A link attenuation closed-loop adaptive compensation method of an open optical layer system

Abstract

The application relates to the field of optical transmission equipment and discloses a link attenuation closed-loop adaptive compensation method of an open optical layer system, which comprises the following steps: performing multi-dimensional state sensing and data preprocessing on the open optical layer equipment, constructing a three-layer attenuation model to realize intelligent identification of deterioration sources, distributing compensation amounts through a multi-device collaborative algorithm after priority sorting and resource constraint checking, and then performing smooth execution in stages and completing closed-loop verification. The application can accurately distinguish attenuation types, dynamically adaptively compensate, and suppress power mutation, solves the problems of compensation lag, low precision and easy imbalance of traditional schemes, and significantly improves system transmission stability and operation and maintenance efficiency.

Description

Technical Field

[0001] This invention relates to the field of optical communication equipment, and in particular to a closed-loop adaptive compensation method for link attenuation in an open optical layer system. Background Technology

[0002] Open optical transmission systems, as the mainstream solution for data center interconnection, offer advantages such as miniaturized equipment, high board integration, and standardized open interfaces. Current technologies often employ fixed-gain amplification or manual adjustment of a single VOA for link attenuation compensation, relying solely on open-loop control of single-point optical power. This lacks joint monitoring and dynamic identification of multiple attenuation sources, including fiber aging, environmental temperature and humidity fluctuations, and loose connectors. Traditional solutions struggle to distinguish between fixed attenuation, slowly varying attenuation, and rapidly varying attenuation, easily leading to overcompensation, compensation lag, or power imbalance between channels, impacting system OSNR and service stability. Furthermore, compensation processes are often one-time adjustments, lacking closed-loop verification and long-term optimization capabilities, making them unsuitable for adapting to the dynamic characteristics of open optical layer links.

[0003] To address the shortcomings of existing technologies, this patent proposes a closed-loop adaptive compensation method for link attenuation in open optical layer systems. Through multi-dimensional joint sensing of link status, hierarchical attenuation modeling and identification of degradation root causes, multi-device collaborative compensation allocation, phased smooth execution, and closed-loop optimization, it can accurately distinguish different types of link attenuation and dynamically allocate compensation amounts. It prioritizes passive adjustment to reduce the impact of noise, while simultaneously achieving real-time verification of compensation effects and self-optimization of algorithm parameters. Summary of the Invention

[0004] This invention overcomes the shortcomings of the prior art and provides a closed-loop adaptive compensation method for link attenuation in an open optical layer system.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: The first aspect of this invention provides a link attenuation closed-loop adaptive compensation method for an open optical layer system, comprising the following steps: Multi-dimensional link status perception is performed on open optical layer system equipment, and core equipment data is collected and preprocessed simultaneously. Within the central data control unit, a link attenuation hierarchical model is constructed based on the pre-processed core data of the target sub-devices, and the root causes of degradation are intelligently identified through the link attenuation hierarchical model to obtain the link attenuation identification result data packet. Based on the data packets of the link attenuation identification results, dynamic sorting of compensation priorities and verification of resource constraints are performed. Based on the final compensation priority sequence, as well as the maximum allowable adjustment amount and adjustment step size of each compensation item, the multi-device collaborative attenuation compensation amount allocation processing is carried out in the target device. Based on the compensation instructions of the output port corresponding to the target sub-device, perform phased smooth compensation and closed-loop verification of the compensation effect.

[0006] Furthermore, in a preferred embodiment of the present invention, the step of performing multi-dimensional link status perception on the open optical layer system equipment, and simultaneously collecting core equipment data and performing data preprocessing, specifically includes: Identify open optical layer system equipment that requires link attenuation closed-loop adaptive compensation and designate it as the target equipment; On the target device, the sub-devices and ports to be detected are determined. The sub-devices to be detected have their own detection modules, and the built-in detection modules directly read the core data of the sub-devices to be detected and mark the sub-devices to be detected as the target sub-devices. Run the target device and control all target sub-devices on the target device to collect corresponding core data within a fixed time period to achieve target sub-device data collection time alignment; The core data of the target sub-device includes optical power data, link quality data, and system status data, and the core data of the target sub-device is marked with a timestamp during the acquisition process. A central data control unit is introduced to control the data connection between the target sub-device and the central data control unit, and a sliding window anomaly removal algorithm is introduced into the central data control unit; The sliding window anomaly removal algorithm is used to perform instantaneous anomaly removal on the core data of the target sub-device. The instantaneous anomaly removal method is to compare the power values ​​of the data in the sliding window. If the power value of the core data of the target sub-device is less than the average power value of all data in the sliding window, the core data of the corresponding target sub-device is removed. The core data of the target sub-device after instantaneous anomaly removal is processed to standardize the data format and output the preprocessed core data of the target sub-device.

[0007] Furthermore, in a preferred embodiment of the present invention, the step of constructing a link attenuation hierarchical model based on preprocessed target sub-device core data within the central data control unit, and intelligently identifying the root causes of degradation through the link attenuation hierarchical model to obtain a link attenuation identification result data packet, specifically involves: Within the central data control unit, a raw dataset of link attenuation in the open optical layer system is constructed based on the preprocessed core data of the target sub-devices. By introducing historical data networks, the attenuation characteristics of the open optical layer system link are determined, and based on the attenuation characteristics of the open optical layer system link, the link attenuation is divided into a three-layer structure, including a fixed attenuation layer, a slowly varying attenuation layer, and a rapidly varying attenuation layer. Based on the three-layer structure of link attenuation, a hierarchical model of link attenuation is constructed, in which the fixed attenuation layer serves as the base layer of the model, and the slowly varying attenuation layer and the rapidly varying attenuation layer serve as the dynamic layers of the model. Based on historical data networks, the factory calibration parameters of the target sub-device are obtained. In the baseline layer, the factory calibration parameters of the target sub-device are used as the baseline value. In the dynamic layer of the model, dynamic layer model input features are introduced and extracted from the original dataset of optical layer system link attenuation using the time-series feature extraction method. By combining the input features of the dynamic layer model, intelligent identification of the root causes of degradation is performed within the central data control unit.

[0008] Furthermore, in a preferred embodiment of the present invention, the step of combining the input features of the dynamic layer model to perform intelligent identification of degradation root causes within the central data control unit to obtain the link attenuation identification result data packet specifically includes: A sliding window difference algorithm is introduced into the central data control unit to perform temporal difference operations on the input features of the dynamic layer model, and output the temporal features of the slow decay layer and the fast decay layer. The temporal features of the slow decay layer and the fast decay layer include the corresponding decay rate, fluctuation amplitude and duration. In the preprocessing of the target sub-device core data, link quality data and system status data are obtained, and fuzzy logic discrimination rules are introduced to match the factory calibration parameters, slow attenuation layer timing characteristics and fast attenuation layer timing characteristics of the target sub-device with the link quality data and system status data respectively. Output feature matching results, and determine the root cause of degradation corresponding to attenuation at each layer based on the feature matching results, including fast attenuation compensation, channel-level slow attenuation compensation, line-level slow attenuation compensation, and fixed attenuation calibration. Based on the root causes of degradation corresponding to each layer of attenuation, and combined with the factory calibration parameters of the target sub-device, the timing characteristics of the slow attenuation layer and the timing characteristics of the fast attenuation layer, the data is packaged and encapsulated into a link attenuation identification result data packet, which is then transmitted to the central data control unit for storage.

[0009] Furthermore, in a preferred embodiment of the present invention, the step of dynamically sorting compensation priorities and verifying resource constraints based on the data packets of the link attenuation identification results specifically includes: Within the central data control unit, a compensation priority determination principle is preset, wherein the priority in the compensation priority determination principle, from high to low, is as follows: fast attenuation compensation, channel-level slow attenuation compensation, line-level slow attenuation compensation, and fixed attenuation calibration. Based on the data packets of the link attenuation identification results, the compensation items corresponding to different degradation root causes are output; Based on the aforementioned compensation priority determination principle, an initial compensation priority sequence is generated, and the hardware parameters of the target sub-device and the compensation items corresponding to different deterioration root causes are retrieved and matched with the initial compensation priority sequence for resource constraint verification. Specifically, during resource constraint matching verification, compensation items that fail the resource constraint matching verification are removed, while compensation items that pass the resource constraint matching verification are retained, and then reordered to generate the final compensation priority sequence. Simultaneously, the maximum allowable adjustment amount and adjustment step size of each compensation item are output and transmitted to the central data control unit for storage.

[0010] Furthermore, in a preferred embodiment of the present invention, the multi-device collaborative attenuation compensation allocation processing in the target device based on the final compensation priority sequence, the maximum allowable adjustment amount of each compensation item, and the adjustment step size specifically includes: In the central data control unit, a nonlinear collaborative allocation algorithm is introduced, and in the target sub-device, the hardware parameters of the corresponding output port of the target sub-device are obtained and combined to perform compensation amount matching calculation. According to the final compensation priority sequence, compensation amounts are allocated to the output ports corresponding to the target sub-devices, and the compensation amounts are controlled to remain within the maximum allowable adjustment amount and adjustment step size of each compensation item, thereby generating compensation instructions for the output ports corresponding to the target sub-devices.

[0011] Furthermore, in a preferred embodiment of the present invention, the step of performing phased smooth compensation and closed-loop verification of compensation effect based on the compensation instruction of the output port corresponding to the target sub-device specifically includes: Within the central data control unit, compensation instructions for the output ports corresponding to the target sub-devices are sent to each target sub-device for compensation operation according to the final compensation priority sequence. The compensation operation includes a pre-compensation phase and a main compensation phase. First, a pre-compensation phase is executed. In this phase, the pre-compensation step size is set based on the allowable adjustment step size of each compensation item. When the compensation command of the output port corresponding to the target sub-device is output in the target sub-device, it is executed according to the pre-compensation step size. After the pre-compensation phase, the main compensation phase is executed. The main compensation phase is to complete all compensation operations. After the main compensation phase, the latest optical power data, link quality data and system status data transmitted back by the target sub-device are received in real time in the central data control unit and labeled as real-time transmitted data. Set the standard feedback data and calculate the deviation value between it and the real-time feedback data. If the deviation value remains within the preset range, the compensation is considered complete. If the deviation value is not within the preset range, the compensation operation of each target sub-device will be re-executed until the deviation value is maintained within the preset range.

[0012] This invention addresses the technical deficiencies in the background technology and offers the following advantages: It performs multi-dimensional state perception and data preprocessing on open optical layer devices, constructs a three-layer attenuation model to intelligently identify the root causes of degradation, and after priority sorting and resource constraint verification, allocates compensation amounts through a multi-device collaborative algorithm, then performs phased smooth execution and completes closed-loop verification. This invention can accurately distinguish attenuation types, dynamically adaptively compensate, and suppress power surges, solving problems such as compensation lag, low accuracy, and susceptibility to imbalance in traditional solutions, significantly improving system transmission stability and operational efficiency. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained from these drawings without creative effort.

[0014] Figure 1 A flowchart of a closed-loop adaptive compensation method for link attenuation in an open optical layer system is shown. Figure 2 A flowchart illustrating the method for outputting link attenuation identification result data packets is shown. Detailed Implementation

[0015] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0016] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0017] Figure 1 A flowchart of a link attenuation closed-loop adaptive compensation method for an open optical layer system is shown, including the following steps: Multi-dimensional link status perception is performed on open optical layer system equipment, and core equipment data is collected and preprocessed simultaneously. Within the central data control unit, a link attenuation hierarchical model is constructed based on the pre-processed core data of the target sub-devices, and the root causes of degradation are intelligently identified through the link attenuation hierarchical model to obtain the link attenuation identification result data packet. Based on the data packets of the link attenuation identification results, dynamic sorting of compensation priorities and verification of resource constraints are performed. Based on the final compensation priority sequence, as well as the maximum allowable adjustment amount and adjustment step size of each compensation item, the multi-device collaborative attenuation compensation amount allocation processing is carried out in the target device. Based on the compensation instructions of the output port corresponding to the target sub-device, perform phased smooth compensation and closed-loop verification of the compensation effect.

[0018] Furthermore, in a preferred embodiment of the present invention, the step of performing multi-dimensional link status perception on the open optical layer system equipment, and simultaneously collecting core equipment data and performing data preprocessing, specifically includes: Identify open optical layer system equipment that requires link attenuation closed-loop adaptive compensation and designate it as the target equipment; On the target device, the sub-devices and ports to be detected are determined. The sub-devices to be detected have their own detection modules, and the built-in detection modules directly read the core data of the sub-devices to be detected and mark the sub-devices to be detected as the target sub-devices. Run the target device and control all target sub-devices on the target device to collect corresponding core data within a fixed time period to achieve target sub-device data collection time alignment; The core data of the target sub-device includes optical power data, link quality data, and system status data, and the core data of the target sub-device is marked with a timestamp during the acquisition process. A central data control unit is introduced to control the data connection between the target sub-device and the central data control unit, and a sliding window anomaly removal algorithm is introduced into the central data control unit; The sliding window anomaly removal algorithm is used to perform instantaneous anomaly removal on the core data of the target sub-device. The instantaneous anomaly removal method is to compare the power values ​​of the data in the sliding window. If the power value of the core data of the target sub-device is less than the average power value of all data in the sliding window, the core data of the corresponding target sub-device is removed. The core data of the target sub-device after instantaneous anomaly removal is processed to standardize the data format and output the preprocessed core data of the target sub-device.

[0019] It should be noted that, firstly, the optical layer equipment requiring attenuation compensation needs to be selected, and then the sub-devices and ports with detection capabilities must be located on it. These sub-devices include OA boards, OLA trunk boards, OLP protection boards, MUX passive multiplexing / splitting devices, and system function modules. At each of these points, the built-in monitoring module of the device is activated, requiring no additional hardware installation; data is directly read from the board's internal registers. Subsequently, all boards collect data at a unified time to ensure time alignment. Optical power data is the real-time input / output optical power for each port; link quality data includes OTDR (total fiber attenuation, event location, loss value, dead zone information) and OCM (channel power, center wavelength, signal-to-noise ratio for each wavelength). System status data includes OSC, VGA, VOA, and APR status, respectively: communication normal / abnormal, bit error rate; current gain value, operating mode; current attenuation value; triggered / not triggered, alarm type.

[0020] Subsequently, all data from sub-devices are uploaded to the central unit, and a sliding window algorithm is used to remove outliers. This introduces a lightweight, real-time algorithm into optical layer preprocessing, which is simple, efficient, and has the advantages of strong anti-interference, low computational load, and no occupation of service resources. The average power is calculated within the sliding window, and data below the average is identified as transient jitter and removed. Units, precision, and format are standardized to output clean and well-organized data.

[0021] Furthermore, in a preferred embodiment of the present invention, the step of dynamically sorting compensation priorities and verifying resource constraints based on the data packets of the link attenuation identification results specifically includes: Within the central data control unit, a compensation priority determination principle is preset, wherein the priority in the compensation priority determination principle, from high to low, is as follows: fast attenuation compensation, channel-level slow attenuation compensation, line-level slow attenuation compensation, and fixed attenuation calibration. Based on the data packets of the link attenuation identification results, the compensation items corresponding to different degradation root causes are output; Based on the aforementioned compensation priority determination principle, an initial compensation priority sequence is generated, and the hardware parameters of the target sub-device and the compensation items corresponding to different deterioration root causes are retrieved and matched with the initial compensation priority sequence for resource constraint verification. Specifically, during resource constraint matching verification, compensation items that fail the resource constraint matching verification are removed, while compensation items that pass the resource constraint matching verification are retained, and then reordered to generate the final compensation priority sequence. Simultaneously, the maximum allowable adjustment amount and adjustment step size of each compensation item are output and transmitted to the central data control unit for storage.

[0022] It should be noted that the compensation order is pre-set in the central data control unit to address the most dangerous and service-impacting abrupt attenuation first, followed by gradual attenuation, and finally fixed calibration. Based on the attenuation type identified by the hierarchical model (fast change / slow channel change / slow line change / fixed), the required compensation items are automatically matched to achieve accurate compensation, no redundant operations, and reduced system malfunction probability. Subsequently, an initial compensation order is generated according to priority, and the hardware capabilities of the target sub-devices (such as maximum adjustment range and safety thresholds) are retrieved to verify whether the compensation is within the equipment's tolerance range. Specifically, during resource constraint matching verification, compensation items that fail the resource constraint matching verification are eliminated. These failures include compensation exceeding the VOA adjustable attenuation range, compensation gain exceeding the VGA preset gain range, single-board output optical power exceeding the rated upper limit after compensation, compensation operation triggering APR protection malfunction, and compensation process occupying OSC communication bandwidth exceeding the threshold. Finally, compensation items that exceed hardware capabilities, are unsafe, or are unexecutable are removed, leaving only legal and executable items. These items are then reordered, and the maximum allowable adjustment value for each compensation item and the amount that can be adjusted at each step are given and stored in the central data control unit.

[0023] Furthermore, in a preferred embodiment of the present invention, the multi-device collaborative attenuation compensation allocation processing in the target device based on the final compensation priority sequence, the maximum allowable adjustment amount of each compensation item, and the adjustment step size specifically includes: In the central data control unit, a nonlinear collaborative allocation algorithm is introduced, and in the target sub-device, the hardware parameters of the corresponding output port of the target sub-device are obtained and combined to perform compensation amount matching calculation. According to the final compensation priority sequence, compensation amounts are allocated to the output ports corresponding to the target sub-devices, and the compensation amounts are controlled to remain within the maximum allowable adjustment amount and adjustment step size of each compensation item, thereby generating compensation instructions for the output ports corresponding to the target sub-devices.

[0024] It should be noted that a nonlinear collaborative allocation algorithm is enabled in the central data control unit, and the hardware capability parameters of the target sub-device output port (such as maximum adjustment range, safety threshold, and step size limit) are read simultaneously to match the compensation requirements with the hardware capabilities. Specifically, following the allocation principle of "prioritizing passive adjustment, followed by active amplification," the total attenuation compensation is dynamically allocated to the VOA adjustable attenuator, VGA optical amplification module, and OLP optical line protection module. For fast attenuation compensation items, priority is given to the line-side VOA adjustable attenuator, with an allocation ratio of 85%~95% of the total compensation. The remaining compensation is allocated to the VGA optical amplifier module for fine-tuning. For channel-level slow attenuation compensation items, they are allocated to the OA board channel-side VOA adjustable attenuator, with precise allocation of compensation for each wavelength to ensure that the power deviation of each channel is controlled within ±0.3dB. For line-level slow attenuation compensation items, they are allocated to the VGA optical amplifier module of the OA-L, OA-H, or OLA board, with step-by-step allocation according to the gain range, and the single-step gain adjustment not exceeding 1dB. For fixed attenuation calibration items, the associated attenuation of the MUX passive device and the connector loss are fine-tuned only during initial system deployment or periodic maintenance, with a single calibration amount not exceeding 0.5dB.

[0025] Finally, the total compensation amount is allocated to the corresponding ports in sequence to ensure that high-priority faults are handled first. During the allocation process, each adjustment step is strictly constrained, not exceeding the maximum allowable adjustment amount, nor less than or greater than the specified adjustment step size, to ensure smoothness, safety, and controllability.

[0026] Furthermore, in a preferred embodiment of the present invention, the step of performing phased smooth compensation and closed-loop verification of compensation effect based on the compensation instruction of the output port corresponding to the target sub-device specifically includes: Within the central data control unit, compensation instructions for the output ports corresponding to the target sub-devices are sent to each target sub-device for compensation operation according to the final compensation priority sequence. The compensation operation includes a pre-compensation phase and a main compensation phase. First, a pre-compensation phase is executed. In this phase, the pre-compensation step size is set based on the allowable adjustment step size of each compensation item. When the compensation command of the output port corresponding to the target sub-device is output in the target sub-device, it is executed according to the pre-compensation step size. After the pre-compensation phase, the main compensation phase is executed. The main compensation phase is to complete all compensation operations. After the main compensation phase, the latest optical power data, link quality data and system status data transmitted back by the target sub-device are received in real time in the central data control unit and labeled as real-time transmitted data. Set the standard feedback data and calculate the deviation value between it and the real-time feedback data. If the deviation value remains within the preset range, the compensation is considered complete. If the deviation value is not within the preset range, the compensation operation of each target sub-device will be re-executed until the deviation value is maintained within the preset range.

[0027] It should be noted that the central data control unit, according to the previously determined final compensation priority sequence, issues compensation commands to the output ports of the corresponding target sub-devices to begin compensation execution. This ensures that high-priority faults are compensated first, reducing business risks. Subsequently, compensation is divided into two steps: a pre-compensation phase and a main compensation phase. First, small-step pre-compensation is used, performing minor adjustments within the hardware-allowed safe step size to observe the system response. This verifies that the compensation direction is being adjusted completely, preventing over-compensation or reverse compensation. After the pre-compensation is confirmed to be normal, full compensation is executed, supplementing the remaining compensation amount step by step to achieve the target power. After compensation is completed, the latest optical power data, link quality data, and system status data of the target sub-devices are immediately collected to form real-time feedback data. This data is then compared with standard data, and interpolation is used to determine whether the compensation was successful.

[0028] Figure 2 The flowchart of the method for outputting link attenuation identification result data packets is shown, including the following steps: S202: Within the central data control unit, a link attenuation hierarchical model is constructed based on the core data of the preprocessed target sub-devices, and the root causes of degradation are intelligently identified through the link attenuation hierarchical model to obtain the link attenuation identification result data packet.

[0029] S204: Combining the input features of the dynamic layer model, intelligent identification of the root causes of degradation is performed in the central data control unit to obtain the data packet of link attenuation identification result.

[0030] Furthermore, in a preferred embodiment of the present invention, the step of constructing a link attenuation hierarchical model based on preprocessed target sub-device core data within the central data control unit, and intelligently identifying the root causes of degradation through the link attenuation hierarchical model to obtain a link attenuation identification result data packet, specifically involves: Within the central data control unit, a raw dataset of link attenuation in the open optical layer system is constructed based on the preprocessed core data of the target sub-devices. By introducing historical data networks, the attenuation characteristics of the open optical layer system link are determined, and based on the attenuation characteristics of the open optical layer system link, the link attenuation is divided into a three-layer structure, including a fixed attenuation layer, a slowly varying attenuation layer, and a rapidly varying attenuation layer. Based on the three-layer structure of link attenuation, a hierarchical model of link attenuation is constructed, in which the fixed attenuation layer serves as the base layer of the model, and the slowly varying attenuation layer and the rapidly varying attenuation layer serve as the dynamic layers of the model. Based on historical data networks, the factory calibration parameters of the target sub-device are obtained. In the baseline layer, the factory calibration parameters of the target sub-device are used as the baseline value. In the dynamic layer of the model, dynamic layer model input features are introduced and extracted from the original dataset of optical layer system link attenuation using the time-series feature extraction method. By combining the input features of the dynamic layer model, intelligent identification of the root causes of degradation is performed in the central data control unit to obtain the data packet of link attenuation identification result.

[0031] It should be noted that, firstly, a complete dataset is constructed to collect three types of effective data: optical power, link quality, and system status. Then, historical link operation data accumulated from the network is retrieved to summarize the system attenuation variation patterns. The total link loss is divided into a three-layer attenuation architecture, including a fixed attenuation layer, a slowly varying attenuation layer, and a rapidly varying attenuation layer. The aim is to achieve physical isolation between static inherent loss and dynamic random loss in the model, ensuring that they do not interfere with each other's feature extraction. This results in high model fitting accuracy and reduces misjudgments caused by interference between different types of loss.

[0032] Furthermore, in a preferred embodiment of the present invention, the step of combining the input features of the dynamic layer model to perform intelligent identification of degradation root causes within the central data control unit to obtain the link attenuation identification result data packet specifically includes: A sliding window difference algorithm is introduced into the central data control unit to perform temporal difference operations on the input features of the dynamic layer model, and output the temporal features of the slow decay layer and the fast decay layer. The temporal features of the slow decay layer and the fast decay layer include the corresponding decay rate, fluctuation amplitude and duration. In the preprocessing of the target sub-device core data, link quality data and system status data are obtained, and fuzzy logic discrimination rules are introduced to match the factory calibration parameters, slow attenuation layer timing characteristics and fast attenuation layer timing characteristics of the target sub-device with the link quality data and system status data respectively. Output feature matching results, and determine the root cause of degradation corresponding to attenuation at each layer based on the feature matching results, including fast attenuation compensation, channel-level slow attenuation compensation, line-level slow attenuation compensation, and fixed attenuation calibration. Based on the root causes of degradation corresponding to each layer of attenuation, and combined with the factory calibration parameters of the target sub-device, the timing characteristics of the slow attenuation layer and the timing characteristics of the fast attenuation layer, the data is packaged and encapsulated into a link attenuation identification result data packet, which is then transmitted to the central data control unit for storage.

[0033] It should be noted that the central unit employs a sliding window differential algorithm to process the dynamic layer input features, separating slowly changing and rapidly changing attenuation data. Three key parameters are extracted: rate of change, fluctuation amplitude, and duration. Preprocessed link quality and system status data are retrieved, and combined with factory baseline parameters and the fast / slow changing time-series characteristics, multi-dimensional feature matching and comparison are performed using preset fuzzy logic rules. The significance lies in the ability to cross-validate the causes of loss from multiple dimensions, avoiding misjudgments based on a single parameter. The output feature matching results are used to determine the root cause of degradation corresponding to each layer of attenuation: when the reference layer parameters match the link quality data and there is no dynamic time-series characteristic fluctuation, it is determined to be fixed attenuation, corresponding to the inherent loss of the optical fiber, the insertion loss of the MUX passive device, and the inherent loss of the connector; when the time-series characteristics of the slowly varying attenuation layer match the ambient temperature and humidity and the aging trend of the optical fiber, it is determined to be slowly varying attenuation, corresponding to the fluctuation of ambient temperature and humidity, long-term aging of the optical fiber, and the gradual loss of device performance; when the time-series characteristics of the rapidly varying attenuation layer show short-period, large-amplitude fluctuations and match the system vibration and connector loosening alarms, it is determined to be rapidly varying attenuation, corresponding to the micro-bending deformation of the optical fiber, external vibration and impact, and sudden loss caused by temporary loosening of the connector.

[0034] Finally, based on the feature matching results, four types of degradation compensation items are identified: fast degradation, slow channel degradation, slow line degradation, and fixed calibration. The factory parameters, fast / slow degradation characteristics, and degradation classification information are then summarized and uniformly packaged into a link attenuation identification data packet, which is stored in the central data control unit for subsequent data retrieval.

[0035] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A closed-loop adaptive compensation method for link attenuation in an open optical layer system, characterized in that, Includes the following steps: Multi-dimensional link status perception is performed on open optical layer system equipment, and core equipment data is collected and preprocessed simultaneously. Within the central data control unit, a link attenuation hierarchical model is constructed based on the pre-processed core data of the target sub-devices, and the root causes of degradation are intelligently identified through the link attenuation hierarchical model to obtain the link attenuation identification result data packet. Based on the data packets of the link attenuation identification results, dynamic sorting of compensation priorities and verification of resource constraints are performed. Based on the final compensation priority sequence, as well as the maximum allowable adjustment amount and adjustment step size of each compensation item, the multi-device collaborative attenuation compensation amount allocation processing is carried out in the target device. Based on the compensation instructions of the output port corresponding to the target sub-device, perform phased smooth compensation and closed-loop verification of the compensation effect.

2. The link attenuation closed-loop adaptive compensation method for an open optical layer system according to claim 1, characterized in that, The process of performing multi-dimensional link status awareness on open optical layer system devices, and simultaneously collecting core device data and performing data preprocessing, specifically includes: Identify open optical layer system equipment that requires link attenuation closed-loop adaptive compensation and designate it as the target equipment; On the target device, the sub-devices and ports to be detected are determined. The sub-devices to be detected have their own detection modules, and the built-in detection modules directly read the core data of the sub-devices to be detected and mark the sub-devices to be detected as the target sub-devices. Run the target device and control all target sub-devices on the target device to collect corresponding core data within a fixed time period to achieve target sub-device data collection time alignment; The core data of the target sub-device includes optical power data, link quality data, and system status data, and the core data of the target sub-device is marked with a timestamp during the acquisition process. A central data control unit is introduced to control the data connection between the target sub-device and the central data control unit, and a sliding window anomaly removal algorithm is introduced into the central data control unit; The sliding window anomaly removal algorithm is used to perform instantaneous anomaly removal on the core data of the target sub-device. The instantaneous anomaly removal method is to compare the power values ​​of the data in the sliding window. If the power value of the core data of the target sub-device is less than the average power value of all data in the sliding window, the core data of the corresponding target sub-device is removed. The core data of the target sub-device after instantaneous anomaly removal is processed to standardize the data format and output the preprocessed core data of the target sub-device.

3. The link attenuation closed-loop adaptive compensation method for an open optical layer system according to claim 1, characterized in that, Within the central data control unit, a link attenuation hierarchical model is constructed based on preprocessed target sub-device core data. This model is then used for intelligent identification of degradation root causes, yielding a link attenuation identification result data packet. Specifically: Within the central data control unit, a raw dataset of link attenuation in the open optical layer system is constructed based on the preprocessed core data of the target sub-devices. By introducing historical data networks, the attenuation characteristics of the open optical layer system link are determined, and based on the attenuation characteristics of the open optical layer system link, the link attenuation is divided into a three-layer structure, including a fixed attenuation layer, a slowly varying attenuation layer, and a rapidly varying attenuation layer. Based on the three-layer structure of link attenuation, a hierarchical model of link attenuation is constructed, in which the fixed attenuation layer serves as the base layer of the model, and the slowly varying attenuation layer and the rapidly varying attenuation layer serve as the dynamic layers of the model. Based on historical data networks, the factory calibration parameters of the target sub-device are obtained. In the baseline layer, the factory calibration parameters of the target sub-device are used as the baseline value. In the dynamic layer of the model, dynamic layer model input features are introduced and extracted from the original dataset of optical layer system link attenuation using the time-series feature extraction method. By combining the input features of the dynamic layer model, intelligent identification of the root causes of degradation is performed in the central data control unit to obtain the data packet of link attenuation identification result.

4. The link attenuation closed-loop adaptive compensation method for an open optical layer system according to claim 3, characterized in that, The process of combining the input features of the dynamic layer model and performing intelligent identification of the root causes of degradation within the central data control unit to obtain the link attenuation identification result data packet is as follows: A sliding window difference algorithm is introduced into the central data control unit to perform temporal difference operations on the input features of the dynamic layer model, and output the temporal features of the slow decay layer and the fast decay layer. The temporal features of the slow decay layer and the fast decay layer include the corresponding decay rate, fluctuation amplitude and duration. In the preprocessing of the target sub-device core data, link quality data and system status data are obtained, and fuzzy logic discrimination rules are introduced to match the factory calibration parameters, slow attenuation layer timing characteristics and fast attenuation layer timing characteristics of the target sub-device with the link quality data and system status data respectively. Output feature matching results, and determine the root cause of degradation corresponding to attenuation at each layer based on the feature matching results, including fast attenuation compensation, channel-level slow attenuation compensation, line-level slow attenuation compensation, and fixed attenuation calibration. Based on the root causes of degradation corresponding to each layer of attenuation, and combined with the factory calibration parameters of the target sub-device, the timing characteristics of the slow attenuation layer and the timing characteristics of the fast attenuation layer, the data is packaged and encapsulated into a link attenuation identification result data packet, which is then transmitted to the central data control unit for storage.

5. The link attenuation closed-loop adaptive compensation method for an open optical layer system according to claim 1, characterized in that, The process of dynamically sorting compensation priorities and verifying resource constraints based on the data packets derived from link attenuation identification results is as follows: Within the central data control unit, a compensation priority determination principle is preset, wherein the priority in the compensation priority determination principle, from high to low, is as follows: fast attenuation compensation, channel-level slow attenuation compensation, line-level slow attenuation compensation, and fixed attenuation calibration. Based on the data packets of the link attenuation identification results, the compensation items corresponding to different degradation root causes are output; Based on the aforementioned compensation priority determination principle, an initial compensation priority sequence is generated, and the hardware parameters of the target sub-device and the compensation items corresponding to different deterioration root causes are retrieved and matched with the initial compensation priority sequence for resource constraint verification. Specifically, during resource constraint matching verification, compensation items that fail the resource constraint matching verification are removed, while compensation items that pass the resource constraint matching verification are retained, and then reordered to generate the final compensation priority sequence. Simultaneously, the maximum allowable adjustment amount and adjustment step size of each compensation item are output and transmitted to the central data control unit for storage.

6. The link attenuation closed-loop adaptive compensation method for an open optical layer system according to claim 1, characterized in that, The process of allocating multi-device collaborative attenuation compensation in the target device based on the final compensation priority sequence, the maximum allowable adjustment amount of each compensation item, and the adjustment step size is as follows: In the central data control unit, a nonlinear collaborative allocation algorithm is introduced, and in the target sub-device, the hardware parameters of the corresponding output port of the target sub-device are obtained and combined to perform compensation amount matching calculation. According to the final compensation priority sequence, compensation amounts are allocated to the output ports corresponding to the target sub-devices, and the compensation amounts are controlled to remain within the maximum allowable adjustment amount and adjustment step size of each compensation item, thereby generating compensation instructions for the output ports corresponding to the target sub-devices.

7. The link attenuation closed-loop adaptive compensation method for an open optical layer system according to claim 1, characterized in that, The step of performing phased smooth compensation and closed-loop verification of the compensation effect based on the compensation command of the output port corresponding to the target sub-device is as follows: Within the central data control unit, compensation instructions for the output ports corresponding to the target sub-devices are sent to each target sub-device for compensation operation according to the final compensation priority sequence. The compensation operation includes a pre-compensation phase and a main compensation phase. First, a pre-compensation phase is executed. In this phase, the pre-compensation step size is set based on the allowable adjustment step size of each compensation item. When the compensation command of the output port corresponding to the target sub-device is output in the target sub-device, it is executed according to the pre-compensation step size. After the pre-compensation phase, the main compensation phase is executed. The main compensation phase is to complete all compensation operations. After the main compensation phase, the latest optical power data, link quality data and system status data transmitted back by the target sub-device are received in real time in the central data control unit and labeled as real-time transmitted data. Set the standard feedback data and calculate the deviation value between it and the real-time feedback data. If the deviation value remains within the preset range, the compensation is considered complete. If the deviation value is not within the preset range, the compensation operation of each target sub-device will be re-executed until the deviation value is maintained within the preset range.

Images