A remote monitoring and coordination system based on intelligent networking

Abstract

The application discloses a kind of remote monitoring cooperation systems based on intelligent networking, it is related to cooperative control technical field, including: protocol analysis uniformity module, the original protocol message uploaded by analysis heterogeneous terminal, generate standard event metadata;Arrange scheduling module, to the linkage execution sequence generated based on standard event metadata and policy library Injection execution token and issue action instruction, feedback exception along rollback edge trigger remedial sub-process;Parameter intelligent setting module, based on execution trajectory log generation running portrait, based on running portrait obtains policy parameter set;Intelligent cooperative processing module, according to policy parameter set execution digital signature verification, certificate rolling rotation, deterministic transmission control and video adaptive adjustment;The application realizes the automatic response and efficient control for the coupling of multiple types of events in intelligent networking remote monitoring scene by the organic combination of strategy analysis, arrangement scheduling, parameter intelligent setting and intelligent cooperative processing module.

Description

Technical Field

[0001] The present invention relates to the technical field of collaborative control, and more specifically, to a remote monitoring collaboration system based on intelligent networking. Background Art

[0002] In intelligent networking remote monitoring systems, as the digital levels of industrial parks, smart cities, and critical infrastructure continue to improve, the diversity and complexity of remote monitoring objects have increased significantly. The monitoring system not only needs to detect various types of events in real time, but also trigger cross-device and cross-network linkage control processes according to event characteristics to ensure the continuity and security of production operations. Traditional remote monitoring mostly relies on static configuration and manual intervention to execute control instructions and handle exceptions, and it is difficult to balance real-time performance, stability, and security in the face of emergencies, multi-task concurrency, and complex network conditions. In addition, in the face of multi-dimensional strategies such as certificate management, link transmission optimization, and video stream adjustment, existing technologies generally adopt fixed parameter settings and lack the ability of dynamic optimization based on the operating state.

[0003] However, in the actual application of intelligent networking remote monitoring systems, the lack of a unified strategy parsing and topology sorting mechanism for the linkage control process choreography easily leads to disorder in the execution order and untimely abnormal rollback during multi-terminal collaboration; in addition, policy parameters are usually set based on manual experience and cannot be dynamically adjusted in combination with the system operation trajectory, resulting in the disconnection of certificate rotation, deterministic transmission, and video adaptive strategies from the actual operating state.

[0004] In summary, the problems existing in the prior art directly restrict the applicability and reliability of remote monitoring systems in high-security, high-real-time, and high-concurrency scenarios. In view of this, the present invention proposes a remote monitoring collaboration system based on intelligent networking to solve the above problems. Summary of the Invention

[0005] In order to overcome the above defects of the prior art and achieve the above objectives, the present invention provides the following technical solution: A remote monitoring collaboration system based on intelligent networking, comprising:

[0006] A protocol parsing and normalization module, configured to parse and process the original protocol messages uploaded by heterogeneous terminals to obtain standard event metadata;

[0007] An orchestration and scheduling module, based on the standard event metadata and a pre-constructed policy library, generates a linkage control choreography diagram and forms a linkage execution sequence through topology sorting; injects execution tokens into each node in the linkage execution sequence and issues action instructions, records the execution trajectory log, and triggers a remedial sub-process mechanism along the rollback edge when the terminal feedback is timeout or fails.

[0008] The parameter intelligent setting module generates a running profile based on the execution trajectory log, inputs the running profile into the strategy parameter setting model, and outputs a corresponding set of strategy parameters; the set of strategy parameters includes certificate rotation strategy, deterministic transmission strategy, video adaptive strategy, and log sealing parameters;

[0009] The intelligent collaborative processing module performs digital signature verification, rolling rotation, deterministic transmission control, and video adaptive adjustment operations based on a set of strategy parameters, and hashes and stores the execution records and status results generated by the operations.

[0010] Furthermore, the method for performing digital signature verification, rolling rotation, deterministic transmission control, and video adaptive adjustment operations based on a set of policy parameters, and for hash-signing and sealing the execution records and state results generated by the operations, includes:

[0011] When a loading, update, or call request is detected, digital signature verification is performed on the gateway plugin, orchestration script, and device certificate. Objects that pass verification are set to the enabled state, and objects that fail verification are set to the isolated state.

[0012] For device certificates, rolling rotation is performed according to the certificate rotation policy;

[0013] According to the preset tagging rules, the control instructions in the linkage execution sequence are mapped to the deterministic channel. The priority of the control instructions in the deterministic transmission strategy is used as the queue priority of the deterministic channel. The control jitter threshold in the deterministic transmission strategy is used as the maximum allowable delay jitter of the deterministic channel. The upper limit of the control delay is used as the maximum transmission delay constraint of the deterministic channel.

[0014] Based on the bitrate and frame rate ranges in the video adaptive strategy, the video stream bitrate and frame rate are maintained, and the downgrading strategy is executed when the remaining bandwidth is lower than the downgrading bandwidth threshold.

[0015] The execution records and status results generated during the execution process are summarized into an operation log file. The operation log file is hash-signed and sealed, and stored according to the storage path in the log sealing parameters.

[0016] Furthermore, the method for implementing rolling rotation based on the certificate rotation strategy includes:

[0017] The rotation scheduler calculates the next rotation time based on the most recent effective time and the rotation cycle; it then calculates the pre-distribution start time based on the next rotation time and the overlapping effective window.

[0018] The parallel time window is constructed by taking the pre-distribution start time as the starting point of the parallel time window and the next rotation time as the ending point of the parallel time window.

[0019] After entering the parallel time window, the parameter validator sends a new device certificate to the target device, and the security verifier performs a handshake verification on the new device certificate;

[0020] If the verification is successful, the new device certificate will be marked as valid, and the old device certificate will remain valid until the next rotation time. When the next rotation time arrives, the parameter activator will switch the old device certificate to the new device certificate, and the security activator will revoke the old device certificate, ending the current rotation.

[0021] If the verification fails, the invalid mark of the new device certificate is written back, a failure rollback is executed, the use of the new device certificate is terminated, the old device certificate remains valid, and the current number of failures is recorded. If the current number of failures is less than or equal to the number of failure rollbacks, the rotation scheduler will re-formulate the rotation plan in the next cycle. If the current number of failures is greater than the number of failure rollbacks, the security protector will put the corresponding device into the isolation list and trigger the rollback edge and remediation sub-process.

[0022] Furthermore, the method for implementing the downshift strategy includes:

[0023] Adjust the bitrate to the lower limit of the bitrate range, and adjust the frame rate to the lower limit of the frame rate range;

[0024] After the bitrate and frame rate are adjusted, the link bandwidth status is continuously monitored. If the remaining bandwidth exceeds the downgrade bandwidth threshold and the duration exceeds the preset duration threshold, the scheduler will automatically call the upgrade process to restore the frame rate and bitrate to the range.

[0025] Furthermore, methods for generating runtime profiles based on execution trajectory logs include:

[0026] Calculate the execution time, waiting time, and confirmation time for each node in the execution trajectory log, and record them as the first profile parameters;

[0027] Calculate the trigger delay for each connection in the execution trajectory log and record it as the second profile parameter;

[0028] For nodes with rollback edges in the execution trajectory log, calculate the rollback trigger delay, the execution time of the remedial subprocess, and the time interval from the completion of the remediation to the recovery of the main process, and record them as the third profile parameter;

[0029] The success rate, rollback trigger rate, and recovery success rate of the instructions are calculated from the number of successful nodes, rollback nodes, successful recovery nodes, and total number of nodes counted in the execution trajectory log, and recorded as the fourth profile parameter.

[0030] The first profile parameter, the second profile parameter, the third profile parameter, and the fourth profile parameter are collected and encapsulated according to a preset data structure to construct a running profile.

[0031] Furthermore, the method for generating a linkage control orchestration diagram and forming a linkage execution sequence through topological sorting includes:

[0032] The policy parser reads the event type-event level-device identifier ternary semantic structure from the standard event metadata and retrieves the matching business rules from the pre-built policy library; the event type-event level-device identifier ternary semantic structure and the business rules are then used to construct the retrieval results.

[0033] The search results are parsed into node-edge descriptions, and a linkage control orchestration diagram with start node, condition gateway, parallel convergence and end node is generated according to the specifications of the business process modeling language.

[0034] The scheduler assigns a corresponding session identifier to the linkage control orchestration diagram, performs topological sorting on the linkage control orchestration diagram, and obtains the linkage execution sequence.

[0035] Furthermore, the method of injecting execution tokens and issuing action instructions to each node in the coordinated execution sequence, recording execution trajectory logs, and triggering a remedial sub-process mechanism along the rollback edge when the terminal reports a timeout or failure includes:

[0036] An execution token is injected into the node of the linked execution sequence, and an action command is issued to the corresponding target terminal according to the linked execution sequence; the execution token carries token operation parameters, including node start time, timeout time, and retry count.

[0037] The scheduler writes the node status, edge status, and execution token migration records of the linkage control orchestration graph into the execution trajectory log, and then synchronizes the execution trajectory log to the security protector after binding it with the session identifier.

[0038] The scheduler creates a status checker for nodes that enter the running state. The status checker continuously collects terminal feedback. If the execution token of a node does not receive a successful confirmation before the timeout period, the rollback edge connected to the corresponding node in the linkage control orchestration diagram is immediately triggered, and the remedial subprocess mechanism is invoked.

[0039] Furthermore, the remedial sub-process mechanism is as follows:

[0040] For each node with a rollback flag, trigger conditions are pre-bound in the policy library; the trigger conditions include action execution failure, timeout threshold reaching, and device offline.

[0041] Draw an independent sub-flowchart for each triggering condition. The sub-flowchart includes a start event, a remedial action node, and an end event. The sub-flowchart is connected to the rollback edge.

[0042] When the scheduler detects that any triggering condition is met during runtime, it suspends the subsequent execution of the main process at the current node and transfers the execution token to the corresponding remedial subprocess for execution.

[0043] After the sub-process is completed, the output status value determines whether to return to the main process to continue execution, terminate the current session directly, or recursively trigger a remedial sub-process.

[0044] Furthermore, the method for obtaining the standard event metadata includes:

[0045] The protocol mapping gateway receives raw protocol messages from heterogeneous terminals;

[0046] The protocol mapping gateway calls a parser that matches the protocol type based on the protocol type of the original protocol message, and checks the handshake validity, payload integrity and sequence continuity through the protocol state machine; it extracts command type, alarm level and device identification key fields from the original protocol message and aggregates them to form a protocol field set;

[0047] The protocol mapping gateway calls a field-semantic lookup table to map each protocol field in the protocol field set to a three-element semantic structure of event type-event level-device identifier, thus obtaining a standard semantic record; the current timestamp is appended to the standard semantic record and encapsulated together with the index pointer pointing to the original protocol message to form an event metadata draft;

[0048] The protocol mapping gateway performs sequence number continuity detection and circular cache rearrangement on the event metadata draft, removes duplicate frames and identifies lost frames, and generates standard event metadata.

[0049] Furthermore, the heterogeneous terminal includes a camera, sensors, and a voice terminal.

[0050] Compared with existing technologies, the remote monitoring and collaborative system based on intelligent network proposed in this invention has the following technical effects and advantages:

[0051] This invention achieves automated response and efficient control for coupled multi-type events in intelligent connected remote monitoring scenarios by organically combining strategy parsing, orchestration scheduling, intelligent parameter setting, and intelligent collaborative processing modules.

[0052] Specifically, this invention utilizes standard event metadata and a pre-built strategy library to generate a linkage control orchestration diagram. Through topology sorting and execution token mechanisms, it ensures the orderly execution of multiple terminals and tasks, as well as anomaly rollback handling. Based on execution trajectory logs, it constructs a runtime profile and dynamically sets the strategy parameter set using a strategy parameter setting model, enabling device certificate rotation, deterministic transmission, and video adaptive strategies to highly match the actual operating state. During the execution phase, through digital signature verification, rolling rotation, jitter and latency constraint control, video bitrate and frame rate adjustment, and the dynamic application of downgrading strategies, it significantly improves the real-time performance, stability, and security of the remote monitoring and collaboration system under multi-task concurrency and complex network conditions. Simultaneously, this invention ensures the integrity and traceability of operation records through hash signatures and log sealing mechanisms, thereby meeting the comprehensive requirements of security, reliability, and efficient collaboration. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of a remote monitoring and collaborative system based on intelligent network connectivity according to Embodiment 1 of the present invention;

[0054] Figure 2 This is a flowchart of a remote monitoring and collaborative method based on intelligent network connectivity according to Embodiment 2 of the present invention;

[0055] Figure 3 This is a flowchart of the method for performing rolling rotation based on a certificate rotation strategy according to Embodiment 1 of the present invention;

[0056] Figure 4 This is a flowchart of the method for generating a runtime profile based on execution trajectory logs according to Embodiment 1 of the present invention. Detailed Implementation

[0057] The technical solutions of the embodiments of the present invention will be described in detail, clearly, and completely below with reference to the accompanying drawings. It should be particularly noted that the specific embodiments described below are only for better illustrating and explaining the technical solutions of the present invention, and are intended to enable those skilled in the art to better understand and implement the present invention, and should not be construed as limiting the scope of protection of the present invention. Without departing from the spirit and substance of the present invention, those skilled in the art can modify, adjust, or make equivalent substitutions based on the content disclosed in the present invention, and these should all be considered within the scope of protection of the present invention.

[0058] Example 1:

[0059] Please see Figure 1 As shown in the figure, this embodiment discloses a remote monitoring and collaborative system based on intelligent network connectivity, including a protocol parsing and normalization module, an orchestration and scheduling module, a parameter intelligent setting module, and an intelligent collaborative processing module. Each module is connected by wired and / or wireless means to realize data transmission.

[0060] The protocol parsing and normalization module is used to parse and process the original protocol messages uploaded by heterogeneous terminals to obtain standard event metadata; the heterogeneous terminals include cameras, sensors and voice terminals.

[0061] The methods for obtaining the standard event metadata include:

[0062] The protocol mapping gateway receives raw protocol messages from heterogeneous terminals;

[0063] The protocol mapping gateway calls a parser that matches the protocol type based on the protocol type of the original protocol message, and checks the handshake validity, payload integrity and sequence continuity through the protocol state machine; it extracts command type, alarm level and device identification key fields from the original protocol message and aggregates them to form a protocol field set;

[0064] The protocol mapping gateway calls a field-semantic lookup table to map each protocol field in the protocol field set to a three-element semantic structure of event type-event level-device identifier, thus obtaining a standard semantic record; the current timestamp is appended to the standard semantic record and encapsulated together with the index pointer pointing to the original protocol message to form an event metadata draft;

[0065] The protocol mapping gateway performs sequence number continuity detection and circular cache rearrangement on the event metadata draft, removes duplicate frames and identifies lost frames, and generates standard event metadata.

[0066] It should be noted that the standard event metadata comes with a transaction token, which can accurately locate each link in the processing chain during auditing or backtracking. The protocol mapping gateway pushes the standard event metadata to the system event bus using a zero-copy method. At this point, the original protocol messages have been transformed layer by layer into structured, low-latency, and traceable unified event metadata, providing a reliable data foundation for subsequent event orchestration, timing alignment, and security assurance.

[0067] It should be noted that the remote monitoring and collaborative system involved in this invention supports multiple communication protocols to achieve unified access and semantic parsing of heterogeneous terminal devices, such as cameras, sensors, and voice devices. For example, in a preferred embodiment of this invention, the protocol types include, for instance, ONVIF, GB / T 28181, SIP, and MQTT protocols.

[0068] The ONVIF protocol is an open interface standard for network video devices, widely used for access control and media streaming of video terminals such as surveillance cameras and encoders. Based on a Web Services architecture, it uses SOAP / XML format to implement functions such as device discovery, configuration management, event subscription, and video stream access, making it suitable for typical video surveillance scenarios.

[0069] GB / T 28181 is a protocol issued by the Standardization Administration of China, specifying the "Technical Requirements for Information Transmission, Exchange, and Control of Security Video Surveillance Network Systems." Its core objective is to achieve a unified access and reporting mechanism for video surveillance systems in fields such as public security and urban management. Based on the SIP and RTP technology framework, this protocol defines functional processes such as device registration, directory lookup, real-time video playback, video recording playback, and alarm uploading.

[0070] SIP (Signaling over IP) is a signaling protocol primarily used to establish, manage, and terminate multimedia session connections such as voice and video. Its protocol mechanism is based on a request / response model and is widely used in VoIP phones, video conferencing systems, and alarm intercom terminals. In this invention, the SIP protocol is mainly responsible for establishing the control link of voice communication devices, such as initiating calls, answering calls, and hanging up.

[0071] MQTT is a lightweight publish / subscribe messaging protocol widely used for data reporting and command issuance by IoT terminals. This protocol uses TCP connections and features low bandwidth, low power consumption, and high real-time performance, making it particularly suitable for high-frequency data communication between sensor devices and edge computing nodes. In this invention, MQTT primarily handles the data uploading responsibility for environmental monitoring, status detection, and other sensing devices, providing the monitoring system with refined and continuous operational status input.

[0072] In summary, ONVIF, GB / T 28181, SIP, and MQTT respectively cover four typical communication scenarios: video acquisition, video networking, voice interaction, and sensor data uploading. This invention, through a unified access and semantic mapping mechanism, effectively breaks down the barriers between these protocols, enabling collaborative sensing and control across protocol terminals, and providing technical support for building an efficient, secure, and scalable remote monitoring and collaborative system.

[0073] For example, in this invention, the field-semantic lookup table is shown in Table 1.

[0074] Table 1 Fields - Semantic Reference Table

[0075]

[0076] The orchestration and scheduling module generates a linkage control orchestration diagram based on standard event metadata and a pre-built policy library, and forms a linkage execution sequence through topological sorting; it injects execution tokens into each node in the linkage execution sequence and issues action instructions, records the execution trajectory log, and triggers a remedial sub-process mechanism along the rollback edge when the terminal reports a timeout or failure.

[0077] Methods for generating a linkage control orchestration diagram and forming a linkage execution sequence through topological sorting include:

[0078] The policy parser reads the event type-event level-device identifier ternary semantic structure from the standard event metadata and retrieves matching business rules from the pre-built policy library; it constructs the event type-event level-device identifier ternary semantic structure and business rules into the retrieval results; specifically, the retrieval results include the event type-event level-device identifier ternary semantic structure and the business rules that match the event type-event level-device identifier ternary semantic structure.

[0079] The search results are parsed into node-edge descriptions, and a linkage control orchestration diagram with start nodes, condition gateways, parallel merging, and end nodes is generated according to the specifications of the business process modeling language. The business process modeling language is a graphical process modeling language primarily used to describe the process logic of "who does what under what conditions" within an enterprise or system, and is applied in scenarios such as process automation, business orchestration, and process engine execution.

[0080] The scheduler assigns a corresponding session identifier to the linkage control orchestration diagram, performs topological sorting on the linkage control orchestration diagram, and obtains the linkage execution sequence.

[0081] The methods for injecting execution tokens and issuing action instructions to each node in the coordinated execution sequence, recording execution trajectory logs, and triggering a remedial sub-process mechanism along the rollback edge when the terminal reports a timeout or failure include:

[0082] An execution token is injected into each node in the linked execution sequence, and action instructions are issued to the corresponding target terminal according to the linked execution sequence. The execution token carries token operation parameters, including node start time, timeout time, and retry count. Injecting execution tokens into nodes ensures that each node can be independently tracked in its internal state machine. For example, in this invention, the topology sorting algorithm can be either Depth-First Search (DFS) or Breadth-First Search (BFS).

[0083] The scheduler writes the node states, edge states, and execution token migration records of the linkage control orchestration diagram into the execution trajectory log, and then binds the execution trajectory log with the session identifier to synchronize it to the security protector. This ensures that multimodal data can be accurately reproduced according to the linkage control orchestration diagram during replay, forensics, or AI linkage analysis, thereby achieving automation, determinism, and traceability of event-level linkage and realizing highly reliable multi-terminal collaborative control in remote monitoring scenarios.

[0084] The scheduler creates a status checker for nodes entering the running state. The status checker continuously collects terminal feedback. If the execution token of a node does not receive a successful confirmation before the timeout period, the rollback edge connected to the corresponding node in the linkage control orchestration diagram is immediately triggered, and the remedial sub-process mechanism is invoked. At the same time, an exception label is recorded in the diagram to prevent subsequent nodes from continuing to execute in an unknown state. If the node completes execution successfully, the execution token is released and the next node is run, ensuring that the process proceeds smoothly without exceptions.

[0085] The strategy library is constructed as follows:

[0086] During the deployment phase, based on historical alarm data, industry standards, and on-site operation and maintenance experience, a set of triggering conditions covering the three-element semantic structure of "event type - event level - device identifier" is determined.

[0087] Each triggering condition is assigned a corresponding strategy number and stored in the main strategy index table using a hash index, forming a strategy retrieval entry point.

[0088] The business scenarios corresponding to the triggering conditions are categorized to obtain corresponding policy types. These policy types include common layer policies, priority layer policies, and customized layer policies. Common layer policies address general alarms for all devices, priority layer policies address pre-defined critical security events, and customized layer policies address pre-defined specific devices or areas. For example, critical security events might include a rapid rise in temperature inside a server rack or continuous impact sounds; specified devices might include explosion-proof PTZ cameras, biometric access controllers, and multi-gas composite sensors; and specified areas might include hazardous materials warehouses, substation control rooms, and data center server rooms.

[0089] After all policy types pass consistency checks and conflict detection, the main policy index table and policies at each level are persistently written to the policy library. The completed policy library supports both exact matching and wildcard or range matching, providing a data foundation for the policy parser's fast retrieval.

[0090] The method for constructing the business rules includes:

[0091] For each policy number, obtain the policy type corresponding to the policy number, and obtain the list of linked devices involved in the policy type; the list of linked devices includes, for example, cameras, voice terminals, and access control devices.

[0092] The logical dependencies between each linked device are described by node-connection relationships, forming a linkage action template. Each node in the linkage action template includes an action type, timeout threshold, number of retries and rollback flag, and each connection includes parallel execution constraints or sequential execution constraints.

[0093] Business process modeling is used to compile the linked action templates into an executable flowchart data structure; the flowchart data structure and the corresponding strategy number are used to form business rules, and the business rules are written into the business rule table. The business rule table and the strategy number in the strategy master index table form a one-to-one mapping relationship.

[0094] The business rule table is persistently written to the policy library. That is, the complete business rule entity is loaded into the policy library, and together with the corresponding trigger conditions, they form a rule orchestration unit that can be retrieved by the policy parser, laying the foundation for the automatic generation of subsequent linkage control processes.

[0095] It should be noted that when describing the logical dependencies between various linked devices using node-connection relationships, a node represents a specific linked action, and the nodes form an execution path through directed connections.

[0096] Action type refers to the actual control behavior or operation target corresponding to this node. Examples include: starting a video preview, initiating a voice call, issuing a door access control closing command, or switching to a backup camera. The action type determines the control interface and target device type that the system needs to call, and is the basic unit definition for process execution.

[0097] The timeout threshold specifies the maximum allowed time for the system to wait for a response or completion from the target device after issuing an action command. If no successful execution feedback is received within the maximum allowed time, the action is considered a failure. The timeout threshold is used to ensure the determinism of process execution and prevent the entire process from being blocked due to device disconnection or malfunction.

[0098] The retry count refers to the maximum number of times the system allows an action command to automatically retrieve after the initial execution failure or timeout. For example, a retry count of 2 means the system can try a maximum of 3 times, i.e., 1 initial attempt plus 2 retries. The retry mechanism is used to improve the reliability of critical actions in unstable link scenarios.

[0099] The rollback flag indicates whether a predefined remedial procedure or rollback operation should be triggered when the node fails. For example, if the "close access control" action fails and is marked as rollbackable, the system will automatically initiate alternative procedures such as "reverse alarm" or "enable physical isolation." The rollback flag plays a crucial role in ensuring the fault tolerance of the linkage process.

[0100] Parallel execution constraints mean that multiple successor nodes pointed to by a connection can start simultaneously without waiting for the previous node to complete. For example, a "high temperature alarm" node can simultaneously trigger two actions: "camera rotation and focusing" and "voice warning broadcast," improving response efficiency.

[0101] Sequential execution constraints mean that the next node connected to a connection can only start after the current node has successfully completed, ensuring that actions are executed in a strict logical order. For example, the "remote door opening" operation can only be executed after "identifying the person entering or exiting" to prevent accidental triggering.

[0102] It's important to note that a condition gateway is a process control node used to branch execution paths according to business rules in a linkage control orchestration diagram. When a linkage control process reaches a condition gateway, the system reads the context information input by the current node, including key fields such as event type, event level, and device identifier, and then evaluates the Boolean condition expressions attached to the gateway. Only connections where the condition evaluates to "true" are activated and continue to the downstream node; the rest are ignored. With the help of condition gateways, the system can automatically select differentiated linkage strategies based on different levels of urgency or device location within the same event triggering scenario, thereby achieving conditional branch execution and mutually exclusive path control of the process. At the implementation level, the condition gateway is parsed by the scheduler's built-in decision engine. The decision logic is bound to business rules during the flowchart data structure generation stage, ensuring that branch selection can be completed without external intervention at runtime.

[0103] Parallel convergence is a type of synchronization node used in the linkage control orchestration diagram to reconverge multiple concurrent execution paths previously split by the parallel gateway into a single main line. The parallel convergence node waits for all its upstream concurrent branches to report "execution complete" status before triggering downstream nodes to continue execution. If any concurrent branch is not yet complete or a timeout occurs, the convergence node will remain blocked. Through parallel convergence, the system can ensure that multiple independent actions are successfully completed before proceeding to subsequent steps. These independent actions include, for example, simultaneously rotating multiple cameras, simultaneously issuing voice broadcasts and SMS alarms, triggering critical control commands, or archiving and reviewing data. The runtime logic of parallel convergence is implemented through the scheduler's token management mechanism. The scheduler assigns an independent token to each concurrent path, and the node only releases a new token to advance the process when all tokens arrive at the convergence node, ensuring the integrity and consistency of the coordinated actions.

[0104] The remedial sub-process mechanism is as follows:

[0105] For each node with a rollback flag, trigger conditions are pre-bound in the policy library; the trigger conditions include action execution failure, timeout threshold reaching, and device offline.

[0106] In the modeler for business process modeling and annotation specifications, an independent sub-flowchart is drawn for each triggering condition. The sub-flowchart includes a start event, a remedial action node, and an end event. The sub-flowchart is connected to the rollback edge. The sub-flowcharts after modeling are compiled into an executable format.

[0107] When the scheduler detects that any triggering condition is met during runtime, it suspends the subsequent execution of the main process at the current node and transfers the execution token to the corresponding remedial subprocess for execution.

[0108] After a sub-process is completed, the system selects to return to the main process to continue execution, terminate the current session directly, or recursively trigger a remedial sub-process based on the output status value, thereby forming a closed-loop fault-tolerant control.

[0109] For example, when the output status value is 1, the main process is returned to continue execution; when the output status value is 2, the current session is terminated directly; and when the output status value is 3, the remedial sub-process mechanism is recursively triggered.

[0110] For example, in this invention, taking the "second dialing" remedial sub-process as an example: when the "initiate voice call" node does not receive a connection confirmation from the called party within 10 seconds and the retries are exhausted, the scheduler immediately triggers the sub-process. The sub-process generates a new call request from the starting node, calls the SIP terminal to perform a second automatic dialing; if it fails again, it enters the backup contact list and sequentially tries the authorized second contact; if the dialing is successful, the sub-process outputs a "success" status token at the end event, allowing the main process to continue along the normal path. Taking the "backup camera switching" remedial sub-process as another example: when the main camera's pan-tilt-zoom fails to rotate and returns an error code, the sub-process starting node immediately searches the list of cameras marked "rollback" in the same area, selects the one with the highest overlap of viewing angles, and performs video stream switching; after the switching is completed, it returns a "successful replacement" status to the main process, ensuring that the monitoring screen is not interrupted. Through the above-described methods and examples, the remedial sub-process mechanism of this invention can automatically intervene when critical actions fail, significantly improving the fault self-healing capability and operational reliability of the remote monitoring collaborative system.

[0111] The intelligent parameter setting module generates a runtime profile based on the execution trajectory log, inputs the runtime profile into the strategy parameter setting model, and outputs a corresponding set of strategy parameters. The set of strategy parameters includes a certificate rotation strategy, a deterministic transmission strategy, a video adaptive strategy, and log archiving parameters. The certificate rotation strategy includes a rotation period, an overlapping effective window, and a failure rollback count. The deterministic transmission strategy includes controlling command priority, controlling jitter threshold, and controlling latency upper limit. The video adaptive strategy includes a bitrate range, a frame rate range, and a downsampling bandwidth threshold. The log archiving parameters include a storage path.

[0112] like Figure 4 As shown, the methods for generating runtime profiles based on execution trajectory logs include:

[0113] For each node in the execution trajectory log, calculate the execution duration, waiting duration, and confirmation duration, and record them as the first profile parameter; the execution duration is the node departure time minus the node entry time; the waiting duration is the time between the arrival of the execution token and the issuance of the action; the confirmation duration is the time between the issuance of the action and successful confirmation.

[0114] The trigger delay is calculated for each connection in the execution trajectory log and recorded as the second profile parameter; the trigger delay is the duration from the start to the end of the connection.

[0115] For nodes with rollback edges in the execution trajectory log, calculate the rollback trigger delay, the execution time of the remedial subprocess, and the time interval from the completion of the remediation to the recovery of the main process, and record them as the third profile parameter;

[0116] The success rate of instructions, the rollback trigger rate, and the recovery success rate are calculated from the number of successful nodes, the number of rollback nodes, the number of successful recovery nodes, and the total number of nodes counted in the execution trajectory log, and are recorded as the fourth profile parameter. The success rate of instructions is obtained by dividing the number of successful nodes by the total number of nodes, the rollback trigger rate is obtained by dividing the number of rollback nodes by the total number of nodes, and the recovery success rate is obtained by dividing the number of successful recovery nodes by the number of rollback nodes.

[0117] The first profile parameter, the second profile parameter, the third profile parameter, and the fourth profile parameter are collected and encapsulated according to a preset data structure to construct a running profile.

[0118] It should be noted that the data structure may include, for example, a hierarchical JSON structure and a key-value index structure. Specifically, the hierarchical JSON structure uses the session identifier as the top-level key, and sets the first profile parameter, second profile parameter, third profile parameter, and fourth profile parameter as four sub-objects. Each sub-object stores the corresponding duration, latency, and statistical values ​​according to the node identifier or connection relationship identifier. The hierarchical JSON structure has the advantages of strong self-description and the ability to quickly locate fields directly through the key path. The key-value index structure uses "session identifier + parameter type + element identifier" as a composite key, stores the values ​​of each profile parameter as values, and establishes a secondary index in the key space, supporting fast retrieval by session, by node, or by parameter type. The key-value index structure is suitable for high-concurrency, low-latency real-time strategy calculation scenarios.

[0119] The training method for the policy parameter setting model includes:

[0120] A strategy parameter setting dataset is pre-constructed, which includes CL group strategy parameter setting data and the strategy parameter set corresponding to the CL group strategy parameter setting data, where CL is a positive integer; the strategy parameter setting data includes running profiles; the strategy parameter setting dataset is divided into a training set and a validation set, the training set is used for learning the parameters of the strategy parameter setting model, and the validation set is used to monitor the generalization performance and overfitting of the strategy parameter setting model in real time.

[0121] A node-output graph neural network architecture is adopted as the policy parameter setting model. The policy parameter setting data is standardized and vectorized before being input into the node-output graph neural network architecture, which consists of an input layer, hidden layers, and an output layer. Each hidden layer uses a non-linear activation function to extract features, and the output layer uses a sigmoid activation function to obtain the probability distribution corresponding to each policy parameter set. Finally, the policy parameter set corresponding to the highest probability is taken as the prediction result of the policy parameter setting model. During training, the cross-entropy loss function is used as the optimization objective, and a gradient descent-type optimization algorithm is used to update the network weights. An early stopping strategy is set: when the prediction accuracy on the validation set reaches or exceeds a preset threshold, the policy parameter setting model is determined to have converged and training is terminated.

[0122] It should be noted that the relationship between the operational profile and the policy parameter set lies in the fact that the operational profile comprehensively depicts the linkage execution performance, link status, and recovery efficiency under a specific session. This information directly reflects the support effect of the current policy parameters on system operation goals such as low latency, high reliability, and security compliance. The first, second, third, and fourth profile parameters included in the operational profile all have corresponding influence relationships with the setting values ​​of the certificate rotation policy, deterministic transmission policy, video adaptive policy, and log sealing parameters.

[0123] For example, when the running profile shows that the control command trigger latency is close to or exceeds the predetermined control latency limit, it indicates that the current control command priority or control jitter threshold setting is insufficient. For instance, the control latency limit can be set to 50ms. When the rollback trigger rate or the execution time of the remedial subprocess is too high, the rotation cycle or overlapping effective window needs to be adjusted to reduce the execution failure rate caused by key expiration. For instance, the upper limit corresponding to the rollback trigger rate can be set to 8%, meaning that a rollback trigger rate exceeding 8% is considered too high. The upper limit corresponding to the execution time of the remedial subprocess can be set to 3s, meaning that a remedial subprocess execution time exceeding 3s is considered too high. When the trigger latency or the time interval between remediation completion and main process recovery is too high, the bitrate range, frame rate range, or bandwidth reduction threshold needs to be adjusted to optimize link utilization efficiency. For instance, the upper limit corresponding to the trigger latency can be set to 120ms, meaning that a trigger latency exceeding 120ms is considered too high. The upper limit corresponding to the main process recovery time interval can be set to 1.5s, meaning that a main process recovery time interval exceeding 1.5s is considered too high.

[0124] Based on the operational profile, the strategy parameter setting model can learn the mapping relationship between different operational states and the optimal strategy parameter combination from historical data through pattern recognition and regression prediction. In a new session, it can deduce the set of strategy parameters that are suitable for the current operational environment based on the real-time generated operational profile, thereby achieving adaptive optimization and closed-loop adjustment of parameters, and ensuring that the system continuously meets the predetermined performance and security goals in a variable environment.

[0125] It should be further noted that, for example, in the preferred embodiment of the present invention, two scenarios—an industrial park scenario and a high-security, low-latency scenario—are used to fully disclose the implementable configuration of the policy parameter set. During specific deployment, the policy parameter setting model adaptively adjusts according to the operational profile.

[0126] In the industrial park scenario, this corresponds to medium concurrency and general security levels. For the certificate rotation strategy, the rotation period is set to 72 hours, the overlap window is set to 15 minutes, and the number of failure rollbacks is set to 2. For the deterministic transmission strategy, the control command priority is set to 6, with priority levels ranging from 0 to 7 (higher values ​​indicate higher priority). The jitter threshold is set to ±5ms, and the upper limit of control latency is set to 50ms. For the video adaptive strategy, the bitrate range is set to 0.8-3.0Mbps (inclusive), the frame rate range is set to 12-25fps (inclusive), and the bandwidth downgrading threshold is set to 1.2Mbps (triggered when the remaining link bandwidth is less than 1.2Mbps). For log archiving parameters, the storage path is set to worm: / / cluster / audit / std / .

[0127] In high-security, low-latency scenarios, corresponding to critical areas and stringent service level agreements, the certificate rotation strategy is configured as follows: Rotation period is set to 24 hours, overlap window to 30 minutes, and failure rollback count to 3. For deterministic transmission, control command priority is set to 7, jitter threshold to ±3ms, and maximum latency to 30ms. For adaptive video, bitrate range is set to 0.5-1.5Mbps (inclusive), frame rate range to 10-20fps (inclusive), and bandwidth downgrading threshold to 0.8Mbps (triggered when remaining link bandwidth < 0.8Mbps). For log archiving parameters, the storage path is set to worm: / / cluster / audit / secure / .

[0128] The intelligent collaborative processing module performs digital signature verification, rolling rotation, deterministic transmission control, and video adaptive adjustment operations based on a set of strategy parameters, and hashes and stores the execution records and status results generated by the operations.

[0129] The method for performing digital signature verification, rolling rotation, deterministic transmission control, and video adaptive adjustment operations based on a set of policy parameters, and for hashing and storing the execution records and state results generated by the operations, includes:

[0130] When a loading, update, or call request is detected, the security protector performs digital signature verification on the gateway plugin, orchestration script, and device certificate. Objects that pass verification are set to the enabled state, while objects that fail verification are set to the isolated state.

[0131] For device certificates, rolling rotation is performed according to the certificate rotation policy;

[0132] According to the preset tagging rules, the control commands in the linkage execution sequence are mapped to the deterministic channel, and deterministic transmission control is performed on the deterministic channel. The queue priority of the deterministic channel is set to the priority of the control commands in the deterministic transmission strategy; non-control commands retain their original queue priorities; the control jitter threshold in the deterministic transmission strategy is used as the maximum allowable delay jitter of the deterministic channel, and the upper limit of the control delay in the deterministic transmission strategy is used as the maximum transmission delay constraint of the deterministic channel, thereby ensuring the real-time performance and stability of the control commands during transmission.

[0133] It should be noted that, for example, in this invention, the tagging rule is as follows: when the event level in the standard event metadata is ≥L3 and the event type belongs to one of {emergency shutdown, valve shutdown, access control lock}, the control commands in the linkage execution sequence are mapped to the deterministic channel according to the port P2Q mapping rule. The P2Q mapping rule, i.e., the port-to-queue mapping rule, refers to the process by which a network device, after receiving a data packet from an ingress port, maps the packet to the corresponding output queue based on the priority identifier carried by the packet, such as the Ethernet PCP value, the IP packet DSCP value, or other service classification identifiers. The P2Q mapping rule is usually pre-set in the port configuration of switches, gateways, or routers. Different priority identifiers correspond to different queue numbers, thereby achieving traffic classification and differentiated scheduling.

[0134] For video adaptive adjustment operations, based on the bitrate and frame rate ranges set in the video adaptive strategy, the bitrate of the video stream is kept within the bitrate range, and the frame rate of the video stream is kept within the frame rate range; during the monitoring of remaining bandwidth of the link, if it is detected that the current remaining bandwidth is lower than the downgrading bandwidth threshold in the video adaptive strategy, the downgrading strategy is executed.

[0135] The execution records and status results generated from digital signature verification, rolling rotation, deterministic transmission control, and video adaptive adjustment operations are summarized to form an operation log file. This operation log file is then hash-signed and sealed to ensure its immutability and traceability during transmission and storage. After sealing, the operation log file is stored according to the storage path specified in the log sealing parameters.

[0136] like Figure 3 As shown, the methods for performing rolling rotation based on the certificate rotation strategy include:

[0137] The rotation scheduler calculates the next rotation time based on the most recent effective time and the rotation cycle; it then calculates the pre-distribution start time based on the next rotation time and the overlapping effective window.

[0138] The parallel time window is constructed by taking the pre-distribution start time as the starting point of the parallel time window and the next rotation time as the ending point of the parallel time window.

[0139] After entering the parallel time window, the parameter validator sends a new device certificate to the target device, and the security verifier performs a handshake verification on the new device certificate;

[0140] If the verification is successful, the new device certificate will be marked as valid, and the old device certificate will remain valid until the next rotation time. When the next rotation time arrives, the parameter activator will switch the old device certificate to the new device certificate, and the security activator will revoke the old device certificate, ending the current rotation.

[0141] If the verification fails, the invalid mark of the new device certificate is written back, a failure rollback is executed, the use of the new device certificate is terminated, the old device certificate remains valid, and the current number of failures is recorded. If the current number of failures is less than or equal to the number of failure rollbacks, the rotation scheduler will re-formulate the rotation plan in the next cycle. If the current number of failures is greater than the number of failure rollbacks, the security protector will put the corresponding device into the isolation list and trigger the rollback edge and remediation sub-process.

[0142] The method for calculating the next rotation time includes:

[0143] LHSJ next =SXSJ last +LHZQ;

[0144] Among them, LHSJ next SXSJ last The most recent effective date is indicated by LHZQ, which represents the rotation period.

[0145] The method for calculating the pre-distribution start time includes:

[0146] YFF start =LHSJ next -SJCK overlap ;

[0147] Among them, YFF start SJCK is the pre-distribution start time. overlap This is an overlapping window.

[0148] The method for implementing the downshift strategy includes:

[0149] Adjust the bitrate to the lower limit of the bitrate range, and adjust the frame rate to the lower limit of the frame rate range;

[0150] After the bitrate and frame rate are adjusted, the link bandwidth status is continuously monitored. If the remaining bandwidth exceeds the downgrade bandwidth threshold and the duration exceeds the preset duration threshold, the scheduler will automatically call the upgrade process to restore the frame rate and bitrate to the range.

[0151] Example 2:

[0152] Please see Figure 2 As shown, this embodiment provides a remote monitoring and collaborative method based on intelligent network connectivity, including:

[0153] Parse and process the raw protocol messages uploaded by heterogeneous terminals to obtain standard event metadata;

[0154] Based on standard event metadata and a pre-built strategy library, a linkage control orchestration diagram is generated and a linkage execution sequence is formed by topological sorting; execution tokens are injected into each node in the linkage execution sequence and action instructions are issued, and the execution trajectory log is recorded. When the terminal reports a timeout or failure, a remedial sub-process mechanism is triggered along the rollback edge.

[0155] A runtime profile is generated based on the execution trajectory log, and the runtime profile is input into the strategy parameter setting model to output the corresponding strategy parameter set; the strategy parameter set includes certificate rotation strategy, deterministic transmission strategy, video adaptive strategy and log sealing parameters;

[0156] Based on the set of policy parameters, digital signature verification, rolling rotation, deterministic transmission control, and video adaptive adjustment operations are performed, and the execution records and status results generated by the operations are hashed, signed, and sealed.

[0157] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0158] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A remote monitoring and collaborative system based on intelligent network connectivity, characterized in that, include: The protocol parsing and normalization module is used to parse and process the original protocol messages uploaded by heterogeneous terminals to obtain standard event metadata; The orchestration and scheduling module generates a linkage control orchestration diagram based on standard event metadata and a pre-built strategy library, and forms a linkage execution sequence through topological sorting; it injects execution tokens into each node in the linkage execution sequence and issues action instructions, records the execution trajectory log, and triggers a remedial sub-process mechanism along the rollback edge when the terminal reports a timeout or failure. The parameter intelligent setting module generates a running profile based on the execution trajectory log, inputs the running profile into the strategy parameter setting model, and outputs a corresponding set of strategy parameters. The set of strategy parameters includes a certificate rotation strategy, a deterministic transmission strategy, a video adaptive strategy, and log sealing parameters. The deterministic transmission strategy includes control command priority, control jitter threshold, and control latency upper limit. The intelligent collaborative processing module performs digital signature verification, rolling rotation, deterministic transmission control, and video adaptive adjustment operations based on a set of strategy parameters, and hash-signs and stores the execution records and status results generated by the operations. Specifically, it includes: When a loading, update, or call request is detected, digital signature verification is performed on the gateway plugin, orchestration script, and device certificate. Objects that pass verification are set to the enabled state, and objects that fail verification are set to the isolated state. For device certificates, rolling rotation is performed according to the certificate rotation policy; According to the preset tagging rules, the control instructions in the linkage execution sequence are mapped to the deterministic channel. The priority of the control instructions in the deterministic transmission strategy is used as the queue priority of the deterministic channel. The control jitter threshold in the deterministic transmission strategy is used as the maximum allowable delay jitter of the deterministic channel. The upper limit of the control delay is used as the maximum transmission delay constraint of the deterministic channel. Based on the bitrate and frame rate ranges in the video adaptive strategy, the video stream bitrate and frame rate are maintained, and the downgrading strategy is executed when the remaining bandwidth is lower than the downgrading bandwidth threshold. The execution records and status results generated during the execution process are summarized into an operation log file. The operation log file is hash-signed and sealed, and stored according to the storage path in the log sealing parameters. Methods for implementing rolling certificate rotation based on a certificate rotation strategy include: The rotation scheduler calculates the next rotation time based on the most recent effective time and the rotation cycle; it then calculates the pre-distribution start time based on the next rotation time and the overlapping effective window. The parallel time window is constructed by taking the pre-distribution start time as the starting point of the parallel time window and the next rotation time as the ending point of the parallel time window. After entering the parallel time window, the parameter validator sends a new device certificate to the target device, and the security verifier performs a handshake verification on the new device certificate; If the verification is successful, the new device certificate will be marked as valid, and the old device certificate will remain valid until the next rotation time. When the next rotation time arrives, the parameter activator will switch the old device certificate to the new device certificate, and the security activator will revoke the old device certificate, ending the current rotation. If the verification fails, the invalid mark of the new device certificate is written back, a failure rollback is executed, the use of the new device certificate is terminated, the old device certificate remains valid, and the current number of failures is recorded. If the current number of failures is less than or equal to the number of failure rollbacks, the rotation scheduler will re-formulate the rotation plan in the next cycle. If the current number of failures is greater than the number of failure rollbacks, the security protector will put the corresponding device into the isolation list and trigger the rollback edge and remediation sub-process.

2. The remote monitoring and collaborative system based on intelligent network connectivity according to claim 1, characterized in that, The method for implementing the downshift strategy includes: Adjust the bitrate to the lower limit of the bitrate range, and adjust the frame rate to the lower limit of the frame rate range; After the bitrate and frame rate are adjusted, the link bandwidth status is continuously monitored. If the remaining bandwidth exceeds the downgrade bandwidth threshold and the duration exceeds the preset duration threshold, the scheduler will automatically call the upgrade process to restore the frame rate and bitrate to the range.

3. The remote monitoring and collaborative system based on intelligent network connectivity according to claim 1, characterized in that, Methods for generating runtime profiles based on execution trajectory logs include: Calculate the execution time, waiting time, and confirmation time for each node in the execution trajectory log, and record them as the first profile parameters; Calculate the trigger delay for each connection in the execution trajectory log and record it as the second profile parameter; For nodes with rollback edges in the execution trajectory log, calculate the rollback trigger delay, the execution time of the remedial subprocess, and the time interval from the completion of the remediation to the recovery of the main process, and record them as the third profile parameter; The success rate, rollback trigger rate, and recovery success rate of the instructions are calculated from the number of successful nodes, rollback nodes, successful recovery nodes, and total number of nodes counted in the execution trajectory log, and recorded as the fourth profile parameter. The first profile parameter, the second profile parameter, the third profile parameter, and the fourth profile parameter are collected and encapsulated according to a preset data structure to construct a running profile.

4. The remote monitoring and collaborative system based on intelligent network connectivity according to claim 1, characterized in that, Methods for generating a linkage control orchestration diagram and forming a linkage execution sequence through topological sorting include: The policy parser reads the event type-event level-device identifier ternary semantic structure from the standard event metadata and retrieves the matching business rules from the pre-built policy library; the event type-event level-device identifier ternary semantic structure and the business rules are then used to construct the retrieval results. The search results are parsed into node-edge descriptions, and a linkage control orchestration diagram with start node, condition gateway, parallel convergence and end node is generated according to the specifications of the business process modeling language. The scheduler assigns a corresponding session identifier to the linkage control orchestration diagram, performs topological sorting on the linkage control orchestration diagram, and obtains the linkage execution sequence.

5. The remote monitoring and collaborative system based on intelligent network connectivity according to claim 1, characterized in that, The methods for injecting execution tokens and issuing action instructions to each node in the coordinated execution sequence, recording execution trajectory logs, and triggering a remedial sub-process mechanism along the rollback edge when the terminal reports a timeout or failure include: An execution token is injected into the node of the linked execution sequence, and an action command is issued to the corresponding target terminal according to the linked execution sequence; the execution token carries token operation parameters, including node start time, timeout time, and retry count. The scheduler writes the node status, edge status, and execution token migration records of the linkage control orchestration graph into the execution trajectory log, and then synchronizes the execution trajectory log to the security protector after binding it with the session identifier. The scheduler creates a status checker for nodes that enter the running state. The status checker continuously collects terminal feedback. If the execution token of a node does not receive a successful confirmation before the timeout period, the rollback edge connected to the corresponding node in the linkage control orchestration diagram is immediately triggered, and the remedial subprocess mechanism is invoked.

6. A remote monitoring and collaborative system based on intelligent network connectivity according to claim 5, characterized in that, The remedial sub-process mechanism is as follows: For each node with a rollback flag, trigger conditions are pre-bound in the policy library; the trigger conditions include action execution failure, timeout threshold reaching, and device offline. Draw an independent sub-flowchart for each triggering condition. The sub-flowchart includes a start event, a remedial action node, and an end event. The sub-flowchart is connected to the rollback edge. When the scheduler detects that any triggering condition is met during runtime, it suspends the subsequent execution of the main process at the current node and transfers the execution token to the corresponding remedial subprocess for execution. After the sub-process is completed, the output status value determines whether to return to the main process to continue execution, terminate the current session directly, or recursively trigger a remedial sub-process.

7. The remote monitoring and collaborative system based on intelligent network connectivity according to claim 1, characterized in that, The methods for obtaining the standard event metadata include: The protocol mapping gateway receives raw protocol messages from heterogeneous terminals; The protocol mapping gateway calls a parser that matches the protocol type based on the protocol type of the original protocol message, and checks the handshake validity, payload integrity and sequence continuity through the protocol state machine; it extracts command type, alarm level and device identification key fields from the original protocol message and aggregates them to form a protocol field set; The protocol mapping gateway calls a field-semantic lookup table to map each protocol field in the protocol field set to a three-element semantic structure of event type-event level-device identifier, thus obtaining a standard semantic record; the current timestamp is appended to the standard semantic record and encapsulated together with the index pointer pointing to the original protocol message to form an event metadata draft; The protocol mapping gateway performs sequence number continuity detection and circular cache rearrangement on the event metadata draft, removes duplicate frames and identifies lost frames, and generates standard event metadata.

8. The remote monitoring and collaborative system based on intelligent network connectivity according to claim 1, characterized in that, The heterogeneous terminal includes a camera, sensors, and a voice terminal.

Images