Self-diagnosis method and system for errors of stored program control exchange

Abstract

The invention relates to the technical field of stored program control exchange self-diagnosis, and provides a stored program control exchange error self-diagnosis method and system, and the method comprises the steps: starting a diagnosis task for a working module corresponding to an error correction event in response to a diagnosis trigger signal; on the basis of the started diagnosis task, collecting instantaneous operation parameters which are related to the error correction event and are not smoothed; performing time correlation on the instantaneous operation parameter and the error correction event, and storing the instantaneous operation parameter as context information of the error correction event; continuously analyzing the occurrence frequency of the error correction event, and identifying the performance attenuation trend of the working module corresponding to the error correction event in combination with the instantaneous operation parameter in the context information; and when the performance attenuation trend is identified, sending out an early warning containing a working module corresponding to the error correction event and potential root information. The method has the effect of improving the early warning capability and accuracy of fault diagnosis of the stored program control exchange.

Description

Technical Field

[0001] This invention relates to the technical field of self-diagnosis of program-controlled exchanges, and specifically to a method and system for self-diagnosis of errors in program-controlled exchanges. Background Technology

[0002] As core equipment in modern communication networks, the stable operation of PBXs is crucial for ensuring the quality of communication services. However, during long-term operation, PBXs may experience signal transmission interruptions or connection abnormalities due to gradual hardware aging or minor software logic defects. Traditional maintenance methods often rely on periodic manual inspections, which are not only slow in response but also difficult to accurately pinpoint the root cause of complex and deeply rooted faults. Especially under complex operating environments such as high loads, slow component performance degradation, and limitations of the diagnostic system itself, existing diagnostic methods struggle to effectively identify early and hidden hardware performance degradation, potentially leading to service interruptions and misjudgments by maintenance personnel.

[0003] Self-diagnostic systems face a dual challenge in identifying this gradual degradation. First, their fault judgment thresholds are typically set based on capacitors operating within their normal performance range. When power supply ripple slightly exceeds limits due to capacitor degradation but hasn't reached a hard alarm level, the system may fail to recognize it as an anomaly. Second, if the polling reads of the diagnostic task (whose timeliness is already affected by high load) happen to miss the brief spike in ripple, or if the read data is processed by the system's internal smoothing algorithm before entering the diagnostic logic, these subtle performance degradation signals are more easily masked. Smoothing algorithms are typically used to filter out transient noise and prevent false alarms, but in this case, they may also mistake a persistent but small increase in ripple for normal fluctuations, thus causing the system to lose its accurate ability to detect this gradual physical degradation.

[0004] This persistent but unidentified localized power supply ripple causes long-term, slight electrical stress on the high-speed digital circuitry within the switching matrix chip. Although the chip design incorporates certain power supply tolerances, prolonged operation in a slightly degraded power supply environment can affect the integrity of its internal signals. Under certain packet patterns, this can lead to extremely low-probability bit errors. To address these sporadic errors, modern switching matrix chips commonly integrate Error Correction Code (ECC) mechanisms. When these sporadic bit errors occur, the ECC mechanism can detect and correct them in real time at the hardware level, ensuring data integrity is restored before forwarding. However, systems typically log these "correctable error" events as low-priority "notification" or "information" level logs. These logs may only contain information about the time and location of the error, but lack sufficient contextual information, such as the specific flow pattern at the time of the error, the chip's internal temperature, or the precise voltage fluctuations of the power rails. More importantly, in a high-throughput PBX, the system generates massive amounts of routine operation logs. These sporadic, independent ECC correction records can easily be lost in the huge log stream and fail to be identified by system or maintenance personnel as clear warning signs of impending hardware failure. They are treated as tolerable events in normal operation rather than early signs of potential problems.

[0005] As the performance of electrolytic capacitors further degrades, local power supply ripple continues to worsen, leading to a gradual increase in the frequency of sporadic bit errors within the switching matrix chip. Eventually, at some point, the error rate exceeds the corrective capacity of the ECC mechanism, or it erupts in a concentrated burst within a very short period, making it impossible for the ECC to handle all errors in a timely manner. This directly results in packet processing delays or packet drops, triggering higher-level Quality of Service (QoS) alarms. However, the fault codes for these QoS alarms are often generic, such as being mapped by default to "network congestion," "link jitter," or "packet loss." When operations personnel receive these generic alarms, lacking a clear association with underlying hardware errors, they prioritize troubleshooting network-level factors such as network configuration, link status, routing problems, or abnormal external traffic based on experience. This conventional troubleshooting logic based on generic alarms causes operations personnel to miss the root cause of the problem—the gradual degradation of the switch's internal hardware performance. Such diagnostic biases not only delay the location and handling of the true root cause of the fault, but may also lead to unnecessary network configuration adjustments. They may even temporarily alleviate symptoms but fail to solve the underlying problems, thus creating potential risks for future service interruptions.

[0006] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0007] This application discloses a method and system for self-diagnosing errors in a program-controlled exchange, which aims to solve the problems that existing diagnostic methods are unable to effectively identify early and hidden hardware performance degradation in complex operating environments, as well as the slow response and difficulty in accurately finding the root cause of problems in traditional maintenance methods, leading to service interruptions and misjudgments by maintenance personnel.

[0008] The technical solution of this application is as follows: Firstly, this application discloses a method for self-diagnosing errors in a program-controlled exchange, specifically including the following steps: When a data transmission error is corrected, the error correction event corresponding to the data transmission error is identified as a diagnostic trigger signal; In response to the diagnostic trigger signal, a diagnostic task is initiated for the working module corresponding to the error correction event; Based on the diagnostic tasks initiated, collect unsmoothed transient runtime parameters related to error correction events; The instantaneous running parameters are correlated with the error correction event in time and stored as context information for the error correction event; By continuously analyzing the frequency of this error correction event and combining it with the instantaneous operating parameters in the context information, the performance degradation trend of the working module corresponding to the error correction event can be identified. When this performance degradation trend is detected, an alert is issued containing information about the working module corresponding to the bug fix event and the potential root cause.

[0009] This technical solution overcomes the traditional diagnostic system's neglect of correctable errors. By identifying error correction events as diagnostic trigger signals and combining them with unsmoothed instantaneous operating parameters for in-depth analysis, it effectively identifies early and hidden hardware performance degradation trends. This solves the problem of existing technologies failing to detect potential faults in a timely manner and can issue early warnings containing potential root cause information, providing maintenance personnel with more accurate fault location information.

[0010] Secondly, this application also discloses a self-diagnostic system for errors in a program-controlled exchange, used to perform self-diagnostic tests on errors in a program-controlled exchange, specifically including: The trigger signal recognition module is used to identify the error correction event corresponding to the data transmission error as a diagnostic trigger signal when the data transmission error is corrected. The diagnostic task initiation module is used to initiate a diagnostic task for the working module corresponding to the error correction event in response to the diagnostic trigger signal. The runtime parameter acquisition module is used to acquire unsmoothed instantaneous runtime parameters related to error correction events based on the diagnostic tasks that are initiated. The time correlation processing module is used to correlate the instantaneous running parameters with the error correction event in time, and store it as the context information of the error correction event; The performance degradation identification module is used to continuously analyze the frequency of occurrence of the error correction event and, in combination with the instantaneous operating parameters in the context information, identify the performance degradation trend of the working module corresponding to the error correction event. The warning information sending module is used to issue a warning containing information about the working module corresponding to the error correction event and the potential root cause when the performance degradation trend is detected.

[0011] This application provides a system that can implement the above-mentioned method through this technical solution. Through modular design, it ensures timely identification of error correction events, accurate acquisition of instantaneous operating parameters, effective storage of context information, intelligent identification of performance degradation trends, and accurate transmission of early warning information, thereby realizing early and accurate diagnosis and early warning of potential faults in program-controlled exchanges.

[0012] Beneficial Effects: The self-diagnostic method for errors in a program-controlled exchange disclosed in this application identifies error correction events generated when data transmission errors are corrected as diagnostic trigger signals. This overcomes the traditional diagnostic system's neglect of correctable errors and solves the problem in existing technologies where these events are buried in the log stream and fail to be identified as clear warning signals of impending hardware failure. In response to the diagnostic trigger signal, this method initiates a diagnostic task for the working module corresponding to the error correction event and collects unsmoothed instantaneous operating parameters related to the error correction event at a higher frequency than conventional polling. These parameters include power supply quality parameters, signal transmission quality parameters, and environmental parameters. This high-frequency, unsmoothed parameter collection effectively avoids the problem of traditional diagnostic systems losing weak performance degradation signals due to reduced resource scheduling priority or smoothing algorithms, enabling the system to capture early, hidden signs of hardware performance degradation.

[0013] Furthermore, this application correlates the collected instantaneous operating parameters with error correction events over time and stores this information as contextual information, providing a rich data foundation for subsequent in-depth analysis. By continuously analyzing the frequency of error correction events and combining the instantaneous operating parameters in the contextual information, this application can identify the performance degradation trend of the corresponding working module. This comprehensive analysis method overcomes the limitations of existing technologies that rely solely on general alarms for troubleshooting, avoiding misjudgment or delayed fault location by maintenance personnel due to a lack of clear underlying hardware error alarms. When a performance degradation trend is identified, the system issues an early warning containing information about the corresponding working module and potential root causes of the error correction event, providing maintenance personnel with accurate fault location information and enabling timely intervention. This effectively solves the problems of existing technologies failing to detect potential faults in a timely manner, leading to service interruptions and misjudgments by maintenance personnel. In summary, this application, through innovative diagnostic triggering mechanisms, high-frequency instantaneous parameter collection, contextual information correlation, and comprehensive performance degradation trend identification, significantly improves the early warning capability and accuracy of fault diagnosis in program-controlled exchanges, effectively ensuring the stable operation of the communication network. Attached Figure Description

[0014] Figure 1 This is a flowchart of a method for self-diagnosing errors in a program-controlled exchange, as described in one embodiment of the present invention. Figure 2 This is a flowchart of a method for self-diagnosing errors in a program-controlled exchange, according to another embodiment of the present invention. Figure 3 This is a system block diagram of a self-diagnostic system for errors in a program-controlled exchange, according to another embodiment of the present invention. Explanation of reference numerals in the attached figures: 1. Program-controlled exchange error self-diagnosis system; 11. Trigger signal identification module; 12. Diagnostic task startup module; 13. Operating parameter acquisition module; 14. Time correlation processing module; 15. Performance degradation identification module; 16. Early warning information sending module. Detailed Implementation

[0015] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0016] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0017] This application proposes a self-diagnosis method for errors in a program-controlled exchange, combined with... Figure 1 As shown, it includes the following steps: S1, When a data transmission error is corrected, the error correction event corresponding to the data transmission error is identified as a diagnostic trigger signal; S2, in response to the diagnostic trigger signal, initiates a diagnostic task for the working module corresponding to the error correction event; S3, based on the initiated diagnostic task, collects unsmoothed transient operating parameters related to error correction events; S4. The instantaneous running parameters are correlated with the error correction event over time and stored as the context information of the error correction event. S5 continuously analyzes the frequency of error correction events and, in conjunction with instantaneous operating parameters in the context information, identifies the performance degradation trend of the working module corresponding to the error correction event. S6, when a performance degradation trend is detected, issues an alert containing information about the working module corresponding to the error correction event and the potential root cause.

[0018] To facilitate understanding of the self-diagnosis method for errors in program-controlled exchanges proposed in this application, the following explains the key terms and implementation environment involved.

[0019] A program-controlled exchange is a communication device that uses program control to perform functions such as telephone call connection and data exchange. Internally, it typically includes multiple working modules such as a switching matrix, circuit boards, and power supply modules. This application focuses on program-controlled exchanges as the target object for error self-diagnosis.

[0020] Error correction events refer to events in which a program-controlled exchange detects errors during data transmission and corrects them through its internal error correction mechanism, such as the event generated when the error correction code (ECC) mechanism inside the switching matrix chip detects and corrects bit errors. These events are usually logged as low-priority operation logs by the system, but this application uses them as an important signal source for triggering diagnostics.

[0021] A diagnostic trigger signal is a signal used to initiate a diagnostic process. In this application, when an error correction event is detected, the error correction event is identified as a diagnostic trigger signal to indicate that further diagnostics need to be performed on the relevant working module.

[0022] A working module refers to a hardware or software unit within a program-controlled exchange that has independent functions and operational capabilities, including but not limited to switching matrix chips, circuit boards, and power supply units.

[0023] Instantaneous operating parameters refer to raw operating data collected at a specific point in time that has not been smoothed or averaged. These parameters reflect the actual operating status of the equipment over a short time scale and may include power supply quality parameters, signal transmission quality parameters, and environmental parameters.

[0024] Context information refers to the set of runtime data that is temporally associated with an error correction event. In this application, instantaneous runtime parameters are temporally associated with an error correction event and stored as the context information of that error correction event to support subsequent diagnostic analysis.

[0025] The performance degradation trend refers to the gradual deterioration of the performance of a working module over time, which can be manifested as an increase in the frequency of error correction events and abnormal fluctuations in operating parameters.

[0026] Early warning refers to alarm information issued before an actual failure occurs, used to alert maintenance personnel to potential risks and guide them to take preventative measures.

[0027] The implementation environment of this application is typically a diagnostic system or independent monitoring unit deployed inside a program-controlled exchange, which has the ability to collect, process, store and analyze operational data.

[0028] This application proposes a self-diagnostic method for errors in a program-controlled exchange. Its core lies in identifying the performance degradation trend of the working module based on error correction events and issuing an early warning before the fault occurs.

[0029] First, when a data transmission error occurs and is corrected during the operation of the PBX, the corresponding error correction event is identified as a diagnostic trigger signal. Specifically, the system log stream can be monitored in real time by a log parser. When preset identification information related to error correction is detected, the event is identified as an error correction event and further used as a diagnostic trigger signal. For example, a time window or quantity threshold can be set, triggering diagnosis when the number of error correction events reaches a preset condition within a unit of time, or directly triggering diagnosis when any error correction event occurs.

[0030] Secondly, in response to the diagnostic trigger signal, the diagnostic task for the corresponding working module of the error correction event is initiated. The diagnostic system determines the corresponding working module based on the source information carried by the error correction event, and ensures that the diagnostic program for that working module is executed in a timely manner by sending diagnostic instructions or increasing the scheduling priority.

[0031] Subsequently, after the diagnostic task is initiated, instantaneous operating parameters associated with error correction events are acquired. The acquisition employs a higher polling frequency than usual, and the acquired data is not smoothed. For example, short-term abnormal fluctuations can be captured by directly reading power quality parameters, signal transmission quality parameters, and local environmental parameters from hardware registers at millisecond or microsecond frequencies using a high-speed data acquisition unit.

[0032] Next, the collected instantaneous operating parameters are correlated with the corresponding error correction events in time and stored as context information for those events. Specifically, instantaneous operating parameters within a preset time window before and after the error correction event are selected based on the event occurrence time and stored in the diagnostic database along with the event identifier to ensure that subsequent analysis can accurately reproduce the operating state at the time of the error.

[0033] Subsequently, the frequency of error correction events is continuously analyzed, and combined with instantaneous operating parameters in the context information, the performance degradation trend of the corresponding working module is identified. The system can perform time-series statistics on error correction events and compare them with historical baselines, while also analyzing the changes in associated instantaneous operating parameters. When the frequency of error correction events is found to be continuously increasing, accompanied by abnormal trend changes in operating parameters, it is determined that the working module has a performance degradation trend, even if the parameters have not yet reached the traditional alarm threshold.

[0034] Finally, when a performance degradation trend is identified, an early warning message is generated and output. The warning message must at least include the identifier of the working module experiencing performance degradation, and provide potential root cause suggestions based on contextual information. The warning message can be output through the network management system, alarm platform, or other operation and maintenance interfaces to guide operation and maintenance personnel to intervene promptly and perform preventative maintenance operations.

[0035] Optionally, based on the initiated diagnostic task, the steps of collecting unsmoothed transient runtime parameters related to error correction events include: Based on the diagnostic task initiated, transient operational parameters related to error correction events are collected at a higher frequency than regular polling. These transient operational parameters include power supply quality parameters, signal transmission quality parameters, and environmental parameters.

[0036] Specifically, collecting instantaneous operating parameters at a higher frequency than the regular polling frequency refers to the system temporarily increasing the frequency of collecting operating parameters for the target working module after the diagnostic task is started. Compared to the data collection at the second level in the regular polling mode, the collection frequency can be increased to the millisecond or microsecond level during the diagnostic task to capture transient abnormal fluctuations that occur within a short period of time. In this way, it is possible to avoid the small anomalies that are masked by time averaging during the regular polling process, thus providing a more refined data foundation for the early identification of performance degradation.

[0037] Instantaneous operating parameters directly reflect the operating status of the working module at the moment of diagnostic triggering and the impact of the external environment. Specifically, power supply quality parameters may include instantaneous changes in input and output voltage, current ripple, power supply noise, and instantaneous power consumption, used to assess the impact of power supply stability on the performance of the working module; signal transmission quality parameters may include signal-to-noise ratio, instantaneous bit error rate, packet loss rate, transmission delay, and signal strength, used to assess the health of the data communication link over a short timescale; environmental parameters may include local temperature, humidity, vibration, or air pressure of the working module, used to assess the impact of changes in the external environment on the module's stability. The high-frequency acquisition of these multiple parameters enables the diagnostic system to comprehensively characterize the instantaneous operating characteristics of the working module from multiple dimensions, including power supply, signal, and environment.

[0038] In some preferred embodiments, the data processing module in a program-controlled exchange is used as an example. Assume that after long-term operation, the internal power supply regulation circuit of this data processing module exhibits slight aging. In normal polling mode, because the collected data undergoes time averaging, short-term fluctuations in the output voltage may still appear within the normal range, making it difficult to detect anomalies in a timely manner. However, when a data transmission error occurs and is corrected, and this error correction event is identified as a diagnostic trigger signal, the system will immediately initiate a diagnostic task for the data processing module.

[0039] After the diagnostic task is initiated, the operational parameter acquisition module collects the module's instantaneous operational parameters at a significantly higher frequency than conventional polling (e.g., once every 100 microseconds). These parameters include the instantaneous values ​​of its power input voltage, output voltage, and ripple coefficient. Simultaneously, it acquires signal transmission quality parameters such as the instantaneous bit error rate and signal strength at the data output port, and synchronously obtains local temperature information within the module. This high-frequency acquisition method ensures that even extremely short-lived voltage drops or momentary signal quality degradations can be accurately recorded and correlated with corresponding error correction events.

[0040] For example, milliseconds before an error correction event occurs, the system can detect a momentary drop in the data processing module's output voltage exceeding a preset small threshold, or a brief increase in the bit error rate. These unsmoothed instantaneous operating parameters, combined with changes in the frequency of error correction events, can be used to identify a gradual deterioration trend in the power supply regulator circuit's performance. In this way, the system can issue early warnings before a module experiences a significant failure, guiding maintenance personnel to conduct timely inspections or replacements, thus preventing further escalation of the fault.

[0041] Optional, combined Figure 2As shown, the steps by which S5 continuously analyzes the frequency of error correction events and, in conjunction with instantaneous runtime parameters in the context information, identifies the performance degradation trend of the working module corresponding to the error correction event include: S51 establishes an operating status reference range for the power supply quality parameters and signal transmission quality parameters of the working module corresponding to the error correction event; the operating status reference range is initialized based on the historical operating data of the program-controlled exchange under different load levels and ambient temperature conditions. S52 adjusts the operating status reference range in real time based on the current real-time load and ambient temperature of the program-controlled exchange; S53, compare the current instantaneous operating parameters with the operating state reference interval, determine whether the instantaneous operating parameters exceed the upper or lower limit of the operating state reference interval, and obtain the interval comparison result; S54, when the interval comparison result indicates an excess, the instantaneous operating parameter is marked as an abnormal fluctuation; S55: When the frequency of error correction events increases and abnormal fluctuations occur, identify the performance degradation trend of the working module corresponding to the error correction event.

[0042] Specifically, the operating status reference range refers to a dynamically acceptable range set for the key operating parameters of the working module (such as power supply quality parameters and signal transmission quality parameters). This reference range is not fixed, but is initialized based on historical operating data of the PBX under different load levels and ambient temperature conditions to reflect the parameter fluctuation characteristics of the system under various normal operating conditions. For example, during the system design phase or after a long period of stable operation, a large amount of data can be collected to analyze the normal fluctuation range of parameters such as power supply voltage, current ripple, signal-to-noise ratio, and bit error rate under light load, medium load, heavy load, and different ambient temperatures (such as low temperature, normal temperature, and high temperature), and an initial reference range can be established accordingly.

[0043] The real-time adjustment of the operating status reference range can be understood as dynamically correcting or selecting the most suitable reference range based on the current real-time load and ambient temperature of the PBX. For example, when the PBX is under high load and the ambient temperature is high, the normal fluctuation range of some parameters may be wider than when it is under low load and low temperature. In this case, the reference range will be widened accordingly to avoid misjudging normal fluctuations caused by environmental changes as abnormalities. Conversely, in a stable, low-load environment, the reference range may be narrowed to improve the sensitivity of anomaly detection.

[0044] In practical applications, the current instantaneous operating parameters are compared with the operating state reference range to determine whether the current parameter value deviates from the fluctuation range considered normal under the current operating conditions. When the interval comparison result indicates that the instantaneous operating parameter exceeds the upper or lower limit of the operating state reference range, the parameter is marked as abnormal fluctuation. This abnormal fluctuation is different from a simple parameter exceeding the limit; it is a "true" abnormality after considering the influence of the current operating environment.

[0045] Ultimately, an increase in the frequency of bug fix events, coupled with abnormal fluctuations, is recognized as a performance degradation trend for the corresponding working module. This means that a single increase in frequency or a single abnormal fluctuation in a parameter is insufficient to trigger a judgment of performance degradation; only a combination of both can more accurately indicate the actual decline in module performance.

[0046] In the self-diagnosis method for errors in a PBX (Publicly Controlled Switchgear), when the operating status reference range is adjusted in real time based on the current real-time load and ambient temperature of the PBX, there is a direct and dynamic correspondence between the adjustment amount and the current real-time load and ambient temperature. This correspondence is not a simple linear change, but is determined after modeling the historical operating behavior of the PBX under different operating conditions.

[0047] The operating status reference interval is determined during the initialization phase based on historical operating data of the PBX under different load levels and ambient temperatures. This historical operating data characterizes the normal fluctuation range of key operating parameters of the PBX under various typical operating conditions. These key operating parameters include at least power supply quality parameters and signal transmission quality parameters. The resulting operating status reference interval constitutes a multi-dimensional behavioral baseline reflecting the normal operating status of the PBX and can be stored as a set of intervals or a parameter model corresponding to each operating condition.

[0048] During the actual operation of the PBX, the system acquires the current load level and ambient temperature in real time and adjusts the operating status reference range based on this operating condition. Specifically, when historical operating data is stored in discrete operating condition format, the system matches the current operating condition with preset typical operating conditions and selects the operating status reference range corresponding to the operating condition that is closest to or consistent with the current operating condition, thereby realizing the switching of the reference range. When historical operating data supports continuous modeling, the system dynamically calculates the upper and lower limits of the operating status reference range based on the parameter model established by historical data, combined with the current real-time load and ambient temperature, so that the reference range is continuously adjusted as the operating condition changes.

[0049] During the aforementioned adjustments, the adjustment amount of the operating status reference range is manifested as the dynamic change in the reference range boundary due to variations in the current real-time load and ambient temperature. When the real-time load increases, the normal fluctuation range of power supply quality parameters and signal transmission quality parameters increases accordingly due to the increased workload of the internal components of the PBX, and the system appropriately widens the operating status reference range. When the real-time load is low, the operating status reference range is tightened accordingly to improve the sensitivity to abnormal changes. The adjustment method for the operating status reference range based on ambient temperature changes is similar. When the ambient temperature rises, considering the impact of changes in electronic component performance on operating parameters, the system expands the operating status reference range accordingly; when the ambient temperature is low, it tightens the range.

[0050] By using the above methods, the operating status reference range is kept consistent with the current actual operating conditions of the program-controlled exchange. This allows for a more accurate distinction between normal parameter fluctuations caused by changes in operating conditions and abnormal parameter changes caused by hardware performance degradation or malfunctions during subsequent diagnostic processes. This reduces misjudgments and omissions, and improves the reliability of error self-diagnosis.

[0051] Optionally, the steps to continuously analyze the frequency of error correction events and, in conjunction with instantaneous runtime parameters in the context information, identify the performance degradation trend of the working module corresponding to the error correction event include: Obtain the historical fluctuation range of several operating parameters of the working module corresponding to the error correction event under different load levels and ambient temperature conditions; the several operating parameters include the occurrence frequency of the error correction event, power supply quality parameters, signal transmission quality parameters, and environmental parameters; The current load level and ambient temperature of the PBX are obtained in real time, and the current normal fluctuation range of the operating parameters is determined and adjusted based on the current load level and ambient temperature, combined with the historical fluctuation range. The frequency of continuously monitored error correction events is compared with the corresponding adjusted normal fluctuation range to determine whether the frequency of error correction events exceeds the adjusted normal fluctuation range, and the result of the frequency exceeding the normal fluctuation range is obtained. Continuously monitor power supply quality parameters, signal transmission quality parameters, and environmental parameters, and compare each parameter with its adjusted normal fluctuation range to obtain the judgment result of parameter exceeding the range; When the frequency exceeds the judgment result and the parameter exceeds the judgment result, indicating that any parameter exceeds the corresponding normal fluctuation range, the performance degradation trend of the working module corresponding to the error correction event is identified.

[0052] Specifically, to improve the accuracy of performance degradation trend identification, the historical fluctuation range of the working module corresponding to the error correction event is first obtained under different load levels and ambient temperature conditions. The historical fluctuation range is obtained based on statistical analysis of long-term operating data of the PBX under various operating conditions, and is used to characterize the normal fluctuation range of various operating parameters under specific load and ambient temperature combinations. Operating parameters include at least the frequency of error correction events, power supply quality parameters, signal transmission quality parameters, and environmental parameters. Since load level and ambient temperature significantly affect the operating status of the working module, the normal fluctuation range of each parameter is usually wider under heavy load or high temperature conditions compared to light load or normal temperature conditions.

[0053] Real-time acquisition of the current load level and ambient temperature of the PBX is the foundation for dynamically adjusting the judgment criteria. Based on the real-time collected current operating conditions, the system selects the parameter fluctuation range corresponding to the operating condition from a pre-established historical fluctuation range dataset and uses it as the current normal fluctuation range. Therefore, parameter anomaly judgment is no longer based on a fixed threshold, but on a dynamic range that changes with operating conditions, thereby improving the adaptability and accuracy of the judgment results to complex operating environments.

[0054] In practical applications, the system continuously monitors the frequency of error correction events and compares their real-time values ​​with the dynamically adjusted normal fluctuation range to determine if any exceedances are possible. Simultaneously, instantaneous operating parameters such as power supply quality, signal transmission quality, and environmental parameters are monitored and compared with their corresponding dynamic normal fluctuation ranges. Only when the frequency of an error correction event exceeds its current normal fluctuation range, and at least one other key operating parameter also exceeds its corresponding dynamic normal fluctuation range, does the system determine that the working module corresponding to the error correction event exhibits a performance degradation trend. This multi-dimensional, dynamic threshold-based joint judgment mechanism effectively avoids misjudgments caused by short-term fluctuations in a single parameter or changes in environmental conditions.

[0055] In the step of "determining and adjusting the current normal fluctuation range of operating parameters based on the current load level and ambient temperature, combined with the historical fluctuation range", the adjustment amount has a dynamic correspondence with the current real-time load level and ambient temperature. This correspondence is established based on the historical operating data of the program-controlled exchange under different operating conditions.

[0056] The historical fluctuation range is established based on long-term operating data of the PBX under different load levels and ambient temperature conditions. This operating data characterizes the normal fluctuation characteristics of the operating parameters of each working module under healthy conditions. These operating parameters include at least power supply quality parameters, signal transmission quality parameters, environmental parameters, and the frequency of error correction events. The system statistically analyzes this operating data to establish the normal fluctuation range for each operating parameter under different combinations of load levels and ambient temperatures, thus forming a set of historical fluctuation ranges for subsequent diagnostics. This set of historical fluctuation ranges can be stored as an interval mapping table corresponding to load-temperature conditions, or as a parameter model.

[0057] During the actual operation of the PBX, the system acquires the current load level and ambient temperature in real time, and adjusts the current normal fluctuation range of the operating parameters based on this operating condition. This adjustment is not a matter of adding or subtracting fixed values, but rather, based on the current operating condition, selecting or deriving a normal fluctuation range that matches the current operating condition from the historical fluctuation range set. The adjustment amount is expressed as the change in the upper and lower limits of the current normal fluctuation range relative to the fluctuation range under the baseline operating condition.

[0058] When the current real-time load level and ambient temperature match the preset typical operating conditions in the historical fluctuation range set, the system directly selects the historical fluctuation range under the corresponding typical operating condition as the current normal fluctuation range. In this case, the adjustment amount reflects the difference between the historical fluctuation range and the benchmark fluctuation range. When the current operating condition does not completely match any typical operating condition, the system calculates and processes the historical fluctuation ranges corresponding to adjacent operating conditions based on the parameter model established by historical operating data, and derives the normal fluctuation range under the current operating conditions.

[0059] During the aforementioned adjustments, the impact of the current real-time load level and ambient temperature on the normal fluctuation range is reflected in the dynamic changes of the interval boundaries. As the load level increases, the processing intensity of the working module increases, power consumption and internal temperature rise rise, and the normal fluctuation amplitude of some operating parameters increases accordingly. The system then widens the adjustment of the normal fluctuation range. When the load level is low, the system tightens the normal fluctuation range accordingly to improve the sensitivity of anomaly detection. The adjustment of the normal fluctuation range based on changes in ambient temperature is similar. When the ambient temperature rises, considering the impact of changes in electronic component performance on the stability of operating parameters, the system expands the upper and / or lower limits of the normal fluctuation range accordingly. When the ambient temperature is low, the system tightens the adjustment.

[0060] In addition, the synergistic effect between load level and ambient temperature has been considered in the establishment of the historical fluctuation range. When both increase simultaneously, the adjustment range of the system to the normal fluctuation range is greater than that under the condition of a single factor change, so that the determined current normal fluctuation range can truly reflect the parameter change characteristics under complex operating conditions.

[0061] In this way, the system can dynamically determine the normal fluctuation range that best matches the actual operating state under different operating conditions. In the subsequent error self-diagnosis process, it can more accurately distinguish between normal parameter fluctuations caused by changes in operating conditions and parameter deviations caused by hardware performance degradation or abnormalities, thereby reducing the probability of misjudgment and missed judgment and improving the reliability of diagnostic results.

[0062] In the error self-diagnosis method of PBX (Publicly Controlled Switchgear), determining and adjusting the current normal fluctuation range of operating parameters based on the current load level and ambient temperature, combined with historical fluctuation ranges, is a key step in achieving accurate identification of performance degradation. This step ensures that the judgment of abnormal operating parameters fully reflects the actual operating conditions of the PBX, avoiding the misjudgment of normal parameter fluctuations caused by changes in load or environmental conditions as performance degradation.

[0063] When the operating parameter is a signal transmission quality parameter, its current normal fluctuation range is adjusted as follows. The signal transmission quality parameter characterizes the data forwarding capability and signal stability of the PBX (Publicly Controlled Switching Exchange), and its normal fluctuation characteristics are significantly affected by load level and ambient temperature. Based on historical operating data, the system pre-establishes historical fluctuation ranges for the signal transmission quality parameter under different load levels and ambient temperatures. During the real-time operation of the PBX, the system acquires the current load level and ambient temperature, and, in conjunction with the historical fluctuation range, dynamically determines the normal fluctuation range of the signal transmission quality parameter under the current operating conditions. This determination process can be achieved through operating condition matching, interval calculation, or modeling based on historical data, allowing the determined normal fluctuation range to adjust with changes in operating conditions. When monitoring the signal transmission quality parameter, the system compares the real-time collected parameter values ​​with the dynamically adjusted normal fluctuation range; only when the parameter value exceeds this range is it determined to be an abnormal fluctuation.

[0064] When the operating parameters are environmental parameters, the adjustment method for their current normal fluctuation range is similar to the method described above. Environmental parameters reflect the local operating environment status of each working module within the program-controlled exchange, and their normal fluctuation range is also affected by the overall load level and external ambient temperature. Based on historical operating data, the system establishes historical fluctuation ranges for the environmental parameters of each working module under different load levels and external ambient temperatures. During real-time operation, the system acquires the current load level and external ambient temperature, and, in conjunction with the historical fluctuation ranges, dynamically determines the normal fluctuation range of the environmental parameters of each working module under the current operating conditions. Through dynamic adjustment of the normal fluctuation range, reasonable changes in environmental parameters under high load or high temperature operating conditions can be correctly identified and will not be misjudged as abnormal.

[0065] In some preferred embodiments, the data processing module in a program-controlled exchange is used as an example for illustration. It is assumed that the historical normal fluctuation range of the error correction event frequency of this data processing module under low load and normal temperature conditions is 0 to 5 times per hour, while under high load and high temperature conditions, the historical normal fluctuation range of this frequency is 0 to 15 times per hour; simultaneously, the voltage fluctuation rate in its power supply quality parameters has a historical normal fluctuation range of ±2% under low load and normal temperature conditions, and ±5% under high load and high temperature conditions.

[0066] During a certain monitoring period, the system detected that the PBX was operating under high load and high ambient temperature. The system first obtains the current load level and ambient temperature in real time, and accordingly adjusts the current normal fluctuation range of the error correction event frequency to 0 to 15 times per hour, and adjusts the current normal fluctuation range of the power supply quality parameters to ±5%.

[0067] During this monitoring period, the system continuously detected that the frequency of error correction events gradually increased from 5 times per hour to 12 times per hour. Although the frequency showed an upward trend, it remained within the adjusted normal fluctuation range. Therefore, the system did not determine that the frequency exceeded the limit and would not conclude that there was performance degradation based on this.

[0068] In another implementation scenario, if the frequency of error correction events increases from 5 times per hour to 18 times per hour, exceeding the adjusted normal fluctuation range (0 to 15 times / hour), and at the same time, the voltage fluctuation rate of the power supply quality parameter increases from ±3% to ±6%, also exceeding the adjusted normal fluctuation range (±5%), then the system will simultaneously meet both the "frequency of occurrence exceeds" and "critical operating parameters exceed" judgment conditions, thereby identifying that the data processing module has a performance degradation trend and generating corresponding warning information.

[0069] By utilizing the above method and combining the normal fluctuation range of dynamic adjustment with the joint judgment of multiple parameters, the performance degradation trend of working modules can be identified more accurately under complex load and environmental conditions, effectively reducing the risk of false alarms or false alarms caused by relying on a single indicator.

[0070] Optionally, when the frequency of occurrence exceeds the judgment result and the above parameter exceedance judgment result indicates that any parameter exceeds the corresponding normal fluctuation range, the steps to identify the performance degradation trend of the working module corresponding to the error correction event include: Obtain the historical correlation range of several operating parameters of the working module corresponding to the error correction event under different load levels and ambient temperature conditions; Calculate the instantaneous correlation between several current operating parameters; Determine whether the instantaneous correlation continues to deviate from the historical correlation range; When the instantaneous correlation continuously deviates from the historical correlation range and the frequency of error correction events increases, identify the performance degradation trend of the working module corresponding to the error correction event.

[0071] Specifically, obtaining the historical correlation range of several operating parameters of the working module corresponding to the error correction event under different load levels and ambient temperature conditions refers to establishing a baseline model of the degree of correlation between various operating parameters (such as power supply quality parameters, signal transmission quality parameters, environmental parameters, and the frequency of error correction events) through long-term monitoring and data analysis under normal system conditions. This historical correlation range may include, but is not limited to, statistical indicators such as Pearson correlation coefficient, Spearman rank correlation coefficient, and mutual information, used to quantify the linear or nonlinear relationship between parameters. These ranges are obtained through training and learning based on historical operating data of the PBX under different load levels (e.g., low load, medium load, high load) and ambient temperature conditions (e.g., low temperature, normal temperature, high temperature) to reflect the inherent behavioral patterns of the system under healthy conditions.

[0072] Calculating the instantaneous correlation between several current operating parameters refers to the system periodically collecting instantaneous operating parameters within the current moment or a short time window during real-time monitoring, and calculating the correlation index between these real-time parameters using the same method as when establishing historical correlation ranges. For example, data can be collected every certain period of time (such as 1 minute or 5 minutes), and the correlation between each parameter within the current time window can be calculated.

[0073] Furthermore, determining whether instantaneous correlation continuously deviates from the historical correlation range involves comparing the real-time calculated instantaneous correlation with a pre-established historical correlation range. If the instantaneous correlation indicator (e.g., the correlation coefficient between one parameter pair and another) exceeds its normal fluctuation range observed in historical data, and this deviation is not sporadic but persists for a period of time (e.g., N consecutive sampling periods) or reaches a certain deviation magnitude, then a continuous deviation is considered to exist. This continuous deviation may indicate a change in the operating mechanism within or between modules.

[0074] Therefore, when instantaneous correlation continuously deviates from the historical correlation range and the frequency of error correction events increases, the system will comprehensively judge and identify the performance degradation trend of the working module corresponding to the error correction event. This means that even if some individual parameters have not significantly exceeded their absolute thresholds, but their relationship patterns with other parameters change abnormally, combined with the increase in the frequency of error correction events, it can more sensitively and accurately capture potential performance degradation.

[0075] Optionally, when the instantaneous correlation continuously deviates from the historical correlation range and the frequency of error correction events increases, the steps to identify the performance degradation trend of the working module corresponding to the error correction event include: Obtain the historical correlation range of the operating parameters of each working module inside the program-controlled exchange under different load levels and ambient temperature conditions; Calculate the instantaneous correlation between the operating parameters of each working module; Determine whether the instantaneous correlation between the operating parameters of each working module continuously deviates from the historical correlation range; When the instantaneous correlation between the operating parameters of each working module continues to deviate from the historical correlation range, and the frequency of error correction events increases, identify the performance degradation trend of the working module corresponding to the error correction event.

[0076] Specifically, obtaining the historical correlation range of operating parameters of various working modules within a PBX under different load levels and ambient temperatures refers to the system establishing a historical correlation model of operating parameters (including but not limited to the frequency of error correction events, power supply quality parameters, signal transmission quality parameters, and environmental parameters) of all key working modules (e.g., CPU module, memory module, switching matrix module, interface module, power supply module, etc.) under different operating conditions (such as low load, medium load, high load, and different ambient temperature ranges) through long-term operational data analysis. These historical correlation ranges can be represented as intervals of correlation coefficients, eigenvalue ranges of the covariance matrix, or more complex nonlinear correlation patterns, with the aim of providing a benchmark for subsequent real-time monitoring.

[0077] The calculation of the instantaneous correlation between the operating parameters of each working module can be understood as the system continuously collecting the operating parameters of each working module within the program-controlled exchange during real-time operation, and using statistical methods (such as Pearson correlation coefficient, Spearman rank correlation coefficient, mutual information, etc.) to calculate the degree of interrelationship of these parameters within the current time window. This instantaneous correlation reflects the dynamic relationship between the parameters of each module under the current system state.

[0078] In practical applications, determining whether the instantaneous correlation between the operating parameters of each working module continuously deviates from the historical correlation range involves comparing the instantaneous correlation between the operating parameters of each working module, calculated in real time, with a pre-established historical correlation range. If the instantaneous correlation continuously exceeds or falls below the historical range, or if its change pattern significantly deviates from the historical pattern, it is determined to be a continuous deviation. This continuous deviation may indicate some abnormal change in the system's internal structure, function, or environment.

[0079] Optionally, when it is determined that there is a transient correlation that continuously deviates from the historical correlation range, and the frequency of error correction events is increasing, the steps to identify the performance degradation trend of the working module corresponding to the error correction event include: Based on the collected operating parameters of each working module, calculate the instantaneous correlation between parameters within each working module, as well as the instantaneous correlation between corresponding parameters between different working modules; The instantaneous correlation between parameters within each working module and the instantaneous correlation between corresponding parameters between different working modules are compared with the preset historical correlation range to determine whether there is a continuous deviation. When at least two working modules simultaneously exhibit transient correlations that continuously deviate from the historical correlation range, and the frequency of error correction events increases, a comprehensive evaluation is conducted based on the degree of deviation of the transient correlation of each working module, the magnitude of the increase in the frequency of error correction events, and the changes in the correlation between modules, to obtain comprehensive evaluation information. Based on comprehensive evaluation information, the performance degradation can be distinguished as either the performance degradation of a single module or the synergistic performance degradation of multiple modules, and the corresponding performance degradation trend can be identified.

[0080] Specifically, during the operation of the PBX, the system continuously collects operating parameters from each working module, including power supply quality parameters, signal transmission quality parameters, and environmental parameters. Based on the collected operating parameters, the system calculates the instantaneous correlation between parameters within each working module, such as the correlation between processor load and memory utilization within the module. Simultaneously, the system also calculates the instantaneous correlation between corresponding parameters between different working modules, such as the correlation between the output signal quality of module A and the input signal quality of module B. Instantaneous correlations can be quantified using statistical analysis methods, including but not limited to Pearson correlation coefficient, Spearman's rank correlation coefficient, or mutual information index.

[0081] The preset historical correlation range refers to the correlation baseline established through long-term operation monitoring and statistical analysis when the program-controlled exchange is in normal operation. It characterizes the normal correlation range between parameters within each working module and between corresponding parameters of different working modules. The historical correlation range is typically modeled hierarchically, incorporating different load levels and ambient temperature conditions, to reflect the normal fluctuation characteristics of correlation under different operating conditions. Comparing the instantaneous correlation calculated in real time with the corresponding historical correlation range can be used to determine whether there is a continuous deviation. A continuous deviation refers to the instantaneous correlation consistently exceeding the upper or lower limit of the historical correlation range for multiple consecutive sampling periods or within a preset time window, indicating that the operating status of the working module or between modules may have changed abnormally.

[0082] In practical applications, when the system detects that at least two working modules simultaneously exhibit a sustained deviation from their historical correlation range in instantaneous correlation, and the frequency of error correction events also shows an increasing trend, the system will initiate a comprehensive evaluation process. This comprehensive evaluation considers multiple dimensions of judgment factors, including the deviation magnitude of the instantaneous correlation of each working module, the degree of increase in the frequency of error correction events, and changes in the correlation structure between different working modules. For example, the parameter correlation between previously highly correlated modules may significantly decrease, or new correlations may appear between previously weakly correlated or uncorrelated modules. By performing weighted analysis or multi-dimensional joint evaluation of the above factors, comprehensive evaluation information reflecting the overall operating status of the system is generated.

[0083] Based on comprehensive evaluation information, the system further distinguishes the types of performance degradation to determine whether it is an independent performance degradation of a single working module or a synergistic performance degradation caused by the interaction of multiple working modules. When the correlation of internal parameters of a working module deviates significantly, while the correlation between it and other modules does not change significantly, it can be identified as an independent performance degradation of that module. When the correlation of internal parameters of multiple working modules deviates continuously, and the correlation structure between modules undergoes significant and unexpected changes, it is identified as a synergistic performance degradation caused by the combined effect of multiple modules. Through the above differentiation mechanism, the root causes of performance degradation leading to an increase in the frequency of error correction events can be identified more accurately, providing a basis for subsequent fault location and maintenance decisions.

[0084] In the error self-diagnosis method of program-controlled exchanges, when it is determined that there is a continuous deviation of the instantaneous correlation from the historical correlation range, the continuous deviation means that the instantaneous correlation is not only briefly exceeding or falling below the historical correlation range at a single moment, but remains in a deviation state or repeatedly deviates within a certain time span, thus indicating that the deviation is not caused by instantaneous noise or occasional fluctuations, but reflects an early sign of changes in the internal operating state of the system.

[0085] In practical implementation, the determination of persistent deviation can be based on a comprehensive consideration of the magnitude of the instantaneous correlation deviation and the characteristics of the deviation over time. First, when the instantaneous correlation is compared with the corresponding historical correlation range, if its value exceeds the upper or lower limit of the historical correlation range, a deviation can be identified. To avoid misjudgment caused by measurement errors or minor random fluctuations, persistent deviation requires that the deviation have a identifiable magnitude.

[0086] Secondly, the determination of persistent deviation is related to the continuity or repetition of the deviation state over time. In one implementation, persistent deviation is determined when the instantaneous correlation is outside the historical correlation range for multiple consecutive sampling periods, where the number of consecutive sampling periods can be preset according to the system's operating characteristics. In another implementation, if the number or proportion of instantaneous correlation deviations exceeds a corresponding threshold within a preset time window, it can be considered persistent deviation even if the deviations are not strictly continuous.

[0087] In addition, if the instantaneous correlation remains relatively stable within the deviation range after a deviation occurs, or shows a trend of gradually moving away from the historical correlation range without recovering to the historical correlation range in a short period of time, this can also be used as a basis for judging the continuous deviation.

[0088] For example, in a certain working module, if the historical correlation range between the power supply quality parameter and the signal transmission quality parameter under normal operating conditions is a preset range, and the system detects that the instantaneous correlation value is lower than the lower limit of the historical correlation range multiple times, and this situation occurs repeatedly within multiple consecutive sampling periods or within a preset time window, then the system can determine that the working module has a situation where the instantaneous correlation continuously deviates from the historical correlation range.

[0089] By comprehensively judging the deviation magnitude, as well as the deviation's continuity, frequency characteristics, and changing trends over time, the system can effectively distinguish between transient anomalies caused by random disturbances and persistent anomalies reflecting performance degradation trends. This improves the accuracy of judgment results during error self-diagnosis, reduces false alarms, and provides a reliable basis for subsequent performance degradation analysis and early warning.

[0090] In some preferred embodiments, a programmable switch comprising modules A, B, and C is used as an example for illustration. Data transmission and control logic are interconnected among modules A, B, and C. During a certain operational phase, the system detects a continuous increase in the frequency of error correction events. Simultaneously, the system discovers through real-time calculations that the correlation of internal parameters of module A (e.g., the correlation between processor load and data cache hit rate) continuously deviates from its historical normal range, and a similar deviation is observed in the correlation of internal parameters of module B (e.g., the correlation between signal processing unit input bandwidth and output delay).

[0091] Furthermore, the system calculates the correlation of data transmission rates between module A and module B, finding that this correlation also deviates significantly from the historical baseline. At this point, the system generates a comprehensive evaluation result based on the degree of deviation of the instantaneous correlation of each module, the increase in the frequency of error correction events, and the changes in the correlation structure between modules. When the comprehensive evaluation result indicates a large deviation in the internal correlation between module A and module B, and a significant change in their interrelationship (e.g., from strong correlation to weak correlation, or from weak correlation to strong correlation), the system classifies this performance degradation as multi-module collaborative performance degradation. For example, this situation might be caused by abnormalities in shared resources (such as power systems or clock signals), or by a chain reaction in downstream modules triggered by an upstream module failure. Conversely, when only the internal parameter correlation of module A deviates significantly, while the correlation between module B and between module A and module B remains relatively stable, the system classifies it as independent performance degradation of module A. Through this judgment mechanism, the diagnostic system can more accurately distinguish the nature of performance degradation, thereby guiding maintenance personnel to specifically check shared resources or specific working modules, improving the efficiency and accuracy of fault diagnosis.

[0092] Optionally, when at least two working modules simultaneously exhibit transient correlations that continuously deviate from their historical correlation range, and the frequency of error correction events increases, a comprehensive evaluation is conducted based on the degree of deviation of the transient correlation of each working module, the magnitude of the increase in the frequency of error correction events, and the changes in the correlation between modules. The steps to obtain comprehensive evaluation information include: Obtain the instantaneous load level and ambient temperature of the current operating status of the PBX; Based on the instantaneous load level and ambient temperature, select a weight configuration that matches the current operating status from the preset weight configuration set; The deviation of instantaneous correlation and the increase in error correction event frequency are multiplied by the corresponding weight values ​​in the selected weight configuration to obtain the weighted deviation of instantaneous correlation and the increase in error correction event frequency. For working modules whose instantaneous correlation continuously deviates from the historical correlation range and whose error correction event frequency increases, the weighted instantaneous correlation deviation degree and error correction event frequency increase of each module, as well as the change in correlation between modules, are summed to obtain their respective comprehensive performance degradation index. All of the comprehensive performance degradation indices are used as comprehensive evaluation information. Based on all the comprehensive performance degradation indices in the comprehensive evaluation information, the performance degradation is distinguished as either the performance degradation of a single module or the synergistic performance degradation of multiple modules, and the corresponding performance degradation trend is identified.

[0093] Specifically, during the comprehensive evaluation, the system first obtains the instantaneous load level and ambient temperature corresponding to the current operating status of the PBX. The instantaneous load level characterizes the workload of the PBX at the current moment, and may include indicators such as call processing volume, data traffic, or processor utilization. The ambient temperature is obtained in real time through temperature sensors deployed inside the PBX or in the cabinet, reflecting the impact of external environmental conditions on the equipment's operating status. These parameters together constitute the basic information reflecting the current operating condition of the PBX.

[0094] The preset weight configuration set is a set of weight combinations corresponding to different operating conditions. Each weight combination describes the relative importance of various evaluation indicators in performance degradation determination under specific load levels and ambient temperature conditions. The weight configuration set can be generated based on historical operating data statistical analysis, expert experience modeling, or machine learning training results, and is updated during system deployment or operation to ensure that evaluation indicators such as the degree of instantaneous correlation deviation, the rate of increase in error correction event frequency, and changes in correlation between modules can be assigned reasonable weights under different operating conditions.

[0095] In practical applications, the system selects the weight combination that best matches the current operating state from a preset weight configuration set based on the real-time instantaneous load level and ambient temperature. Matching can be done using exact matching or adjacent interval matching. For example, when the current operating state corresponds to high load and high ambient temperature, the system selects a weight configuration optimized for high load and high temperature operating scenarios.

[0096] Subsequently, the system multiplies the deviation of the instantaneous correlation and the increase in the frequency of error correction events by the corresponding weight values ​​in the selected weight configuration, respectively, to obtain the weighted deviation of the instantaneous correlation and the weighted increase in the frequency of error correction events. This weighting process allows indicators that have a more significant impact on performance degradation under the current operating conditions to have a higher weight in the comprehensive evaluation, thereby enhancing the sensitivity of the evaluation results to the actual operating conditions.

[0097] Furthermore, the system combines the weighted instantaneous correlation deviation, the weighted error correction event frequency increase, and the change in correlation between modules to generate a comprehensive performance degradation index, for example, through summation. The comprehensive performance degradation index is a comprehensive indicator used to quantify the overall performance degradation status of the working modules.

[0098] Finally, the system determines the type and trend of performance degradation based on the comprehensive performance degradation index. Specifically, a performance degradation threshold can be preset. When the comprehensive performance degradation index exceeds this threshold, performance degradation is determined to exist. Furthermore, by combining the magnitude of the comprehensive performance degradation index and the relative contribution of each weighted item, the system distinguishes whether the performance degradation is an independent performance degradation of a single working module or a collaborative performance degradation caused by the mutual influence of multiple working modules, and determines whether the degradation trend is accelerated or slow.

[0099] In some preferred embodiments, the example is a program-controlled exchange operating at high load and with an ambient temperature of 35°C during a certain operating period. The system first obtains the current instantaneous load level and ambient temperature, and then selects the weight configuration that best matches this operating state from the weight configuration set. For example, in this weight configuration, the weight of the instantaneous correlation deviation is 0.6, the weight of the increase in the frequency of error correction events is 0.4, and the weight of the change in correlation between modules is 1.

[0100] Assume that the instantaneous correlation deviation of a certain working module is 0.8, the increase in error correction event frequency is 0.5, and the change in correlation between modules is 0.2. Based on this, the system calculates the weighted instantaneous correlation deviation as 0.8 × 0.6 = 0.48 and the weighted increase in error correction event frequency as 0.5 × 0.4 = 0.20. Combining the above weighted results with the change in correlation between modules, the system obtains a comprehensive performance degradation index of 0.88.

[0101] When the preset performance degradation threshold is 0.7, the system determines that the working module has performance degradation because the comprehensive performance degradation index of 0.88 exceeds this threshold. Furthermore, by analyzing the contribution ratio of each weighted item to the comprehensive performance degradation index, and combining this with the specific pattern of changes in the correlation between modules, the system can determine that the performance degradation is that of a single working module, and because the comprehensive performance degradation index is high, its degradation trend is relatively rapid.

[0102] Through the above methods, the system can combine the real-time operating conditions of the program-controlled exchange to conduct a dynamic and targeted comprehensive assessment of performance degradation, thereby improving the accuracy and reliability of the diagnostic results.

[0103] Optionally, the step of distinguishing between performance degradation of a single module or synergistic performance degradation of multiple modules based on all the comprehensive performance degradation indices in the comprehensive evaluation information, and identifying the corresponding performance degradation trend, includes: The comprehensive performance degradation index in the comprehensive evaluation information is compared with a preset threshold to obtain the evaluation result of the performance degradation type; Based on the evaluation results, the performance degradation can be distinguished as either the performance degradation of a single module or the synergistic performance degradation of multiple modules. When the performance degradation of multiple modules is identified, the patterns and time series of changes in the degree of correlation between the relevant modules are analyzed to identify the core module or correlation path that leads to the degradation of the correlation. Based on the identified core modules or associated paths, identify the performance degradation trend of the working modules corresponding to the error correction events.

[0104] Specifically, the preset threshold can be a pre-set performance degradation threshold. When the comprehensive performance degradation index exceeds this threshold, it is determined that there is performance degradation. Furthermore, by combining the magnitude of the comprehensive performance degradation index and the relative contribution of each weighting item, it is determined whether the performance degradation is an independent performance degradation of a single working module or a collaborative performance degradation formed by the mutual influence of multiple working modules, and at the same time, it is determined whether the degradation trend is accelerated degradation or slow degradation.

[0105] For example, if the evaluation results show that the correlation between parameters within a module deviates significantly, while the correlation with other modules does not change significantly, it may indicate a performance degradation of a single module. Conversely, if there are significant and mutually influential deviations in the correlation between parameters and between multiple modules, it is more likely to indicate a synergistic performance degradation.

[0106] Furthermore, when identifying coordinated performance degradation across multiple modules, a deeper analysis of the patterns and time series of changes in the degree of correlation between related modules is needed to more accurately pinpoint the problem. This can include analyzing which modules' correlations strengthen or weaken, whether these changes are instantaneous or continuous, and the order in which they occur. Through these analyses, the core module playing a crucial role in the coordinated degradation, or the correlation path leading to degradation propagation, can be identified. For example, if the performance parameter changes of module A always precede the correlation changes of modules B and C, then module A is likely the core module. Therefore, based on the identified core module or correlation path, the performance degradation trend of the working module corresponding to the error correction event can be more accurately identified. For example, if the power supply module is identified as the core module, the root cause of the error correction event may be related to power supply instability; if a data bus is identified as the correlation path, it may be related to data transmission integrity or timing issues.

[0107] In some preferred embodiments, a specific example is given below. Suppose a program-controlled exchange system includes a power supply module, a CPU module, a memory module, and a data exchange module. Over a period of time, the frequency of error correction events reported by the system increases.

[0108] First, the system calculates the instantaneous correlation between parameters within each module based on its operating parameters, such as the output voltage stability of the power supply module, the temperature and load of the CPU module, the read / write latency of the memory module, and the packet loss rate of the data exchange module. For example, it calculates the correlation between the output voltage and ripple coefficient of the power supply module, and the correlation between the temperature and load of the CPU module. Simultaneously, it also calculates the instantaneous correlation between corresponding parameters across different modules, such as the correlation between the output voltage of the power supply module and the load of the CPU module, or the correlation between the read / write latency of the memory module and the packet loss rate of the data exchange module.

[0109] Next, these calculated instantaneous correlation levels are compared with a preset historical correlation range. It is assumed that under normal operating conditions, there is a strong positive correlation between the power module's output voltage and the CPU module's load, and this correlation fluctuates within a specific range. If this positive correlation is detected to be continuously weakening and deviating from the historical correlation range, this is considered an anomaly.

[0110] When the system detects an increase in the frequency of error correction events and finds that, for example, the correlation between internal parameters of the power module continues to deviate from the historical range, and the correlation between the CPU module and the data exchange module also continues to deviate from the historical range, the system will make a comprehensive evaluation based on the magnitude of these deviations, the increase in the frequency of error correction events, and the changes in the correlation between modules.

[0111] Based on the evaluation results, the system can distinguish between performance degradation of a single module and synergistic performance degradation of multiple modules. For example, if the evaluation results show that the correlation between the internal parameters of the power supply module deviates the most, and this deviation precedes the changes in the correlation between other modules, it may be initially judged as a single performance degradation of the power supply module. However, if the evaluation results show that the correlation between the power supply module, CPU module, and data exchange module all show significant and mutually influential deviations, it will be identified as synergistic performance degradation of multiple modules.

[0112] When a cooperative performance degradation is identified, the system further analyzes the patterns and time series of changes in the correlation between related modules. For example, the system might find that a decrease in the output voltage stability of the power supply module initially alters the correlation of its internal parameters, subsequently affecting the power supply to the CPU module, which in turn changes the correlation between the CPU module's load and temperature, ultimately impacting the performance of the data exchange module, manifested as an increase in packet loss rate. Through this time series analysis, the system can identify the power supply module as the core module causing the cooperative degradation, and the power supply-CPU-data exchange path as a critical correlation.

[0113] Ultimately, based on the identified core modules or related paths, the system can more accurately identify the performance degradation trends of the working modules corresponding to error correction events and issue warnings containing information on potential root causes such as power module issues and unstable power supply. This allows maintenance personnel to directly inspect and repair the power module, rather than blindly troubleshooting across multiple modules.

[0114] In the self-diagnosis method for errors in program-controlled exchanges, although the evaluation result or comprehensive performance degradation index is ultimately expressed as a single numerical value, this value is not an isolated indicator. Instead, it is a comprehensive quantitative result formed by weighting multiple factors reflecting abnormal operating conditions. This comprehensive quantitative result can reflect the nature of performance degradation, and its determination is based on the relative contribution relationship and numerical distribution characteristics of the weighted components that constitute the value.

[0115] The comprehensive evaluation information is obtained based on at least the following factors: the degree of deviation in the instantaneous correlation of each working module, the rate of increase in the frequency of error correction events, and the changes in the correlation between different working modules. When calculating the comprehensive performance degradation index, the system selects corresponding weight values ​​from a preset weight configuration set based on the current instantaneous load level and ambient temperature of the PBX, performs weighted processing on the above factors, and aggregates the weighted components to obtain the comprehensive performance degradation index.

[0116] Since the comprehensive performance degradation index is composed of multiple weighted components, it not only characterizes the overall severity of performance degradation but also reflects the specific nature of the degradation by analyzing the contribution of each weighted component. When the main contribution of the comprehensive performance degradation index comes from the instantaneous correlation deviation component corresponding to a specific working module, while the deviation components of other working modules and the correlation change components between modules account for a low proportion, the system can determine that the performance degradation is mainly manifested as the performance degradation of a single working module. This situation indicates that the internal operating state of this working module is abnormal, but it has not yet had a significant impact on other modules or the collaborative relationship between modules.

[0117] Conversely, when the main contribution of the overall performance degradation index comes from the instantaneous correlation deviation component of multiple working modules, or when the correlation change component between modules accounts for a high proportion of the overall performance degradation index, the system can determine that the performance degradation is a coordinated performance degradation of multiple working modules. In this case, the performance degradation is not only reflected in the abnormal operating state within a single module, but also in the change in the interaction relationship between multiple modules, indicating that a coordinated anomaly has occurred in the system-level operating state.

[0118] Among them, the component of inter-module correlation change is of great significance in distinguishing collaborative performance degradation. Since the essence of collaborative performance degradation lies in the abnormal change of the operating relationship or cooperation mode between multiple working modules, when this component contributes significantly to the comprehensive performance degradation index, it can serve as an important basis for identifying collaborative performance degradation.

[0119] In this way, even if the evaluation results or the comprehensive performance degradation index are presented in the form of a single numerical value, the system can still distinguish the type of performance degradation by combining the composition of the value, and provide a basis for locating the working modules or inter-module relationships involved in the performance degradation, thereby improving the accuracy and interpretability of the error self-diagnosis results.

[0120] This application also discloses a program-controlled exchange error self-diagnosis system, used to perform error self-diagnosis of program-controlled exchanges, combined with... Figure 3 As shown, the self-diagnostic system 1 for program-controlled exchanges includes: The trigger signal identification module 11 is used to identify the error correction event corresponding to the data transmission error as a diagnostic trigger signal when the data transmission error is corrected. The diagnostic task initiation module 12 is used to initiate a diagnostic task for the working module corresponding to the error correction event in response to a diagnostic trigger signal. The runtime parameter acquisition module 13 is used to acquire instantaneous runtime parameters related to error correction events without smoothing based on the diagnostic task that is initiated. The time correlation processing module 14 is used to correlate instantaneous running parameters with error correction events in time and store them as context information of the error correction events. The performance degradation identification module 15 is used to continuously analyze the frequency of error correction events and, in combination with the instantaneous operating parameters in the context information, identify the performance degradation trend of the working module corresponding to the error correction event. The warning information sending module 16 is used to issue a warning containing information about the working module and potential root causes of the error correction event when a performance degradation trend is detected.

[0121] To better understand the self-diagnostic system for errors in a program-controlled exchange proposed in this application, the following will provide a detailed description of each module involved.

[0122] First, the trigger signal identification module identifies the error correction event corresponding to the data transmission error as a diagnostic trigger signal when the data transmission error is corrected. This module filters and judges the recorded error correction event information to determine whether to trigger the diagnostic process. The specific logic and implementation of trigger signal identification have been described in the preceding embodiments and will not be repeated here.

[0123] It should be noted that the trigger signal identification module can be implemented in several ways. One implementation is as a standalone software service deployed within the operating system of the PBX (Publicly Controlled Switchboard). It receives error correction event information in real time by subscribing to system logs or message queues. This module may include an event filtering unit and a rule determination unit, used to analyze error correction events according to preset trigger rules. These rules can include at least the error type, frequency of occurrence, and source module, thereby identifying and diagnosing the trigger signal. Another implementation is integrated into the hardware monitoring unit, directly capturing error correction events by listening to hardware interrupt signals or register state changes and converting them into diagnostic trigger signals. Furthermore, this module can provide a configuration interface, allowing maintenance personnel to dynamically adjust trigger rules and thresholds to adapt to different operating environments and diagnostic needs.

[0124] The diagnostic task initiation module is used to initiate diagnostic tasks for the corresponding working modules in response to diagnostic trigger signals. The specific process and implementation of diagnostic task initiation have been described in the foregoing embodiments and will not be repeated here.

[0125] In a practical implementation, the diagnostic task initiation module can be implemented as a task scheduler. This scheduler receives diagnostic trigger signals from the trigger signal identification module and, based on the error correction event source information contained in the signal, queries a predefined diagnostic task mapping relationship to determine the specific diagnostic program or script to be initiated. Subsequently, the scheduler can send a start command to the diagnostic agent or hardware control unit within the PBX to activate the corresponding diagnostic task. Alternatively, the diagnostic task initiation module can also adopt an event-driven microservice architecture. Upon receiving a diagnostic trigger signal, it publishes a start message to the diagnostic interface of the target working module via a message bus, thereby triggering the execution of the diagnostic program within the target module. This module can also have task queue management capabilities to support the scheduling and orderly execution of concurrent diagnostic requests.

[0126] The runtime parameter acquisition module is used to collect instantaneous runtime parameters related to error correction events, without smoothing, after the diagnostic task is started. The basic logic and method of runtime parameter acquisition have been described in the aforementioned implementation and will not be repeated here.

[0127] In one implementation, the operating parameter acquisition module can be implemented as a high-speed data acquisition unit, directly connected to sensors, hardware registers, or dedicated measurement interfaces within the programmable exchange. After the diagnostic task is initiated, it acquires data at a programmable high sampling rate, reaching millisecond or microsecond levels, ensuring that the acquired instantaneous operating parameters fully reflect short-term fluctuation characteristics without any smoothing or averaging processing. In another implementation, the operating parameter acquisition module can be implemented as a software agent deployed within each working module. Upon receiving the diagnostic task start command, it acquires instantaneous operating parameters in real time by accessing the module's internal performance counters, status registers, or diagnostic ports. This module supports multiple data formats and transmission protocols to adapt to the acquisition needs of different types of operating parameters.

[0128] The time correlation processing module is used to correlate the collected instantaneous operating parameters with the corresponding error correction events, and store the correlation results as context information of the error correction events. The basic logic and method of time correlation processing have been described in the previous implementation, and will not be repeated here.

[0129] In a practical implementation, the time correlation processing module can be implemented as a data processing unit. This unit receives instantaneous operating parameter data from the operating parameter acquisition module and error correction event information from the trigger signal identification module, and assigns high-precision timestamps to the data and events. This module can match and bind instantaneous operating parameters and error correction events based on a preset time window (e.g., a few milliseconds before and after the error correction event occurs). Alternatively, the time correlation processing module can also be implemented as a stream processing engine. This engine uses a sliding time window or an event timestamp alignment mechanism to correlate data streams and event streams in real time, and structures the correlation results into a unified context information record before sending it to the storage system.

[0130] The performance degradation identification module continuously analyzes the frequency of error correction events and, in conjunction with instantaneous operating parameters in the context information, identifies the performance degradation trend of the corresponding working module. The overall logic of performance degradation identification has been described in the aforementioned implementation and will not be repeated here.

[0131] In one implementation, the performance degradation identification module can be implemented as a data analysis engine. This engine continuously receives error correction event frequency data and stored context information, and processes the data based on a pre-defined analysis algorithm. The analysis algorithm may include statistical trend analysis algorithms, anomaly detection algorithms, or machine learning models, used to comprehensively evaluate the correlation between changes in error correction event frequency and fluctuations in instantaneous operating parameters, thereby determining whether the working module exhibits a performance degradation trend. Alternatively, the performance degradation identification module can be implemented as a rule-based reasoning system. This system performs logical reasoning on error frequency and parameter fluctuations based on predefined expert rules to identify potential performance degradation states. This module can also support dynamic adjustments to the analysis model or rules to improve the accuracy and adaptability of the identification.

[0132] The early warning information sending module is used to generate and send an early warning containing information about the working module corresponding to the error correction event and potential root cause information when a performance degradation trend is detected. The basic logic of early warning information sending has been described in the aforementioned implementation and will not be repeated here.

[0133] In practical implementation, the early warning information sending module can be implemented as a message notification service. It generates structured early warning messages based on the early warning instructions output by the performance degradation identification module and sends these messages through various notification channels. These channels may include sending alarms to the network management system via SNMP protocol, sending notifications via email or SMS, or pushing early warning information to the operation and maintenance platform or work order system via an interface. Alternatively, the early warning information sending module can also be implemented as a visualization unit, presenting early warning information to operation and maintenance personnel in the form of a graphical interface or dashboard, and providing further diagnostic suggestions or operational guidance. This module can also be configured with alarm suppression and escalation strategies to avoid duplicate alarms and ensure that critical early warnings are handled promptly.

[0134] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for self-diagnosing errors in a program-controlled exchange, characterized in that, Includes the following steps: When a data transmission error is corrected, the error correction event corresponding to the data transmission error is identified as a diagnostic trigger signal; In response to the diagnostic trigger signal, a diagnostic task is initiated for the working module corresponding to the error correction event; Based on the diagnostic tasks initiated, collect unsmoothed transient runtime parameters related to error correction events; The instantaneous operating parameters are associated with the error correction event over time and stored as context information of the error correction event. By continuously analyzing the frequency of occurrence of the error correction events and combining the instantaneous operating parameters in the context information, the performance degradation trend of the working module corresponding to the error correction event can be identified. When the performance degradation trend is detected, an alert is issued containing information about the working module corresponding to the bug correction event and the potential root cause.

2. The method for self-diagnosing errors in a program-controlled exchange according to claim 1, characterized in that, The steps of the startup-based diagnostic task, which involve collecting unsmoothed transient operating parameters related to error correction events, include: Based on the initiated diagnostic task, instantaneous operating parameters related to error correction events, without smoothing, are collected at a higher frequency than regular polling. These instantaneous operating parameters include power supply quality parameters, signal transmission quality parameters, and environmental parameters.

3. The method for self-diagnosing errors in a program-controlled exchange according to claim 1, characterized in that, The step of continuously analyzing the frequency of occurrence of the error correction events and, in conjunction with the instantaneous operating parameters in the context information, identifying the performance degradation trend of the working module corresponding to the error correction event includes: For the power supply quality parameters and signal transmission quality parameters of the working module corresponding to the error correction event, an operating status reference range is established; the operating status reference range is initialized based on the historical operating data of the program-controlled exchange under different load levels and ambient temperature conditions; The operating status reference range is adjusted in real time based on the current real-time load and ambient temperature of the program-controlled exchange; The current instantaneous operating parameters are compared with the operating state reference interval to determine whether the instantaneous operating parameters exceed the upper or lower limit of the operating state reference interval, and the interval comparison result is obtained. When the interval comparison result indicates an excess, the instantaneous operating parameter is marked as an abnormal fluctuation; When the frequency of the error correction event increases and abnormal fluctuations occur, identify the performance degradation trend of the working module corresponding to the error correction event.

4. The method for self-diagnosing errors in a program-controlled exchange according to claim 1, characterized in that, The step of continuously analyzing the frequency of occurrence of the error correction events and, in conjunction with the instantaneous operating parameters in the context information, identifying the performance degradation trend of the working module corresponding to the error correction event includes: Obtain the historical fluctuation range of several operating parameters of the working module corresponding to the error correction event under different load levels and ambient temperature conditions; the several operating parameters include the occurrence frequency of the error correction event, power supply quality parameters, signal transmission quality parameters, and environmental parameters; The current load level and ambient temperature of the PBX are obtained in real time, and the current normal fluctuation range of the operating parameters is determined and adjusted based on the current load level and ambient temperature, combined with the historical fluctuation range. The frequency of continuously monitored error correction events is compared with the corresponding adjusted normal fluctuation range to determine whether the frequency of the error correction events exceeds the adjusted normal fluctuation range, and the result of the frequency exceeding the normal fluctuation range is obtained. The power supply quality parameters, signal transmission quality parameters, and environmental parameters are continuously monitored, and each parameter is compared with its adjusted normal fluctuation range to obtain the result of the parameter exceeding the range. When the occurrence frequency exceeds the judgment result, and the parameter exceeds the judgment result, indicating that any parameter exceeds the corresponding normal fluctuation range, the performance degradation trend of the working module corresponding to the error correction event is identified.

5. The method for self-diagnosing errors in a program-controlled exchange according to claim 4, characterized in that, When the occurrence frequency exceeds the judgment result, and the parameter exceeds the judgment result, indicating that any parameter exceeds the corresponding normal fluctuation range, the steps for identifying the performance degradation trend of the working module corresponding to the error correction event include: Obtain the historical correlation range of several operating parameters of the working module corresponding to the error correction event under different load levels and ambient temperature conditions; Calculate the instantaneous correlation between several current operating parameters; Determine whether the instantaneous correlation continuously deviates from the historical correlation range; When the instantaneous correlation continues to deviate from the historical correlation range, and the frequency of error correction events increases, the performance degradation trend of the working module corresponding to the error correction event is identified.

6. The method for self-diagnosing errors in a program-controlled exchange according to claim 5, characterized in that, When the instantaneous correlation continuously deviates from the historical correlation range, and the frequency of error correction events increases, the steps for identifying the performance degradation trend of the working module corresponding to the error correction event include: Obtain the historical correlation range of the operating parameters of each working module inside the program-controlled exchange under different load levels and ambient temperature conditions; Calculate the instantaneous correlation between the operating parameters of each working module; Determine whether the instantaneous correlation between the operating parameters of each working module continuously deviates from the historical correlation range; When it is determined that there is a transient correlation that continuously deviates from the historical correlation range, and the frequency of the error correction event increases, the performance degradation trend of the working module corresponding to the error correction event is identified.

7. The method for self-diagnosing errors in a program-controlled exchange according to claim 6, characterized in that, When it is determined that there is a transient correlation that continuously deviates from the historical correlation range, and the frequency of the error correction event is increasing, the steps for identifying the performance degradation trend of the working module corresponding to the error correction event include: Based on the collected operating parameters of each working module, calculate the instantaneous correlation between parameters within each working module, as well as the instantaneous correlation between corresponding parameters between different working modules; The instantaneous correlation between parameters within each working module and the instantaneous correlation between corresponding parameters between different working modules are compared with the preset historical correlation range to determine whether there is a continuous deviation. When at least two working modules simultaneously exhibit transient correlations that continuously deviate from the historical correlation range, and the frequency of error correction events increases, a comprehensive evaluation is conducted based on the degree of deviation of the transient correlation of each working module, the magnitude of the increase in the frequency of error correction events, and the changes in the correlation between modules, to obtain comprehensive evaluation information. Based on the comprehensive evaluation information, the performance degradation can be distinguished as either the performance degradation of a single module or the synergistic performance degradation of multiple modules, and the corresponding performance degradation trend can be identified.

8. The method for self-diagnosing errors in a program-controlled exchange according to claim 7, characterized in that, When at least two working modules simultaneously exhibit transient correlations that consistently deviate from their historical correlation range, and the frequency of error correction events increases, a comprehensive evaluation is conducted based on the degree of deviation in transient correlation for each working module, the magnitude of the increase in the frequency of error correction events, and the changes in correlation between modules. The steps to obtain comprehensive evaluation information include: Obtain the instantaneous load level and ambient temperature of the current operating status of the PBX; Based on the instantaneous load level and ambient temperature, select a weight configuration that matches the current operating state from a preset weight configuration set; The deviation of the instantaneous correlation and the increase in the frequency of error correction events are multiplied by the corresponding weight values ​​in the selected weight configuration to obtain the weighted deviation of the instantaneous correlation and the increase in the frequency of error correction events. For working modules whose instantaneous correlation continuously deviates from the historical correlation range and whose error correction event frequency increases, the weighted instantaneous correlation deviation degree and error correction event frequency increase of each module, as well as the change in correlation between modules, are summed to obtain their respective comprehensive performance degradation index. All of the comprehensive performance degradation indices are used as comprehensive evaluation information. Based on all the comprehensive performance degradation indices in the comprehensive evaluation information, the performance degradation is distinguished as either the performance degradation of a single module or the synergistic performance degradation of multiple modules, and the corresponding performance degradation trend is identified.

9. A method for self-diagnosing errors in a program-controlled exchange according to claim 8, characterized in that, The step of distinguishing between performance degradation of a single module or synergistic performance degradation of multiple modules based on all the comprehensive performance degradation indices in the comprehensive evaluation information, and identifying the corresponding performance degradation trend, includes: The comprehensive performance degradation index in the comprehensive evaluation information is compared with a preset threshold to obtain the evaluation result of the performance degradation type; Based on the evaluation results, the performance degradation can be distinguished as either the performance degradation of a single module or the synergistic performance degradation of multiple modules. When the performance degradation of multiple modules is identified, the patterns and time series of changes in the degree of correlation between the relevant modules are analyzed to identify the core module or correlation path that leads to the degradation of the correlation. Based on the identified core modules or associated paths, identify the performance degradation trend of the working modules corresponding to the error correction events.

10. A self-diagnostic system for errors in a program-controlled exchange, used to perform self-diagnostic tests for errors in a program-controlled exchange, characterized in that, include: The trigger signal recognition module is used to identify the error correction event corresponding to the data transmission error as a diagnostic trigger signal when the data transmission error is corrected. A diagnostic task initiation module is used to initiate a diagnostic task for the working module corresponding to the error correction event in response to the diagnostic trigger signal. The runtime parameter acquisition module is used to acquire unsmoothed instantaneous runtime parameters related to error correction events based on the diagnostic tasks that are initiated. The time association processing module is used to associate the instantaneous running parameters with the error correction event in time, and store them as context information of the error correction event; The performance degradation identification module is used to continuously analyze the occurrence frequency of the error correction event and, in combination with the instantaneous operating parameters in the context information, identify the performance degradation trend of the working module corresponding to the error correction event. The early warning information sending module is used to issue an early warning containing information about the working module corresponding to the error correction event and the potential root cause when the performance degradation trend is detected.

Images