An industrial field heterogeneous network converged communication terminal

Abstract

This application discloses a converged communication terminal for heterogeneous networks in industrial settings, belonging to the fields of communication technology and smart chips. By semantically parsing data packets in heterogeneous industrial networks and combining physical link quality monitoring with transmission scheduling queue status, it achieves deep integration of protocol semantics, transmission scheduling, and link status. It can dynamically generate scheduling priority correction values ​​and migration urgency assessment values, thereby determining gating time slot reallocation strategies and link migration execution strategies. By adjusting the TSN gating list and executing sequential migration of status, it solves the problems of resource mismatch for high-urgency control commands and link switching interruptions under traditional fragmented architectures, ensuring the determinism and continuity of industrial control services.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a converged communication terminal for heterogeneous networks in industrial settings. Background Technology

[0002] With the advancement of Industry 4.0, industrial sites often contain devices using multiple communication protocols such as Modbus, Profinet, and EtherCAT, creating information silos. To achieve interconnection of heterogeneous devices, related technologies typically employ protocol conversion modules to uniformly encapsulate data from different protocols and perform deterministic transmission via Time-Sensitive Networking (TSN), while configuring multimode links to ensure communication reliability.

[0003] However, in related technologies, the three stages of protocol conversion, TSN scheduling, and multi-link management operate independently. The protocol conversion module only focuses on data format uniformity and is unaware of the data's importance in the control loop; the TSN scheduling module only allocates time slots based on a preset gating list and is unaware of the control semantics of the transmitted data; the link management module only performs switching based on link quality and is unaware of the upper-layer protocol session status. This fragmented architecture results in high-urgency control commands failing to obtain transmission resources matching their real-time requirements during mixed traffic surges or link quality fluctuations. Protocol session interruptions during link switching cause data loss, making it difficult to guarantee the determinism of industrial control services.

[0004] Therefore, there is an urgent need for a converged communication method that can integrate the three dimensions of protocol semantics, transmission scheduling, and link status to improve the end-to-end guarantee capability of heterogeneous networks in industrial fields for control services. Summary of the Invention

[0005] This application provides an industrial field heterogeneous network converged communication terminal, the technical solution of which is as follows: On the one hand, a method for converged communication of heterogeneous networks in industrial fields is provided, wherein the terminal includes a processor and a memory, and the processor is configured to perform the following steps: Semantic parsing is performed on data packets received from heterogeneous protocol ports to extract the protocol semantic attributes of data objects in the data packets. Quality monitoring is performed on the physical links carrying the data objects to extract the link quality status. The protocol semantic attributes, the queuing status of the data objects in the current transmission scheduling queue, and the link quality status are associated as data records. Based on the coupling relationship between protocol semantic attributes and queuing status in the data records, a scheduling priority correction value is generated, and based on the coupling relationship between protocol semantic attributes and link quality status in the data records, a migration urgency assessment value is generated. Based on the scheduling priority correction amount and the current queuing state of the data object, a gating slot reallocation strategy is determined, and based on the migration urgency assessment value and the protocol semantic attributes, a link migration execution strategy matching the protocol semantic attributes is determined. The TSN gating list is adjusted based on the gating time slot reallocation strategy, and the physical link carrying the data object is subjected to state continuity migration based on the link migration execution strategy. Attached Figure Description

[0006] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0007] Figure 1 This is a flowchart of a heterogeneous network converged communication method for industrial sites provided in an embodiment of this application; Figure 2 This is a flowchart of another industrial field heterogeneous network converged communication method provided in the embodiments of this application; Figure 3 This is a flowchart of another industrial field heterogeneous network converged communication method provided in the embodiments of this application; Figure 4 This is a flowchart of another industrial field heterogeneous network converged communication method provided in the embodiments of this application; Figure 5 This is a flowchart of another industrial field heterogeneous network converged communication method provided in the embodiments of this application. Detailed Implementation

[0008] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0009] In this application, the terms "first," "second," etc., are used to distinguish identical or similar items with essentially the same function. It should be understood that there is no logical or temporal dependency between "first," "second," and "nth," nor are there any restrictions on quantity or execution order.

[0010] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, data stored, data displayed, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0011] In industrial field heterogeneous network converged communication technologies, the three components—protocol conversion, time-sensitive networking (TSN) scheduling, and multi-link management—operate independently. The protocol conversion module is unaware of the data's importance within the control loop, the TSN scheduling module is unaware of the control semantics of the transmitted data, and the link management module is unaware of the upper-layer protocol session status. This fragmented architecture leads to situations where, under mixed traffic surges or link quality fluctuations, high-urgency control commands cannot obtain transmission resources matching their real-time requirements. Protocol session interruptions during link switching cause data loss, making it difficult to guarantee the determinism of industrial control services.

[0012] To address this issue, this application proposes a converged communication terminal for heterogeneous networks in industrial settings. This terminal performs semantic parsing on data packets received from heterogeneous protocol ports, extracting the protocol semantic attributes of data objects within the data packets. It also monitors the quality of the physical links carrying the data objects, extracting the link quality status, and associating the protocol semantic attributes, the queuing status of the data object in the current transmission scheduling queue, and the link quality status as data records. Based on the coupling relationship between the protocol semantic attributes and queuing status in the data records, a scheduling priority correction value is generated. Simultaneously, based on the coupling relationship between the protocol semantic attributes and the link quality status, a migration urgency assessment value is generated. Based on the scheduling priority correction value and the current queuing status of the data object, a gating time slot reallocation strategy is determined. Furthermore, based on the migration urgency assessment value and the protocol semantic attributes, a link migration execution strategy matching the protocol semantic attributes is determined. The TSN gating list can be adjusted based on the gating time slot reallocation strategy, and the physical links carrying the data objects can be migrated sequentially based on the link migration execution strategy, thereby achieving deep integration and collaborative management of protocol semantics, transmission scheduling, and link status.

[0013] For ease of understanding, the following explains some key terms in this embodiment: Heterogeneous protocol port: refers to an interface that receives data streams from different industrial communication protocols (such as Modbus, Profinet, EtherCAT, etc.).

[0014] Data message: refers to the basic data unit transmitted in a network, which includes information such as control commands and sensor data.

[0015] Data object: refers to the smallest unit of information carried in a data message that has a specific function or meaning, such as a sensor reading or an actuator command.

[0016] Protocol semantic attributes: refers to a set of attributes that describe the deeper meaning of a data object in an industrial control system, such as its functional role, importance, and real-time requirements.

[0017] Physical link: refers to the actual physical channel that carries the transmission of data objects, such as Ethernet cable, wireless channel, etc.

[0018] Link quality status: refers to a set of indicators that describe the current transmission performance and reliability of a physical link, such as bandwidth, latency, packet loss rate, and bit error rate.

[0019] Transmission scheduling queue: refers to a buffer in which data objects wait to be scheduled before data transmission, and is usually managed according to priority.

[0020] Queuing status: refers to information such as the current position of the data object in the transmission scheduling queue, waiting time, and queue depth.

[0021] Data records: refer to data structures that associate and encapsulate information such as protocol semantic attributes, queuing status, and link quality status for subsequent decision analysis.

[0022] Scheduling priority adjustment amount: refers to the value dynamically calculated based on the protocol semantic attributes and queuing status of the data object, used to adjust its priority in the transmission scheduling queue.

[0023] Migration urgency assessment value: refers to the numerical value that assesses the urgency of migrating the link carried by the data object based on the protocol semantic attributes of the data object and the quality status of the physical link.

[0024] Gated time slot reallocation strategy: refers to the adjustment scheme of the gating list in a time-sensitive network (TSN) to dynamically allocate or reallocate transmission time slots for data objects.

[0025] Link migration execution strategy: This refers to the specific plan for selecting a backup link and performing data object transmission switching when link quality degradation or high urgency is detected.

[0026] TSN gate list: refers to the list of scheduling rules in Time-Sensitive Networking (TSN) used to control the passage of data frames through switch ports within a specific time window.

[0027] State-based migration: refers to a migration method that ensures the continuity of upper-layer protocol sessions and avoids data loss or service interruption by saving and restoring the protocol communication state of data objects during physical link switching.

[0028] The terminal in this application includes a processor and memory, see [link to application]. Figure 1 The processor is configured to perform the following operations: 101. Perform semantic parsing on data packets received from heterogeneous protocol ports to extract the protocol semantic attributes of data objects in the data packets.

[0029] Specifically, the terminal can pre-define a static rule base containing mappings between different data packet types and fixed protocol semantic attributes. The corresponding protocol semantic attributes are retrieved by querying this rule base. For example, for data packets from a specific sensor, their protocol semantic attributes are fixedly marked as "high real-time, control data." As another implementation, the terminal can perform pattern matching on the payload content of the data packets, identifying predefined keywords or data structures, and inferring the protocol semantic attributes of the data objects based on the matched patterns. Furthermore, the terminal can parse the protocol header information of the data packets, such as source address, destination address, and function code. Based on the combination of these header fields and pre-define semantic rules, the protocol semantic attributes of the data objects are extracted. In some embodiments, the processor can be implemented as a smart chip.

[0030] 102. Perform quality monitoring on the physical links carrying data objects and extract the link quality status.

[0031] For example, a terminal can periodically send probe packets to the physical link and measure the round-trip time and packet loss rate of the packets, using these measurements as a direct reflection of the link quality status. Alternatively, the terminal can monitor data packets flowing through the physical link, statistically analyzing their transmission success rate, retransmission count, etc., using these statistics to assess the transmission quality of the physical link and using this data as the link quality status. The terminal can also obtain hardware-level link status information, such as signal strength, bit error rate, and connection speed, from the physical layer interface chip or network adapter, directly using these hardware indicators as the link quality status.

[0032] 103. Associate the protocol semantic attributes, the queuing status of the data object in the current transmission scheduling queue, and the link quality status as data records.

[0033] Specifically, the terminal can encapsulate the extracted protocol semantic attributes, the queuing status of the data object in the current transmission scheduling queue, and the link quality status into a unified data structure. This data structure can contain fields to store this information and be associated with a unique identifier. Alternatively, the terminal can store this information as different fields in an in-memory database or file system, with each data object corresponding to a record containing a reference or copy of its protocol semantic attributes, queuing status, and link quality status.

[0034] 104. Based on the coupling relationship between protocol semantic attributes and queuing status in data records, generate scheduling priority correction amount.

[0035] For example, the terminal can pre-set a simple set of rules: if the protocol semantic attribute of a data object indicates that it is "high priority," a fixed positive correction amount is directly generated. If the queuing status shows that the queue depth exceeds a certain threshold, a positive correction amount is also generated; the two are added together to obtain the total correction amount. As another implementation, the terminal can define a linear function that takes the protocol semantic attribute and queuing status as input and calculates the scheduling priority correction amount through a weighted summation. Furthermore, the terminal can maintain a lookup table, where the index is composed of combinations of protocol semantic attributes and queuing status; the pre-calculated scheduling priority correction amount is retrieved by querying this table.

[0036] 105. Based on the coupling relationship between protocol semantic attributes and link quality status in data records, generate a migration urgency assessment value.

[0037] Specifically, the terminal can set one or more link quality thresholds. When a certain indicator in the link quality status exceeds the preset threshold, a fixed migration urgency assessment value is directly generated, such as "high urgency." Protocol semantic attributes can be used as the basis for selecting the trigger threshold. Alternatively, the terminal can assign different weights to the protocol semantic attributes and link quality status, and then calculate the migration urgency assessment value through a weighted average. The terminal can also employ a fuzzy logic system, using protocol semantic attributes and link quality status as fuzzy inputs, and generating the migration urgency assessment value through fuzzy inference rules.

[0038] 106. Determine the gating time slot reallocation strategy based on the scheduling priority correction amount and the current queuing status of the data object.

[0039] For example, the terminal can preset some exchangeable time slot pairs. When the scheduling priority adjustment indicates that a data object needs higher priority, the terminal selects a time slot occupied by idle or low-priority data from the preset list and allocates it to the current data object. The queuing status can be used as the basis for selecting an idle time slot. As another implementation, the terminal can dynamically adjust the length of the existing gated time slots based on the scheduling priority adjustment and the current queuing status of the data object. In addition, the terminal can also identify unallocated idle time slots in the TSN gating list. When the scheduling priority adjustment indicates that a data object needs to be transmitted immediately, the terminal inserts the data object into the nearest idle time slot.

[0040] 107. Based on the migration urgency assessment value and protocol semantic attributes, determine the link migration execution strategy that matches the protocol semantic attributes.

[0041] Specifically, the terminal can preset multiple link migration execution strategies, such as "fast handover," "smooth handover," and "delayed handover." Based on the migration urgency assessment value and protocol semantic attributes, the terminal selects one of these preset strategies. Alternatively, the terminal can define a generic migration execution strategy template containing adjustable parameters. Based on the migration urgency assessment value and protocol semantic attributes, the terminal dynamically adjusts these parameters to generate a customized link migration execution strategy. The terminal can also maintain a list of backup links and select the most suitable target backup link from the list based on the migration urgency assessment value and protocol semantic attributes.

[0042] 108. Adjust the TSN gating list based on the gating time slot reallocation strategy.

[0043] For example, the terminal can regenerate a complete TSN gating list based on a defined gating time slot reallocation policy, and then send this new gating list to the TSN switch to replace the old list. Alternatively, the terminal can generate only the parts of the gating list that need modification based on the gating time slot reallocation policy, and then send these incremental updates to the TSN switch, allowing the switch to locally modify its gating list.

[0044] 109. Perform state-sequential migration of physical links carrying data objects based on link migration execution strategies.

[0045] Specifically, during link migration, the terminal can simply retransmit the currently transmitting data object on the new link. Alternatively, the terminal can record the sequence number of the current data object transmission before the link migration. After switching to the new physical link, it can send a synchronization request with that sequence number to inform the peer to start receiving data from that sequence number onwards. The terminal can also notify upper-layer applications of the link migration event. Upon detecting the link switch, the upper-layer application can re-initiate data transmission from the most recent successful point based on its own retry mechanism.

[0046] This application achieves deep integration of protocol semantics, transmission scheduling, and link status by semantically parsing data packets in heterogeneous industrial networks and combining physical link quality monitoring with transmission scheduling queue status. It can dynamically generate scheduling priority correction values ​​and migration urgency assessment values, thereby determining gating time slot reallocation strategies and link migration execution strategies. By adjusting the TSN gating list and executing state-sequential migration, it solves the problems of resource mismatch for high-urgency control commands and link switching interruptions in traditional fragmented architectures, ensuring the determinism and continuity of industrial control services.

[0047] In some of the solutions mentioned above in this application, semantic parsing is proposed to extract the protocol semantic attributes of the data objects in the data packets received from heterogeneous protocol ports, so as to associate data records and support subsequent scheduling and migration decisions. However, in this process, the static parsing method cannot dynamically capture the changes in the urgency of the data objects and the protocol session state, which makes it impossible to extract key semantic information in a timely manner when emergency control instructions appear, thereby affecting the accuracy of resource allocation and the continuity of link switching.

[0048] To address this, this application further proposes a method for semantic parsing of data packets received from heterogeneous protocol ports, see [link to relevant documentation]. Figure 2 The method includes: 201. Identify the industrial communication protocol type followed by the data message, and determine the protocol state machine instance corresponding to the data message based on the industrial communication protocol type.

[0049] 202. Drive the protocol state machine instance to parse the data message, extract the functional role label and dynamic urgency level of the data object in the data message, and use them as the basic fields of the protocol semantic attributes.

[0050] 203. When the dynamic urgency level exceeds the level threshold, extract the protocol communication state checkpoint of the data object from the current state of the protocol state machine instance, and supplement the protocol communication state checkpoint as an extended field of the protocol semantic attribute.

[0051] For example, when identifying the industrial communication protocol type followed by a data packet and determining the corresponding protocol state machine instance based on that protocol type, the aim is to provide the correct context and parsing logic for subsequent data parsing. The processor can pre-store a mapping table between protocol types and packet header features (such as port number, protocol ID, magic words, etc.). When a data packet is received, the processor identifies its industrial communication protocol type by checking specific header fields of the packet and comparing them with the mapping table. Once the protocol type is determined, the corresponding protocol state machine instance is loaded and initialized from a pre-loaded protocol state machine template library. Alternatively, heterogeneous protocol ports of the terminal can be pre-configured to bind specific industrial communication protocols. For example, one port might be dedicated to receiving Modbus TCP packets, and another port might be used to receive EtherCAT packets. When a data packet enters from a specific port, the processor directly identifies the protocol type based on the port's pre-defined binding relationship and activates or creates the corresponding protocol state machine instance.

[0052] When driving the protocol state machine instance to parse the data packet and extract the functional role labels and dynamic urgency levels of the data objects as the base fields of the protocol's semantic attributes, this step leverages the dynamic characteristics of the protocol state machine to deeply parse the data packet and obtain its core semantic information. The protocol state machine instance processes the data packet byte-by-byte or field-by-field according to its internally defined parsing logic. During parsing, the state machine triggers corresponding actions based on its current state and input data, such as searching a preset semantic annotation rule base. This rule base can contain the mapping relationship between data object address ranges and functional role labels and baseline urgency levels. When the state machine parses a specific data object, it queries the rule base to obtain its functional role label and baseline urgency level. The dynamic urgency level can be obtained by adjusting the baseline urgency level in conjunction with real-time operating status (such as system load and device health status). Furthermore, the protocol state machine instance can also call pre-registered semantic parsing callback functions when parsing data packets. These callback functions are triggered at different layers of the protocol stack (such as the application layer) and are responsible for extracting functional role labels (e.g., by looking up the node ID of the data dictionary or OPC UA) and initial urgency levels from the parsed data structures. The calculation of dynamic urgency levels can be performed by a separate semantic analysis module, which receives the raw data and initial urgency parsed from the protocol state machine and makes real-time adjustments based on the current system context information (such as control cycles and device fault warnings).

[0053] When the dynamic urgency level exceeds the threshold, the protocol communication state checkpoint for the data object is extracted from the current state of the protocol state machine instance, and this checkpoint is added as an extended field of the protocol semantic attributes. This step provides a proactive protection mechanism for high-urgency data. The protocol state machine instance can be designed with built-in checkpoint recording functionality. When an instruction indicating that the dynamic urgency level exceeds the threshold is received, the checkpoint module inside the state machine is activated, recording the sequence number, acknowledgment number, incomplete transaction ID, and related context data (such as sliding window information in TCP / IP, transaction ID and function code status in Modbus). This information is packaged into a protocol communication state checkpoint. Alternatively, the processor can run a separate protocol state monitoring module that continuously monitors the internal state changes of the protocol state machine instance. When the dynamic urgency level reaches the trigger condition, the monitoring module immediately sends a request to the protocol state machine instance to obtain its current key state variables (e.g., the currently processed message ID, the timeout timer waiting for a response, the session ID, and specific state flags of the application layer protocol). These key state variables are combined into protocol communication state checkpoints and attached to the protocol semantic properties of the data object.

[0054] The above technical solution enables dynamic and refined semantic parsing of data packets from heterogeneous industrial networks. By identifying protocol types and driving corresponding protocol state machine instances, accurate parsing of different protocol data is achieved, avoiding the limitations of general parsing methods. Dynamically extracting functional role labels and dynamic urgency levels allows for real-time perception of changes in the importance and urgency of data within the control loop, providing precise semantic input for subsequent scheduling priority correction. More importantly, when high-urgency data is detected, protocol communication state checkpoints are conditionally extracted and supplemented, providing crucial information for the sequential migration of the physical link carrying the data object's execution state, thus avoiding protocol session interruptions and data loss during link switching or interruption. This dynamic and forward-looking semantic parsing mechanism enhances the end-to-end guarantee capability of heterogeneous industrial networks for control services, especially under complex conditions such as mixed traffic surges or link quality fluctuations, ensuring that high-urgency control commands receive transmission resources matching their real-time requirements and guaranteeing communication continuity.

[0055] In some of the embodiments described above in this application, a driver protocol state machine instance is proposed to parse data packets to extract functional role labels and dynamic urgency levels, so as to accurately extract the semantic attributes of data objects in heterogeneous network converged communication. However, in its implementation process, if only static baseline values ​​are relied upon without considering the real-time dynamic changes of protocol states, the extracted attributes may not accurately reflect the actual urgency and functional role of the data, thereby affecting the accuracy of subsequent scheduling and migration decisions.

[0056] To address this, this application further proposes a method to drive the protocol state machine instance to parse the data packet and extract the functional role labels and dynamic urgency levels of the data objects within the data packet. This includes: querying a semantic annotation rule base based on the industrial communication protocol type followed by the data packet and the address range of the data objects in the data packet to obtain the baseline functional role labels and baseline urgency levels of the data objects; obtaining the historical state transition frequency of the protocol state machine instance before the current parsing cycle, where the historical state transition frequency is the ratio of the number of state transitions of the protocol state machine instance within a preset observation window to the duration of the preset observation window; correcting the baseline urgency level based on the historical state transition frequency to obtain the dynamic urgency level; and determining whether to change the baseline functional role label based on the historical state transition frequency to obtain the functional role label.

[0057] For example, when performing semantic parsing on data packets received from heterogeneous protocol ports, the semantic annotation rule base is queried based on the industrial communication protocol type followed by the data packet and the address range of the data objects in the data packet to obtain the baseline functional role label and baseline urgency level of the data objects. This semantic annotation rule base can be pre-configured; for example, it can be stored in a mapping table or database to store the data functional roles (such as control commands, status feedback, alarm information, etc.) and their default urgency levels corresponding to specific address ranges (such as register addresses, variable IDs, etc.) under different protocol types (such as Modbus, Profinet, EtherCAT, etc.) and their default urgency levels. Another implementation method is to load an XML or JSON configuration file during initialization. This file defines in detail the correspondence between protocol types, address ranges, and baseline functional role labels and baseline urgency levels. When parsing data packets, the processor performs a fast matching query based on the protocol identifier in the packet header and the address information in the data payload.

[0058] The historical state transition frequency of the protocol state machine instance prior to the current parsing cycle is obtained. The historical state transition frequency is an indicator of the protocol state machine's activity level and dynamic changes in behavior patterns. It is defined as the ratio of the number of state transitions within a preset observation window to the duration of the preset observation window. For example, the processor can maintain a circular buffer to record all state transition events and their timestamps that occurred within a past period (i.e., the preset observation window). In each parsing cycle, the processor traverses this buffer, counts the total number of state transitions, and divides it by the preset observation window duration (e.g., 1 second, 5 seconds, or 10 seconds) to calculate the historical state transition frequency. Alternatively, the protocol state machine instance can integrate a counter and timer module. The counter increments whenever a state transition occurs. The timer triggers periodically, dividing the counter value by the window duration at the end of each preset observation window to obtain the frequency, and then resetting the counter.

[0059] The baseline urgency level is adjusted based on the acquired historical state transition frequency to obtain the dynamic urgency level. The baseline functional role label is then determined based on the historical state transition frequency to obtain the functional role label. Various strategies can be employed to adjust the urgency level. For example, when the historical state transition frequency exceeds a certain preset threshold, it indicates abnormally active protocol behavior or a high-load state. In this case, the baseline urgency level can be linearly or non-linearly increased to reflect the higher urgency of the data object in the current environment. Specifically, a correction function f(frequency, base_urgency) = dynamic_urgency can be defined, which increases the dynamic urgency level to different degrees based on the frequency growth trend. For changing the functional role label, one or more frequency thresholds can be set. When the historical state transition frequency exceeds a specific threshold, it may mean that the protocol state machine has entered an abnormal or special operating mode. In this case, the baseline functional role label can be changed to a specific functional role label associated with this abnormal mode (e.g., from "regular data" to "abnormal alarm" or "diagnostic data") according to preset rules to more accurately reflect the actual function of the data object.

[0060] The above technical solution introduces a dynamic sensing and utilization mechanism for the historical state transition frequency of protocol state machine instances. This solves the problem that relying solely on a static semantic annotation rule base in heterogeneous network converged communication may lead to inaccurate reflection of real-time dynamic changes in the functional role labels and urgency levels of data objects. By quantifying the activity level and behavioral patterns of protocol states and using this as a basis to correct the baseline urgency level, it ensures that the urgency assessment of data objects is more closely aligned with the actual operating conditions in industrial settings. This avoids the underestimation of important data due to static low urgency during abnormal protocol behavior or high loads. Dynamically adjusting functional role labels based on historical state transition frequency enables the identification and response to abnormal or special patterns in protocol states, thereby more accurately identifying the actual functions of data objects and preventing static labels from becoming ineffective in dynamic environments. This dynamic and adaptive semantic parsing capability provides more accurate and real-time input for subsequent TSN gating time slot reallocation strategies and link migration execution strategies. It enhances the terminal's ability to guarantee high-urgency control commands during mixed traffic surges or link quality fluctuations, and avoids data loss caused by protocol session interruptions, thus enhancing the determinism of industrial control services.

[0061] In some of the embodiments described above in this application, a method is proposed to modify the baseline urgency level based on the historical state transition frequency in order to dynamically adjust the urgency level. However, in its implementation, since the historical state transition frequency is not compared with the steady-state transition frequency during the steady-state operation phase, the modification process lacks accurate quantification of abnormal acceleration of state transitions. This results in the urgency level adjustment not being sensitive enough to real-time network fluctuations and failing to accurately reflect the actual urgent needs of data objects, thereby affecting the accuracy of subsequent scheduling priority and link migration decisions.

[0062] To address this, this application further proposes a step to revise the baseline urgency level based on historical state transition frequencies to obtain a dynamic urgency level. This step includes: comparing the historical state transition frequencies with the steady-state transition frequencies of the protocol state machine instance during steady-state operation to obtain the degree of frequency deviation. When the frequency deviation indicates that the historical state transition frequency is higher than the steady-state transition frequency, an urgency increase is determined based on the frequency deviation, and the sum of the baseline urgency level and the urgency increase is used as the dynamic urgency level. When the frequency deviation indicates that the historical state transition frequency is not higher than the steady-state transition frequency, the baseline urgency level is used as the dynamic urgency level.

[0063] The historical state transition frequency refers to the ratio of the number of state transitions of a protocol state machine instance within a preset observation window to the duration of that window. It reflects the activity or volatility of the data object's communication behavior. This frequency can be obtained by continuously monitoring state change events of the protocol state machine instance, counting the number of state changes within a sliding time window, and then dividing by the window's duration. Another implementation method is to use a combination of event counters and timers. Each time a state transition occurs, the counter is incremented, and at the end of each fixed period, the counter value is divided by the period duration, and then the counter is reset. A protocol state machine instance is an independent state management logic unit created and running within the terminal for a specific industrial communication protocol followed by a specific data message or data object. It is responsible for tracking and managing the communication state of the data object at the protocol level, such as connection establishment, data transmission, and error handling. This instance can be a software object implemented through object-oriented programming, with one instance corresponding to each data object or communication session. Alternatively, it can be a hardware logic circuit designed based on finite state automata theory for efficient processing of high-speed industrial protocols. The steady-state transition frequency during steady-state operation refers to the average or typical state transition frequency exhibited by a protocol state machine instance under normal and stable operating conditions in an industrial field network. This frequency serves as a benchmark value to measure whether the current historical state transition frequency is abnormal. It can be obtained by long-term monitoring of operating data under fault-free, low-load conditions, statistically calculating the average state transition frequency as a preset value. Alternatively, it can be adaptively calculated and updated dynamically when no abnormal events are detected over a long period (e.g., hours or days). Frequency deviation is an indicator that quantifies the difference between the historical state transition frequency and the steady-state transition frequency, used to assess the degree of abnormality in the current communication state. This deviation can be calculated by dividing the absolute difference between the historical state transition frequency and the steady-state transition frequency by the steady-state transition frequency to obtain the relative deviation rate. Alternatively, statistical methods, such as calculating the standard deviation multiple between the two, can be used to more accurately reflect the significance of the deviation. The urgency increase is an incremental value dynamically calculated based on the frequency deviation, used to increase the baseline urgency level of the data object when an abnormal state transition is detected. This urgency level can be pre-configured using a lookup table, corresponding to different urgency increments based on varying frequency deviation ranges. Alternatively, a non-linear function, such as an exponential or piecewise linear function, can be used to map the frequency deviation to an incremental urgency increment, allowing for finer adjustments. The baseline urgency level is the initial or default urgency assigned to a data object under normal communication conditions, typically based on its functional importance within the industrial control system or a preset business priority. This level can be a fixed value predefined for a specific data object or data type within a semantic annotation rule base.Alternatively, the urgency level can be obtained from a pre-defined level mapping table based on the functional role labels of the data objects. The dynamic urgency level represents the urgency of a data object after real-time adjustments, comprehensively considering both the baseline importance of the data object and real-time fluctuations in the current communication status. This level is a crucial basis for subsequent scheduling priority and link migration decisions. It is calculated as the sum of the baseline urgency level and the urgency increase, or it can be achieved through a more complex weighted average or decision tree model that comprehensively evaluates the baseline urgency level, the urgency increase, and other relevant factors.

[0064] The above technical solution enables precise quantification of abnormal state transitions in protocol state machine instances, thereby achieving dynamic and sensitive adjustment of the urgency level of data objects. For example, comparing the historical state transition frequency with the steady-state transition frequency of the protocol state machine instance during steady-state operation yields the degree of frequency deviation. This step utilizes the steady-state frequency as an objective reference benchmark, avoiding the subjectivity and inaccuracy that may arise from relying solely on historical frequencies for correction, thus making the identification of abnormal accelerations in state transitions more accurate. When the frequency deviation indicates that the historical state transition frequency is higher than the steady-state transition frequency, the urgency level is increased based on the frequency deviation, and the sum of the benchmark urgency level and the urgency level increase is used as the dynamic urgency level. This allows the urgency level of relevant data objects to be increased promptly and reasonably when abnormal fluctuations occur in the industrial field (such as precursors to equipment failure or unstable control loops) leading to frequent protocol state machine transitions. This results in higher priority and faster response in subsequent TSN gating time slot reallocation and link migration decisions. Conversely, when the frequency deviation indicates that the historical state transition frequency is not higher than the steady-state transition frequency, the baseline urgency level is used as the dynamic urgency level. This avoids over-adjusting the urgency level under normal or low transition frequencies, ensuring the rational allocation of system resources and the stability of the evaluation. Overall, by introducing the steady-state transition frequency as a calibration benchmark, this application makes the evaluation of the dynamic urgency level more closely match the real-time operating status of the industrial site, improves the terminal's perception and response speed to high-urgency control commands, and thus optimizes the determinism and reliability of heterogeneous network converged communication.

[0065] In some of the embodiments described above in this application, a method is proposed to determine whether to change the baseline functional role label based on the frequency of historical state transitions, so as to dynamically adjust the functional role of data objects to reflect their importance in the control loop. However, in its implementation, if the label change is decided solely based on the frequency without considering whether the current state type of the protocol state machine instance is abnormal, it may lead to erroneous changes to the functional role label during normal high-frequency state transitions, resulting in system resource allocation errors or decreased stability, and affecting the reliability of industrial communication.

[0066] To address this, this application further proposes a step for determining whether to change the baseline functional role label based on the historical state transition frequency, thereby obtaining the functional role label. This step includes: when the historical state transition frequency is higher than a preset frequency threshold, obtaining the state type of the protocol state machine instance within the current parsing cycle; determining whether the state type belongs to the abnormal state type set; when the state type belongs to the abnormal state type set, changing the baseline functional role label to the functional role label associated with the abnormal state type set, and using it as the functional role label; when the state type does not belong to the abnormal state type set, keeping the baseline functional role label unchanged, and using it as the functional role label.

[0067] For example, when the frequency of historical state transitions exceeds a preset frequency threshold, the state type of the protocol state machine instance within the current parsing cycle is obtained. This step aims to set a trigger condition: only when the state transition activity of the protocol state machine instance is abnormally frequent will its specific state type be further checked. This avoids unnecessary complex judgments during normal or low-frequency state transitions, thereby optimizing system resource utilization. Obtaining the state type within the current parsing cycle is to obtain the real-time runtime context of the protocol state machine instance, providing a basis for subsequent anomaly judgment. For example, a counter can be used to continuously monitor the number of state transitions of the protocol state machine instance within a preset observation window and calculate the historical state transition frequency. When this frequency exceeds a preset frequency threshold (e.g., determined by empirical values, statistical analysis, or machine learning models), the system triggers a query for the current state of the protocol state machine instance. When the protocol state machine instance performs a parsing task, it maintains an internal state variable that indicates its current state (e.g., waiting for instructions, performing operations, error handling, etc.), which the system directly reads. Alternatively, this can also be implemented through an event-driven mechanism. Each time a protocol state machine instance transitions to a new state, it triggers an event. The system listens for these events and counts their occurrences within a sliding time window. Once the frequency of these events exceeds a frequency threshold, the system sends a query request to the protocol state machine instance to retrieve its current state identifier. This state identifier can be a predefined enumeration value or a string, representing a specific phase of the protocol's operation.

[0068] The core anomaly detection mechanism involves determining whether the current state type belongs to the set of abnormal state types. By comparing the currently acquired protocol state type with a predefined set of abnormal state types, it's possible to accurately determine whether the protocol state machine instance is in an unexpected or potentially unstable state. This provides crucial decision-making support for subsequent changes to functional role labels, ensuring the rationality and accuracy of such changes. For example, the set of abnormal state types can be pre-stored in a configuration table or database, containing all protocol state identifiers considered abnormal. Upon receiving the current state type of the protocol state machine instance, the system performs a set lookup operation to check if the state type exists in the set of abnormal state types. For the Modbus protocol, a state type such as "response timeout," "CRC error," or "illegal function code" might be defined as an abnormal state. Furthermore, the set of abnormal state types can be maintained through dynamic learning or expert systems. Protocol states related to system failures, performance degradation, or security events can be identified based on historical operational data and dynamically added to the set of abnormal state types. During the judgment, the system calls a state classification module, which, based on a rule engine or pattern matching algorithm, matches the current state type with patterns in the set to determine whether it belongs to an anomaly.

[0069] When a state type belongs to the abnormal state type set, the baseline functional role label is changed to the functional role label associated with the abnormal state type set, serving as the functional role label. When it is confirmed that the protocol state machine instance is in an abnormal state, this step immediately updates the baseline functional role label of the data object to the label reflecting the abnormal situation. This allows the data object to be given higher priority or specific processing strategies in subsequent transmission scheduling and link migration decisions to cope with abnormal situations and ensure the continuity and reliability of critical services. For example, each abnormal state type in the abnormal state type set can be pre-associated with a specific functional role label (e.g., "emergency control command", "fault diagnosis data", "alarm information", etc.). When it is determined that the current state type belongs to the abnormal set, the system directly obtains the corresponding functional role label from the association mapping and assigns it to the functional role label of the data object. If the abnormal state is "equipment failure", the functional role label may be changed to "fault alarm". Alternatively, it can be implemented through a rule-based decision engine. When an abnormal state type is detected, the decision engine dynamically generates or selects a new functional role label according to preset rules (e.g., IF state type = X AND protocol type = YTHEN functional role label = Z). These rules can be configured according to the specific business needs and safety levels of the industrial site to ensure that the most appropriate functional roles are assigned in different abnormal scenarios.

[0070] Conversely, when the state type does not belong to the set of abnormal state types, the baseline functional role label remains unchanged and is used as the functional role label. This step ensures that the functional role label of the data object is not incorrectly modified when the protocol state machine instance is in normal operation (even if the state transition frequency is high). This maintains the stability, predictability, and accuracy of resource allocation during normal operation, avoiding resource waste or unnecessary scheduling adjustments due to misjudgment. For example, this is a default operation. If the result of the previous step is "not belonging to the set of abnormal state types," the system directly skips the label change operation and continues to use the previously determined baseline functional role label as the functional role label of the data object. Alternatively, a "functional role label" variable can be maintained and assigned the baseline functional role label in the initial stage. When the state is determined to be non-abnormal, the value of this variable remains unchanged, i.e., the baseline functional role label is used. This ensures that the label is modified only when an anomaly is clearly detected; otherwise, the original, verified label is retained.

[0071] The above technical solution, by introducing a state type checking mechanism, solves the problem of misjudgment caused by ignoring state anomalies when dynamically changing functional role labels. It ensures that label changes only occur in truly abnormal scenarios, thereby improving the accuracy of functional role identification and the robustness of system decisions. Specifically, when the frequency of historical state transitions exceeds a preset frequency threshold, the state type of the protocol state machine instance within the current parsing cycle is obtained. This step uses a frequency threshold as a trigger condition, initiating the check only when state transitions are abnormally frequent, avoiding redundant processing during low-frequency or no-change periods, and reducing system overhead. Obtaining the current state type provides real-time basis for subsequent anomaly judgment, ensuring that decisions are based on the latest protocol state. Determining whether the state type belongs to the abnormal state type set involves matching based on a preset anomaly set. This provides clear anomaly criteria for change decisions, preventing subjective or random judgments and enhancing reliability. When the state type belongs to the abnormal state type set, the baseline functional role label is changed to the functional role label associated with the abnormal state type set. Changing the label in confirmed abnormal states allows the functional roles of data objects to promptly reflect their urgent needs, optimizing subsequent resource allocation. When the state type does not belong to the abnormal state type set, the baseline functional role label remains unchanged, and the original label is maintained in non-abnormal states, avoiding unnecessary changes during normal high-frequency operations and ensuring system consistency and stability. Furthermore, combined with the above-mentioned scheme of correcting the baseline urgency level and determining functional role labels based on the frequency of historical state transitions, this application further refines the logic for determining functional role labels. By introducing a judgment on the current state type of the protocol state machine instance when the frequency is abnormally high, and combining it with a preset abnormal state type set, it is possible to more accurately identify whether a data object is truly in an abnormal situation requiring special handling. This mechanism avoids misjudging anomalies solely based on high frequency, thus preventing unnecessary resource preemption or scheduling adjustments, and ensuring the accuracy and effectiveness of TSN gating list adjustments and link migration execution strategies. For example, in some industrial protocols, high-frequency state transitions may be normal working modes (such as rapid data acquisition) rather than abnormal ones. By distinguishing between normal high-frequency and abnormal high-frequency, this application makes the change of functional role labels more intelligent and accurate, thereby improving the end-to-end guarantee capability of the entire heterogeneous network converged communication terminal for industrial control services, especially its response capability and stability in complex, variable, or faulty scenarios.

[0072] In some of the solutions mentioned above in this application, the physical link carrying the data object is monitored for quality, and the link quality status is extracted to evaluate the physical link status and support subsequent link migration decisions. However, in this process, the monitoring process does not consider the semantic attributes of the data object, resulting in fixed monitoring parameters that cannot be adaptively adjusted. For high-urgency data, it may not be able to capture link quality degradation in time, thereby affecting the accuracy and response speed of link switching and causing the risk of protocol session interruption or data loss.

[0073] To address this, this application further proposes a method for quality monitoring of physical links carrying data objects and extracting link quality status. The method includes: obtaining the protocol semantic attributes of the data object after semantic parsing; determining a quality monitoring sensitivity parameter for the physical link based on the dynamic urgency level in the protocol semantic attributes; collecting physical layer transmission quality indicators for the physical link based on the quality monitoring sensitivity parameter; and generating the link quality status based on the physical layer transmission quality indicators.

[0074] For example, when acquiring the protocol semantic attributes determined after semantic parsing of a data object, these attributes represent the deep meaning and characteristics of the data object at the industrial communication protocol level, such as its functional role, urgency, and communication status. Acquiring these attributes allows subsequent link quality monitoring to perceive the business importance of the data and thus perform differentiated processing. One implementation is that after completing the semantic parsing of the data packet, the terminal processor stores the parsed protocol semantic attributes (such as functional role labels, dynamic urgency levels, and protocol communication status checkpoints) in memory and provides an interface for the quality monitoring module to call. Another implementation is that when parsing the data packet, the terminal processor transmits the extracted protocol semantic attributes to the link quality monitoring module in real time through an internal message queue or shared data structure, ensuring that the monitoring module can obtain the latest semantic information.

[0075] When determining the quality monitoring sensitivity parameter for a physical link based on the dynamic urgency level in the protocol's semantic attributes, the dynamic urgency level reflects the real-time importance of the data object in the current system state. The quality monitoring sensitivity parameter is a key indicator guiding physical link quality monitoring behavior; it determines the monitoring frequency, accuracy, or scope of indicators of interest. Adjusting the sensitivity parameter according to the dynamic urgency level allows for more rational allocation of monitoring resources, ensuring that high-urgency data receives more timely and detailed monitoring. One implementation method is to pre-define a mapping table or function to map different dynamic urgency levels to different quality monitoring sensitivity parameter values. For example, a higher urgency level corresponds to a larger sensitivity parameter value, potentially meaning a shorter sampling period, a stricter degradation threshold, or more comprehensive indicator collection. Another implementation method is to use an adaptive algorithm to dynamically calculate the optimal quality monitoring sensitivity parameter based on the dynamic urgency level and the current network load. For example, when the volume of high-urgency data is large, the sensitivity can be appropriately increased to ensure the quality of critical data transmission. When low-urgency data is predominant, the sensitivity can be appropriately decreased to conserve system resources.

[0076] When collecting physical layer transmission quality indicators (PMIs) of a physical link based on the quality monitoring sensitivity parameter, these PMIs are fundamental data for measuring physical link performance, such as bit error rate (BER), packet loss rate (PCR), latency, jitter, and signal strength. The quality monitoring sensitivity parameter guides the collection method for these indicators, ensuring that the collection process meets the urgency requirements of the data object. One implementation is that the monitoring module configures the physical layer interface chip or network adapter according to the determined quality monitoring sensitivity parameter, and collects data at a set sampling frequency, sampling precision, or specific indicator set. For example, high sensitivity might trigger BER statistics every millisecond, while low sensitivity might only trigger PCR statistics once per second. Another implementation is that the monitoring module obtains transmission quality statistics generated in real time by the physical layer hardware by querying the underlying driver or hardware registers. Here, the quality monitoring sensitivity parameter can guide the monitoring module to filter, aggregate, or selectively read these raw statistics to focus on the indicators most relevant to the current data urgency.

[0077] When generating the link quality status based on the physical layer transmission quality indicators, the link quality status is an abstract representation derived from a comprehensive evaluation of the collected physical layer transmission quality indicators. It reflects the current health status and future trends of the physical link. Generating the link quality status aims to provide a high-level, easily understandable evaluation result for subsequent scheduling and migration decisions. One implementation involves comparing multiple collected physical layer transmission quality indicators with preset health thresholds, and then, based on the number, severity, or duration of indicators exceeding the thresholds, using methods such as weighted averaging or fuzzy logic to generate a comprehensive link quality status that includes the current quality score and quality evolution trend. Another implementation involves using a machine learning model, taking historical physical layer transmission quality indicator data as input, training the model to predict the future quality trend of the link, and combining this prediction with current indicators to generate a link quality status containing predictive information.

[0078] By integrating protocol semantic attributes into the link quality monitoring process through the above technical solutions, the problems of fixed monitoring parameters and inability to adapt to changes in data urgency in traditional monitoring methods are solved. For example, by acquiring the protocol semantic attributes of data objects, the monitoring process can perceive the business importance of the data, thereby achieving differentiated monitoring. Determining the quality monitoring sensitivity parameters based on dynamic urgency levels allows for dynamic adjustment of monitoring frequency and accuracy according to the real-time urgency of data objects, ensuring that high-urgency data can be detected promptly and accurately when link degradation occurs, avoiding protocol session interruptions or data loss due to untimely monitoring. For low-urgency data, the monitoring sensitivity can be appropriately reduced, thereby optimizing system resource utilization. The link quality status is generated based on the collected physical layer transmission quality indicators, providing a more accurate and business-aware basis for subsequent gating time slot reallocation and link migration decisions, improving the end-to-end guarantee capability and communication reliability of heterogeneous networks in industrial fields for control services.

[0079] In some of the embodiments described above in this application, quality monitoring of the physical link carrying the data object is proposed to extract the link quality status. However, in its implementation, the quality monitoring may lack specificity and fail to effectively identify link degradation indicators that have a real impact on the transmission of the data object, resulting in wasted resources or ignoring key fluctuations, which affects the accuracy of subsequent link migration decisions.

[0080] To address this, this application further proposes a method for collecting physical layer transmission quality indicators (PMIs) of physical links based on quality monitoring sensitivity parameters, and generating link quality status based on these PMIs. This includes: determining the sampling period for the physical link based on quality monitoring sensitivity parameters, and collecting multiple PMIs of the physical link according to the sampling period; obtaining the functional role labels of the data objects determined after semantic parsing; determining the tolerance thresholds for fluctuations in various PMIs of the data objects based on the functional role labels; marking the indicators whose fluctuation amplitude exceeds the corresponding tolerance thresholds as effective degradation indicators; and generating link quality status based on the effective degradation indicators.

[0081] For example, in the above scheme, the sampling period for the physical link is determined based on the quality monitoring sensitivity parameter, and multiple physical layer transmission quality indicators of the physical link are collected according to the sampling period. The quality monitoring sensitivity parameter is determined based on the dynamic urgency level of the data object, and its role is to guide the physical link quality monitoring process, enabling it to adaptively adjust according to the importance or real-time requirements of the data object. For example, for data objects with a high dynamic urgency level, a higher quality monitoring sensitivity parameter can be set, thereby determining a shorter sampling period to achieve more frequent and granular link quality monitoring. Conversely, for data objects with a low dynamic urgency level, a lower quality monitoring sensitivity parameter can be set, thereby determining a longer sampling period to reduce monitoring overhead. The sampling period refers to the time interval for collecting physical layer transmission quality indicators of the physical link. It can be determined by consulting a preset mapping table, which maps different quality monitoring sensitivity parameter values ​​to corresponding sampling period lengths, or by using a dynamic calculation model to calculate the optimal sampling period in real time based on factors such as the quality monitoring sensitivity parameter and the current network load. Physical layer transmission quality metrics are various parameters that reflect the transmission performance of the physical link, such as bit error rate, packet loss rate, latency, jitter, signal strength, and signal-to-noise ratio. These metrics can be collected through the terminal's internal hardware interface (such as network card driver, PHY chip register) or through the network protocol stack (such as SNMP protocol).

[0082] Obtain the functional role label of the data object after semantic parsing. The functional role label is the specific function or category identifier of the data object in the industrial control system, such as "control command," "sensor data," "status feedback," or "alarm information." This label is determined during semantic parsing of data packets received from heterogeneous protocol ports, based on information such as the type of industrial communication protocol followed by the data packet, the address range of the data object in the data packet, and the historical state transition frequency of the protocol state machine instance. The functional role label provides important semantic context for subsequently determining the data object's tolerance to link quality fluctuations.

[0083] Based on functional role labels, the tolerance thresholds for data objects to fluctuations in various physical layer transmission quality indicators are determined. The tolerance threshold refers to the maximum acceptable range of fluctuations or degree of degradation in various physical layer transmission quality indicators during data object transmission. Data objects with different functional role labels have varying tolerances for link quality fluctuations. For example, control commands may have extremely low tolerance for latency and packet loss rates, while some non-real-time sensor data may have higher tolerances for these indicators. The tolerance thresholds can be determined through a pre-defined configuration table or rule base, which associates different functional role labels with the tolerance thresholds for various physical layer transmission quality indicators. Alternatively, machine learning models can be used, trained based on historical data and expert experience, to dynamically generate tolerance thresholds applicable to specific functional role labels.

[0084] Among multiple physical layer transmission quality indicators, those whose fluctuation amplitude exceeds the corresponding tolerance threshold are marked as valid degradation indicators. Fluctuation amplitude refers to the degree of deviation between the currently collected physical layer transmission quality indicator value and the indicator's baseline or normal range. By comparing the fluctuation amplitude with the tolerance threshold, degradation indicators that truly negatively impact data object transmission can be identified, thus filtering out "noise" indicators that, while exhibiting fluctuations, remain within acceptable limits. For example, if a control command has a latency tolerance threshold of 5ms, and the current link latency fluctuation reaches 10ms, then the latency is marked as a valid degradation indicator. If the latency fluctuation is only 2ms, it is not marked.

[0085] Link quality status is generated based on effective degradation indicators. Link quality status is a comprehensive assessment of the current transmission performance of a physical link. It is no longer a raw, unfiltered set of physical layer indicators, but focuses on degradation information that has a real impact on the transmission of specific data objects. Link quality status can include information such as the current quality score and quality evolution trends. For example, a comprehensive quality score can be calculated based on the number, severity, and duration of effective degradation indicators through weighted summation or fuzzy logic reasoning. By analyzing the changing trends of effective degradation indicators, the future direction of link quality can be predicted.

[0086] The above technical solutions address the lack of specificity in traditional quality monitoring. By determining the sensitivity parameters for quality monitoring based on the dynamic urgency level of data objects and adjusting the sampling cycle accordingly, the monitoring frequency can adaptively adjust based on the importance of the data. This avoids resource waste caused by over-monitoring low-urgency data while ensuring timely and detailed monitoring of high-urgency data. By acquiring the functional role tags of data objects and determining the tolerance thresholds for various physical layer transmission quality indicators based on these tags, fine-grained matching of different types of data transmission needs is achieved, avoiding misjudgments that may result from using uniform thresholds. By marking indicators with fluctuations exceeding the tolerance thresholds as effective degradation indicators and generating link quality status based on these indicators, the link quality status accurately reflects degradation that has a real impact on data object transmission, filtering out irrelevant fluctuations. This provides more accurate and reliable input for subsequent scheduling priority correction and link migration decisions, improving the end-to-end assurance capability of heterogeneous network converged communication terminals in industrial sites for control services.

[0087] In some of the embodiments described above in this application, a link quality status is generated based on effective degradation indicators to assess the degree of degradation of physical links. However, in its implementation, only identifying effective degradation indicators while ignoring their duration and changing trends may lead to incomplete link quality assessment and inability to accurately predict degradation evolution, thereby affecting the timeliness and accuracy of subsequent migration decisions.

[0088] To address this, this application further proposes a method for generating the link quality status based on effective degradation indicators, specifically including the following steps: obtaining the degradation duration and degradation change slope of the effective degradation indicator over multiple consecutive sampling periods; determining the degradation trend level of the effective degradation indicator based on the degradation duration and degradation change slope; and jointly evaluating the degradation trend level and the number of effective degradation indicators to generate the link quality status including the current quality score and quality evolution trend.

[0089] When obtaining the duration and slope of degradation of an effective degradation index over multiple consecutive sampling periods, the duration of degradation refers to the length of time a physical layer transmission quality index remains in a degraded state after being marked as an effective degradation index. Its purpose is to reflect the persistence of degradation phenomena and distinguish between instantaneous fluctuations and persistent issues. In practical implementation, a timer or timestamp can be maintained for each effective degradation index. The timer is started when the index first degrades and updated when it recovers or continues to degrade, thus calculating its duration over consecutive sampling periods. Another implementation method is to record the state of the effective degradation index in each sampling period and count the number of consecutive degradation occurrences by backtracking historical sampling records, then multiplying this by the sampling period duration to obtain the duration. The slope of degradation refers to the rate of change of the degree of degradation of the effective degradation index over consecutive sampling periods. Its purpose is to reveal the accelerating or decelerating characteristics of the degradation trend, helping to predict the direction of degradation evolution. In practice, this can be achieved by performing linear regression analysis on the values ​​of the effective degradation index over the most recent N sampling periods. The slope of the regression line is the degradation change slope, where N is a positive integer. Alternatively, a difference calculation can be used, which calculates the ratio of the difference in degradation values ​​between the current sampling period and the previous one or several sampling periods to the time interval, to approximate the change slope.

[0090] When determining the degradation trend level of an effective degradation indicator based on its duration and slope, this degradation trend level is a quantitative classification of the severity and evolution characteristics of a single effective degradation indicator. Its function is to abstract complex information about degradation duration and slope into easily understood and processed levels, thereby distinguishing different types of degradation trends such as short-term fluctuations, slow deterioration, and rapid deterioration. In implementation, a multi-dimensional lookup table can be pre-defined. The input to this table is the degradation duration and slope, and the output is the corresponding degradation trend level, such as slight degradation, moderate degradation, severe degradation, and critical degradation. Another implementation method is to define a series of piecewise functions or decision rules to dynamically calculate and assign the corresponding degradation trend level based on whether the degradation duration exceeds a specific threshold and the sign and magnitude of the degradation slope.

[0091] When jointly evaluating the degradation trend level and the number of effective degradation indicators to generate a link quality status that includes the current quality score and the quality evolution trend, this joint evaluation refers to comprehensively considering the degradation trend level of individual effective degradation indicators as well as the total number of all current effective degradation indicators to obtain a comprehensive judgment on the overall health status of the physical link. Its purpose is to avoid evaluation bias caused by focusing on only a single dimension and to more accurately reflect the overall degradation status of the link. In implementation, a weighted summation method can be used, assigning different weights to different levels of degradation trends and accumulating them together with the number of effective degradation indicators to obtain a comprehensive quality score. Alternatively, a rule-based expert system or decision tree model can be constructed, with the degradation trend level of each effective degradation indicator and the total number of effective degradation indicators as input, and the link quality status as output. The current quality score is a quantitative representation of the overall health status of the physical link at the current moment. Its purpose is to provide an immediate and comparable link quality indicator. For example, a full score can be set, and then corresponding scores can be deducted based on the joint evaluation results to obtain the current quality score. The quality evolution trend is a prediction of the direction of quality change of the physical link over a future period. Its role is to provide forward-looking information for subsequent link migration decisions. For example, if the degradation trend levels of multiple effective degradation indicators are generally high and their numbers continue to increase, the quality evolution trend may be judged as "rapid deterioration". Conversely, if the degradation trend level is low and the number is decreasing, it may be judged as "stabilizing" or "improving".

[0092] The above technical solutions enable a more comprehensive and dynamic assessment of physical link degradation. For example, by acquiring the duration and slope of degradation changes of effective degradation indicators, the persistence and rate of degradation can be analyzed in depth over time, avoiding the limitations of relying solely on single-point data or instantaneous states for judgment. This allows for a more accurate capture of the evolution of link degradation. Determining the degradation trend level based on the duration and slope of degradation changes quantifies the severity and acceleration characteristics of degradation, effectively distinguishing between short-term fluctuations and long-term deterioration, providing a more refined basis for subsequent decision-making. Jointly evaluating the degradation trend level and the number of effective degradation indicators comprehensively considers the synergistic impact of multiple degradation factors. This results in a link quality status that not only includes the current quality score but also provides a prediction of the future direction of quality change. This multi-dimensional link quality status improves the accuracy and foresight of link quality assessment, providing a more reliable and timely decision-making basis for subsequent link migration execution strategies, thereby effectively ensuring the deterministic transmission of industrial control services.

[0093] In some of the solutions mentioned above in this application, a scheduling priority correction amount is generated based on the coupling relationship between the protocol semantic attributes and queuing status in the data record to optimize the scheduling priority. However, in its implementation, since the latency change trend of the data object and the urgency of the associated device are not considered, it may lead to inaccurate prediction of timeout risk or insufficient scheduling coordination.

[0094] To address this, this application further proposes a method for generating scheduling priority correction quantities based on the coupling relationship between protocol semantic attributes and queuing states in the data record, see [link to relevant documentation]. Figure 3 The specific steps include: 301. Obtain the queuing delay sequence of the data object in multiple consecutive scheduling cycles, and determine the delay change trend of the data object based on the queuing delay sequence.

[0095] 302. Compare the trend of the delay change with the maximum tolerable delay of the data object in the data record to obtain the estimated timeout risk of the data object.

[0096] 303. Based on the estimated timeout risk, the dynamic urgency level of the data object in the data record, and the dynamic urgency level of other data objects in the data record that have the same associated device identifier as the data object, generate the scheduling priority correction amount.

[0097] In the above technical solution, obtaining the queuing delay sequence of the data object in multiple consecutive scheduling cycles refers to the system continuously monitoring and recording the length of time a specific data object waits for transmission in the transmission scheduling queue, forming a time-ordered set of delay data. For example, this can be achieved by recording the enqueue timestamp when the data object enters the queue and the dequeue timestamp when it leaves the queue; the difference between the two is the queuing delay within that scheduling cycle. Storing these delay values ​​in chronological order forms the queuing delay sequence. Another implementation method is to maintain a fixed-size circular buffer for each data object. After each scheduling cycle ends, the queuing delay value of the data object in the current cycle is stored in the buffer. When the buffer is full, the oldest data is overwritten, thus always maintaining the queuing delay records of the most recent consecutive scheduling cycles.

[0098] Determining the latency trend of a data object based on its queuing latency sequence refers to identifying the pattern or direction of queuing latency changes over time by analyzing historical queuing latency data. For example, linear regression analysis can be used to fit the queuing latency sequence, and the slope of the fitted line can be used to determine whether the latency is increasing, decreasing, or remaining stable. Alternatively, moving averages or exponential smoothing methods can be used to process the queuing latency sequence, and the short-term latency trend can be determined by comparing the current average latency with the average latency of the previous period.

[0099] Comparing the latency trend with the maximum tolerable latency of the data object in the data record yields the estimated timeout risk for that data object. This involves quantifying the probability or severity of a timeout based on the dynamic changes in the data object's latency and its acceptable maximum latency limit. For example, the queuing latency of a data object in a future scheduling cycle can be predicted based on the latency trend. This predicted value is then compared with the maximum tolerable latency. If the predicted value is close to or exceeds the maximum tolerable latency, a corresponding estimated timeout risk is generated. This value can be a percentage or a risk level. Alternatively, a risk function can be defined that comprehensively considers the current queuing latency, the latency trend (such as the latency growth rate), and the maximum tolerable latency, outputting a continuous risk score. A higher score indicates a greater timeout risk.

[0100] Based on the estimated timeout risk, the dynamic urgency level of the data object in the data record, and the dynamic urgency levels of other data objects with the same associated device identifier in the data record, the scheduling priority adjustment amount is generated. This means comprehensively considering the potential timeout risk of the data object itself, its own urgency level, and the urgency levels of other associated data objects to calculate a value used to adjust its scheduling priority. For example, a multi-dimensional lookup table can be designed. The input of this table includes the estimated timeout risk, the dynamic urgency level of the data object, and the dynamic urgency levels of other associated data objects, and the output is the corresponding scheduling priority adjustment amount. This lookup table can be pre-configured according to the actual needs and experience of the industrial site. Another implementation method is to use a weighted summation approach, assigning different weights to the three factors, and using their weighted sum as the scheduling priority adjustment amount. The weights can be dynamically adjusted according to the actual application scenario to reflect the degree of influence of different factors on the scheduling priority.

[0101] By incorporating the latency trend analysis of data objects, the above technical solution can more accurately predict the queuing latency of data objects in future scheduling cycles. Combined with their maximum tolerable latency, a timeout risk prediction is dynamically generated, thus avoiding biases caused by relying solely on the current queuing status and improving the predictability of scheduling decisions. When generating scheduling priority corrections, not only the dynamic urgency level and timeout risk prediction of the data object itself are considered, but also the dynamic urgency level of other data objects with the same associated device identifier are integrated. This allows the scheduler to perceive and respond to the overall urgency status of associated devices when adjusting priorities, achieving device-level collaborative scheduling optimization. This collaborative mechanism helps avoid situations where a single high-urgency data object obtains resources, causing delays in other critical data objects of its associated devices. Therefore, under mixed traffic surges or link quality fluctuations, it ensures that high-urgency control commands can obtain transmission resources matching their real-time requirements, guaranteeing the determinism of industrial control services and solving the problems of inaccurate timeout risk prediction and insufficient scheduling collaboration.

[0102] In some of the solutions described above in this application, a timeout risk prediction is generated by comparing the latency change trend with the maximum tolerable latency of the data object. This prediction is then used to generate the scheduling priority correction amount. However, in this process, a simple comparison between the trend and the tolerable latency may not accurately capture the dynamic characteristics of latency changes, such as nonlinear fluctuations in the latency growth rate or the real-time impact of the current queuing status. This results in inaccurate timeout risk predictions, which in turn affects the accuracy of scheduling priority correction and makes it impossible to respond to the transmission needs of high-urgency data objects in a timely manner.

[0103] To address this, this application further proposes comparing the latency change trend with the maximum tolerable latency of the data object in the data record to obtain a timeout risk estimate for the data object. Specifically, this includes: determining the latency change slope of the data object within the current scheduling period based on the latency change trend; calculating the remaining latency margin of the data object based on the maximum tolerable latency and the current queuing latency corresponding to the current scheduling period in the queuing latency sequence; and generating a timeout risk estimate of the corresponding magnitude when the latency growth rate indicated by the latency change slope will consume the remaining latency margin within a preset number of scheduling periods.

[0104] For example, when determining the slope of the delay change for a data object within the current scheduling cycle, this slope characterizes the rate at which the queuing delay of the data object changes over time. It can be determined in several ways. For instance, based on the queuing delay sequence of the data object across multiple consecutive scheduling cycles, a linear regression analysis method can be used to calculate the slope of the fitted straight line of the delay sequence as the delay change slope. Alternatively, the difference between the queuing delay of the current scheduling cycle and that of the previous one or several scheduling cycles can be calculated and divided by the corresponding time interval to obtain the instantaneous or average rate of change of the delay as the delay change slope. This slope can intuitively reflect whether the queuing delay is accelerating, decelerating, or remaining stable.

[0105] When calculating the remaining latency tolerance margin for a data object, this margin refers to the difference between the data object's maximum tolerable latency and its current queuing latency. The calculation method is typically: Remaining latency tolerance margin = Maximum tolerable latency - Current queuing latency. For example, if a data object's maximum tolerable latency is 100ms and its current queuing latency is 30ms, then its remaining latency tolerance margin is 70ms. This margin quantifies the length of time the data object can continue to wait without timeout.

[0106] When generating a timeout risk estimate of a corresponding magnitude, the preset scheduling period is a configurable parameter used to set a forward-looking prediction window. For example, it can be set to 5 scheduling periods in the future. When the latency growth rate of a data object is predicted by the latency change slope, and combined with the remaining tolerance latency margin, the system calculates that the queuing latency will exceed the maximum tolerable latency at a future point in time (e.g., within the preset scheduling period), and then generates the corresponding timeout risk estimate. This timeout risk estimate can be a discrete risk level (e.g., low, medium, high) or a continuous value, and its magnitude is proportional to the predicted urgency of the timeout. For example, if a timeout is predicted to occur within one scheduling period, a high-risk estimate is generated. If a timeout is predicted to occur within five scheduling periods, a medium-risk estimate is generated.

[0107] The above technical solution addresses the problem of inaccurate timeout risk prediction by introducing dynamic calculations of the latency change slope and the remaining tolerance latency margin, ensuring that scheduling priority adjustments better align with real-time transmission needs. Specifically, the latency change slope is determined based on the latency change trend, capturing the instantaneous rate of latency growth. This allows risk prediction to reflect dynamic changes rather than static averages, avoiding prediction deviations caused by ignoring acceleration or deceleration trends. The remaining tolerance latency margin is calculated based on the maximum tolerable latency and the current queuing latency. This combines the actual waiting time of the data object with the tolerance limit, quantifying the remaining buffer time and providing a more accurate margin benchmark, ensuring that the prediction is based on the latest state rather than historical data. When the growth rate indicated by the latency change slope will consume the remaining margin within a preset period, a corresponding magnitude of timeout risk prediction is generated. This predicts future risk thresholds through the correlation between the slope and the margin, generating quantitative values ​​rather than qualitative judgments, enabling the scheduling system to proactively identify and respond to potential timeout events. Overall, these features work together to improve the accuracy and timeliness of risk prediction, providing a reliable basis for adjusting scheduling priorities. This enables high-urgency control commands to obtain transmission resources that match their real-time requirements, effectively ensuring the determinism of industrial control operations.

[0108] In response, this application further proposes to generate a scheduling priority correction amount to optimize data transmission scheduling, in order to solve the problem that when generating the scheduling priority correction amount, the urgency status of other data objects with the same associated device identifier as the data object is ignored, which may cause the correction amount to fail to accurately reflect the overall scheduling needs of the device.

[0109] This application proposes a method for generating scheduling priority correction amounts, which includes: determining a basic correction component for a data object based on its dynamic urgency level and estimated timeout risk; obtaining the percentage of data objects with the same associated device identifier whose dynamic urgency level exceeds a preset threshold; determining a coordination enhancement coefficient based on this percentage; and using the product of the basic correction component and the coordination enhancement coefficient as the scheduling priority correction amount.

[0110] For example, when determining the basic correction component for a data object, it is necessary to obtain the dynamic urgency level and timeout risk estimate of the data object. The dynamic urgency level indicates the real-time importance and urgency of the data object in the current industrial control operation. It can be dynamically calculated based on factors such as the type of industrial communication protocol followed by the data message, the functional role label of the data object, and the historical state transition frequency of the protocol state machine instance. For example, the dynamic urgency level of a control command may be higher than that of a monitoring data. The timeout risk estimate assesses the degree of risk that the data object may exceed its maximum tolerable delay in the current transmission scheduling queue. This value can be predicted based on parameters such as the queuing delay sequence of the data object, the delay change trend, and the maximum tolerable delay. For example, when the delay growth rate of the data object indicates that it will exhaust its remaining tolerable delay in a short period of time, the timeout risk estimate will be increased accordingly. The basic correction component is an initial scheduling priority adjustment calculated for the characteristics of an individual data object (i.e., its dynamic urgency level and timeout risk estimate), aiming to reflect the urgency and transmission needs of the data object itself. In one implementation, the dynamic urgency level and the estimated timeout risk can be mapped to an initial base correction component using a lookup table or a pre-defined nonlinear function. For example, a two-dimensional lookup table can be defined, taking discrete intervals of the dynamic urgency level and the estimated timeout risk as input and outputting the corresponding base correction component. In another implementation, a weighted summation method can be used, assigning weights to the dynamic urgency level and the estimated timeout risk, and then summing the weighted values ​​to obtain the base correction component. For example, the base correction component = W1 * dynamic urgency level + W2 * estimated timeout risk, where W1 and W2 are pre-defined weights.

[0111] When obtaining the percentage of data objects with the same associated device identifier whose dynamic urgency level exceeds a preset threshold, the "same associated device identifier" refers to the fact that in an industrial setting, multiple data objects may originate from or serve the same physical device (e.g., a PLC, a sensor group, or an actuator). These data objects share a unique device identifier, indicating that they are logically related. The preset threshold is a critical value used to determine whether a data object is in a "high urgency" state. Only when the dynamic urgency level of a data object exceeds this threshold is it considered urgent data requiring special attention. The percentage of data objects refers to the ratio of the number of data objects with the same associated device identifier whose dynamic urgency level exceeds the preset threshold to the total number of all related data objects of that device. This percentage reflects the overall urgency level of the device. In one implementation, the terminal processor can maintain a device-data object mapping table, periodically traversing all data objects under the device associated with the current data object, calculating their dynamic urgency levels and comparing them with the preset threshold to calculate the percentage. In another implementation, an event-driven mechanism can be used to trigger a counter update when the dynamic urgency level of a data object changes and exceeds a preset threshold, and to calculate the proportion of high-urgency data objects on that device in real time.

[0112] When determining the synergy enhancement coefficient based on the proportion of data objects, the synergy enhancement coefficient is a multiplicative factor used to amplify or reduce the base correction component to reflect the overall urgency of the devices associated with the current data object. When the overall urgency of the devices is high (i.e., the proportion of high-urgency data objects is high), the coefficient increases, thereby further enhancing the scheduling priority of the current data object. In one implementation, the proportion of data objects can be mapped to the synergy enhancement coefficient using a preset function (e.g., a linear function, an exponential function, or a piecewise function). For example, when the proportion is below a certain value, the coefficient is 1. When the proportion is above a certain value, the coefficient increases linearly or exponentially with the proportion. In another implementation, the synergy enhancement coefficient can be dynamically adjusted based on the changing trend of the proportion of data objects over multiple consecutive evaluation periods. For example, when the proportion continues to rise, the coefficient is increased according to a first increasing rate. When the proportion continues to fall, the coefficient is decreased according to a second decreasing rate.

[0113] The product of the base correction component and the coordination enhancement coefficient is used as the scheduling priority correction amount. The scheduling priority correction amount is a quantified value used to adjust the scheduling priority of a data object in the TSN gating list. This value comprehensively considers the urgency of the data object itself, the timeout risk, and the overall urgency of the device to which it belongs.

[0114] The above technical solution enables a more comprehensive and accurate assessment of the actual scheduling priority requirements of data objects. It not only focuses on the urgency of individual data objects but also considers the overall operational status of the equipment to which they belong, avoiding the problem of neglecting global emergencies due to local optimization. For example, a basic correction component is determined based on the dynamic urgency level and timeout risk estimate of the data object. This allows the scheduling priority correction amount to directly reflect the urgency and transmission risk of the individual data object, providing a solid foundation for subsequent collaborative enhancement. By obtaining the proportion of data objects with the same associated device identifier whose dynamic urgency level exceeds a preset threshold, this application can quantitatively assess the overall urgency status of the equipment related to the current data object. When a large number of high-urgency data objects exist in a device, it indicates that the device may be in an abnormal or critical operating state, requiring higher priority resource allocation. A collaborative enhancement coefficient is determined based on this proportion of data objects and multiplied by the basic correction component to obtain the scheduling priority correction amount. This multiplication method allows the collaborative enhancement coefficient to amplify the basic correction component when the overall urgency of the equipment is high, thereby increasing the scheduling priority of the current data object. This enables high-urgency control commands to obtain transmission resources that better match their real-time requirements when faced with mixed traffic surges or link quality fluctuations, thereby effectively improving the end-to-end guarantee capability of heterogeneous networks in industrial sites for control services and ensuring the determinism and reliability of industrial control services.

[0115] In some of the solutions mentioned above in this application, a basic correction component is proposed to be determined based on the estimated timeout risk and dynamic urgency level to generate the scheduling priority correction amount. However, in this process, the functional role label of the data object is not considered, which may cause the basic correction component to fail to accurately reflect the specific functional requirements of the data object, thereby affecting the accuracy and adaptability of scheduling optimization.

[0116] To address this, this application further proposes a method for determining the basic correction component of a data object based on its dynamic urgency level and estimated timeout risk. Specifically, this method includes: performing a non-linear mapping between the dynamic urgency level and the estimated timeout risk of the data object to obtain a joint risk factor; obtaining the functional role labels of the data objects in the data records; determining the upper limit of the correction coefficient based on the functional role labels; and using the product of the joint risk factor and the upper limit of the correction coefficient as the basic correction component.

[0117] The dynamic urgency level reflects the real-time importance or urgency of the data object in the current industrial control operations, and its value changes dynamically based on factors such as system status, business processes, or external events. The timeout risk estimate assesses the probability or severity of a data object experiencing a timeout (i.e., exceeding the maximum tolerable delay) in the current transmission scheduling queue. This value is typically calculated based on the data object's queuing delay sequence, delay variation trend, and maximum tolerable delay. A nonlinear mapping is applied between the dynamic urgency level and the timeout risk estimate to transform these two input variables into a single output variable—a joint risk factor—through a nonlinear functional relationship. This nonlinear mapping captures the complex and non-simple interactions between the input variables. For example, when the urgency or risk reaches a certain critical point, its impact on the joint risk factor may increase dramatically, allowing the joint risk factor to more accurately reflect the current urgency and timeout risk of the data object. For instance, a piecewise function or lookup table can be used for the nonlinear mapping. A two-dimensional lookup table is predefined, with discrete intervals of the dynamic urgency level and the timeout risk estimate as inputs, and the corresponding joint risk factor as the output. Alternatively, activation functions such as the Sigmoid function, ReLU function, and Tanh function, or polynomial functions, can be used for nonlinear mapping. The dynamic urgency level and the estimated timeout risk are used as inputs to the neural network, and a joint risk factor is output through one or more hidden layers after nonlinear transformation.

[0118] These functional role tags identify the specific functional role or business attribute of a data object within an industrial control system. Examples include "safety instructions," "real-time control instructions," "status monitoring data," or "alarm information." These tags reflect the semantic importance and business priority of the data object within the entire control loop. The functional role tag of a data object in a data record can be obtained by querying a pre-defined semantic annotation rule base based on information such as the protocol type and address range of the data message during semantic parsing. Alternatively, machine learning or pattern recognition technologies can be used to analyze and train on historical data messages to automatically identify the functional role characteristics of different data objects and generate corresponding functional role tags.

[0119] Determining the upper limit of the correction coefficient based on the functional role label is to limit or standardize the range of values ​​for the basic correction component. This upper limit ensures that when correcting scheduling priorities, excessive correction due to high urgency or risk assessment will not lead to over-correction, thereby affecting the normal transmission of other important data objects or causing an imbalance in system resource allocation. Different functional role labels correspond to different business importance and resource requirements; therefore, their upper limit correction coefficients should also differ. For example, a mapping table between functional role labels and upper limit correction coefficients can be maintained. When the functional role label of a data object is obtained, the corresponding upper limit correction coefficient can be obtained by directly querying this mapping table. Alternatively, the upper limit correction coefficient can be dynamically calculated based on the type of functional role label. For example, the upper limit correction coefficient can be determined through weighted summation or a multi-factor evaluation model, considering multiple dimensions such as the business level to which the functional role label belongs, real-time requirements, and impact on system stability.

[0120] The product of the joint risk factor and the upper limit of the correction coefficient is used as the base correction component. This calculation method combines the real-time risk of a data object (reflected by the joint risk factor) with the maximum correction range allowed by the functional role (reflected by the upper limit of the correction coefficient). This means that even if a data object has a high joint risk factor, its base correction component will not exceed the upper limit that its functional role can bear, thus avoiding over-scheduling or resource bias. The joint risk factor allows the correction component to dynamically reflect the current urgency and timeout risk.

[0121] The above technical solutions enable more accurate determination of the basic correction components of data objects, thus solving the problem of ignoring functional context in scheduling priority correction. For example, a nonlinear mapping is performed between the dynamic urgency level of a data object and the estimated timeout risk to obtain a joint risk factor. This captures the complex nonlinear relationship between urgency and risk, avoiding deviations caused by simple linear calculations and ensuring that the risk factor better reflects the dynamic changes in actual industrial control scenarios. By obtaining the functional role labels of data objects in data records, the functional role information of data objects in the control closed loop is introduced, providing additional semantic context and compensating for the shortcomings of relying solely on urgency and risk. The upper limit of the correction coefficient is determined based on the functional role labels, and reasonable upper limit constraints are set according to the functional roles to prevent the correction component from exceeding the functional requirements, ensuring the rationality and controllability of the correction. The product of the joint risk factor and the upper limit of the correction coefficient is used as the basic correction component. Combined with risk factors and functional constraints, a correction value that reflects both real-time risk and functional characteristics is generated. This allows the generation of scheduling priority corrections to consider not only the real-time urgency and timeout risk of data objects, but also their inherent functional importance within the industrial control system, thereby improving the accuracy and adaptability of scheduling priority corrections. This more refined and semantically aware scheduling priority correction enables high-urgency control commands to obtain transmission resources better suited to their real-time requirements during mixed traffic surges or link quality fluctuations in heterogeneous industrial networks. This optimizes resource allocation and ultimately enhances the end-to-end guarantee capability of heterogeneous industrial networks for control services.

[0122] In some of the embodiments described above in this application, a collaborative enhancement coefficient is proposed to be determined based on the proportion of data objects to adjust the scheduling priority correction amount. However, in its implementation, the determination of the collaborative enhancement coefficient fails to fully consider the current trend of the proportion sequence, which may result in the coefficient adjustment rate not being able to match the actual needs in a timely manner when the proportion of data objects continues to rise or fall, thus affecting scheduling efficiency.

[0123] To address this, this application further proposes a method for determining a synergy enhancement coefficient based on the proportion of data objects, comprising: obtaining a sequence of the proportion of data objects over multiple consecutive evaluation periods; determining the current trend of the proportion of data objects based on the sequence; increasing the synergy enhancement coefficient by a first growth rate when the current trend indicates a continuous increase in the proportion of data objects; and decreasing the synergy enhancement coefficient by a second deceleration rate, wherein the second deceleration rate is less than the first growth rate, when the current trend indicates a continuous decrease in the proportion of data objects.

[0124] For example, to obtain the percentage sequence of the number of data objects over multiple consecutive evaluation periods, either or a combination of the following two methods can be used: One method is for the system to periodically count the number of data objects in the current transmission scheduling queue whose dynamic urgency level exceeds a preset threshold, calculate their proportion of the total number of data objects, and store this proportion in a sliding window buffer. This buffer maintains the proportion values ​​for the most recent N evaluation periods, thus forming a percentage sequence. Another method is to record the calculated percentage value of the number of data objects in each evaluation period in a time-series database. When trend analysis is needed, the database is queried and the proportion values ​​for a specified consecutive evaluation period are extracted to construct the percentage sequence. Obtaining the percentage sequence provides historical data support for subsequent analysis of the dynamic changes in the percentage of the number of data objects.

[0125] To determine the current trend of the proportion of data objects based on this proportion sequence, various methods can be used. For example, linear regression analysis can be performed on the proportion sequence, and the trend can be determined by calculating the slope of the regression line: a positive slope indicates an upward trend, a negative slope indicates a downward trend, and a slope close to zero indicates a stable trend. Alternatively, a moving average comparison method can be used, comparing the proportion of data objects in the current evaluation period with the moving average of the previous N evaluation periods. If the current value is higher than the moving average, it is determined to be an upward trend; if it is lower, it is determined to be a downward trend. Furthermore, the average values ​​of the initial and final stages in the proportion sequence can be compared. If the average value of the final stage is higher than that of the initial stage, it is determined to be an upward trend; conversely, it is determined to be a downward trend. These methods can identify the dynamic direction of change in the proportion of data objects.

[0126] When the current trend indicates a continuous increase in the proportion of data objects, this application increases the synergy enhancement coefficient according to a first incremental rate to quickly respond to the increase in system load. For example, a fixed increment method can be used, adding a preset fixed value (i.e., the first incremental rate) to the current synergy enhancement coefficient in each adjustment cycle. Alternatively, a proportional increment method can be used, increasing the synergy enhancement coefficient by a certain percentage based on the first incremental rate. Furthermore, different first incremental rates can be dynamically selected based on the steepness of the upward trend; for example, when the upward trend is very obvious, a larger incremental rate is used to enhance the synergy effect more quickly. This rapid increase mechanism enables a rapid increase in the scheduling priority of high-urgency data objects when they surge, ensuring the real-time performance of critical business operations.

[0127] When the current trend indicates a continuous decline in the proportion of the data object's quantity, this application reduces the coordination enhancement coefficient according to a second deceleration rate to avoid over-adjustment and resource waste. For example, a fixed reduction method can be used, subtracting a preset fixed value (i.e., the second deceleration rate) from the current coordination enhancement coefficient in each adjustment cycle. Alternatively, a proportional reduction method can be used, reducing the coordination enhancement coefficient by a certain percentage based on the second deceleration rate. During the reduction process, a minimum threshold can be set to ensure that the coordination enhancement coefficient does not fall below this threshold, thus maintaining basic coordination scheduling capabilities. This reduction mechanism allows priority resources to be gradually released when the load decreases, avoiding system oscillations caused by overreaction.

[0128] It is worth noting that the second deceleration rate is smaller than the first acceleration rate. This design is one of the key innovations of this application. Its purpose is to enable a rapid response and a quick increase in the collaboration enhancement coefficient when the proportion of high-urgency data objects rises, in order to cope with emergencies and ensure that critical data is processed in a timely manner. Conversely, when the proportion of high-urgency data objects decreases, the system reduces the collaboration enhancement coefficient at a relatively slower rate. This helps avoid prematurely lowering priorities due to short-term fluctuations, thereby maintaining system stability and reserving a buffer time for possible recurrence of upward trends, preventing frequent oscillations in critical states.

[0129] The above technical solution intelligently adjusts the coordination enhancement coefficient based on the dynamic trend of the proportion of data objects. When the proportion of high-urgency data objects continues to rise, the coordination enhancement coefficient can be rapidly increased. This provides a stronger coordination priority boost for data objects with the same associated device identifier when generating scheduling priority corrections, ensuring that critical control commands still receive sufficient transmission resources under mixed traffic surges. Conversely, when the proportion of high-urgency data objects continues to decline, the system can gradually reduce the coordination enhancement coefficient, avoiding over-allocation of resources, thereby optimizing overall scheduling efficiency and reducing unnecessary priority preemption, allowing other non-urgent data to also obtain reasonable transmission opportunities. This dynamic, asymmetric adjustment mechanism solves the problem of mismatched coefficient adjustment rates in traditional methods, improving the end-to-end guarantee capability of heterogeneous network converged communication terminals in industrial fields for control services under complex operating conditions. Especially when dealing with sudden high-urgency data traffic, it can provide more stable and efficient deterministic transmission services.

[0130] In some of the solutions mentioned above in this application, a migration urgency assessment value is generated based on data records to determine the link migration execution strategy. However, in this process, if only the link quality status is considered and protocol semantic attributes such as the dynamic urgency level of data objects are ignored, the migration urgency assessment may be inaccurate, and high-urgency control commands may not be effectively prioritized. This may result in protocol session interruption or data loss during link switching, affecting the determinism of industrial control services.

[0131] To address this, this application proposes a method for generating migration urgency assessment values ​​based on the coupling relationship between protocol semantic attributes and link quality status in data records. (See [link to relevant documentation]). Figure 4 The method includes: 401. Obtain the current quality score and quality evolution trend of the physical link from the link quality status recorded in the data.

[0132] 402. Correlate the current quality score with the quality evolution trend to determine the urgency of the basic migration of physical links.

[0133] 403. Based on the dynamic urgency level of data objects in the protocol semantic attributes of data records, the basic migration urgency is modified to generate a migration urgency assessment value.

[0134] For example, extracting the current quality score and quality evolution trend of a physical link from the link quality status data records aims to extract key indicators about the physical link performance from the stored data records. The current quality score reflects the instantaneous performance status of the physical link at a given moment, such as a comprehensive evaluation value including bit error rate, packet loss rate, and latency. The quality evolution trend describes the dynamic direction of these performance indicators over time, such as whether they continue to deteriorate, improve, or remain stable. Obtaining this information is the basis for link migration decisions, providing a comprehensive and dynamic view of the link's health status. For example, physical layer transmission quality indicators (such as signal-to-noise ratio, signal strength, and retransmission count) can be collected periodically from network interface cards or dedicated hardware sensors, and the current quality score can be generated after weighted averaging or processing by machine learning models. The quality evolution trend can be obtained by performing linear regression analysis or exponential smoothing on the quality scores over multiple consecutive time windows. Alternatively, the current quality score can be calculated by monitoring the performance counters of network devices (such as switches and routers) to obtain data such as port error rate, bandwidth utilization, and jitter, and combining this data with preset thresholds and rules. Quality evolution trends can be predicted by comparing the differences between current and historical scores, and by combining time series analysis methods (such as moving averages and ARIMA models) to predict the future direction and rate of change.

[0135] The basic migration urgency of a physical link is determined by correlating the current quality score with its evolution trend. This step aims to quantify the initial urgency of migrating the link by comprehensively considering both its immediate performance and future trends. Relying solely on the current quality score may not accurately predict the future condition of the link, while incorporating the evolution trend provides a more comprehensive risk assessment, preventing responses only when the link is about to deteriorate and improving the predictability of migration decisions. The basic migration urgency is a preliminary assessment that does not consider data semantics. For example, a function can be designed that takes the current quality score as the primary input and the quality evolution trend as a correction factor. The basic migration urgency increases when the current quality score is low (poor performance) and the quality evolution trend indicates continued deterioration. Conversely, the basic migration urgency also increases when the current quality score is acceptable but the evolution trend shows rapid deterioration. This function can be a linear weighted sum or a non-linear mapping, such as through a lookup table or piecewise function. Alternatively, a state machine model can be constructed to classify the current quality rating into multiple levels such as "Excellent", "Good", "Medium", and "Poor", and the quality evolution trend into "Stable", "Slowly Deteriorating", "Rapidly Deteriorating", and "Improving". Based on the combination of the current quality rating level and the quality evolution trend, a predefined basic migration urgency level (such as "Low", "Medium", "High", "Very High") can be directly mapped.

[0136] Based on the dynamic urgency level of data objects in the protocol semantic attributes of data records, the basic migration urgency is adjusted to generate a migration urgency assessment value. This step aims to incorporate the business importance of the data packet itself (i.e., the dynamic urgency level) into the link migration decision. The basic migration urgency only reflects the physical condition of the link, while the dynamic urgency level reflects the criticality of the data object in the industrial control closed loop. By adjusting the basic migration urgency, it can be ensured that high-urgency data objects receive higher migration priority when link quality deteriorates, thereby guaranteeing the deterministic transmission of critical services. For example, a correction function can be defined, which takes the basic migration urgency and dynamic urgency level as inputs. When the dynamic urgency level is high, the correction function will amplify the basic migration urgency with a larger weight or multiplication factor, thus increasing its migration urgency assessment value. Conversely, for data objects with low urgency, the correction effect may be small or even non-existent. This correction can be linear or non-linear, such as amplifying or reducing using exponential or logarithmic functions. Alternatively, a tiered correction strategy can be adopted, selecting different correction coefficients or correction curves according to different dynamic urgency levels. For example, for data objects with "extremely high" urgency, even if the underlying migration urgency is not high, a large correction factor might be used to elevate them to a higher level. For data objects with "normal" urgency, only minor corrections might be made, or corrections might only be applied when the underlying migration urgency reaches a certain threshold. The corrected result is the migration urgency assessment value.

[0137] By correlating the current quality score and quality evolution trend of the physical link with the above technical solution, a comprehensive assessment of the link's current status and future trends is conducted to determine a preliminary basic migration urgency. Furthermore, a dynamic urgency level for data objects is introduced to correct this basic migration urgency. This correction mechanism ensures that the generated migration urgency assessment not only reflects the objective performance of the physical link but also incorporates the criticality and real-time requirements of the data objects in industrial control operations. Therefore, when link quality deteriorates, high-urgency control commands receive higher migration priority, prompting the system to more promptly and proactively find and switch to a more reliable physical link. This avoids critical data transmission interruptions or delays caused by link degradation, ensuring the determinism of industrial control operations, preventing data loss due to protocol session interruptions, and enhancing the end-to-end guarantee capability of heterogeneous network converged communication terminals for control operations.

[0138] In some of the solutions mentioned above in this application, the current quality score is correlated with the quality evolution trend to determine the basic migration urgency of the physical link, so as to assess the urgency of migration when the link quality deteriorates. However, in its implementation, relying solely on the simple correlation between the score and the trend may lead to an insufficiently refined assessment, failing to quantify the acceleration factor of degradation and the remaining quality margin, thereby causing a bias in the judgment of migration urgency. For example, it may fail to capture risks in time when the link quality deteriorates rapidly, or over-trigger migration when there is sufficient remaining buffer time, affecting the accuracy and efficiency of link switching.

[0139] To address this, this application further proposes a method to correlate the current quality score with the quality evolution trend to determine the fundamental migration urgency of the physical link. This includes: determining a quality degradation acceleration factor for the physical link based on the quality evolution trend; comparing the current quality score with a quality degradation threshold to obtain the remaining quality margin of the physical link; and determining the fundamental migration urgency of the physical link based on the ratio of the quality degradation acceleration factor to the remaining quality margin.

[0140] For example, when determining the quality degradation acceleration factor of a physical link, this factor characterizes the rate of change of the physical link's quality degradation speed, i.e., the "trend" of the degradation trend. Its role is to capture the dynamics of link quality deterioration in order to predict potential link failures or sharp performance degradation earlier. One implementation is to perform a quadratic difference calculation on the quality evolution trend (i.e., the rate of quality change) collected at multiple consecutive time points, and divide by the corresponding time interval to obtain the quality degradation acceleration factor. For example, the rate of quality change over the most recent three or more sampling periods can be obtained, and then the acceleration can be estimated using quadratic or higher-order difference. Another implementation is to collect physical link quality data over a period of time, perform curve fitting on this data (e.g., using polynomial or exponential fitting), and then calculate the second derivative of the fitted curve to obtain the quality degradation acceleration factor. This method can effectively smooth data noise and provide a more stable acceleration estimate.

[0141] When determining the remaining quality margin of a physical link, this margin refers to the difference between the current physical link's quality score and a preset quality degradation threshold. The quality degradation threshold is typically set as the critical point where link performance has severely degraded or is about to become unusable. The role of the remaining quality margin is to quantify the degree of degradation the link can withstand before reaching an unusable state, providing a buffer time or degradation space for migration decisions. One implementation is to directly subtract the preset quality degradation threshold from the current quality score; the difference is the remaining quality margin. For example, if the quality score range is 0-100, the degradation threshold is 20, and the current score is 60, then the remaining margin is 40. Another implementation is to divide the difference between the current quality score and the quality degradation threshold by the total quality score range or the degradation threshold itself, expressing the remaining margin as a percentage. This facilitates standardized comparisons across different quality scoring systems.

[0142] The basic migration urgency of a physical link is determined by the ratio of the quality degradation acceleration factor to the remaining quality margin. Basic migration urgency is a comprehensive indicator that combines the rate of link quality deterioration with the remaining degradation tolerance of the link, thus quantifying the urgency of migrating the current physical link. Its purpose is to provide a more comprehensive and dynamic basis for migration decisions, avoiding the bias of judging solely based on current quality or degradation trends, thereby improving the accuracy and timeliness of migration decisions. One implementation is to directly divide the quality degradation acceleration factor by the remaining quality margin to obtain the basic migration urgency. To avoid abnormal results due to a zero or excessively small divisor, the remaining quality margin can be appropriately smoothed or a minimum value can be set. Another implementation is to predefine a lookup table or rule set, directly mapping different combinations of the quality degradation acceleration factor and the remaining quality margin to different basic migration urgency levels or values. This method allows for more flexible definition of nonlinear relationships to adapt to different industrial application scenarios.

[0143] The above technical solution addresses the problem of inaccurate assessment by introducing a quality degradation acceleration factor and remaining quality margin, and dynamically quantifying the basic migration urgency using their ratio. Specifically, the quality degradation acceleration factor is determined based on the quality evolution trend, capturing the rate of link quality deterioration and enabling earlier prediction of future risks. The remaining quality margin is obtained by comparing the current quality score with the quality degradation threshold, reflecting the link's tolerable degradation margin and providing a migration buffer time. The basic migration urgency is determined based on the ratio of the quality degradation acceleration factor to the remaining quality margin. This ratio integrates degradation dynamics and tolerance; a high ratio accelerates migration decisions to cope with rapid degradation, while a low ratio delays migration to avoid resource waste. This ensures that the migration urgency assessment more closely reflects the actual link status, enabling more accurate assessment of physical link migration urgency during mixed traffic surges or link quality fluctuations. This avoids migration decision biases caused by inaccurate assessments, improves the perception and response capabilities of heterogeneous network converged communication terminals in industrial settings to changes in link status, and ensures the determinism of industrial control services.

[0144] In some of the embodiments described above in this application, a migration urgency assessment value is generated based on the coupling relationship between protocol semantic attributes and link quality status to determine the link migration execution strategy. However, in its implementation, the correction of the basic migration urgency lacks comprehensive consideration of the functional role labels of data objects and dynamic urgency levels, resulting in the assessment value not accurately reflecting the actual migration needs and affecting the timeliness and effectiveness of link migration.

[0145] To address this, this application further proposes a method to modify the basic migration urgency based on the dynamic urgency level of data objects in the protocol semantic attributes of data records, thereby generating a migration urgency assessment value. This modification process includes: obtaining the functional role label of the data object from the protocol semantic attributes of the data record; determining the modification strategy type based on the dynamic urgency level and the functional role label; and modifying the basic migration urgency based on the modification strategy type to generate a migration urgency assessment value.

[0146] For example, when retrieving the functional role label of a data object from the protocol semantic attributes of a data record, this functional role label characterizes the specific role or importance of the data object in the industrial control closed loop, such as sensor data, control commands, alarm information, or configuration parameters. This functional role label can be obtained in several ways. For instance, during the data packet semantic parsing stage, its functional role label can be identified and extracted based on a preset semantic annotation rule base and the address range of the data object in the data packet. This label can be further dynamically adjusted based on the historical state transition frequency of the protocol state machine instance to reflect its role changes in a specific operating state. Alternatively, its predefined functional role in the industrial control system can be obtained by querying the device configuration file or business process definition associated with the data object.

[0147] When determining the correction strategy type based on dynamic urgency level and functional role label, the aim is to comprehensively consider the real-time nature and functional importance of the data object to select the most appropriate correction method. One implementation is to construct a multidimensional lookup table or decision tree, whose input is the dynamic urgency level and functional role label, and whose output is a preset correction strategy type identifier. For example, when the dynamic urgency level is high and the functional role label is "control command," the strategy type might be determined as "accelerated correction." When the dynamic urgency level is low and the functional role label is "log information," the strategy type might be determined as "gradual correction." Another implementation is to calculate a comprehensive strategy judgment value, which combines the weighting factor of the dynamic urgency level and the sensitivity factor of the functional role label, and then compare this judgment value with a series of preset thresholds to determine the specific correction strategy type.

[0148] When adjusting the basic migration urgency based on the adjustment strategy type to generate a migration urgency assessment value, this step applies the selected adjustment strategy to the initially calculated basic migration urgency to generate the assessment value. For example, depending on the determined adjustment strategy type, a corresponding adjustment function can be selected from a preset function library, such as a linear adjustment function, a non-linear upward convex adjustment function, or a non-linear downward convex adjustment function. The basic migration urgency is used as input to these functions to calculate the migration urgency assessment value. When the adjustment strategy type is non-linear upward convex adjustment, a gradual adjustment can be obtained in the low-value range of the basic migration urgency, while an accelerated adjustment can be obtained in the high-value range to cope with high-urgency data. Conversely, when the adjustment strategy type is non-linear downward convex adjustment, an accelerated adjustment can be obtained in the low-value range, while a gradual adjustment can be obtained in the high-value range. Alternatively, a rule-based adjustment engine can be used, activating a specific set of adjustment rules based on the adjustment strategy type. These rules define how to adjust according to the current value of the basic migration urgency and the semantic attributes of the data object to generate the migration urgency assessment value.

[0149] The above technical solution comprehensively considers the functional role labels and dynamic urgency levels of data objects, thereby more accurately determining the type of correction strategy for basic migration urgency. This allows the generated migration urgency assessment value to more accurately reflect the actual migration needs of data objects, avoiding biases that may result from a single assessment dimension. For example, for control commands with high dynamic urgency levels, even slight degradation in link quality will accelerate the increase in migration urgency assessment value, prompting the system to take timely link migration measures. Conversely, for routine monitoring data of lower importance, even fluctuations in link quality will not over-amplify the migration urgency assessment value, avoiding unnecessary link migration. This semantically aware correction mechanism enhances the timeliness and effectiveness of link migration decisions, improves the end-to-end assurance capability of heterogeneous networks in industrial settings for critical control services, and solves the problem that the lack of comprehensive consideration of the functional role labels and dynamic urgency levels of data objects in basic migration urgency correction leads to assessment values ​​that cannot accurately reflect actual migration needs, thus affecting the timeliness and effectiveness of link migration.

[0150] In some of the embodiments described above in this application, a method for determining the type of correction strategy to correct the urgency of basic migration is proposed. However, in its implementation, due to the lack of sufficient consideration of the synergistic effect between functional role labels and dynamic urgency levels, the selection of correction strategies may not be accurate enough, and it may not be able to adaptively adjust to the characteristics of different data objects, thereby affecting the accuracy and timeliness of link migration decisions.

[0151] To address this, this application further proposes a method for determining the correction strategy type based on dynamic urgency levels and functional role labels, including: determining a basic strategy sensitivity value corresponding to the data object based on the functional role label; determining an urgency amplification factor corresponding to the data object based on the dynamic urgency level; calculating the product of the basic strategy sensitivity value and the urgency amplification factor to obtain a strategy type determination value; and determining the correction strategy type from multiple preset correction strategy types based on a comparison between the strategy type determination value and a preset determination threshold.

[0152] For example, the functional role label is a specific responsibility or classification identifier for a data object in an industrial control system. Its function is to reflect the importance of the data, real-time requirements, and the degree of impact on system stability. For instance, functional role labels can be predefined as "sensor data," "actuator commands," "diagnostic information," or "configuration parameters," etc. These labels can be manually set or automatically classified using machine learning models based on characteristics such as data source, data type, or communication protocol. The basic strategy sensitivity value is a quantitative indicator reflecting the inherent sensitivity of data objects with specific functional role labels to link migration strategy adjustments. Its function is to provide differentiated strategy adjustment benchmarks for data objects with different functional roles. For example, a value can be preset for each functional role label based on expert experience or system design specifications; for instance, actuator commands may have a higher sensitivity value, while diagnostic information may have a lower one. Alternatively, historical data analysis can be used to statistically analyze the degree of impact of different functional role data on system performance when link quality changes, thereby quantifying their sensitivity.

[0153] The dynamic urgency level represents the real-time urgency of a data object at the current moment, reflecting the immediate priority and timeliness requirements of data transmission. For example, the dynamic urgency level can be dynamically calculated based on the protocol semantic attributes of the data packet (such as timestamps and priority fields), or comprehensively evaluated by combining system operating status (such as equipment failures and abnormal alarms) and the functional role labels of the data object. The urgency amplification factor is a multiplier factor that dynamically adjusts the policy sensitivity based on the dynamic urgency level of the data object. Its function is to give higher-urgency data objects greater weight or more aggressive policies in policy selection. For example, a mapping function can be defined to map the dynamic urgency level to an amplification factor within a positive range from 0; the higher the urgency, the larger the amplification factor. Alternatively, a piecewise function can be used to set different amplification factors for different urgency level ranges; for example, 1 for normal urgency, 1.5 for medium urgency, and 2 for high urgency.

[0154] The strategy type determination value is a comprehensive indicator used to determine the correction strategy type after comprehensively considering functional role labels and dynamic urgency levels. Its function is to quantify the comprehensive demand of the data object for the link migration strategy in the current state. For example, it can be obtained by multiplying the basic strategy sensitivity value by the urgency amplification factor using a simple product method. Alternatively, a weighted summation or a more complex nonlinear combination method can be used. The preset determination threshold is a critical value used to divide the strategy type determination value into different intervals. Its function is to discretize continuous determination values ​​into a finite number of correction strategy types. For example, multiple thresholds can be manually set according to the demand and tolerance for different strategy types. Or, the optimal threshold can be automatically determined by clustering historical data using machine learning algorithms. These multiple preset correction strategy types are predefined strategies used to correct the basic migration urgency. Their function is to provide diverse correction methods to adapt to the needs of different scenarios and data objects. For example, these strategy types can include linear correction functions, nonlinear upward convex correction functions, or nonlinear downward convex correction functions, or rule-based correction strategies, such as using conservative correction when the determination value is below the threshold of 1. When the judgment value is between threshold 1 and threshold 2, a moderate correction is applied. When the judgment value is higher than threshold 2, an aggressive correction is applied.

[0155] The above technical solution enables precise determination of the correction strategy type, thus solving the problem of inaccurate correction strategy selection caused by insufficient consideration of the synergistic effect of functional role labels and dynamic urgency levels. For example, by determining the basic strategy sensitivity value based on functional role labels, the inherent sensitivity of the data object's role characteristics in industrial control to strategy changes can be reflected, ensuring that strategy selection can adapt to the specific needs of different roles (such as sensors or actuators). Determining the urgency amplification factor based on the dynamic urgency level can amplify the impact of the data object's real-time urgency on the strategy, enabling the strategy to respond to urgency fluctuations. Calculating the product of the basic strategy sensitivity value and the urgency amplification factor yields the strategy type determination value, capturing the synergistic effect between role sensitivity and urgency amplification, generating a comprehensive index. The correction strategy type is determined from multiple preset correction strategy types based on the comparison result of the strategy type determination value and a preset determination threshold. This quantitative selection of the strategy through the preset threshold ensures that the correction strategy adapts to the actual characteristics of the data object, avoids strategy selection bias, and improves the accuracy and timeliness of link migration decisions. This strategy determination method, which comprehensively considers both static functional roles and dynamic urgency, enables more intelligent and flexible link migration decisions, effectively ensuring the certainty and reliability of industrial control operations.

[0156] In some of the solutions mentioned above in this application, a correction strategy type is determined based on dynamic urgency level and functional role label to correct the basic migration urgency. However, in this process, there is a lack of specific correction methods for different strategy types, which leads to insufficient accuracy in the migration urgency assessment and an inability to adapt to the urgency and functional role changes of different data objects, thereby affecting the timeliness and accuracy of the link migration decision.

[0157] To address this, this application further proposes a method for modifying the basic migration urgency based on the modification strategy type to generate a migration urgency assessment value. Specifically, this includes: when the modification strategy type is the first type, using a linear modification function to modify the basic migration urgency to obtain a migration urgency assessment value; when the modification strategy type is the second type, using a nonlinear upward convex modification function to modify the basic migration urgency, resulting in a gradual modification in the low-value region and an accelerated modification in the high-value region, thus obtaining a migration urgency assessment value; and when the modification strategy type is the third type, using a nonlinear downward convex modification function to modify the basic migration urgency, resulting in an accelerated modification in the low-value region and a gradual modification in the high-value region, thus obtaining a migration urgency assessment value.

[0158] The correction strategy type refers to predefined categories of correction logic or algorithms based on the dynamic urgency level and functional role labels of the data object, used to guide how to adjust the basic migration urgency. For example, the correction strategy type can be an enumeration value (such as "default," "aggressive," or "conservative"), or an index pointing to a specific function in a pre-defined correction function library. Its purpose is to select the most suitable mathematical model to refine the assessment of migration urgency based on the characteristics of the data object.

[0159] The first strategy type typically represents a general or default correction method, suitable for scenarios where changes in migration urgency are relatively stable or predictable. For example, the first strategy type can be used for routine monitoring data or control commands with extremely high non-real-time requirements.

[0160] A linear correction function is a mathematical function whose output has a simple linear relationship with its input, such as f(x) = ax + b. Here, a is the slope and b is the intercept. In practical applications, a and b can be preset constants or dynamically adjusted according to system operating conditions or data object types. For example, a fixed scaling factor can be set to amplify or reduce the basic migration urgency, or a fixed offset can be added. This function provides a stable and predictable correction effect, ensuring that the migration urgency assessment value changes proportionally with the basic migration urgency.

[0161] The second strategy type is applicable to scenarios that are highly sensitive to the assessment of migration urgency and require accelerated response when the urgency is relatively high. For example, for critical closed-loop control instructions or safety interlock signals, they may not need to be migrated immediately when the link quality deteriorates slightly, but when the deterioration trend is obvious or reaches a certain level, migration needs to be triggered promptly.

[0162] The non-linear upward convex correction function is a mathematical function whose curve bends upward, showing that when the input value is low, the growth rate of the output value is slow (gentle correction), and when the input value is high, the growth rate of the output value accelerates (accelerated correction). For example, it can take the form of an exponential function f(x)=e^(kx)-1 or a power function f(x)=x^n (n>1), etc. Through this function, when the basic migration urgency is at a low level, the corrected migration urgency assessment value grows slowly, avoiding unnecessary premature migration. When the basic migration urgency reaches a high level, the corrected assessment value will increase rapidly, thus accelerating the triggering of link migration to ensure the timely guarantee of critical services.

[0163] The third strategy type is applicable to scenarios that are highly sensitive to potential risks and require accelerated response when the urgency is low. For example, for some diagnostic data or warning information, even a slight fluctuation in the link quality may indicate potential failures, and the system needs to pay quick attention and evaluate the possibility of migration. However, when the link quality is already very poor, excessive accelerated correction may lead to system instability.

[0164] The non-linear downward convex correction function is a mathematical function whose curve bends downward, showing that when the input value is low, the growth rate of the output value is fast (accelerated correction), and when the input value is high, the growth rate of the output value slows down (gentle correction). For example, it can take the form of a logarithmic function f(x)=log(kx+1) or a power function f(x)=x^n (0<n<1), etc. Through this function, when the basic migration urgency is at a low level, the corrected migration urgency assessment value will increase rapidly, prompting a quick response to early deterioration signs. When the basic migration urgency reaches a high level, the growth of the corrected assessment value tends to be gentle, avoiding excessive correction in extreme cases, thus maintaining system stability.

[0165] The above technical solution allows for fine-tuning of the basic migration urgency by selecting the most suitable correction function based on the dynamic urgency level of the data object and the correction strategy type determined by the functional role label. This differentiated correction mechanism ensures that the generated migration urgency assessment value more accurately reflects the actual migration needs of the data object. For example, a linear correction function provides stable and predictable correction in general cases. A nonlinear upward convex correction function prevents premature migration of high-urgency data at low urgency levels, while accelerating correction at high urgency levels, thus ensuring critical business operations while avoiding resource waste. A nonlinear downward convex correction function enables rapid response to potential risks, accelerating correction at low urgency levels and maintaining stability at high urgency levels, preventing overreaction. This precise and adaptive migration urgency assessment improves the timeliness and accuracy of link migration decisions, solving the problems of high-urgency control commands failing to obtain transmission resources matching their real-time requirements under mixed traffic surges or link quality fluctuations in heterogeneous networks in industrial settings, as well as data loss caused by protocol session interruptions during link switching, thereby comprehensively enhancing the deterministic assurance capabilities of industrial control services.

[0166] In some of the embodiments described above in this application, a gating time slot reallocation strategy based on scheduling priority correction amount and current queuing state of data object is proposed to optimize transmission scheduling resource allocation. However, in its implementation, time slot preemption may be directly triggered without evaluating the protocol semantic attributes of the preempted data object, resulting in the unexpected interruption of the transmission of high urgency control instructions or the aggravation of resource conflicts.

[0167] In response, this application further proposes a method for determining the gating time slot reallocation strategy, see [link to relevant documentation]. Figure 5 The method includes the following steps: 501. Obtain the queue depth and gating time slot configuration information of each priority queue in the current transmission scheduling queue from the queuing status of the data record.

[0168] 502. Compare the scheduling priority adjustment amount with the preemption trigger threshold. When the scheduling priority adjustment amount exceeds the preemption trigger threshold, determine the target time slot that can be preempted based on the queue depth and gating time slot configuration information.

[0169] 503. Based on the protocol semantic attributes of the original bearer data object currently assigned to the target time slot, determine the preemptibility condition of the target time slot, and generate a gated time slot reallocation strategy containing the target time slot identifier and the data object identifier when the preemptibility condition is met.

[0170] For example, when retrieving the queue depth and gating slot configuration information of each priority queue in the current transmission scheduling queue from the queuing status of data records, the queuing status of data records refers to a data set formed by semantically parsing received data packets in an industrial heterogeneous network converged communication terminal, associating the protocol semantic attributes of the data object, its queuing status in the current transmission scheduling queue, and the link quality status. Specifically, the queuing status describes the waiting status of the data object in the transmission queue. The current transmission scheduling queue is a buffer used to temporarily store data objects to be sent, typically divided into multiple priority queues based on the priority of the data objects. The queue depth of each priority queue represents the amount of data or data objects currently waiting to be transmitted in each priority queue, and is a key indicator for measuring queue congestion and system load. It can be obtained by real-time querying of the queue status register inside the TSN switch or scheduler, or by periodically receiving status reports from the queue management module. The gating slot configuration information refers to a pre-defined time scheduling table in the TSN network, which specifies which priority queues are allowed to send data within a specific time period. This information is typically read from the TSN switch's configuration table or obtained from the locally stored Gate Control List (GCL), which defines in detail the on / off state of each time slot and its corresponding priority.

[0171] When comparing the scheduling priority adjustment amount with the preemption trigger threshold, and determining the target time slot that can be preempted based on queue depth and gating slot configuration information when the scheduling priority adjustment amount exceeds the preemption trigger threshold, the scheduling priority adjustment amount is dynamically calculated based on the protocol semantic attributes and queuing status of the data object, and is used to adjust the priority of the data object in transmission scheduling. The higher the adjustment amount, the more urgent the data object's need for transmission resources. The preemption trigger threshold is a preset critical value used to determine whether the data object has sufficient urgency to trigger a time slot preemption operation. This threshold can be configured according to the real-time requirements of industrial control services, the importance of data types, or the overall system performance goals. For example, it can be set to a fixed value or dynamically adjusted according to network load or system operating mode. Determining the target time slot that can be preempted refers to identifying those time slots that can be reassigned to more urgent data objects from the currently allocated time slots after the preemption trigger conditions are met. This process comprehensively considers queue depth (to assess the impact of preemption on existing data flows) and gating slot configuration information (to understand the current allocation status of time slots). For example, time slots with low current load, relatively low priority, or whose original data objects can be interrupted can be initially screened. Alternatively, by analyzing the gating time slot configuration, time slots currently occupied by non-critical data flows can be identified.

[0172] In the process of determining the preemptibility conditions of a target time slot based on the protocol semantic attributes of the original bearer data objects currently allocated to that time slot, and generating a gated time slot reallocation strategy containing the target time slot identifier and data object identifier when the preemptibility conditions are met, the protocol semantic attributes of the original bearer data objects currently allocated to that target time slot refer to the semantic information of the data objects currently being transmitted in that time slot that needs to be acquired and analyzed before considering preempting a time slot. These protocol semantic attributes include, but are not limited to, the functional role label, dynamic urgency level, and protocol communication status checkpoints of the data objects, which reflect the importance and real-time requirements of the data objects in the industrial control system. Determining the preemptibility conditions of the target time slot is a crucial step in ensuring that time slot preemption operations will not negatively impact critical business operations. This determination is based on the evaluation of the protocol semantic attributes of the original bearer data objects. For example, if the dynamic urgency level of the original bearer data object is lower than a preset preemption allowable threshold, or if its functional role label indicates that the data stream can be interrupted or delayed, then the preemptibility conditions are considered met. Conversely, if the original data object has a high urgency or critical functional role, then the time slot may not be allowed to be preempted. Generating a gated time slot reallocation policy that includes the target time slot identifier and the data object identifier means creating an explicit instruction or configuration information when all preemption conditions are met. This policy explicitly specifies the unique identifier of the time slot to be preempted (target time slot identifier) ​​and the unique identifier of the new data object to occupy that time slot (data object identifier). This policy can be a data structure containing parameters such as time slot ID, data object ID, new time slot length, and start time, or it can be a control command that can be directly issued to the TSN switch to update its gating list.

[0173] The above technical solution introduces the judgment of the protocol semantic attributes of the original data objects currently carried by the target time slot when determining the gated time slot reallocation strategy. For example, when the scheduling priority adjustment reaches the preemption trigger threshold and time slot reallocation is required, the system does not blindly preempt; instead, it obtains the queue depth and gated time slot configuration information of each priority queue in the current transmission scheduling queue, providing real-time basis for subsequent decisions. Furthermore, based on the protocol semantic attributes of the original data objects currently allocated to the target time slot, it determines whether the target time slot meets the preemptibility conditions. For example, by evaluating the dynamic urgency level or functional role label of the original data objects, it avoids preempting time slots carrying high-urgency control instructions or critical business data. This mechanism solves the problem in traditional TSN scheduling where direct preemption may lead to unexpected interruptions in the transmission of high-urgency control instructions or exacerbated resource conflicts. By conditionally and intelligently reallocating time slots, critical data streams can obtain the necessary transmission resources even under mixed traffic surges or network resource constraints, while minimizing the negative impact on existing services. This enhances the end-to-end assurance capability and determinism of control services provided by the converged communication terminal for heterogeneous networks in industrial sites.

[0174] In some of the solutions mentioned above in this application, a target time slot that can be preempted is determined based on queue depth and gating time slot configuration information to optimize gating time slot reallocation. However, in the implementation process, time slots with insufficient remaining duration, high dynamic urgency level of the original data, or insufficient idle buffer capacity may be preempted, resulting in resource waste or data transmission interruption.

[0175] To address this, this application further proposes a specific method for determining preemptible target time slots based on queue depth and gating time slot configuration information, including: determining the remaining duration of each time slot and the originally allocated data object identifier within the current scheduling period based on the gating time slot configuration information; determining the idle buffer capacity of each priority queue based on the queue depth; and determining the time slot whose remaining duration is greater than a preset minimum time slot granularity, whose dynamic urgency level of the data object corresponding to the originally allocated data object identifier is lower than a preset preemption allowable threshold, and whose idle buffer capacity is sufficient to accommodate the originally allocated data object, as the preemptible target time slot.

[0176] For example, when determining the remaining duration of each time slot and the originally allocated data object identifier within the current scheduling period based on the gating time slot configuration information, this step aims to obtain detailed usage information for each time slot within the current Time-Sensitive Network (TSN) scheduling period, including its currently available duration and which data object the time slot was initially allocated to transmit. This provides the foundational data for subsequently determining which time slots can be securely preempted. One implementation is that the processor can access the Gate Control List (GCL) or a copy thereof maintained internally by the TSN switch. The GCL typically contains the start time, duration, and corresponding traffic ID or data stream identifier for each time slot. By comparing the current time with the time slot's start time and combining it with the duration, the remaining duration of each time slot can be calculated. The GCL or related traffic mapping table records which data object or data stream each time slot was allocated to, thus obtaining the originally allocated data object identifier. Another implementation is that the terminal can periodically request the latest gating time slot configuration information from the TSN Network Controller (TNC). As the centralized controller of the TSN network, the TNC manages the time slot allocation for the entire network. After receiving this information, the terminal can dynamically calculate the remaining duration of each time slot based on the current scheduling cycle and the amount of data already transmitted, and directly parse the originally allocated data object identifier from the configuration information.

[0177] When determining the free buffer capacity of each priority queue based on the queue depth, this step is used to assess the available storage space of each priority queue in the current transmission scheduling queue. Understanding the free buffer capacity is crucial to ensuring that preempted data objects have sufficient temporary storage space to avoid data loss due to buffer overflow. One implementation is that the processor can monitor the transmission scheduling queues within the terminal or TSN switch in real time. Each priority queue typically has a preset maximum buffer capacity. By obtaining the current queue depth (i.e., the amount of data already stored), the free buffer capacity of the queue can be obtained by subtracting the current queue depth from the maximum buffer capacity. Another implementation is that the terminal can maintain a queue status management module that continuously tracks the enqueue and dequeue operations of each priority queue. When a data object enters or leaves the queue, the module updates the corresponding queue depth. The free buffer capacity is then calculated as the difference between the preset total buffer size and the current queue depth. Furthermore, different maximum buffer capacities can be configured for different priority queues to adapt to different business needs.

[0178] When a time slot with a remaining duration greater than a preset minimum time slot granularity, a dynamic urgency level corresponding to the originally allocated data object identifier lower than a preset preemption allowable threshold, and sufficient free buffer capacity to accommodate the originally allocated data object is identified as a preemptible target time slot, this step is the core filtering logic. It comprehensively considers three key dimensions: time, urgency, and buffer capacity, to accurately identify time slots that have sufficient time, will not affect high-priority data, and have sufficient buffer space to temporarily store the original data, thereby ensuring the effectiveness and security of the preemption operation. One implementation is that the processor traverses all time slots. For each time slot, it checks whether its remaining duration exceeds a preset minimum time slot granularity (e.g., the transmission time of an Ethernet frame). Based on the originally allocated data object identifier of the time slot, it queries the dynamic urgency level of the data object and compares it with the preset preemption allowable threshold. It checks whether the free buffer capacity of the corresponding priority queue is sufficient to accommodate the original data object. Only when all three conditions are met simultaneously is the time slot marked as a preemptible target time slot. Another approach is to design a multi-dimensional decision matrix or scoring mechanism. Each time slot receives a comprehensive score based on its remaining duration, the dynamic urgency level of the original data object, and the idle buffer capacity of the corresponding queue. Preset minimum time slot granularity, preemption allowance threshold, and buffer capacity requirements are used as hard constraints; time slots that do not meet any of these constraints are directly excluded. For time slots that meet the hard constraints, they can be sorted according to their comprehensive scores, and the time slot with the highest score (i.e., the most suitable for preemption) is selected as the target time slot. For example, the lower the dynamic urgency level, the longer the remaining duration, and the larger the idle buffer capacity, the higher the score.

[0179] Through the above technical solutions, the terminal can accurately identify the target time slots that are truly suitable for preemption, thereby avoiding the negative impact of blind preemption. For example, by obtaining the remaining duration of each time slot and the identifier of the originally allocated data object, only time slots with sufficient transmission time and whose original data objects can be temporarily interrupted are considered, avoiding the incomplete transmission of new data or interruption of original data transmission due to excessively short time slots. By determining the idle buffer capacity of each priority queue, it is ensured that the original data object being preempted has enough space for temporary storage when preemption occurs, preventing data loss or buffer overflow. More importantly, by jointly judging the three key conditions of remaining duration, dynamic urgency level of the original data object, and idle buffer capacity, only time slots with sufficient time, low urgency, and adequate buffer are identified as preemptible targets. This minimizes the impact on existing services while meeting the needs of high-urgency data transmission, improves the efficiency and reliability of TSN-gated time slot reallocation, and ensures the determinism of industrial control services. This refined time slot selection mechanism makes the gated time slot reallocation strategy more intelligent and robust, solving the problem that high-urgency control commands cannot obtain transmission resources that match their real-time requirements when faced with mixed traffic surges or link quality jitter.

[0180] In some of the solutions mentioned above in this application, a link migration execution strategy is proposed based on the migration urgency assessment value and protocol semantic attributes to optimize link switching. However, in this process, if the migration urgency is not precisely quantified and the differences in the functional roles of data objects in industrial control are ignored, the migration strategy may lack pertinence and fail to implement differentiated processing according to the actual urgency and functional requirements of the data objects, which may easily cause session interruption or improper resource allocation during link switching.

[0181] To address this, this application further proposes a method for determining a link migration execution strategy that matches protocol semantic attributes. The method includes: comparing a migration urgency assessment value with a first urgency threshold and a second urgency threshold to determine the migration urgency level corresponding to the data object, wherein the first urgency threshold is less than the second urgency threshold; obtaining the functional role label of the data object from the protocol semantic attributes of the data record; and determining a link migration execution strategy that matches the data object from multiple preset link migration execution strategies based on the migration urgency level and the functional role label.

[0182] For example, the migration urgency assessment value is compared with a first urgency threshold and a second urgency threshold to determine the migration urgency level corresponding to the data object, where the first urgency threshold is less than the second urgency threshold. The migration urgency assessment value is a continuous quantity, reflecting the urgency of the impact of physical link degradation or potential degradation on data transmission. To facilitate policy decision-making, it needs to be transformed into a discrete, actionable "migration urgency level." By setting two different thresholds, namely the first urgency threshold and the second urgency threshold, the assessment value can be divided into at least three level intervals. For example, when the assessment value is below the first urgency threshold, it can be determined as "low urgency level." When the assessment value is between the first and second urgency thresholds, it can be determined as "medium urgency level." When the assessment value is above the second urgency threshold, it can be determined as "high urgency level." This hierarchical processing can more finely reflect the urgency requirement of the data object for link migration, avoiding the coarseness of a single threshold judgment. One implementation is to use a piecewise function or conditional statement to directly map the relative magnitude of the assessment value and the two thresholds to a preset integer or enumerated level value. Another approach is to build a rule-based decision-making module containing a set of predefined rules, such as "If the evaluation value is less than X, the grade is A. If the evaluation value is between X and Y, the grade is B. If the evaluation value is greater than Y, the grade is C," and determine the urgency level by matching these rules.

[0183] The functional role label of a data object is obtained from the protocol semantic attributes of the data record. The functional role label is the business semantic identifier of the data object in the industrial control system; for example, it may be sensor data, actuator commands, controller status information, or diagnostic logs. The purpose of obtaining this label is to fully consider the importance of the data object in the entire control loop, its real-time requirements, and its tolerance for link interruption when formulating link migration strategies, thereby achieving a differentiated strategy based on business awareness. One implementation is to query a pre-configured semantic annotation rule base based on the industrial communication protocol type, address information, or function code in the data packet during the semantic parsing stage, and directly store the identified functional role label in the protocol semantic attributes of the data record for subsequent direct retrieval. Another implementation is to dynamically generate and obtain the functional role label by analyzing the data object's communication pattern, its interaction with other devices, and its position in the control logic, using machine learning models or expert systems for real-time inference.

[0184] Based on migration urgency level and functional role labels, a migration execution strategy matching the data object is determined from multiple preset migration execution strategies. This is a crucial step in combining the urgency of the link status with the business importance of the data object to make strategy decisions. Different migration execution strategies may differ in terms of migration speed, resource reservation, state synchronization mechanism, and data buffering strategy. For example, for data with a high urgency level and an executor command functional role, a fast, lossless migration strategy with reserved bandwidth may be needed. For data with a medium urgency level and a diagnostic log functional role, a smooth migration strategy with lower resource consumption may be chosen. By comprehensively considering these two dimensions, it can be ensured that the selected strategy can respond to the urgency of the link while meeting the business needs of the data object. One implementation is to construct a two-dimensional decision matrix or lookup table, where rows and columns represent migration urgency level and functional role labels, respectively. Each cell is pre-configured with a specific migration execution strategy identifier. The system directly obtains the strategy matching the current data object by querying this matrix. Another implementation is to use a decision algorithm based on priority or weight allocation. Assign corresponding weights or priorities to different migration urgency levels and functional role tags, then calculate a comprehensive score through weighted summation or other combinations, and select the most suitable strategy from the preset strategy set based on the score.

[0185] By discretizing continuous migration urgency assessment values ​​into explicit migration urgency levels and combining them with the functional role labels of data objects, refined and business-aware decision-making on link migration execution strategies is achieved. This method avoids the lack of specificity in migration strategies in traditional solutions, enabling differentiated link migration processing based on the actual urgency of data objects and their functional requirements in industrial control. For example, for high-urgency critical control commands, a fast and lossless migration strategy can be matched to ensure their real-time performance and reliability. For low-urgency non-critical data, a smooth migration strategy with lower resource consumption can be selected. This collaborative decision-making mechanism reduces the risk of session interruption and data loss during link switching, optimizes the allocation of transmission resources, thereby improving the end-to-end guarantee capability of heterogeneous networks in industrial fields for control services and enhancing the determinism and robustness of communication.

[0186] In some of the solutions mentioned above in this application, a link migration execution strategy is proposed based on the migration urgency level and functional role labels to optimize link migration decisions. However, in the implementation process, the differences in functional role labels may lead to insufficiently adaptive strategy selection, making it impossible to dynamically adjust the strategy level according to the specific functional requirements of the data object, thereby affecting the accuracy of migration and the continuity of control services.

[0187] To address this, this application further proposes a method for determining a migration execution strategy that matches the data object from multiple preset migration execution strategies, based on the migration urgency level and the functional role label. Specifically, this includes: determining the strategy upgrade offset corresponding to the data object based on the functional role label; superimposing the migration urgency level and the strategy upgrade offset to obtain the actual strategy determination level of the data object; and selecting the corresponding migration execution strategy from the multiple preset migration execution strategies based on the actual strategy determination level.

[0188] The method of determining the policy upgrade offset corresponding to the data object based on the functional role label aims to adjust the basic migration urgency level according to the functional role label of the data object, so as to more finely reflect the sensitivity and priority requirements of different functional roles to the link migration strategy. This policy upgrade offset is an adjustment factor that provides differentiated policy adjustment capabilities for data objects of different importance. For example, a mapping table can be pre-established to associate different functional role labels (such as "control commands," "status monitoring," "configuration parameters," etc.) with their corresponding policy upgrade offsets, and the corresponding offsets can be obtained by querying this mapping table. Alternatively, a rule-based inference engine can be used to dynamically calculate the policy upgrade offset, for example, defining a series of rules such as "if the functional role label is 'emergency control,' the offset is positive; if it is 'non-critical log,' the offset is negative." These rules can be configured according to the actual business needs of the industrial site and expert experience.

[0189] The migration urgency level is superimposed with the policy upgrade offset to obtain the actual policy judgment level of the data object. This superposition operation combines the original migration urgency level with the policy upgrade offset derived from the functional role label, generating a more comprehensive assessment value that reflects the urgency and business importance of the data object's transmission. This actual policy judgment level is the key basis for determining which link migration execution strategy to adopt. The superposition operation can be simply performed using arithmetic addition, where the actual policy judgment level equals the sum of the migration urgency level and the policy upgrade offset. Alternatively, the superposition operation can employ a weighted average or non-linear function mapping. For example, different weights can be assigned to the migration urgency level and the policy upgrade offset, or a non-linear function can be used to map both to a unified level space to better reflect their interactive influence in a specific scenario.

[0190] Based on the actual policy judgment level, a corresponding link migration execution strategy is selected from multiple preset link migration execution strategies. This step involves selecting the most suitable link migration execution strategy for the current data object and network conditions from a predefined set of strategies, based on the previously calculated actual policy judgment level. Its purpose is to ensure the accuracy and adaptability of link migration decisions, avoiding a one-size-fits-all migration strategy. For example, a strategy selection table or decision tree can be established to map different actual policy judgment level ranges to specific link migration execution strategies. For instance, when the actual policy judgment level falls into a certain range, a "smooth migration strategy" is selected; when it falls into another range, a "fast migration strategy" is selected; and when it falls into a higher range, an "emergency preemptive migration strategy" is selected. Alternatively, a machine learning-based model can be used to implement strategy selection. This model is trained on historical data to learn the complex relationship between the actual policy judgment level and the optimal link migration execution strategy, thereby predicting and outputting the most suitable link migration execution strategy.

[0191] By introducing a strategy upgrade offset and superimposing it on the migration urgency level, a more comprehensive actual strategy determination level is generated through the above technical solution. This makes the determination of the link migration execution strategy more refined and adaptive, fully considering the business importance represented by the functional role labels of data objects, thereby avoiding the problem of inaccurate strategy selection due to differences in functional role labels. This mechanism enables link migration decisions to better match the specific functional requirements of data objects. For example, for high-urgency control instructions, even if their migration urgency level is not high, due to the importance of their functional role labels, the strategy upgrade offset can increase their actual strategy determination level, thus selecting a more aggressive migration strategy. Conversely, for non-critical data, a more conservative strategy may be selected. This dynamic adjustment capability improves the accuracy of link migration execution, effectively ensuring the continuity and determinism of industrial control business during link switching, and avoiding business interruption or data loss due to policy rigidity.

[0192] In some of the embodiments described above in this application, a TSN gating list is adjusted based on a gating time slot reallocation strategy to optimize data transmission scheduling. However, in its implementation, if the time slot parameters of the target time slot are not verified to meet the transmission constraints of the data object, the time slot allocation may not match the transmission requirements of the data object. For example, the time slot length may be insufficient or the start time delay may exceed the tolerance range, which may lead to transmission failure or timeout of critical control commands, affecting the determinism and reliability of the industrial control system.

[0193] In response, this application further proposes steps for adjusting the TSN gating list based on a gating time slot reallocation strategy, including: Obtain the target time slot identifier and the identifier of the data object to be inserted from the gating time slot reallocation strategy.

[0194] Based on the protocol semantic attributes of the data object corresponding to the data object identifier to be inserted, verify whether the time slot parameters corresponding to the target time slot identifier meet the transmission constraints of the data object.

[0195] When the transmission constraint is met, incremental gating list update information is generated for the target time slot identifier, and the incremental gating list update information is sent to the TSN switch to adjust the TSN gating list.

[0196] In this regard, this application elaborates on the technical features of the above-mentioned technical solution.

[0197] The step of "obtaining the target time slot identifier and the identifier of the data object to be inserted from the gated time slot reallocation policy" aims to clarify the time slot to be operated on and the data object to be transmitted. The gated time slot reallocation policy can be a data structure containing unique identifiers of the time slot to be adjusted (e.g., time slot ID, start time, duration) and unique identifiers of the data objects to be inserted (e.g., packet ID, data stream ID). The processor can directly read this information from the data structure as the basic input for subsequent verification and update operations. Alternatively, the policy can be generated in real time by the scheduling algorithm, and the processor can extract the required identifiers by parsing its output instructions.

[0198] The step of "verifying whether the time slot parameters corresponding to the target time slot identifier meet the transmission constraints of the data object based on the protocol semantic attributes of the data object corresponding to the data object to be inserted" is crucial to ensuring the effectiveness of time slot allocation. It utilizes the inherent transmission requirements (i.e., protocol semantic attributes) of the data object to evaluate the suitability of the target time slot. The processor can query the protocol semantic attributes of the data object, such as its functional role label, dynamic urgency level, maximum tolerable delay, and minimum transmission granularity. It obtains the time slot parameters of the target time slot, such as the time slot length and time slot start time. Then, it performs a comparison using a preset set of verification rules. These rules may include, but are not limited to: whether the time slot length meets the minimum transmission granularity requirement of the data object; whether the available time window of the time slot is within the maximum tolerable delay of the data object; or, more complexly, evaluating whether the bandwidth, jitter, and other parameters of the time slot meet its Quality of Service (QoS) requirements based on the dynamic urgency level and functional role label of the data object.

[0199] The step of "generating incremental gating list update information for the target timeslot identifier when the transmission constraint is met" ensures that only validly verified allocations are translated into actual update instructions. Incremental gating list update information is an instruction or data packet used to notify the TSN switch to make partial modifications to its internal gating list. This information may include the identifier of the target timeslot, the new gating state (e.g., open to allow a specific data flow), and other configuration parameters related to the data object to be inserted. After successful verification, the processor generates this update information according to a predefined format based on the target timeslot identifier and the identifier of the data object to be inserted.

[0200] The step of "and sending the incremental gating list update information to the TSN switch to adjust the TSN gating list" is the actual stage of scheduling adjustment. The processor sends the generated incremental gating list update information to the TSN switch through the communication channel established with the TSN switch. This communication channel can be a standardized network management protocol (such as NETCONF / YANG, SNMP) or a TSN-specific control protocol (such as the Central Network Controller (CNC) interface defined by IEEE 802.1Qcc). After receiving the update information, the TSN switch parses and applies this information, thereby dynamically modifying its internal gating list and reallocating data stream transmission time slots.

[0201] The above technical solution introduces a dynamic verification mechanism for the target time slot parameters and data object transmission constraints when adjusting the TSN gating list based on the gating time slot reallocation strategy. For example, after obtaining the target time slot identifier and the identifier of the data object to be inserted from the gating time slot reallocation strategy, the transmission constraints of the data object to be inserted are rigorously verified based on its protocol semantic attributes. This verification process avoids transmission failures or critical control command timeouts caused by mismatches such as insufficient time slot length or time slot start time delays exceeding the tolerance range. Only when the transmission constraints are met is the incremental gating list update information generated and sent to the TSN switch, ensuring that only valid time slot allocations that meet the actual transmission requirements of the data object are executed. This improves the determinism and reliability of industrial control services, preventing high-urgency control commands from failing to obtain transmission resources matching their real-time requirements due to improper time slot allocation during mixed traffic surges or link quality fluctuations, thus guaranteeing the end-to-end guarantee capability of control services in heterogeneous industrial networks.

[0202] In some of the solutions mentioned above in this application, transmission constraints based on protocol semantic attributes are proposed to ensure the effectiveness of scheduling adjustments. However, in this process, there is a lack of specific matching verification between time slot parameters and data object transmission constraints, which may lead to time slot resource allocation not meeting the real-time requirements or transmission granularity requirements of data objects, thereby causing scheduling failure or resource waste.

[0203] To address this, this application further proposes a method to verify whether the time slot parameters corresponding to the target time slot identifier meet the transmission constraints of the data object based on the protocol semantic attributes of the data object corresponding to the identifier of the data object to be inserted. This verification process includes: Obtain the time slot length and time slot start time from the time slot parameters corresponding to the target time slot identifier.

[0204] Obtain the maximum tolerable latency and minimum transmission granularity of the data object from its protocol semantic attributes.

[0205] When the time slot length is greater than or equal to the minimum transmission granularity, and the time difference between the time slot start time and the current time is not greater than the maximum tolerable delay, the transmission constraint condition is considered satisfied.

[0206] Otherwise, it is determined that the transmission constraints are not met.

[0207] For example, during the verification process, it is necessary to obtain the time slot length and time slot start time from the time slot parameters corresponding to the target time slot identifier. This step aims to obtain the basic time attributes of the time slot to be allocated, namely its duration and start time within the scheduling cycle. These parameters are key foundational information for evaluating whether a time slot can carry the transmission of a specific data object. For example, the length and start time of the target time slot can be directly read by querying the Gate Control List (GCL) configuration information maintained by the Time Sensitive Network (TSN) switch or TSN controller. This information is usually expressed in time units (such as microseconds, nanoseconds) or clock cycles. Alternatively, it can be obtained from the pre-configured time slot resource pool management module, which records detailed parameters of all available time slots, including their duration and start position within the scheduling cycle.

[0208] The maximum tolerable latency and minimum transmission granularity of a data object are obtained from its protocol semantic attributes. This step is used to extract the inherent requirements of the data object for transmission performance and resource allocation, serving as the basis for subsequent verification of transmission constraints. Maximum tolerable latency refers to the maximum time delay that a data object can withstand from its creation to processing by the receiver, reflecting its real-time requirements. Minimum transmission granularity refers to the minimum amount of data or the minimum time window required for data object transmission, reflecting its requirements for the continuity of transmission resources. For example, these values ​​can be directly stored as fields in the data record as protocol semantic attributes, and retrieved and associated from a preset policy library or configuration table during semantic parsing of the data object, based on its functional role label, dynamic urgency level, and other information. Alternatively, they can be obtained by querying the industrial control application configuration file or device description file associated with the data object. These files typically define the real-time requirements and data packet size limits for different types of data objects.

[0209] The core verification logic checks whether the transmission constraint is met when the time slot length is greater than or equal to the minimum transmission granularity and the time difference between the time slot start time and the current time is not greater than the maximum tolerable delay. This mechanism ensures that the target time slot meets the transmission requirements of the data object in both time and space. By comparing the obtained time slot length with the minimum transmission granularity of the data object, it is ensured that the time slot capacity is sufficient to carry the data object. The time difference between the time slot start time and the current system time is calculated and compared with the maximum tolerable delay of the data object to ensure that the data object can start transmission within its tolerable delay, avoiding real-time failure due to excessive waiting time. Only when both conditions are met simultaneously is the transmission constraint considered satisfied.

[0210] Otherwise, the transmission constraint is determined to be unsatisfied. This step provides a clear negative result to guide subsequent scheduling decisions. The transmission constraint is determined to be unsatisfied when any of the above conditions (i.e., the slot length is insufficient to support the minimum transmission granularity, or the time difference between the slot start time and the current time exceeds the maximum tolerable delay) is not met. This clear determination helps avoid allocating data objects to inappropriate slots, thereby preventing scheduling failures or resource waste.

[0211] The above technical solution enables precise verification of the effectiveness of the gating time slot reallocation strategy. For example, by obtaining the length and start time of the target time slot and combining it with the maximum tolerable delay and minimum transmission granularity of the data object, a detailed matching judgment can be performed. This verification mechanism ensures that the reallocated time slot is not only long enough to accommodate the minimum transmission granularity of the data object, but also that its start transmission time meets the real-time requirements of the data object, i.e., transmission starts within the maximum tolerable delay of the data object. This avoids scheduling failures, data delays, or resource waste caused by a mismatch between time slot allocation and data object transmission needs, thereby improving the accuracy and reliability of TSN gating list adjustments and ensuring deterministic transmission of industrial control services.

[0212] In some of the solutions described above in this application, a link migration execution strategy is proposed to perform link migration. However, in this process, since the upper-layer protocol session state is not aware, link switching may lead to protocol session interruption and data loss.

[0213] To address this issue, this application proposes a converged communication terminal for heterogeneous networks in industrial settings. Its processor is configured to perform state-continuous migration of the physical link carrying the data object based on a link migration execution strategy. Specifically, this includes: determining a target backup link and a migration triggering time based on the link migration execution strategy; obtaining the protocol communication state checkpoint of the data object from the protocol semantic attributes of the data record; and, when the migration triggering time arrives, sending a state recovery request carrying the protocol communication state checkpoint through the target backup link to continue transmission of the data object from the protocol state recorded at the protocol communication state checkpoint on the target backup link.

[0214] The determination of the target backup link and migration triggering timing based on the link migration execution strategy refers to specifying which backup physical link to switch to and when to initiate the migration process when a link migration requirement occurs, according to a pre-set or dynamically generated link migration execution strategy. This link migration execution strategy can be a set of rules, such as selecting the optimal link from multiple available backup links based on factors like link quality status and protocol semantic attributes of data objects (e.g., dynamic urgency level, functional role labels). For example, the link with the best performance can be selected based on a comprehensive evaluation of its bandwidth, latency, packet loss rate, and other indicators. Alternatively, a backup link that can provide deterministic transmission guarantees can be selected based on the real-time requirements of the data objects. The migration triggering timing can be determined according to different strategy types. For example, migration can be triggered immediately when the quality degradation of the current physical link reaches a preset threshold. Or, migration can be planned and triggered in advance before a link failure is predicted. Alternatively, migration can be triggered within a specific time window when system load is low to minimize the impact on services.

[0215] The protocol communication state checkpoint, obtained from the protocol semantic attributes of the data record, refers to extracting the protocol communication state checkpoint associated with a specific data object from the data record generated after the terminal processor performs semantic parsing on data packets received from heterogeneous protocol ports. This protocol communication state checkpoint is an internal state snapshot of the protocol state machine instance at a specific moment, containing key information about the protocol session, such as sequence number, acknowledgment number, window size, and current transmission stage. As an extended field of the protocol semantic attributes, this checkpoint can be automatically generated and stored by the protocol state machine at critical operation points (such as connection establishment, data transmission, and acknowledgment of receipt), or it can be actively triggered by the system when potential degradation is detected in the link quality. The purpose of obtaining this checkpoint is to accurately restore the context of the protocol session after link migration, avoiding the need to establish a connection from scratch or retransmit large amounts of data.

[0216] When the migration trigger point is reached, the terminal processor sends a state recovery request message carrying the protocol communication state checkpoint through the target backup link. This request allows the data object to continue transmission from the protocol state recorded at the checkpoint on the target backup link. Specifically, once the preset migration trigger point is reached, the terminal processor immediately sends a special state recovery request message to the communication peer or relevant device in the network through the determined target backup link. This request message carries the previously acquired protocol communication state checkpoint information. Upon receiving this request, the receiver uses the checkpoint information to restore its internal protocol state, ensuring it remains consistent with the sender's state before the migration. For example, if the checkpoint records the sequence number and acknowledgment number of a TCP connection, the receiver adjusts its internal state so that subsequent data transmission can seamlessly continue from that sequence number and acknowledgment number without requiring a re-handshake or retransmission of successfully sent data. In this way, the data object can continue transmission from the point of interruption in the protocol session on the new physical link, achieving a smooth continuation of the protocol state.

[0217] The above technical solutions can solve the problem of protocol session interruption and data loss caused by the lack of awareness of the upper-layer protocol session state during link switching. For example, by determining the target backup link and the migration triggering time based on the link migration execution strategy, the link migration operation is executed at the optimal time and on the optimal link, avoiding resource conflicts or delays caused by random switching. The protocol communication state checkpoint of the data object is obtained from the protocol semantic attributes of the data record, and combined with the protocol state information recorded in the protocol semantic attributes, the migration process can accurately locate the breakpoint of the protocol session, avoiding session interruption due to state loss. Furthermore, when the migration triggering time arrives, a state recovery request carrying the protocol communication state checkpoint is sent through the target backup link, realizing the continuation of data object transmission from the protocol state recorded at the checkpoint on the target backup link, thereby seamlessly restoring transmission and preventing data loss and session interruption. This state-continuous migration mechanism enables heterogeneous networks in industrial fields to provide continuous and deterministic transmission guarantees for high-urgency control commands when facing link quality jitter or failure, improving the end-to-end reliability and stability of industrial control services.

[0218] The following example will provide a more detailed explanation of the above technical solution: In an industrial production workshop, a critical industrial robot is performing a high-precision welding task, with its control commands transmitted via industrial Ethernet. Other non-critical sensor data and monitoring video streams also exist on the network. When the robot controller sends a critical welding path adjustment command, this command is received as a data message by a heterogeneous network converged communication terminal in the industrial field.

[0219] The terminal's processor performs semantic parsing on the received data packets. It identifies that the packet conforms to the Profinet protocol and drives a corresponding protocol state machine instance to parse the packet based on this protocol type. During this process, the terminal extracts the functional role label (e.g., "critical motion control command") and dynamic urgency level (e.g., level 5, indicating the highest urgency) of the data object from the packet. Unlike related technologies where the protocol conversion module only focuses on data format uniformity, the terminal analyzes the historical state transition frequency of the protocol state machine instance before the current parsing cycle and compares it with the steady-state transition frequency to obtain the frequency deviation. If the frequency deviation indicates that the historical state transition frequency is higher than the steady-state transition frequency, the urgency increase is determined based on this deviation. The sum of the baseline urgency level and the urgency increase is used as the dynamic urgency level, thus more accurately reflecting the real-time importance of the command. For example, if the robot is in high-speed motion or an abnormal state, its state transition frequency will increase, and the terminal will correspondingly increase the dynamic urgency level of the command. When the dynamic urgency level exceeds a preset threshold, the terminal will also extract the protocol communication status checkpoint of the data object (e.g., the coordinates of the current welding point and welding parameters) from the current state of the protocol state machine instance as an extended field of the protocol semantic attributes to prepare for subsequent link migration.

[0220] The terminal performs quality monitoring on the physical link carrying the welding path adjustment command. Based on the dynamic urgency level of the command (e.g., level 5), the terminal determines the sensitivity parameters for quality monitoring of the physical link, for example, by using a shorter sampling period and a stricter fluctuation tolerance threshold. The terminal collects multiple physical layer transmission quality indicators (e.g., packet loss rate, jitter, signal strength) of the physical link according to the sampling period. Based on the functional role label of the command (“critical motion control command”), the terminal determines the tolerance threshold for fluctuations in each physical layer transmission quality indicator. If an indicator among the multiple physical layer transmission quality indicators whose fluctuation exceeds the corresponding tolerance threshold is marked as a valid degradation indicator (e.g., jitter suddenly increases and exceeds the threshold), the terminal obtains the degradation duration and degradation slope of these valid degradation indicators over multiple consecutive sampling periods, and generates a link quality status including the current quality score and quality evolution trend (e.g., “Current quality score: 70 points, quality evolution trend: moderate degradation, accelerating deterioration”). Unlike related technologies where the link management module only performs switching based on link quality, the terminal's quality monitoring adjusts sensitivity based on the protocol semantic attributes of data objects, ensuring that the link quality of critical services is monitored more precisely.

[0221] The terminal generates a scheduling priority correction amount based on the coupling relationship between protocol semantic attributes and queuing status in the data records. The terminal obtains the queuing delay sequence of the welding path adjustment instruction over multiple consecutive scheduling cycles and determines its delay change trend based on this sequence. This delay change trend is compared with the instruction's maximum tolerable delay to obtain a timeout risk estimate. For example, if the delay growth rate is estimated to consume the remaining margin of tolerable delay in a short time, a high-level timeout risk estimate is generated. Then, the terminal generates a scheduling priority correction amount based on this timeout risk estimate, the instruction's dynamic urgency level, and the dynamic urgency levels of other data objects with the same associated device identifier as the robot. For example, the terminal performs a non-linear mapping between the instruction's dynamic urgency level and the timeout risk estimate to obtain a joint risk factor, and combines this with its functional role label to determine the upper limit of the correction coefficient, calculating the basic correction component. The terminal obtains the percentage of data objects with a dynamic urgency level exceeding a preset threshold among other data objects related to the robot and determines the collaborative enhancement coefficient based on this percentage. The product of the basic correction component and the collaborative enhancement coefficient is used as the scheduling priority correction amount.

[0222] The terminal generates a migration urgency assessment value based on the coupling relationship between protocol semantic attributes and link quality status in data records. The terminal obtains the current quality score and quality evolution trend of the physical link from the link quality status. It correlates the current quality score with the quality evolution trend to determine the basic migration urgency of the physical link. For example, it determines the quality degradation acceleration factor based on the quality evolution trend and calculates the basic migration urgency by combining it with the remaining quality margin. Then, the terminal determines the correction strategy type based on the dynamic urgency level and functional role label (“critical motion control command”) of the instruction. For example, if the dynamic urgency level is high and the functional role label is critical, a nonlinear convex correction function may be used to correct the basic migration urgency, resulting in a gradual correction in the low-value range and an accelerated correction in the high-value range, thereby generating the migration urgency assessment value.

[0223] The terminal determines a gated time slot reallocation strategy based on the scheduling priority correction amount and the current queuing status of the instruction. The terminal obtains the queue depth and gated time slot configuration information of each priority queue in the current transmission scheduling queue from the queuing status. When the scheduling priority correction amount exceeds the preemption trigger threshold, the terminal determines the target time slot that can be preempted based on the queue depth and gated time slot configuration information. For example, a time slot carrying non-critical sensor data, whose remaining duration is greater than the minimum time slot granularity, and whose dynamic urgency level of the original data object is lower than the preset preemption allowable threshold, while having sufficient free buffer capacity to accommodate the original data object, is determined to be a target time slot that can be preempted. The terminal further determines the preemptibility conditions of the target time slot based on the protocol semantic attributes of the original data object currently allocated to the target time slot, and generates a gated time slot reallocation strategy containing the target time slot identifier and the welding path adjustment instruction identifier when the conditions are met. Unlike related technologies where the TSN scheduling module only allocates time slots based on a preset gate list, the terminal can dynamically adjust the TSN gate list according to the real-time semantic importance of the data object and the queuing status, ensuring priority transmission of critical instructions.

[0224] The terminal determines a link migration execution strategy that matches the protocol semantic attributes based on the migration urgency assessment value. The terminal compares the migration urgency assessment value with a first urgency threshold and a second urgency threshold to determine the migration urgency level corresponding to the instruction (e.g., high urgency level). It obtains the functional role label (“critical motion control instruction”) of the instruction from the protocol semantic attributes. Based on the migration urgency level and the functional role label, the terminal determines the link migration execution strategy that matches the instruction from multiple preset link migration execution strategies. For example, if the migration urgency level is high and the functional role label is critical, the strategy of “immediately switching to a high-bandwidth backup link and performing state continuity” may be selected.

[0225] The terminal adjusts the TSN gating list based on the gating time slot reallocation strategy. The terminal obtains the target time slot identifier and the identifier of the data object to be inserted (i.e., the welding path adjustment instruction) from the gating time slot reallocation strategy. Based on the protocol semantic attributes of the instruction, the terminal verifies whether the time slot parameters (e.g., time slot length, start time) corresponding to the target time slot identifier meet the transmission constraints of the instruction (e.g., the time slot length is greater than the minimum transmission granularity, and the time difference between the time slot start time and the current time is not greater than the maximum tolerable delay). When the transmission constraints are met, the terminal generates incremental gating list update information for the target time slot identifier and sends this information to the TSN switch to adjust the TSN gating list, thereby allocating a priority transmission time slot for this critical instruction.

[0226] The terminal performs a state-sequential migration of the physical link carrying the instruction based on the link migration execution strategy. The terminal determines the target backup link (e.g., a redundant fiber optic link) and the migration trigger timing based on the link migration execution strategy. When the migration trigger timing arrives, the terminal obtains the protocol communication state checkpoint of the instruction from the protocol semantic attributes (e.g., previously extracted welding point coordinates and welding parameters). A state recovery request carrying the protocol communication state checkpoint is sent through the target backup link to resume transmission of the instruction from the protocol state recorded at the protocol communication state checkpoint on the target backup link. Unlike related technologies where data loss occurs due to protocol session interruption during link switching, the terminal, through state-sequential migration, ensures the determinism and data integrity of critical control services during link switching, avoiding data loss or control anomalies caused by session interruption.

[0227] Through the above process, the terminal integrates the three dimensions of protocol semantics, transmission scheduling, and link status, enabling deterministic transmission of key welding commands for industrial robots in complex network environments and under link degradation conditions, thereby improving the end-to-end guarantee capability of heterogeneous networks in industrial sites for control services.

[0228] All of the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this application, and will not be described in detail here.

[0229] The above are merely optional embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. An industrial field heterogeneous network converged communication terminal, characterized in that, The terminal includes a processor and a memory, the processor being configured to perform the following steps: Semantic parsing is performed on data packets received from heterogeneous protocol ports to extract the protocol semantic attributes of data objects in the data packets. Quality monitoring is performed on the physical links carrying the data objects to extract the link quality status. The protocol semantic attributes, the queuing status of the data objects in the current transmission scheduling queue, and the link quality status are associated as data records. Based on the coupling relationship between protocol semantic attributes and queuing status in the data records, a scheduling priority correction value is generated, and based on the coupling relationship between protocol semantic attributes and link quality status in the data records, a migration urgency assessment value is generated. Based on the scheduling priority correction amount and the current queuing state of the data object, a gating slot reallocation strategy is determined, and based on the migration urgency assessment value and the protocol semantic attributes, a link migration execution strategy matching the protocol semantic attributes is determined. The TSN gating list is adjusted based on the gating time slot reallocation strategy, and the physical link carrying the data object is subjected to state continuity migration based on the link migration execution strategy.

2. The terminal according to claim 1, characterized in that, The step of semantically parsing data packets received from heterogeneous protocol ports and extracting the protocol semantic attributes of data objects in the data packets includes: Identify the industrial communication protocol type followed by the data packet, and determine the protocol state machine instance corresponding to the data packet based on the industrial communication protocol type; The protocol state machine instance is driven to parse the data packet and extract the functional role label and dynamic urgency level of the data object in the data packet as the basic fields of the protocol semantic attributes; When the dynamic urgency level exceeds the level threshold, the protocol communication state checkpoint of the data object is extracted from the current state of the protocol state machine instance, and the protocol communication state checkpoint is added as an extended field of the protocol semantic attribute.

3. The terminal according to claim 2, characterized in that, The process of driving the protocol state machine instance to parse the data packet and extract the functional role tags and dynamic urgency levels of the data objects in the data packet includes: Based on the industrial communication protocol type followed by the data message and the address range of the data object in the data message, the semantic annotation rule base is queried to obtain the baseline functional role label and baseline urgency level of the data object; Obtain the historical state transition frequency of the protocol state machine instance before the current parsing cycle. The historical state transition frequency is the ratio of the number of state transitions of the protocol state machine instance within a preset observation window to the duration of the preset observation window. The baseline urgency level is corrected based on the historical state change frequency to obtain the dynamic urgency level, and the baseline functional role label is determined based on the historical state change frequency to obtain the functional role label.

4. The terminal according to claim 1, characterized in that, The step of monitoring the quality of the physical link carrying the data object and extracting the link quality status includes: Obtain the protocol semantic attributes of the data object determined after semantic parsing; Based on the dynamic urgency level in the protocol semantic attributes, determine the quality monitoring sensitivity parameters for the physical link; The physical layer transmission quality indicators of the physical link are collected based on the quality monitoring sensitivity parameters, and the link quality status is generated based on the physical layer transmission quality indicators.

5. The terminal according to claim 4, characterized in that, The step of collecting physical layer transmission quality indicators of the physical link based on the quality monitoring sensitivity parameters, and generating the link quality status based on the physical layer transmission quality indicators, includes: The sampling period for the physical link is determined based on the quality monitoring sensitivity parameter, and multiple physical layer transmission quality indicators of the physical link are collected according to the sampling period. Obtain the functional role labels of the data object determined after semantic parsing; Based on the functional role labels, the tolerance threshold for fluctuations in various physical layer transmission quality indicators of the data object is determined. Indicators whose fluctuation amplitude exceeds the corresponding tolerance threshold among the multiple physical layer transmission quality indicators are marked as effective degradation indicators, and the link quality status is generated based on the effective degradation indicators.

6. The terminal according to claim 1, characterized in that, The step of generating a scheduling priority correction amount based on the coupling relationship between protocol semantic attributes and queuing status in the data records includes: Obtain the queuing delay sequence of the data object in multiple consecutive scheduling cycles, and determine the delay change trend of the data object based on the queuing delay sequence; By comparing the latency change trend with the maximum tolerable latency of the data object in the data record, the estimated timeout risk of the data object is obtained. Based on the estimated timeout risk, the dynamic urgency level of the data object in the data record, and the dynamic urgency level of other data objects in the data record that have the same associated device identifier as the data object, the scheduling priority correction amount is generated.

7. The terminal according to claim 1, characterized in that, The process of generating a migration urgency assessment value based on the coupling relationship between protocol semantic attributes and link quality status in the data records includes: The current quality score and quality evolution trend of the physical link are obtained from the link quality status recorded in the data. The current quality score is correlated with the quality evolution trend to determine the basic migration urgency of the physical link; Based on the dynamic urgency level of the data object in the protocol semantic attributes of the data record, the basic migration urgency is modified to generate the migration urgency assessment value.

8. The terminal according to claim 1, characterized in that, The step of determining a link migration execution strategy that matches the protocol semantic attributes based on the migration urgency assessment value and the protocol semantic attributes includes: The migration urgency assessment value is compared with a first urgency threshold and a second urgency threshold to determine the migration urgency level corresponding to the data object, wherein the first urgency threshold is less than the second urgency threshold; Obtain the functional role label of the data object from the protocol semantic attributes of the data record; Based on the migration urgency level and the functional role label, a migration execution strategy matching the data object is determined from multiple preset migration execution strategies.

9. The terminal according to claim 1, characterized in that, The adjustment of the TSN gating list based on the gating time slot reallocation strategy includes: Obtain the target time slot identifier and the identifier of the data object to be inserted from the gated time slot reallocation strategy; Based on the protocol semantic attributes of the data object corresponding to the data object identifier to be inserted, verify whether the time slot parameters corresponding to the target time slot identifier meet the transmission constraints of the data object. When the transmission constraint is met, incremental gating list update information is generated for the target time slot identifier, and the incremental gating list update information is sent to the TSN switch to adjust the TSN gating list.

10. The terminal according to claim 1, characterized in that, The state continuity migration of the physical link carrying the data object based on the link migration execution strategy includes: The target backup link and migration trigger timing are determined based on the link migration execution strategy. Obtain the protocol communication state checkpoint of the data object from the protocol semantic attributes of the data record; When the migration trigger time arrives, a state recovery request carrying the protocol communication state checkpoint is sent through the target backup link to continue the transmission of the data object from the protocol state recorded by the protocol communication state checkpoint on the target backup link.

Images