Scheduling adaptation method for industrial PON (Passive Optical Network) and industrial Ethernet

Abstract

The invention discloses a scheduling adaptation method for an industrial PON and an industrial Ethernet, and the method comprises the steps: constructing a protocol-optical resource mapping semantic model, abstracting the communication demands of different industrial Ethernet protocols into a unified data structure containing a time-sensitive service level, a flow feature and a communication period parameter, mapping to generate corresponding scheduling demand parameters required by the industrial PON management platform; target traffic is sensed in real time based on a port mirroring technology, protocol identification is carried out by using a protocol-optical resource mapping semantic model, the bandwidth demand and transmission time sequence characteristics of industrial Ethernet service flow are further analyzed, the analyzed bandwidth demand and transmission time sequence characteristics are mapped into scheduling demand parameters of an industrial PON, and the scheduling demand parameters of the industrial PON are mapped to the target traffic. And the industrial PON management platform generates and issues a deterministic scheduling instruction based on the received scheduling demand parameters, so that deterministic transmission service is realized. According to the invention, low-delay, low-jitter and high-reliability transmission service can be provided for the multi-protocol industrial equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial communication, in particular to a scheduling adaptation method for industrial PON and industrial Ethernet, and is especially suitable for solving the problem of inaccurate resource allocation and insufficient deterministic guarantee caused by the lack of protocol feature awareness when an industrial PON system carries multi-protocol industrial Ethernet services. BACKGROUND

[0002] With the rapid development of industrial automation and intelligent manufacturing, industrial Ethernet technology (such as PROFINET, EtherCAT, POWERLINK, TSN, etc.) has become the mainstream choice for factory control layer and field layer communication due to its high bandwidth, openness and advantages of integration with IT networks. At the same time, industrial PON (Industrial Passive Optical Network) is widely used to build a new generation of industrial access network due to its all-optical transmission, anti-electromagnetic interference, long-distance coverage and point-to-multipoint topology, etc., to carry the diversified industrial Ethernet services mentioned above.

[0003] However, the current industrial PON system faces a fundamental contradiction in actual deployment: the determinism requirement of upper-layer industrial applications for communication is increasingly stringent, while the underlying PON access mechanism is still essentially based on the "best-effort" mode of statistical multiplexing. Specifically, the existing technology has the following two key bottlenecks:

[0004] First, the lack of protocol awareness capability leads to inaccurate resource allocation. Industrial PON devices usually cannot automatically identify the type of industrial Ethernet protocol carried and its inherent communication characteristics (such as periodicity, payload size, priority level, etc.). Its DBA (Dynamic Bandwidth Allocation) mechanism mostly relies on average traffic or static configuration, which is difficult to adapt to the typical "micro-burst" characteristics of industrial control flow. When the burst traffic exceeds the pre-allocated bandwidth, data packets will be queued or even discarded in the ONU (Optical Network Unit) buffer, causing uncontrollable latency and packet loss, which seriously threatens the stability of the control system.

[0005] Secondly, the scheduling strategy and the service timing characteristics mismatch cause inherent jitter. Even if a certain industrial service flow is allocated with sufficient fixed bandwidth, if the uplink authorization window opening time issued by the OLT (Optical Line Terminal) is not synchronized with the actual arrival time of the data packet of the service flow, the data packet will be forced to wait until the next PON frame (typical frame length is 125us) can be transmitted. The cross-cycle waiting caused by the misalignment of the physical layer TDMA (Time Division Multiple Access) scheduling mechanism and the service layer timing requirement will produce a sawtooth-shaped, periodic queuing delay, which becomes the main source of industrial PON end-to-end jitter, and cannot meet the requirements of microsecond-level synchronous control scenarios.

[0006] Although some studies attempt to improve performance by enhancing the DBA algorithm or introducing a TSN (Time-Sensitive Networking) agent, these solutions are often limited to a single protocol scenario and lack the ability to adapt to multiple manufacturers and multiple protocol heterogeneous industrial devices. More importantly, existing methods fail to establish an automated mapping bridge between the industrial protocol semantics and the optical layer scheduling parameters, resulting in a high dependence on manual experience for network configuration, high operational complexity, and the inability to achieve dynamic closed-loop optimization.

[0007] Therefore, there is an urgent need for an adaptation method that can deeply integrate protocol semantic understanding, real-time traffic feature perception, and dynamic resource scheduling to provide low-latency, low-jitter, and high-reliability transmission services for multi-protocol industrial devices under the PON architecture of a shared medium. SUMMARY

[0008] The purpose of the present application is to overcome the shortcomings of the prior art and provide a scheduling adaptation method for industrial PON and industrial Ethernet, aiming to solve the problem that when the existing industrial PON system carries industrial Ethernet services, it cannot perceive the real-time and periodic characteristics of the protocol layer, resulting in the scheduling parameters of the industrial PON not adapting to the needs of the industrial Ethernet service.

[0009] To achieve the above purpose, the present application provides the following technical solutions:

[0010] A scheduling adaptation method for industrial PON and industrial Ethernet, the method comprising:

[0011] Constructing a protocol-optical resource mapping semantic model, the protocol-optical resource mapping semantic model being used to abstract the communication requirements of different industrial Ethernet protocols into a unified data structure containing time-sensitive service levels, traffic characteristics, and communication cycle parameters, and generating corresponding industrial PON scheduling parameters based on the unified data structure through logical mapping;

[0012] based on the port mirroring technology, real-time sensing of target industrial Ethernet traffic is performed, and the protocol-optical resource mapping semantic model is called to perform protocol identification and time-sensitive service feature extraction on the sensed traffic, so as to identify at least one industrial service flow, and then based on the extracted time-sensitive service feature, the bandwidth requirement and transmission timing characteristic of the industrial service flow are quantified;

[0013] Based on the protocol-optical resource mapping semantic model, the scheduling requirement parameters including T-CONT type, bandwidth configuration parameters and priority parameters are generated in combination with the protocol identification result and the bandwidth requirement and transmission timing characteristic, and then the underlying atomic capability is called through the control interface of the industrial PON management platform to generate and automatically issue the deterministic scheduling instruction based on the scheduling requirement parameters, so as to realize the deterministic transmission service.

[0014] Preferably, the protocol-optical resource mapping semantic model is constructed, specifically including the following steps:

[0015] For each industrial Ethernet protocol, the message field is extracted from its standard document, and a machine-readable field description item is defined for each field;

[0016] Based on the field description item, an extensible protocol-optical resource mapping rule description file is generated, wherein the protocol-optical resource mapping rule description file contains resource adaptation rules, and the resource adaptation rules are used to automatically generate scheduling requirement parameters of the industrial PON management platform according to the parsed field values;

[0017] Based on the protocol-optical resource mapping rule description file, a bit-level parsing engine is constructed, wherein the bit-level parsing engine is configured to perform recursive accurate parsing on the input industrial Ethernet message, and output structured protocol field values;

[0018] The protocol-optical resource mapping rule description file is checked for integrity and consistency by a static checker, wherein the protocol-optical resource mapping rule description file that passes the check and the bit-level parsing engine together constitute the protocol-optical resource mapping semantic model.

[0019] Preferably, the resource adaptation rules include:

[0020] DBA bandwidth mapping rules for dynamically calculating the bandwidth parameters of the industrial PON system according to the parsed protocol field values;

[0021] T-CONT type selection rules for automatically matching the corresponding T-CONT type based on the real-time level of the service flow;

[0022] Flow classification rules for defining service flow identification fingerprints and generating flow classification parameters;

[0023] QoS queue mapping rules are used to establish a mapping relationship between a protocol priority field value and a GemPort queue on a PON side.

[0024] Preferably, the port mirroring-based technology is used to realize real-time sensing of the target industrial Ethernet traffic, including capturing and preprocessing messages constituting the target industrial Ethernet traffic, and the specific steps are as follows:

[0025] The collector management interface of the industrial PON management platform is called to select OLT uplink port mirroring or ONU side user port mirroring as the best capture node based on the port mirroring technology, and map physical interface traffic to a corresponding logical slice channel.

[0026] A packet capture engine is used to capture messages at a line speed and record high-precision arrival timestamps, and multi-level filtering is performed at a packet capture entrance to separate messages belonging to the target industrial Ethernet traffic.

[0027] Preferably, the protocol-optical resource mapping semantic model is called to perform protocol identification and time-sensitive service feature extraction on the sensed traffic to identify at least one industrial service flow, including:

[0028] The protocol-optical resource mapping rule description file is loaded to perform field-level matching verification on the industrial Ethernet messages, calculate a matching score to determine a protocol type, and automatically match a PON service priority according to a real-time level of the protocol;

[0029] The constructed bit-level analysis engine is called to perform deep analysis on the identified messages to extract time-sensitive service features such as a communication period, a payload length, and a priority category, for burst traffic, identify a burst message size and a control service type, introduce a maximum tolerable delay of the service as a constraint variable, and quantify a non-periodic burst payload into an instantaneous peak bandwidth feature satisfying real-time transmission requirements;

[0030] Messages with the same five-tuple are aggregated into one industrial service flow.

[0031] Preferably, the bandwidth requirement and transmission timing characteristics of the industrial service flow are quantified based on the extracted time-sensitive service features, including:

[0032] The communication period is estimated based on a message arrival time sequence of each industrial service flow.

[0033] Based on the estimated communication period, the burst message size of the burst traffic, and the maximum tolerable delay of the service, the peak bandwidth requirement and the average bandwidth requirement per unit time are calculated, and transmission timing characteristic parameters for guiding DBA template selection are generated in combination with the priority category of the service flow.

[0034] Preferably, the step of generating scheduling requirement parameters based on the protocol-optical resource mapping semantic model, combined with the protocol identification results and the bandwidth requirements and transmission timing characteristics, including T-CONT type, bandwidth configuration parameters, and priority parameters, includes:

[0035] Based on the protocol identification results, the priority parameters of the service flow are determined using the protocol-optical resource mapping semantic model;

[0036] Combining the bandwidth requirements and transmission timing characteristics, the target bandwidth value is calculated using the DBA bandwidth mapping rules in the protocol-optical resource mapping semantic model, and the corresponding T-CONT type is determined based on the T-CONT type selection rules.

[0037] The priority parameter, target bandwidth value, and T-CONT type are integrated into the scheduling requirement parameter;

[0038] A configuration suggestion message is constructed, which encapsulates the scheduling requirement parameters, the service flow quintuple, and the device registration information, and marks the service level urgency.

[0039] Preferably, the step of calling the underlying atomic capabilities through the control interface of the industrial PON management platform to generate and automatically issue deterministic scheduling instructions based on the scheduling requirement parameters to achieve deterministic transmission services includes:

[0040] The configuration suggestion message is pushed to the industrial PON management platform according to the service level urgency through the hierarchical reporting channel;

[0041] On the industrial PON management platform side, the configuration suggestion message is parsed to obtain the scheduling requirement parameters, and a predefined DBA template is matched based on the target bandwidth value to obtain the DBA template ID, thereby generating a deterministic scheduling instruction containing the T-CONT type, DBA template ID, and QoS queue mapping relationship determined based on the priority parameter.

[0042] The underlying atomic capabilities are invoked to sequentially execute T-CONT creation, DBA template binding, and QoS queue mapping configuration operations in order to automatically issue the deterministic scheduling instructions.

[0043] Preferably, the method further includes a protocol self-learning step:

[0044] When the message matching score is lower than the preset threshold, the unmatched message fragments and their field offset positions are recorded in the abnormal cache area.

[0045] Based on multiple unknown messages of the same type, clustering is used to generate protocol feature fingerprints, and template incremental description files are automatically generated.

[0046] The template incremental description file is pushed to the administrator for review or automatically merged into the protocol-optical resource mapping rule description file to achieve dynamic adaptation to unknown industrial protocols.

[0047] Preferably, the method further includes an online optimization step for resource adaptation rules:

[0048] Continuously collect the jitter value of service flow transmission after the issuance of deterministic scheduling instructions;

[0049] When the jitter value of the service flow transmission continuously exceeds the tolerance range, the rule calibration mechanism is triggered;

[0050] Based on historical jitter data, the bandwidth calculation formula or T-CONT matching parameter in the resource adaptation rules is dynamically adjusted, and the optimized rule version is persisted to the protocol-optical resource mapping rule description file.

[0051] Compared with the prior art, this application has the following beneficial effects:

[0052] 1. This application establishes a mapping relationship between industrial Ethernet protocol service requirements and industrial PON scheduling parameters, extracts core parameters that can characterize the real-time and deterministic requirements from industrial Ethernet communication, and effectively transforms them into scheduling parameters that can be understood and executed by industrial PON, thereby realizing automated scheduling adaptation between industrial Ethernet and industrial PON. It can provide low latency, low jitter, and high reliability transmission services for multi-protocol industrial equipment under the PON architecture with shared media.

[0053] 2. This application achieves a leap from protocol identification to precise orchestration of optical network resources. This application is not only compatible with mainstream industrial Ethernet protocols such as EtherNet / IP and PROFINET, but also innovatively establishes a mapping model from protocol characteristics to PON atomic configuration (T-CONT type, DBA bandwidth template, GemPort queue). This enables industrial PON systems to automatically calculate and allocate determined time slot resources according to the real-time requirements of service flows (such as IRT / RT), solving the problem of industrial control flow jitter caused by the "best-effort" mode forwarding of traditional PON networks.

[0054] 3. By deeply integrating with the network control layer of the industrial PON management platform, this application uses parameters obtained from real-time packet capture to directly drive the platform to call the underlying atomic capability interfaces (such as automatically creating VLANs and automatically binding DBA templates). This eliminates the dependence on manual static configuration when accessing multi-protocol devices and realizes plug-and-play and adaptive bearer capabilities for industrial all-optical networks. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is a flowchart illustrating a scheduling adaptation method between industrial PON and industrial Ethernet in an embodiment of this application.

[0057] Figure 2 This is a schematic diagram of an industrial PON system architecture in a specific example of this application. Detailed Implementation

[0058] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0059] In the embodiments provided in this application, it should be understood that the disclosed methods and systems can be implemented in other ways. The system embodiments described below are merely illustrative. For example, the division of units and modules is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or modules can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, and can be electrical, mechanical, or other forms.

[0060] In addition, each functional unit in the various embodiments of this application can be integrated into a single processor, or each unit can be a separate device, or two or more units can be integrated into a single device; each functional unit in the various embodiments of this application can be implemented in hardware or in the form of hardware plus software functional units.

[0061] Those skilled in the art will understand that all or part of the steps of the following method embodiments can be implemented by program instructions and related hardware. The aforementioned program instructions can be stored in a computer-readable storage medium. When the program instructions are executed, they perform the steps of the following method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0062] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "a plurality of" or "several" means two or more, unless otherwise explicitly specified.

[0063] See Figures 1-2 This application addresses the issue of multi-protocol adaptation in industrial PON by designing and implementing a scheduling adaptation method for mainstream industrial Ethernet protocols to solve the problem of deriving key communication configurations from configuration software for various protocols. Considering the uncertainty of communication characteristics in dynamic industrial environments, where static configuration alone cannot fully reflect the actual communication behavior of equipment during operation, this application designs and implements a real-time parsing method for industrial Ethernet protocols with protocol identification, parameter extraction, and data reporting functions. This method captures and analyzes real-time Ethernet packets in industrial networks, extracting key communication parameters to provide a real-time sensing foundation for industrial PON systems.

[0064] Figure 1 This is a schematic diagram of the scheduling adaptation method for industrial PON and industrial Ethernet proposed in this application. Addressing the timing mismatch problem between heterogeneous industrial Ethernet protocols (such as PROFINET IRT and EtherCAT) and the TDMA (Time Division Multiple Access) scheduling mechanism of PON networks, this method proposes an adaptive orchestration architecture based on traffic characteristic awareness. Without altering the physical layer transmission mechanism of existing PON standards (ITU-T G.984 / G.9807), this method uses an external probe program to perceive the communication cycle and burst characteristics of industrial service flows in real time, and automatically matches the optimal DBA (Dynamic Bandwidth Allocation) authorization time, thereby eliminating cross-cycle queuing delays.

[0065] like Figure 1 As shown in the figure, this application embodiment provides a scheduling adaptation method for industrial PON and industrial Ethernet, which may include the following steps:

[0066] S1. Construct a protocol-optical resource mapping semantic model. The protocol-optical resource mapping semantic model is used to abstract the communication requirements of different industrial Ethernet protocols into a unified data structure that includes time-sensitive service level, traffic characteristics and communication cycle parameters, and generate corresponding industrial PON scheduling parameters based on this unified data structure through logical mapping.

[0067] S2, based on port mirroring technology, performs real-time sensing of target industrial Ethernet traffic, and calls the protocol-optical resource mapping semantic model to identify the protocol and extract time-sensitive service features of the sensed traffic, so as to identify at least one industrial service flow, and then quantify the bandwidth requirements and transmission timing characteristics of the industrial service flow based on the extracted time-sensitive service features.

[0068] S3, based on the protocol-optical resource mapping semantic model, combines protocol identification results with bandwidth requirements and transmission timing characteristics to generate scheduling requirement parameters including T-CONT type, bandwidth configuration parameters, and priority parameters. Then, through the control interface of the industrial PON management platform, it calls the underlying atomic capabilities to generate and automatically issue deterministic scheduling instructions based on the scheduling requirement parameters to achieve deterministic transmission services.

[0069] This embodiment provides an end-to-end solution from protocol identification to precise orchestration of optical network resources. By constructing a protocol-optical resource mapping semantic model, real-time traffic awareness and feature extraction, and deterministic scheduling instruction generation and issuance, it realizes the automated mapping between industrial Ethernet protocol service requirements and industrial PON scheduling parameters.

[0070] The scheduling and adaptation method for industrial PON and industrial Ethernet in this embodiment is not only compatible with mainstream industrial Ethernet protocols such as EtherNet / IP and PROFINET, but also automatically calculates and allocates specific time slot resources according to the real-time requirements of service flows (such as IRT / RT). This fundamentally solves the problem of industrial control flow jitter caused by the "best-effort" forwarding mode of traditional PON networks, providing low-latency, low-jitter, and highly reliable transmission services for multi-protocol industrial equipment. It also solves the problem of inaccurate resource allocation and insufficient deterministic guarantee caused by the lack of protocol feature awareness when existing industrial PON systems carry industrial Ethernet services.

[0071] Specifically, in this embodiment, through deep protocol awareness based on port mirroring technology, the "micro-burst" characteristics of each industrial business flow can be accurately understood, thereby dynamically calculating and allocating just the right amount of bandwidth, avoiding resource waste or congestion and packet loss caused by traditional static configuration, and fundamentally ensuring the stability of the business.

[0072] By using traffic feature perception and resource adaptation based on the protocol-optical resource mapping semantic model, the uplink authorization timing of PON is proactively adjusted, so that data packets are transmitted as soon as they arrive. This completely eliminates the periodic queuing delay caused by the misalignment of the TDMA mechanism and the service cycle, which is crucial for industrial control scenarios that require microsecond-level synchronization accuracy.

[0073] The entire process requires no manual intervention and can automatically adapt to various heterogeneous industrial protocols, enabling plug-and-play and adaptive carrying of industrial PON networks. This greatly reduces the complexity of network deployment and maintenance, and provides core technical support for building an open and flexible next-generation industrial all-optical network.

[0074] In one embodiment, constructing a protocol-optical resource mapping semantic model specifically includes the following steps:

[0075] For each industrial Ethernet protocol, extract message fields from its standard documentation and define machine-readable field descriptions for each field;

[0076] Based on the field description items, an extensible protocol-optical resource mapping rule description file is generated. The protocol-optical resource mapping rule description file contains resource adaptation rules, which are used to automatically generate scheduling requirement parameters for the industrial PON management platform based on the parsed field values.

[0077] Based on the protocol-optical resource mapping rule description file, a bit-level parsing engine is constructed. The bit-level parsing engine is configured to perform recursive and precise parsing on the input industrial Ethernet packets and output structured protocol field values.

[0078] The protocol-optical resource mapping rule description file is verified for integrity and consistency by a static verifier. The verified protocol-optical resource mapping rule description file and the bit-level parsing engine together constitute the protocol-optical resource mapping semantic model.

[0079] The protocol-optical resource mapping semantic model constructed in this embodiment is the foundation for realizing protocol awareness capabilities. In constructing the protocol-optical resource mapping semantic model, firstly, for each protocol, all key message fields (such as cycle time, payload length, priority identifier, etc.) are extracted from its standard document and precisely defined using unified, machine-readable field description items, such as `name` (field identifier), `bit_offset` (start bit offset), `bit_length` (length), `type` (type), `semantic` (semantic description), and the crucial `pon_attribute_tag` (PON service attribute tag). Next, this field information is abstracted into an extensible description file (such as JSON / YAML). This file not only contains the parsing structure but also resource adaptation rules (adaptation_rules) used to automatically generate scheduling requirement parameters for the industrial PON management platform based on the parsed field values. Then, a bit-level parsing engine is constructed. Its workflow includes loading the model, preprocessing messages, recursively parsing fields (supporting dynamic offsets, expression calculations, and nested structures), and finally outputting structured protocol field values. Finally, a static validator (Model) is used to... Linter performs pre-deployment validation on the protocol-optical resource mapping rule description file, checking for issues such as overlapping fields and circular dependencies.

[0080] The protocol-optical resource mapping semantic model constructed in this embodiment establishes a standardized, scalable, and verifiable "translator" between the diverse upper-layer industrial protocols and the underlying PON resources, making subsequent automated scheduling possible.

[0081] In one embodiment, the resource adaptation rules include:

[0082] DBA bandwidth mapping rules are used to dynamically calculate the bandwidth parameters of industrial PON systems based on the parsed protocol field values.

[0083] The T-CONT type selection rule is used to automatically match the corresponding T-CONT type based on the real-time level of the business flow;

[0084] Flow classification rules are used to define the business flow identification fingerprint and generate flow classification parameters;

[0085] QoS queue mapping rules are used to establish the mapping relationship between protocol priority field values ​​and PON-side GemPort queues.

[0086] Specifically, the DBA bandwidth mapping (`dba_mapping`) rule defines how to calculate DBA parameters based on protocol fields (such as payload size, cycle time), for example, the formula `bandwidth=${fields.PayloadLen}*8*(1000 / ${fields.CycleTime})`; the T-CONT type selection (`tcont_selection`) rule automatically selects the fixed bandwidth Type 1 T-CONT based on the real-time level (such as `if ${fields.RT_Class} == "IRT" then TCONT_Type =1`); the flow classification (`flow_classification`) rule defines the "fingerprint" of the service flow (such as VLAN ID+EtherType), which is used to guide the platform to automatically issue ACL or flow classification rules; the QoS queue mapping (`qos_queue_map`) rule establishes a mapping table between protocol priority and PON-side GemPort queues (Queue0-7).

[0087] These rules directly translate abstract protocol semantics into specific, executable PON configuration instructions, achieving precise and automated mapping from "business requirements" to "network resources" and eliminating the reliance on manual static configuration.

[0088] In one embodiment, real-time sensing of target industrial Ethernet traffic based on port mirroring technology includes capturing and preprocessing the packets constituting the target industrial Ethernet traffic. The specific steps are as follows:

[0089] The collector management interface of the industrial PON management platform is called. Based on port mirroring technology, the OLT uplink port mirror or ONU side user port mirror is selected as the best capture node, and the physical interface traffic is mapped to the corresponding logical slice channel.

[0090] A packet capture engine is used to capture packets at line speed and record their high-precision arrival timestamps. Multi-level filtering is performed at the packet capture entry point to separate packets belonging to the target industrial Ethernet traffic.

[0091] This embodiment, based on port mirroring technology, ensures the data quality and efficiency of subsequent analysis through real-time industrial Ethernet traffic sensing. First, it calls the collector management interface of the industrial PON management platform to automatically scan and select the best capture node (such as the OLT uplink port or ONU-side UNI port with hardware timestamps and burst enhancement buffers), and reads the VLAN-slice mapping table to construct a multi-layer association context of "physical port - logical slice - service type". Then, it uses a kernel bypass mechanism (such as DPDK) to achieve zero-copy packet acquisition and uses a hardware-level micro-burst buffer to smooth the burstiness of the PON uplink TDMA structure, achieving full line rate packet capture and providing nanosecond-level hardware timestamps. At the packet capture entry point, it performs multi-level filtering, including VLAN ID or P-bit-based slice filtering and protocol feature filtering to remove management packets such as OMCI.

[0092] This embodiment accurately identifies the business flow within the target industrial business segment, while ensuring the high-precision timestamps required for time series analysis, laying a solid foundation for subsequent business feature quantification and resource adaptation.

[0093] In one embodiment, a protocol-optical resource mapping semantic model is invoked to perform protocol identification and time-sensitive service feature extraction on the sensed traffic, in order to identify at least one industrial service flow, including:

[0094] Load the protocol-optical resource mapping rule description file, perform field-level matching verification on industrial Ethernet packets, calculate the matching score to determine the protocol type, and automatically match the PON service priority according to the real-time level of the protocol.

[0095] The constructed bit-level parsing engine is invoked to perform deep parsing of the identified packets, extracting time-sensitive service characteristics such as communication period, payload length and priority category. For burst traffic, the size of the burst packets and the control service type are identified, and the maximum tolerable latency of the service is introduced as a constraint variable. The non-periodic burst load is quantified into instantaneous peak bandwidth characteristics that meet the real-time transmission requirements.

[0096] Messages with the same quintuple are aggregated into a single industrial traffic flow.

[0097] In this embodiment, when performing protocol identification and time-sensitive service feature extraction on the sensed traffic using the protocol-optical resource mapping semantic model, each protocol template (i.e., the protocol-optical resource mapping rule description file) in the template library is loaded sequentially, and field-level matching verification is performed on each template. If a template field is mandatory, the corresponding position in the message must be consistent with the template description; if it is optional, a comparison is performed when the built-in conditions are met. By calculating a matching score Score(p) for each template, when the score is greater than a threshold (e.g., 0.9), the protocol type is determined. After determining the protocol type, the service priority (CoS 0-7) of the industrial PON system is automatically matched according to the real-time requirements of the protocol (e.g., RT / IRT level). Next, the bit-level parsing engine is called to perform deep parsing of the message, extracting core parameters such as communication period and payload length; at the same time, delay is used as a constraint variable to quantify non-periodic burst loads into instantaneous peak bandwidth characteristics that meet real-time transmission requirements. Finally, messages with the same five-tuple (source IP / port + destination IP / port + protocol type) are aggregated into an independent industrial service flow.

[0098] The industrial business flow identification process in this embodiment achieves accurate identification and feature quantification of heterogeneous industrial traffic, providing accurate input for subsequent personalized scheduling.

[0099] In one embodiment, quantifying the bandwidth requirements and transmission timing characteristics of the industrial service flow based on the extracted time-sensitive service features includes:

[0100] Communication cycle is estimated based on message arrival time series of each industrial business flow;

[0101] Based on the estimated communication cycle, high-precision message arrival timestamps, burst message size of burst traffic, and maximum service tolerance delay, the peak bandwidth requirement and average bandwidth requirement per unit time are calculated. Combined with the priority category of the service flow, transmission timing characteristic parameters are generated to guide DBA template selection.

[0102] This embodiment addresses the core pain point of mismatch between resource allocation strategies and business characteristics in existing technologies by accurately determining the "burst traffic demand" and "actual bandwidth demand" of industrial business flows based on extracted time-sensitive business characteristics.

[0103] First, if the protocol contains an explicit periodic field (such as cycle_time or heartbeat identifier), it is directly determined to be a periodic flow; otherwise, the five-tuple message is sorted in ascending order of timestamp, the arrival interval sequence of adjacent messages is calculated, and the median difference method is used to estimate the preliminary period. Then, the probe program deployed on the uplink port of the OLT (Optical Line Terminal) or the ONU (Optical Network Unit) side records the arrival times {t1, t2, ..., tn} of N consecutive industrial messages and their corresponding message lengths. Divide the time window into corresponding The logical slice is used to accumulate the length of packets falling within the same periodic window to obtain the periodic traffic value for a single period. Simultaneously, the arrival time deviation between adjacent cycles is calculated to obtain the inherent jitter value of the service flow. Finally, based on the formula... Calculate the guaranteed bandwidth requirement for this service flow, and for bursty service components, based on the maximum allowable transmission time limit (T). limit ), calculate the maximum possible traffic aggregation value Vaggregate and the instantaneous peak bandwidth requirement of the service flow. , Then, it is compared with the guaranteed bandwidth and peak bandwidth configured in the current DBA template. If the guaranteed bandwidth and peak bandwidth requirements exceed the current template's capacity, or the jitter value exceeds the safety tolerance, it is determined as "resource scheduling strategy mismatch".

[0104] The bandwidth requirement and transmission timing characteristics calculation in this embodiment accurately quantifies the supply-demand difference between the burst characteristics of the service flow and the PON bandwidth supply, providing direct data support for automatically matching high-priority T-CONT types and fixed-bandwidth DBA templates.

[0105] In one embodiment, based on the protocol-optical resource mapping semantic model, and combining the protocol identification results with the bandwidth requirements and transmission timing characteristics, scheduling requirement parameters including T-CONT type, bandwidth configuration parameters, and priority parameters are generated, including:

[0106] Based on the protocol identification results, the priority parameters of the service flow are determined using the protocol-optical resource mapping semantic model;

[0107] Combining bandwidth requirements and transmission timing characteristics, the target bandwidth value is calculated using the DBA bandwidth mapping rule in the protocol-optical resource mapping semantic model, and the corresponding T-CONT type is determined based on the T-CONT type selection rule.

[0108] The priority parameter, target bandwidth value, and T-CONT type are integrated into a scheduling requirement parameter.

[0109] Construct a configuration suggestion message, which encapsulates scheduling requirement parameters, service flow 5-tuples and device registration information, and marks the service level urgency.

[0110] This embodiment transforms the protocol identification results and traffic characteristic analysis results from the preceding steps into precise deterministic scheduling instructions that can be directly executed by the PON device, thus completing the transformation from analysis to decision-making.

[0111] Specifically, in this embodiment, resource adjustment suggestions including a bandwidth redundancy factor are automatically generated based on the quantified traffic characteristics. Specifically, by matching fixed-bandwidth DBA templates to periodic, high-real-time services and allocating a guaranteed bandwidth slightly higher than their actual burst peak, it can be ensured that the uplink transmission channel always has sufficient idle time slots to accommodate service bursts, thereby logically eliminating congestion queuing. Simultaneously, the function `MatchDbaTemplate(bandwidth)` is called to automatically match the closest DBA template ID. Finally, the T-CONT type, DBA template ID, and target authorization offset are integrated into a complete deterministic scheduling instruction and encapsulated as scheduling requirement parameters, carrying the service flow quintuple, device registration information, and service level urgency.

[0112] The method in this embodiment generates precise scheduling instructions that can be directly executed by the PON platform, including bandwidth guarantees and priority allocation, effectively ensuring the realization of deterministic transmission services.

[0113] In one embodiment, the control interface of the industrial PON management platform invokes underlying atomic capabilities to generate and automatically issue deterministic scheduling instructions based on the scheduling requirement parameters to achieve deterministic transmission services, including:

[0114] The configuration suggestion message is pushed to the industrial PON management platform according to the urgency of the service level through the hierarchical reporting channel;

[0115] On the industrial PON management platform side, the configuration suggestion message is parsed to obtain the scheduling requirement parameters, and the predefined DBA template is matched based on the target bandwidth value to obtain the DBA template ID, thereby generating a deterministic scheduling instruction containing the T-CONT type, DBA template ID and QoS queue mapping relationship determined based on priority parameters.

[0116] The underlying atomic capabilities are invoked to sequentially execute T-CONT creation, DBA template binding, and QoS queue mapping configuration operations in order to automatically issue the deterministic scheduling instructions.

[0117] Specifically, when the reporting link is restored after an interruption, the configuration suggestion message that was not successfully issued can be retransmitted based on the persistent queue, and the final reliability of the critical scheduling instructions can be guaranteed through the ACK confirmation mechanism.

[0118] This embodiment constructs a highly reliable, low-latency control closed loop, ensuring the effective implementation of the deterministic scheduling strategy.

[0119] Specifically, in this embodiment, the industrial PON system establishes a dedicated communication channel with the industrial PON management platform's control interface. Monitoring metrics can be reported via WebSocket / MQTT, and configuration commands can be issued via RESTful API / RPC interfaces. Reported data is categorized into three levels based on real-time requirements: highest priority (e.g., alarms) bypasses the cache and is pushed directly; high priority (e.g., DBA changes) ensures timeliness; and low priority (e.g., logs) is processed normally. To ensure reliability, a ring buffer is used for short-term high-speed caching, combined with a disk-based persistent queue for long-term backup of important packets. When the reporting link is interrupted and then restored, data that was not successfully reported can be automatically retrieved from the persistent queue, reassembled according to priority, and deduplicated based on message ID. An ACK confirmation and a limited retry mechanism ensure the final reliability of critical status information reporting.

[0120] In one embodiment, the scheduling adaptation method for industrial PON and industrial Ethernet of this application may further include a protocol self-learning step:

[0121] When the message matching score is lower than the preset threshold, the unmatched message fragments and their field offset positions are recorded in the abnormal cache area.

[0122] Based on multiple unknown messages of the same type, clustering is used to generate protocol feature fingerprints, and template incremental description files are automatically generated.

[0123] The template incremental description file can be pushed to the administrator for review or automatically merged into the protocol-optical resource mapping rule description file to achieve dynamic adaptation to unknown industrial protocols.

[0124] This embodiment endows the industrial PON system with the ability to continuously learn and evolve, dynamically expand the protocol library, effectively address the challenges of protocol fragmentation and rapid iteration in the industrial field, significantly improve the versatility and long-term applicability of the industrial PON system, and solve the compatibility issues between proprietary protocols and new protocols in industrial settings.

[0125] Specifically, in this embodiment, for messages that fail to match or whose fields fail verification, the industrial PON system records the unmatched field fragments and their offset positions in the anomaly buffer. By analyzing multiple similar unknown messages, the industrial PON system can cluster to generate new protocol feature fingerprints and automatically generate a template incremental description file (Patch Template). This file can be pushed to the administrator for review, or automatically merged into the main protocol-optical resource mapping rule description file under high confidence.

[0126] In one embodiment, the scheduling adaptation method for industrial PON and industrial Ethernet of this application may further include an online optimization step for resource adaptation rules:

[0127] Continuously collect data on the jitter value of service flow transmission and the fluctuation of message arrival interval after the issuance of deterministic scheduling instructions;

[0128] When the jitter value of the service flow transmission or the fluctuation of the packet arrival interval continues to exceed the tolerance range, the rule calibration mechanism is triggered.

[0129] The bandwidth calculation formula or T-CONT parameter in the resource adaptation rules is dynamically adjusted based on historical jitter data, and the optimized rule version is persisted to the protocol-optical resource mapping rule description file.

[0130] This embodiment, through online closed-loop optimization of resource adaptation rules, can maintain optimal deterministic transmission performance in a long-term and stable manner, overcomes the limitations of static rules in dynamic industrial environments, and realizes the leap from static configuration to dynamic optimization.

[0131] Specifically, the industrial PON management platform, based on network feedback after configuration (such as whether jitter has decreased), reverse-corrects the identification thresholds or calculation parameters in the protocol identification and time-sensitive feature extraction processes and the deterministic scheduling command generation and issuance processes. In particular, the industrial PON system continuously monitors the standard deviation of the service flow packet arrival interval (i.e., the service flow transmission jitter value) after the issuance of scheduling commands. If, due to factors such as equipment aging or environmental changes, the actual monitored jitter value continuously exceeds the preset tolerance, the industrial PON system will trigger a rule calibration mechanism.

[0132] Based on historical jitter data, dynamically adjust the bandwidth calculation formula in the resource adaptation rules (such as introducing a compensation coefficient) or the T-CONT parameter, and persist the optimized rule version.

[0133] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0134] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0135] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0136] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A scheduling adaptation method for industrial PON and industrial Ethernet, characterized in that, The method includes: A protocol-optical resource mapping semantic model is constructed. The protocol-optical resource mapping semantic model is used to abstract the communication requirements of different industrial Ethernet protocols into a unified data structure that includes time-sensitive service level, traffic characteristics and communication cycle parameters, and to generate corresponding industrial PON scheduling parameters based on the unified data structure through logical mapping. Based on port mirroring technology, the target industrial Ethernet traffic is sensed in real time, and the protocol-optical resource mapping semantic model is called to identify the protocol and extract time-sensitive service features of the sensed traffic in order to identify at least one industrial service flow. Then, based on the extracted time-sensitive service features, the bandwidth requirements and transmission timing characteristics of the industrial service flow are quantified. Based on the protocol-optical resource mapping semantic model, and combined with the protocol identification results, bandwidth requirements, and transmission timing characteristics, scheduling requirement parameters including T-CONT type, bandwidth configuration parameters, and priority parameters are generated. Then, through the control interface of the industrial PON management platform, the underlying atomic capabilities are invoked to generate and automatically issue deterministic scheduling instructions based on the scheduling requirement parameters, so as to realize deterministic transmission services.

2. The scheduling and adaptation method for industrial PON and industrial Ethernet according to claim 1, characterized in that, The construction of the protocol-optical resource mapping semantic model specifically includes the following steps: For each industrial Ethernet protocol, extract message fields from its standard documentation and define machine-readable field descriptions for each field; Based on the field description items, an extensible protocol-optical resource mapping rule description file is generated, wherein the protocol-optical resource mapping rule description file contains resource adaptation rules, which are used to automatically generate scheduling requirement parameters for the industrial PON management platform according to the parsed field values. Based on the protocol-optical resource mapping rule description file, a bit-level parsing engine is constructed, wherein the bit-level parsing engine is configured to perform recursive precise parsing on the input industrial Ethernet packets and output structured protocol field values; The protocol-optical resource mapping rule description file is subjected to integrity and consistency verification by a static verifier. The protocol-optical resource mapping rule description file that passes the verification, together with the bit-level parsing engine, constitutes the protocol-optical resource mapping semantic model.

3. The scheduling and adaptation method for industrial PON and industrial Ethernet according to claim 2, characterized in that, The resource adaptation rules include: DBA bandwidth mapping rules are used to dynamically calculate the bandwidth parameters of industrial PON systems based on the parsed protocol field values. The T-CONT type selection rule is used to automatically match the corresponding T-CONT type based on the real-time level of the business flow; Flow classification rules are used to define the business flow identification fingerprint and generate flow classification parameters; QoS queue mapping rules are used to establish the mapping relationship between protocol priority field values ​​and PON-side GemPort queues.

4. The scheduling and adaptation method for industrial PON and industrial Ethernet according to claim 1, characterized in that, The real-time sensing of target industrial Ethernet traffic based on port mirroring technology includes capturing and preprocessing the packets constituting the target industrial Ethernet traffic. The specific steps are as follows: The collector management interface of the industrial PON management platform is called, and based on the port mirroring technology, the OLT uplink port mirror or ONU side user port mirror is selected as the best capture node, and the physical interface traffic is mapped to the corresponding logical slice channel. A packet capture engine is used to capture packets at line speed and record their high-precision arrival timestamps. Multi-level filtering is performed at the packet capture entry point to separate packets belonging to the target industrial Ethernet traffic.

5. The scheduling and adaptation method for industrial PON and industrial Ethernet according to claim 4, characterized in that, The process of invoking the protocol-optical resource mapping semantic model to perform protocol identification and time-sensitive service feature extraction on the sensed traffic is used to identify at least one industrial service flow, including: Load the protocol-optical resource mapping rule description file, perform field-level matching verification on the industrial Ethernet packets, calculate the matching score to determine the protocol type, and automatically match the PON service priority according to the real-time level of the protocol. The constructed bit-level parsing engine is invoked to perform deep parsing of the identified packets, extracting time-sensitive service characteristics such as communication period, payload length and priority category. For burst traffic, the size of the burst packets and the control service type are identified, and the maximum tolerable latency of the service is introduced as a constraint variable. The non-periodic burst load is quantified into instantaneous peak bandwidth characteristics that meet the real-time transmission requirements. Messages with the same quintuple are aggregated into a single industrial traffic flow.

6. The scheduling and adaptation method for industrial PON and industrial Ethernet according to claim 5, characterized in that, The quantification of the bandwidth requirements and transmission timing characteristics of the industrial service flow based on the extracted time-sensitive service features includes: Communication cycle is estimated based on message arrival time series of each industrial business flow; Based on the estimated communication period, the burst packet size of the burst traffic, and the maximum tolerable delay of the service, the peak bandwidth requirement and average bandwidth requirement per unit time are calculated, and combined with the priority category of the service flow, transmission timing characteristic parameters are generated to guide the DBA template selection.

7. The scheduling and adaptation method for industrial PON and industrial Ethernet according to claim 3, characterized in that, Based on the protocol-optical resource mapping semantic model, and combining the protocol identification results with the bandwidth requirements and transmission timing characteristics, scheduling requirement parameters are generated, including T-CONT type, bandwidth configuration parameters, and priority parameters, including: Based on the protocol identification results, the priority parameters of the service flow are determined using the protocol-optical resource mapping semantic model; Combining the bandwidth requirements and transmission timing characteristics, the target bandwidth value is calculated using the DBA bandwidth mapping rules in the protocol-optical resource mapping semantic model, and the corresponding T-CONT type is determined based on the T-CONT type selection rules. The priority parameter, target bandwidth value, and T-CONT type are integrated into the scheduling requirement parameter; A configuration suggestion message is constructed, which encapsulates the scheduling requirement parameters, the service flow quintuple, and the device registration information, and marks the service level urgency.

8. The scheduling and adaptation method for industrial PON and industrial Ethernet according to claim 7, characterized in that, The process of using the control interface of the industrial PON management platform to invoke underlying atomic capabilities to generate and automatically issue deterministic scheduling instructions based on the scheduling requirement parameters, in order to achieve deterministic transmission services, includes: The configuration suggestion message is pushed to the industrial PON management platform according to the service level urgency through the hierarchical reporting channel; On the industrial PON management platform side, the configuration suggestion message is parsed to obtain the scheduling requirement parameters, and a predefined DBA template is matched based on the target bandwidth value to obtain the DBA template ID, thereby generating a deterministic scheduling instruction containing the T-CONT type, DBA template ID, and QoS queue mapping relationship determined based on the priority parameter. The underlying atomic capabilities are invoked to sequentially execute T-CONT creation, DBA template binding, and QoS queue mapping configuration operations in order to automatically issue the deterministic scheduling instructions.

9. The scheduling and adaptation method for industrial PON and industrial Ethernet according to claim 5, characterized in that, The method also includes a protocol self-learning step: When the message matching score is lower than the preset threshold, the unmatched message fragments and their field offset positions are recorded in the abnormal cache area. Based on multiple unknown messages of the same type, clustering is used to generate protocol feature fingerprints, and template incremental description files are automatically generated. The template incremental description file is pushed to the administrator for review or automatically merged into the protocol-optical resource mapping rule description file to achieve dynamic adaptation to unknown industrial protocols.

10. The scheduling adaptation method for industrial PON and industrial Ethernet according to any one of claims 1-9, characterized in that, The method also includes an online optimization step for resource adaptation rules: Continuously collect the jitter value of service flow transmission after the issuance of deterministic scheduling instructions; When the jitter value of the service flow transmission continuously exceeds the tolerance range, the rule calibration mechanism is triggered; Based on historical jitter data, the bandwidth calculation formula or T-CONT matching parameter in the resource adaptation rules is dynamically adjusted, and the optimized rule version is persisted to the protocol-optical resource mapping rule description file.

Images