filtering out time data with incorrect formats.

[0071] Screening result output unit: If After range judgment and format validation, the output passes the signal, triggering... Enter the instantaneous screening module; if it fails, output an invalid signal and send the abnormal information to the synchronization control module, while also recording it in the local log.

[0072] Furthermore, the instantaneous identification module is connected to the time acquisition module, and has a built-in hierarchical threshold setting unit, time difference calculation unit, and threshold comparison unit.

[0073] Hierarchical threshold tuning unit: storing two types of thresholds: basic level and high-precision level. The basic level threshold is set to 10 seconds, which meets the needs of conventional scenarios; the high-precision level threshold is set to 10 milliseconds, which is suitable for high-precision scenarios such as industrial control and supports flexible switching according to the application scenario.

[0074] Time difference calculation unit: Real-time reading of the current time from the time server Through formula The absolute value of the difference is calculated in 10 milliseconds to ensure timely identification.

[0075] Threshold comparison unit: The calculated threshold comparison unit... With the current level If compared, Then the output signal triggers... Enter the long-term cumulative deviation identification module; if it fails, an invalid signal is output to the synchronization control module.

[0076] Furthermore, the long-term cumulative deviation identification module is connected to the time acquisition module and the high-precision timekeeping unit respectively. The high-precision timekeeping unit includes two types: OCXO high-precision temperature-controlled crystal oscillator and rubidium atomic clock. The module has a built-in synchronous timer unit, dynamic threshold calculation unit and deviation comparison unit.

[0077] The long-term cumulative deviation discrimination module is a dual-reference source comparison unit, including a synchronization timer subunit and a deviation dynamic calculation subunit, specifically: The synchronization timer subunit starts when the time server first synchronizes with an external reference. It is based entirely on the inherent frequency of the high-precision timekeeping unit and keeps time independently without relying on an external time source, thus avoiding timing deviations introduced by external interference. The deviation dynamic calculation subunit calculates the absolute value of the difference between the external reference time and the timing time of the synchronization timer subunit in real time, and calls the cumulative deviation threshold dynamically tuned based on frequency stability and lifetime aging coefficient to output the discrimination result. The dual reference source comparison unit enables independent verification between the external time source and the local timekeeping unit, ensuring the anti-interference and accuracy of long-term accumulated deviation identification.

[0078] The synchronous timer subunit is the core timing component of the dual reference source comparison unit. It adopts a hardware-level independent timing architecture and is completely decoupled from the external time source.

[0079] The sub-unit triggers initialization upon successful initial synchronization with the server, precisely calibrating the initial timing to the first valid external reference time. The time calibration error is controlled within 100 nanoseconds. After initialization, the sub-unit operates based on the inherent oscillation frequency of the high-precision timekeeping unit, unaffected by external signal interruptions or interference.

[0080] For the OCXO high-precision temperature-controlled crystal oscillator timing unit, the timing frequency of the sub-unit tracks the crystal oscillator's ±20 PPM stability; for the rubidium atomic clock timing unit, the timing frequency tracks its ±1×10⁻¹¹ stability. The timing data of the sub-unit is updated in real time with a time granularity of 100 nanoseconds, providing a high-precision local reference for deviation calculation.

[0081] The deviation dynamic calculation subunit integrates the difference calculation module and the threshold calling module to realize real-time comparison between external time and local time.

[0082] The difference calculation module receives the output of the time acquisition module. Immediately read the current timing time of the synchronous timer subunit. The read latency is controlled within 1 microsecond; calculations are performed through hardware processing circuits. The calculation accuracy reaches 1 nanosecond, eliminating the time delay error caused by software calculation.

[0083] The threshold call module pre-stores parameters based on the timekeeping unit type. For calling OCXO crystal oscillators Targeting the rubidium atomic clock ;Will and Real-time comparison, if If the signal is clear, the output will be a pass signal; otherwise, the output will be an invalid signal.

[0084] The dual-reference-source independent verification architecture of this sub-unit completely avoids the influence of external time source deception and drift on the identification results, and greatly improves the anti-interference and accuracy of long-term cumulative deviation identification.

[0085] Furthermore, the synchronization control module is connected to three screening modules respectively, and has a built-in three-level progressive verification logic unit, synchronization execution unit, blocking control unit and log recording unit.

[0086] The three-level progressive verification logic unit performs verification in the order of broad-based screening → instantaneous screening → long-term cumulative deviation screening. Only those that pass all three levels of screening consecutively will be verified. Only if it passes the first step is it considered valid. If any step fails, the subsequent process will be terminated. The total verification time is controlled within 100 milliseconds.

[0087] Synchronous execution unit: For valid time, a smooth adjustment rate of 1 millisecond / second is used to update the server's current time. Calibration And update the initial time of the synchronization timer.

[0088] Blocking control unit: For invalid time, it blocks synchronization operation. The control server operates independently based on the timekeeping unit to ensure stable output time.

[0089] Log recording unit: Records information such as synchronization timestamp, identification results, and time difference. The timestamp accuracy is 100 nanoseconds, and data export and traceability are supported.

[0090] This embodiment describes in detail how a three-level progressive discrimination logic accurately intercepts anomalies such as large-scale time jumps, instantaneous fluctuations, and long-term accumulated deviations, dynamically adapts to the full lifecycle accuracy characteristics of the timekeeping unit, blocks invalid time synchronization, and ensures that the time server outputs stable and high-precision standard time in complex interference environments, thereby improving the security and reliability of time synchronization in critical areas.

[0091] Based on Embodiment 1 or 2, this embodiment details the implementation and verification of a time server security protection method for multi-dimensional identification of abnormal times. To verify the effectiveness and reliability of the time server security protection method for multi-dimensional anomaly identification proposed in this application, this implementation verification selected three identical time servers from a power dispatch center as test objects. The servers have built-in OCXO high-precision temperature-controlled crystal oscillator and rubidium atomic clock dual timekeeping units. External reference time sources include three types: BeiDou / GPS dual-mode wireless source, NTP / PTP network source, and IRIG-B fiber optic source. The test period was 30 consecutive days, running 24 hours a day without interruption. The test environment simulated real-world complex scenarios, including various abnormal conditions such as human-induced signal interference, upstream time source failure, and transmission link fluctuations. During the test, core indicators such as the server's time synchronization accuracy, anomaly identification success rate, and independent timekeeping accuracy were monitored in real time. The data collection frequency was 1 time / second, and the data accuracy reached the nanosecond level. All monitoring data was stored in a local 16GB log module to ensure data integrity and traceability. This test has completed more than 100,000 simulations of abnormal operating conditions, covering different test scenarios such as normal interference, extreme boundary value triggering, and multi-time source collaborative anomalies, providing sufficient data samples for indicator statistics.

[0092] The core test indicators for this verification are divided into two categories: time synchronization accuracy and anomaly detection success rate. Time synchronization accuracy includes synchronization accuracy under normal operating conditions and independent timekeeping accuracy under abnormal operating conditions. Anomaly detection success rate includes the detection success rates for three types of anomalies: large-scale time anomalies, instantaneous time anomalies, and long-term cumulative deviation anomalies. To present the test results intuitively, 30 days of test data were statistically analyzed, resulting in Table 1, the statistical results of the core performance test indicators of the time server. In the table, TSSR represents time synchronization accuracy in nanoseconds (ns); TSHR represents independent timekeeping accuracy in microseconds per day (μs / d); SR represents detection success rate in percentage (%); WRA represents large-scale time anomalies; ITA represents instantaneous time anomalies; and LAA represents long-term cumulative deviation anomalies. The simulated operating conditions for the three types of anomalies are time jumps caused by wireless signal interference, time fluctuations caused by brief failures of the upstream network source, and cumulative deviations caused by long-term drift of the fiber optic source, respectively. As shown in Table 1, under normal operating conditions, the synchronization accuracy of the time server equipped with the protection method of this application for the three types of external time sources is consistently within 50 ns, with the best synchronization accuracy for the IRIG-B fiber optic source, averaging 32 ns. Under abnormal operating conditions, the server switches to independent timekeeping mode, and the average independent timekeeping accuracy of the OCXO crystal oscillator is 0.8 μs / d, and the average independent timekeeping accuracy of the rubidium atomic clock is 0.05 μs / d, both better than the design specifications. The success rate of identification for the three types of abnormal operating conditions is above 99.99%, with no valid missed cases. Only a very few identification delays occurred in the extreme boundary value test scenario, fully verifying the high precision and high reliability of the protection method of this application.

[0093] Table 1 Statistical Results of Core Performance Test Indicators for Time Server To visually represent the changing trend of the server's time synchronization accuracy during the 30-day testing period, a BeiDou / GPS wireless source was selected as a typical test object, and a 30-day time synchronization accuracy change trend chart was obtained, as shown below. Figure 8As shown in the figure, during the first 10 days of the test period, the test environment was under normal operating conditions, and the server's synchronization accuracy remained stable between 40-50 ns with minimal fluctuations. From days 11-20, the test environment simulated abnormal conditions caused by wireless signal interference. The server quickly identified the abnormal time source and switched to independent timekeeping mode. During this period, the synchronization accuracy curve remained stable without any abnormal jumps. From days 21-30, the test environment returned to normal operating conditions, and the server switched back to synchronization mode. The synchronization accuracy quickly recovered to normal levels and remained stable without any subsequent fluctuations, fully verifying the stability and adaptability of the protection method in complex operating conditions. It should be noted that during the extreme boundary value test on day 15, one identification delay occurred. This was due to the external time source tampering value being exactly equal to the preset threshold, and the probability of this occurring in actual engineering applications is less than 0.01%.

[0094] To further verify the response speed and processing capability of the protection method under different abnormal operating conditions, this test selected typical abnormal operating condition data from day 15 for detailed analysis, resulting in Table 2, which shows the statistical results of server response performance under typical abnormal operating conditions. In the table, ART represents the abnormal response time in milliseconds (ms); PD represents the processing latency in milliseconds (ms); and RT represents the recovery time in seconds (s). Abnormal operating condition 1 is time jump caused by interference from drone countermeasures equipment to the wireless source; abnormal operating condition 2 is time fluctuation caused by the restart of the upstream server of the network source; and abnormal operating condition 3 is long-term cumulative deviation caused by aging of the fiber optic link. As can be seen from the data in Table 2, the average response time under the three abnormal operating conditions is less than 20ms, and the average processing latency is less than 10ms. After excluding abnormal time sources, the average recovery time of the server switching to independent timekeeping mode is less than 5s. The response speed and processing capability fully meet the real-time requirements of key areas such as power dispatching. At the same time, after restoring normal synchronization, the server's time synchronization accuracy can quickly recover to the normal level without any time abrupt changes, ensuring the stability of the output time. The maximum response time of 21ms in condition 1 of the table was determined to be caused by strong electromagnetic interference, which is a normal fluctuation under extreme test scenarios.

[0095] Table 2 Statistical Results of Server Response Performance under Typical Abnormal Operating Conditions To compare the identification effectiveness of the protection method of this application for three different types of abnormal times, a comparison of the identification response time under the three types of abnormal operating conditions was obtained, such as... Figure 9 As shown, Figure 9 The sub-figure above shows the response time for identifying large-scale time anomalies. Figure 9 Neutron plots represent the instantaneous time anomaly detection response time. Figure 9 The subplot below shows the response time for identifying long-term cumulative deviation anomalies. Figure 9 As can be seen from the subplot, in 100 tests, the average identification response time for large-scale time anomalies was 15ms, the average identification response time for instantaneous time anomalies was 12ms, and the average identification response time for long-term cumulative deviation anomalies was 18ms. The identification response times for all three types of anomalies were less than 20ms, and the dispersion was small. This indicates that the protection method of this application has a fast identification response speed for different types of time anomalies and has good consistency and stability. Figure 9 The two response time peaks that appear in the above subgraph correspond to the strong interference scenarios in the test. They did not affect the final identification results, but only caused minor delay fluctuations.

[0096] To verify the effectiveness of the three-level progressive verification logic of the protection method in this application, a data processing diagram of the three-level verification process is obtained, as follows: Figure 10 As shown, it contains 4 sub-graphs, namely Figure 10 The large-scale screening process in the top left sub-image Figure 10 The instantaneous identification process in the upper right sub-image Figure 10 The process for identifying long-term cumulative deviations in the lower left subgraph. Figure 10 The effective time determination result of the lower right subgraph is from Figure 9 of Figure 10 As shown in the top left sub-graph, the large-scale screening process can filter out approximately 20% of invalid time data, allowing only 80% of the valid data to proceed to the next stage; from Figure 10 As shown in the upper right sub-graph, the instantaneous screening process can filter out approximately 10% of invalid time data, allowing only 70% of valid data to proceed to the next stage; from Figure 10 As shown in the lower left subplot, the long-term cumulative bias screening process can filter out approximately 5% of invalid time data, ultimately allowing only 65% ​​of valid data to pass through the entire screening process; from Figure 10 As can be seen from the lower right sub-figure, the effective time determination result is accurate and error-free. There was only one determination delay in the extreme test, and there were no misjudgments. This fully verifies that the three-level progressive verification logic can achieve multi-level and blind-spot-free interception of abnormal time.

[0097] Therefore, this implementation verification, through 30 days of continuous testing, shows that the success rate of identification is based on a massive number of test samples. The probability of extremely rare deviation scenarios occurring in actual applications is very low, and it can be further optimized through firmware upgrades, fully meeting the stringent requirements for time server security protection in key areas such as power dispatching, communication base stations, and industrial automation.

[0098] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any changes, modifications, substitutions, integrations, and parameter changes made to these embodiments within the spirit and principles of the present invention, without departing from the principles and spirit of the present invention, through conventional substitutions or to achieve the same function, fall within the scope of protection of the present invention.

Claims

1. A method for security protection of time servers that identifies abnormal times from multiple dimensions, characterized in that, include: Obtain an external reference time, the source of which includes at least one of a wireless source, a network time source, or a cable or fiber optic transmission source; The external reference time is sequentially subjected to large-scale discrimination, instantaneous discrimination, and long-term cumulative deviation discrimination; Based on the results of the wide-range screening, instantaneous screening, and long-term cumulative deviation screening, it is determined whether the external reference time is a valid time. Based on the determination of the effective time, the synchronization process between the time server and the external reference time is controlled.

2. The time server security protection method for multi-dimensional identification of abnormal times according to claim 1, characterized in that, The aforementioned large-scale screening of external reference times includes: The time server is preset to accept a minimum time of the device's factory date. The maximum acceptable time is preset to be the equipment's manufacturing date and maximum service life. The sum, the formula is: ; Determine the external reference time Does it meet the requirements? ; If the conditions are met, the external reference time is determined to have passed a wide range of screening; otherwise, it is determined to be an invalid time.

3. The time server security protection method for multi-dimensional identification of abnormal times according to claim 2, characterized in that, The large-scale screening uses a time-axis interval locking mechanism to identify two typical abnormal times that occur after the wireless source is interfered with, specifically: Time jump anomalies that occur after wireless signals are interfered with or spoofed, i.e., errors exceeding the maximum allowable time; Time rollback anomalies caused by wireless positioning device malfunctions, i.e., errors in time below the minimum allowable time; The time axis interval locking mechanism achieves targeted identification of abnormal scenarios of wireless time sources through preset time boundaries.

4. The time server security protection method for multi-dimensional identification of abnormal times according to claim 1, characterized in that, The instantaneous discrimination of the external reference time includes: Get the current time from the time server. Preset time difference threshold ; Calculate the absolute value of the difference between the external reference time and the current time. ; judge Is it less than ; If the time is less than the specified time, the external reference time is determined to have passed instantaneous identification; otherwise, it is determined to be an invalid time.

5. The time server security protection method for multi-dimensional identification of abnormal times according to claim 4, characterized in that, The time difference threshold A tiered configuration strategy is adopted, specifically as follows: At the basic configuration level, It meets the need for discerning temporal continuity in typical scenarios; At the high-precision application level, the accuracy of the timekeeping unit based on the time server will... Scalable down to milliseconds and below, suitable for application scenarios with stringent time synchronization accuracy requirements; The hierarchical configuration strategy enables flexible matching of instantaneous fluctuation identification accuracy under different application scenarios.

6. The time server security protection method for multi-dimensional identification of abnormal times according to claim 1, characterized in that, The long-term cumulative deviation identification of the external reference time includes: When the time server synchronizes with an external reference for the first time, a synchronization timer based on a high-precision timekeeping unit is started. The high-precision timekeeping unit includes an OCXO high-precision temperature-controlled crystal oscillator or a rubidium atomic clock. Obtain the timing time of the synchronization timer. Preset cumulative deviation threshold ; Calculate the absolute value of the difference between the external reference time and the timing time. ; judge Is it less than ; If the time is less than the specified value, the external reference time is determined to be identified through long-term cumulative deviation; otherwise, it is determined to be an invalid time.

7. The time server security protection method for multi-dimensional identification of abnormal times according to claim 6, characterized in that, The cumulative deviation threshold The frequency stability parameters and lifecycle aging coefficient of the high-precision timekeeping unit are dynamically tuned, specifically as follows: For OCXO high-precision temperature-controlled crystal oscillators, first calculate the theoretical deviation value per unit time. ,in For the duration of the time, To ensure crystal oscillator frequency stability, a lifetime aging margin factor is then introduced. ,get ; For rubidium atomic clocks, the tuning is directly based on their inherent frequency stability characteristics. ; The dynamic tuning method is adapted to the accuracy decay characteristics of the timekeeping unit throughout its entire life cycle, enabling accurate identification of accumulated deviations.

8. The time server security protection method for multi-dimensional identification of abnormal times according to claim 1, characterized in that, Construct a three-level progressive verification logic, specifically as follows: S0. External reference times need to be thoroughly screened to filter out invalid times outside the time axis interval; S1. After a large-scale screening process, the system enters the instantaneous screening stage to eliminate small-scale time fluctuation anomalies. S2. After the time spent in the first two stages of screening, the system enters the long-term cumulative deviation screening stage to achieve full-cycle precision control. S3. Only external reference times that pass three levels of continuous screening are determined to be valid times. The progressive verification logic realizes multi-level, all-round interception of abnormal times.

9. A time server security protection system for multi-dimensional identification of abnormal times, characterized in that, include: The time acquisition module acquires an external reference time, the source of which includes at least one of a wireless source, a network time source, or a cable or fiber optic transmission source; A wide-range discrimination module is connected to the time acquisition module and has a built-in time axis interval locking mechanism for performing wide-range discrimination on the external reference time. An instantaneous discrimination module, connected to the time acquisition module, is a threshold tuning unit employing a hierarchical configuration strategy, used to perform instantaneous discrimination on the external reference time; The long-term cumulative deviation discrimination module is connected to the time acquisition module and the high-precision timekeeping unit respectively, and is configured with a dynamically tuned cumulative deviation threshold calculation unit for performing long-term cumulative deviation discrimination on the external reference time. The synchronization control module is connected to the large-scale discrimination module, the instantaneous discrimination module, and the long-term cumulative deviation discrimination module, respectively. It has a built-in three-level discrimination progressive verification logic, which is used to control the synchronization process of the time server based on the results of each discrimination module.

10. The time server security protection system for multi-dimensional identification of abnormal times according to claim 9, characterized in that, The long-term cumulative deviation discrimination module is a dual-reference source comparison unit, including a synchronization timer subunit and a deviation dynamic calculation subunit, specifically: The synchronization timer subunit starts when the time server first synchronizes with an external reference. It is based entirely on the inherent frequency of the high-precision timekeeping unit and keeps time independently without relying on an external time source, thus avoiding timing deviations introduced by external interference. The deviation dynamic calculation subunit calculates the absolute value of the difference between the external reference time and the timing time of the synchronization timer subunit in real time, and calls the cumulative deviation threshold dynamically tuned based on frequency stability and lifetime aging coefficient to output the discrimination result. The dual reference source comparison unit enables independent verification between the external time source and the local timekeeping unit, ensuring the anti-interference and accuracy of long-term accumulated deviation identification.

Images