Automatic routing construction method based on topology information carried by registration message and controller

Abstract

The invention discloses an automatic route construction method based on topology information carried by a registration message and a controller. The method comprises the following steps: an edge gateway obtains topological information of a service device mounted under the edge gateway, packages the topological information, a gateway identifier and a northbound interface IP address of the edge gateway into a UDP registration message, and sends the UDP registration message to a controller; and the controller receives and analyzes the UDP registration message, establishes a mapping relation from the MAC address of the service device to the northbound interface IP address of the edge gateway to which the service device belongs in a global forwarding table, and determines a target edge gateway according to the global forwarding table when an original data frame sent to a target service device is received, thereby realizing accurate forwarding of the data frame. According to the application, the plug and play of the southbound service device and the automatic construction of the route can be realized, the complexity and cost of network operation and maintenance are reduced, the network storm is avoided through accurate forwarding, and the stability of a large-scale industrial network is improved.

Description

Technical Field

[0001] This invention relates to the field of industrial internet communication technology, and more specifically, to an automatic route construction method and controller based on registration messages carrying topology information. Background Technology

[0002] In modern industrial automation and IoT scenarios, network architectures are becoming increasingly complex. Especially in industrial production sites, a large number of operational devices are typically deployed, such as programmable logic controllers (PLCs), temperature and humidity sensors, industrial vision cameras, and automated guided vehicles (AGVs). These operational devices, acting as network terminal nodes, connect to the southbound interface of an edge gateway via Ethernet or other means, and are then uniformly connected to the core control network by the edge gateway.

[0003] With the advancement of flexible manufacturing and intelligent upgrades, adjustments to production line layouts and equipment movements are becoming increasingly frequent. For example, the movement of automated guided vehicles (AGVs) between different work areas, or the addition, removal, or relocation of equipment on the production line, will lead to changes in the physical network access points (i.e., their respective edge gateways) to which the business units are connected. This dynamic topology change presents significant challenges to network management.

[0004] Existing network routing management technologies have the following main shortcomings when dealing with such scenarios:

[0005] 1. Manual configuration is inefficient and error-prone: Traditional network routing, especially for communication across Layer 3 networks (i.e., different network segments), relies on network administrators manually configuring static routes or "next-hop" addresses. When the topology of southbound service devices changes, maintenance personnel must log in to the relevant edge gateways and aggregation layer switches to manually update the routing tables. This process is not only time-consuming and labor-intensive with extremely high maintenance costs, but also highly susceptible to configuration errors due to human negligence in complex network environments, leading to communication interruptions.

[0006] 2. Dynamic routing protocols incur high resource overhead: To achieve automatic route updates, dynamic routing protocols such as Open Shortest Path First (OSB) or Border Gateway Protocol (BGP) can be run on network devices. However, industrial edge gateways are typically resource-constrained embedded devices with limited computing power and memory. Running these complex protocols significantly consumes the gateway's valuable system resources. Furthermore, the periodic protocol messages generated by dynamic routing protocols (such as Hello messages and link state update messages) flood the network, crowding out the already limited industrial control bus bandwidth and potentially affecting the real-time transmission of critical business data.

[0007] 3. Difficulty in discovering cross-network segment devices, posing a broadcast storm risk: When the core layer controller needs to communicate with a service device located on a different network segment, it cannot directly obtain the target service device's MAC address through standard Address Resolution Protocol (ARP) broadcasts, as ARP broadcasts are typically isolated within the local network segment by routers. To address this issue, some solutions may employ network-wide flooding to discover devices, but this can trigger a "broadcast storm" in large-scale networks, severely degrading network performance and stability.

[0008] Therefore, how to provide a low-overhead, high-efficiency routing and data forwarding mechanism that can automatically adapt to dynamic topology changes has become a pressing technical problem in the current industrial internet field. Summary of the Invention

[0009] This invention solves the problems in the prior art, such as difficulty in routing maintenance, large overhead of dynamic protocols, and inability to automatically discover cross-segment devices caused by changes in the topology of edge devices.

[0010] To achieve the above objectives, this invention provides an automatic route construction method and controller based on registration messages carrying topology information, applied to a controller. The method includes:

[0011] Receive at least one UDP registration message sent by an edge gateway. The UDP registration message encapsulates a gateway identifier for identifying the edge gateway, the northbound interface IP address of the edge gateway, and topology information of one or more service devices attached to the edge gateway. The topology information includes the IP address and MAC address of each service device.

[0012] Parse the UDP registration message to extract the gateway identifier, the northbound interface IP address, and the topology information;

[0013] Based on the extracted information, establish or update the mapping relationship between the MAC address of the service device and the northbound interface IP address of the edge gateway to which it belongs in the global forwarding table;

[0014] When a raw data frame destined for a target service device is received, the MAC address of the target service device is queried according to the global forwarding table to determine the northbound interface IP address corresponding to the MAC address;

[0015] The original data frame is encapsulated and forwarded to the edge gateway determined by the northbound interface IP address.

[0016] In a preferred embodiment, the method further includes: setting an effective timer for each mapping relationship in the global forwarding table; when a UDP registration message is received from one of the edge gateways, resetting the effective timers for all mapping relationships associated with that edge gateway; and deleting the mapping relationships whose effective timers have expired from the global forwarding table.

[0017] In a preferred embodiment, the UDP registration message includes a type field indicating the message type; the step of parsing the UDP registration message further includes: determining whether the UDP registration message is a full reporting message or a differential reporting message based on the type field; if it is a full reporting message, replacing all mapping relationships corresponding to the edge gateway in the global forwarding table with the topology information contained in the message; if it is a differential reporting message, incrementally updating or deleting the mapping relationships corresponding to the edge gateway in the global forwarding table based on the topology information contained in the message.

[0018] In a preferred embodiment, the topology information in the differential reporting message further includes an operation code for each entry of a service device; the controller performs an incremental update or deletion operation based on the operation code, wherein the operation code indicates whether the entry of the service device is a new entry or an entry to be deleted.

[0019] In a preferred embodiment, the step of encapsulating the original data frame specifically includes: encapsulating the original data frame using a tunneling protocol, wherein the destination IP address of the encapsulated external packet header is the northbound interface IP address of the edge gateway.

[0020] This invention also provides another automatic route construction method applied to an edge gateway. The method includes:

[0021] Obtain the topology information of one or more service devices connected to its southbound interface, wherein the topology information includes the IP address and MAC address of each service device;

[0022] The topology information, the gateway identifier used to identify the edge gateway, and the northbound interface IP address of the edge gateway itself are encapsulated into a UDP registration message;

[0023] The UDP registration message is sent to a controller through the northbound interface. The UDP registration message is used by the controller to establish or update the mapping relationship between the MAC address of the service device and the IP address of the northbound interface.

[0024] In a preferred embodiment, the step of obtaining the topology information includes: periodically sending Address Resolution Protocol (ARP) request messages through the southbound interface and listening for response messages to actively discover the service devices connected to it and obtain their IP addresses and MAC addresses.

[0025] In a preferred embodiment, the method further includes: maintaining the topology information reported to the controller in a local cache; when a change is detected in the service device mounted under the southbound interface, comparing the current topology information with the topology information maintained in the local cache to generate differential topology information containing only the changed part; encapsulating the differential topology information into the UDP registration message, and setting the type of the UDP registration message to a differential reporting message.

[0026] In a preferred embodiment, detecting a change in the service device includes at least one of the following situations: detecting a new service device by listening to a new ARP response message; or determining that the service device is offline if an ARP response from a recorded service device is not received within a preset number of retries.

[0027] In a preferred embodiment, the step of sending the UDP registration message to the controller is triggered under at least one of the following conditions: triggered according to a preset time period; or triggered immediately when a change in the topology information is detected.

[0028] The beneficial effects of this invention are as follows:

[0029] 1. Automation and Zero Configuration: By actively reporting topology information through the edge gateway, the system enables "plug and play" for southbound service devices and automatic route updates after location changes. The entire network route can be dynamically constructed without any manual intervention, significantly reducing the complexity and cost of network operation and maintenance.

[0030] 2. Low Overhead and High Efficiency: Lightweight UDP registration messages replace complex dynamic routing protocols, avoiding the need to run high-overhead protocol stacks on resource-constrained edge gateways. Furthermore, by introducing a differential reporting mechanism, only heartbeats or minor change information need to be transmitted when the network topology is stable, saving valuable network bandwidth, making it particularly suitable for bandwidth-sensitive environments such as industrial control.

[0031] 3. Suppressing network storms and improving stability: The controller constructs a centralized global forwarding table based on the collected precise topology information, enabling accurate unicast tunnel forwarding of data frames across network segments. This fundamentally eliminates the reliance on ARP broadcasts, avoiding potential network storms and significantly improving the stability and reliability of large-scale industrial networks. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a structural block diagram of an automatic route building system provided in an embodiment of the present invention.

[0034] Figure 2 This is a flowchart of an automatic route construction method for a controller provided by an embodiment of the present invention.

[0035] Figure 3 This is a flowchart of an automatic route construction method for edge gateways provided by an embodiment of the present invention. Detailed Implementation

[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments and the accompanying drawings.

[0038] Example 1

[0039] This embodiment provides a complete automatic route construction system and its working method. The system adopts a two-layer architecture of edge access layer and core control layer to realize "plug and play" access of business devices and automatic route construction in industrial network environment.

[0040] See Figure 1 The system includes a controller (101) in the core control layer and at least one edge gateway (102) in the edge access layer. Each edge gateway (102) connects to one or more service devices (103) through its southbound interface.

[0041] Specifically, the controller (101) can be one or more high-performance servers. Hardware-wise, it includes a high-performance processor for performing complex calculations and data processing, a large-capacity memory for storing data such as the global forwarding table, and at least one physical network interface for communicating with an external network. Software-wise, the controller (101) deploys the control software described in this application, which logically can be divided into modules such as a UDP receiver parser, a global topology database (implemented as a global forwarding table), and a forwarding engine.

[0042] The edge gateway (102) is an intermediate device connecting southbound service devices and northbound controllers, typically an industrial-grade embedded device. Its hardware structure includes a processor for executing the methods described in this application, a memory for storing firmware programs and local topology caches, one or more northbound interfaces for connecting to the core network, and one or more southbound interfaces for connecting service devices. The edge gateway (102) may also include a timer module for periodically triggering topology scanning and packet transmission. In terms of software, it embeds a topology awareness module and a registration packet encapsulation module.

[0043] The service devices (103) are various terminal devices mounted on the edge gateway (102), such as programmable logic controllers, temperature and humidity sensors, industrial vision cameras, and automated guided vehicles. These devices are connected to the southbound interface of the edge gateway (102) via Ethernet interfaces.

[0044] In this embodiment, the collaborative work between the edge gateway (102) and the controller (101) enables automatic route construction and accurate data forwarding. Their interaction logic is not simply data pass-through, but rather forms a control mechanism that separates signaling and data. The edge gateway (102), acting as a distributed "sensing probe," actively "floats" the physical connection status of the underlying network (i.e., the topology information of the service devices) to the controller (101), which serves as the control plane, via a customized UDP registration message. The controller (101) receives and terminates this registration message, parsing it as signaling to update the global routing mapping relationship, thereby guiding the forwarding decisions of subsequent service data frames.

[0045] The methods provided in this application will be described in detail below from the perspectives of the edge gateway and the controller.

[0046] I. Methods on the Edge Gateway Side

[0047] Reference Figure 3 The automatic route construction method applied to the edge gateway (102) has the following detailed steps:

[0048] Step S301: Obtain the topology information of the southbound service device.

[0049] The topology awareness module within the edge gateway (102) is responsible for discovering and recording information about all active service devices (103) directly or indirectly (through a switch) connected to its southbound interface. The topology information includes at least the IP address and MAC address of each service device (103).

[0050] In one specific implementation, this acquisition process can be achieved through active probing. For example, the edge gateway (102) can periodically send Address Resolution Protocol (ARP) request messages to the network segment connected to its southbound interface and listen for ARP response messages in the network. By parsing the received ARP responses, the accurate mapping relationship between the IP address and MAC address of the responding service device can be obtained. This active scanning method can ensure the timeliness of topology information. It should be noted that the use of the ARP protocol is only an example, and those skilled in the art can also use other link layer or network layer discovery protocols to achieve the same function, such as link layer discovery protocols, etc. This application is not limited to this.

[0051] Step S303: Encapsulate the UDP registration message.

[0052] After obtaining the latest topology information, the registration message encapsulation module of the edge gateway (102) constructs a UDP registration message. This message is sent to the pre-configured IP address and listening port of the controller (101) (for example, in this embodiment, the UDP port listened to by the controller is 6677). The payload portion of this message is encapsulated in a compact binary format to reduce network overhead. Its payload structure can be designed as shown in the table below:

[0053]

[0054] Step S304: Generate and encapsulate differential topology information (optional optimization).

[0055] To further reduce unnecessary bandwidth consumption, the edge gateway (102) can implement a differential reporting mechanism. Specifically, the edge gateway (102) maintains a copy of the topology information that has been reported to the controller or its summary information (e.g., a topology fingerprint hash value) in its local memory. When new topology information is obtained through periodic scanning or event triggering, it is compared with the old topology information cached locally.

[0056] If the topology remains unchanged (e.g., the fingerprint hash values ​​of the old and new topologies are consistent), the edge gateway (102) can send only a heartbeat message without a device list (i.e., the device entry field), which contains only the message type and gateway identifier, to indicate to the controller (101) that it is online and to maintain the aging time of the mapping relationship.

[0057] If a topology change is detected, the edge gateway (102) generates a differential topology information that only contains the changed parts. For example, newly discovered devices are marked as "added," and offline devices are marked as "deleted." This differential information is then encapsulated in a UDP registration message, and the message type field is set to "differential reporting."

[0058] The specific conditions for detecting changes in a service device may include (but are not limited to) at least one of the following situations: a new service device is detected by listening to a new ARP response packet that is not recorded in the local cache; or, after sending an ARP request to a recorded service device, if no response is received within a preset timeout period and number of retries, it is determined that the service device is offline.

[0059] Step S305: Send a UDP registration message to the controller.

[0060] After encapsulation, the edge gateway (102) sends the UDP registration message to the controller (101) through its northbound interface. This sending behavior can be triggered by various conditions. For example, it can be triggered according to a preset time period (such as every 30 seconds) to ensure the periodic refresh of routing information on the controller side; it can also be triggered immediately when a change in southbound topology information is detected to ensure the timeliness of routing information updates. This combination of periodic triggering and event triggering takes into account both timeliness and network overhead.

[0061] II. Controller-side methods

[0062] Reference Figure 2 The automatic route construction method applied to controller (101) has the following detailed steps:

[0063] Step S201: Receive the UDP registration message sent by the edge gateway.

[0064] The UDP receiver parser of the controller (101) continuously listens on a preset port (such as 6677) to receive UDP registration messages sent from various edge gateways (102) in the network.

[0065] Step S202: Parse the UDP registration message and extract key information.

[0066] Upon receiving the message, the controller (101) decapsulates it, parses its payload, and extracts the core field information, including: the sender's gateway identifier, the northbound interface IP address of the gateway (which can also be obtained directly from the source IP address of the UDP message), and the list of service device entries reported by the sender.

[0067] Step S203: Execute different update logic according to the message type.

[0068] The controller (101) checks the message type field in the message.

[0069] If the message is determined to be a full report message (e.g., message type 0x01 containing a complete list of device entries), the controller (101) will use the topology information carried in the message to replace all the original mapping relationships corresponding to the gateway identifier in the global forwarding table. This method is suitable for scenarios such as initial gateway registration or periodic full synchronization.

[0070] If the message is determined to be a differential reporting message, the controller (101) incrementally updates or deletes the mapping relationship related to the gateway identifier in the global forwarding table based on the topology information contained in the message. In a more specific implementation, each device entry in the differential message may additionally include an opcode to explicitly indicate whether the entry is "added" or "to be deleted". The controller performs the addition or deletion operation precisely based on this opcode, which is more efficient.

[0071] Step S204: Update the global forwarding table.

[0072] The global forwarding table is a core data structure maintained by the controller (101) to enable fast routing from a target service device to its gateway. This table can be a hash table or an inverted index with the service device's MAC address as the key and its corresponding edge gateway information (e.g., a structure containing {gateway identifier, northbound interface IP}) as the value.

[0073] The controller (101) traverses the device entry list based on the topology information parsed in step S203 and establishes or updates the mapping relationship between "service device MAC address" and "the northbound interface IP address of its respective edge gateway" in the global forwarding table.

[0074] During the update process, conflicts may occur, such as the same MAC address being reported by two different edge gateways (possibly caused by network loops or dual-homing of devices). In this case, the controller (101) can make a decision based on the preset conflict handling logic. For example, it can retain the mapping record with higher priority or the latest one based on the signal quality indicators of the arriving packets, the number of network hops, or a simple "last-in, first-out" principle, thereby realizing the function of multiple transmissions and selective reception.

[0075] Step S205: Aging and maintenance of mapping relationships.

[0076] To address the issue of invalid "ghost" entries appearing in the forwarding table due to abnormal offline status of edge gateways (failure to send deregistration messages), this application introduces an aging and deletion mechanism for mapping relationships. Specifically, the controller (101) associates an expiration timer with each mapping relationship in the global forwarding table. When the controller (101) receives any valid UDP registration message (whether full, differential, or heartbeat) from an edge gateway, it resets the expiration timers for all mapping relationships associated with that edge gateway. The controller periodically checks all timers, and once a timer expires (i.e., no message is received from the corresponding gateway within a preset time), it considers the gateway or its associated devices to be unreachable and immediately deletes the corresponding mapping relationship from the global forwarding table. This mechanism enhances the robustness of the routing table.

[0077] Step S206: Receive the raw data frame destined for the target service device.

[0078] When the controller (101) (or its network) receives a raw data frame (e.g., a SCADA instruction from the monitoring center, which is a standard Layer 2 or Layer 3 data packet) that needs to be forwarded to a specific service device (103), the forwarding engine is triggered.

[0079] Step S207: Query the global forwarding table to determine the next hop.

[0080] The forwarding engine parses the original data frame and extracts its destination MAC address. Then, using this destination MAC address as the key, it queries the global forwarding table.

[0081] If the query is successful, the northbound interface IP address of the edge gateway (102) currently attached to the target service device (103) is successfully obtained.

[0082] If the query fails (i.e., there is no record for the MAC address in the table), it indicates that the controller currently does not know the location of the device. In this case, different methods can be used, such as directly discarding the data packet to prevent network flooding, or temporarily storing it and triggering a passive query mechanism.

[0083] Step S208: Encapsulate and forward the data frame.

[0084] After a successful query, the controller (101) encapsulates the original data frame as a new payload, forming a tunnel message. Specifically, a tunneling protocol (such as VXLAN, GRE, or simple IP-in-UDP encapsulation) can be used to encapsulate the original data frame. In the header of the encapsulated external IP message, the destination IP address is set to the northbound interface IP address of the edge gateway obtained from the global forwarding table in step S207, and the destination port can be set to a predefined service data receiving port. Finally, the controller (101) sends out this encapsulated tunnel message through its physical network interface. The message will be accurately delivered to the target edge gateway (102) via standard IP routing, and then decapsulated by the gateway and forwarded to the final service device (103).

[0085] In summary, this embodiment utilizes a collaborative mechanism between the edge gateway's proactive reporting of southbound topology information and the controller's centralized construction of a global forwarding table. This allows the controller to obtain accurate terminal device location information without performing time-consuming full-network scans. When external data needs to access any service device, the controller can achieve accurate and efficient forwarding across network segments through a single table lookup and tunnel encapsulation, completely resolving a series of problems in traditional solutions, such as cumbersome routing configuration, high resource overhead, and reliance on broadcast discovery.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An automatic route construction method, applied to a controller, characterized in that, include: The system receives UDP registration messages sent by at least one edge gateway. The UDP registration message encapsulates a gateway identifier for identifying the edge gateway, the northbound interface IP address of the edge gateway, and topology information for one or more service devices connected to the edge gateway. The topology information includes the IP address and MAC address of each service device. The system parses the UDP registration message to extract the gateway identifier, the northbound interface IP address, and the topology information. Based on the extracted information, it establishes or updates a mapping relationship in the global forwarding table between the MAC address of the service device and the northbound interface IP address of its corresponding edge gateway. When a raw data frame destined for a target service device is received, the system queries the global forwarding table to find the MAC address of the target service device and determines the northbound interface IP address corresponding to the MAC address. The system then encapsulates and forwards the raw data frame to the edge gateway determined by the northbound interface IP address.

2. The method according to claim 1, characterized in that, Also includes: Set an effective timer for each mapping relationship in the global forwarding table; when a UDP registration message is received from one of the edge gateways, reset the effective timers for all mapping relationships associated with that edge gateway; Remove the mapping relationship of the effective timer timeout from the global forwarding table.

3. The method according to claim 1, characterized in that, The UDP registration message includes a type field indicating the message type; the step of parsing the UDP registration message further includes: determining whether the UDP registration message is a full reporting message or a differential reporting message based on the type field; if it is a full reporting message, replacing all mapping relationships corresponding to the edge gateway in the global forwarding table with the topology information contained in the message; if it is a differential reporting message, incrementally updating or deleting the mapping relationships corresponding to the edge gateway in the global forwarding table based on the topology information contained in the message.

4. The method according to claim 3, characterized in that, The topology information in the differential reporting message also includes an operation code for each entry of a service device; based on the operation code, an incremental update or deletion operation is performed, wherein the operation code indicates whether the entry of the service device is a new entry or an entry to be deleted.

5. The method according to claim 1, characterized in that, The step of encapsulating the original data frame specifically includes: encapsulating the original data frame using a tunneling protocol, wherein the destination IP address of the encapsulated external packet header is the northbound interface IP address of the edge gateway.

6. An automatic route construction method, applied to an edge gateway, characterized in that, include: Obtain the topology information of one or more service devices connected to its southbound interface, wherein the topology information includes the IP address and MAC address of each service device; The topology information, the gateway identifier used to identify the edge gateway, and the northbound interface IP address of the edge gateway itself are encapsulated into a UDP registration message; The UDP registration message is sent to a controller through the northbound interface. The UDP registration message is used by the controller to establish or update the mapping relationship between the MAC address of the service device and the IP address of the northbound interface.

7. The method according to claim 6, characterized in that, The steps for obtaining the topology information include: periodically sending Address Resolution Protocol (ARP) request messages through the southbound interface and listening for response messages to actively discover the service devices connected to it and obtain their IP addresses and MAC addresses.

8. The method according to claim 6, characterized in that, Also includes: Maintain the topology information that has been reported to the controller in the local cache; When a change is detected in the service device mounted under the southbound interface, the current topology information is compared with the topology information maintained in the local cache to generate a differential topology information containing only the changed part; the differential topology information is encapsulated into the UDP registration message, and the type of the UDP registration message is set to differential reporting message.

9. The method according to claim 8, characterized in that, The detection of changes in the service device includes at least one of the following situations: a new service device is detected by listening to a new ARP response message; or the service device is determined to be offline if an ARP response from a recorded service device is not received within a preset number of retries.

10. The method according to claim 6, characterized in that, The step of sending the UDP registration message to the controller is triggered under at least one of the following conditions: triggered according to a preset time period; or triggered immediately when a change in the topology information is detected.

Images