A centralized dynamic routing method for optical networks

Abstract

The application discloses a centralized dynamic routing selection method of an optical network, and the method is composed of a master node control layer and a slave node data transmission layer, wherein the slave node data transmission layer obtains services from terminals; the master node control layer comprises a first message generation and processing module, a route calculation module, a first storage module and a first control module; the slave node data transmission layer comprises a second message generation and processing module, an interface detection module, a second storage module, a wavelength conversion module, an optoelectronic conversion module and a second control module; and a terminal physical layer is a user terminal which uploads service data between users. The method adopts centralized management in the optical network, obtains global topology and all node link states, adjusts the route selection between every two nodes in the region adaptively through real-time monitoring of link cost cost changes, reduces local link congestion of the optical network, and improves global resource utilization.

Description

A centralized dynamic routing method for optical networks Technical Field

[0001] This invention relates to the field of network communication technology, specifically to a centralized dynamic routing method for optical networks. Background Technology

[0002] Optical networks serve as data transport networks for various communication services, with traditional optical networks primarily designed for voice services. In recent years, with the continuous development of modern communication technologies, a large number of personal mobile devices and industrial systems have connected to the Internet, driving exponential growth in traffic driven by video and Internet services. To meet the ever-increasing demand for capacity, this surge in traffic poses a significant challenge to the advanced carrying capacity of the optical layer in optical networks. Furthermore, traditional optical network design and construction methods configure optical channels statically. Because next-generation optical network data services are dynamic and bursty, they are prone to causing local link congestion, creating new demands for dynamic adjustment of optical networks. Currently, optical network controllers and managers are essentially still managed in a distributed manner. Distributed management schemes can effectively reduce local network congestion, but they consume more power and are prone to uneven load distribution on local links. Traditional architectures, due to their rigidity, complexity, cost, and lack of customization, cannot meet the needs of large-scale computing, high resource availability, dynamic infrastructure customization, automation, resilience, overall learning, and other requirements. However, limited by the capabilities of distributed devices and technologies, distributed configuration and maintenance are complex and expensive, restricting openness, scalability, and flexibility.

[0003] To address the above issues, several solutions have emerged. The future management and control architecture of optical systems requires a centralized architecture characterized by agility and flexibility, separating the control plane from the data plane, providing a new approach for centralized network control and dynamic maintenance. Operators can flexibly configure the network according to different user needs. Furthermore, centralized control offers a global network perspective, allowing for centralized distribution of network resources and improving resource utilization efficiency. Centralized routing has become a promising industrial solution, capable of meeting the potential scale and flexibility requirements of optical networks, and providing high-quality, low-latency, resilient, and customized services. Distributed management methods, while effectively reducing local network congestion, suffer from higher device power consumption and are prone to uneven load distribution on local links. Therefore, this invention, aiming to improve network resource utilization and enhance overall network performance, emphasizes research into centralized routing decisions and simplified network management. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a centralized dynamic routing method for optical networks. This method employs a centralized management approach in the optical network, acquiring the global topology and the link status of all nodes. By monitoring changes in link costs in real time, it adaptively adjusts the routing selection between every pair of nodes within the region, reducing local link congestion and improving overall resource utilization.

[0005] The technical solution to achieve the objective of this invention is:

[0006] A centralized dynamic routing method for optical networks is provided, the method being based on the interaction between the master node control layer and the slave node data transmission layer, wherein the slave node data transmission layer obtains services from the terminal.

[0007] The master node control layer includes a first message generation and processing module, a route calculation module, a first storage module, and a first control module. The master node control layer collects the link status of all nodes in the storage area and calculates the routes between each pair of nodes, controls the data transmission routes between nodes in the area, reduces local link congestion, and improves the overall resource utilization.

[0008] The slave node data transmission layer includes a second message generation and processing module, an interface detection module, a second storage module, a wavelength conversion module, a photoelectric conversion module, and a second control module. It transmits data uploaded by the terminal physical layer according to the route calculated by the master node control layer. The terminal physical layer is the user terminal, which uploads business data between users.

[0009] The first and second message generation and processing modules are identical: the messages in this module include Hello messages, LSR link state request messages, LSA link state announcement messages, and LSAck receive response messages. This module is mainly responsible for two aspects: first, generating and sending corresponding messages based on the state changes of this node; second, receiving messages sent by other nodes, identifying the message type (Type), extracting relevant information from the message, and pushing this information to the first and second storage modules for storage. Both master and slave nodes have this functional module, but they differ in function: the messages generated by the master node include Hello messages, LSR link state request messages, and LSAck receive response messages; the messages generated by the slave node include Hello messages, LSA link state announcement messages, and LSR link state request messages.

[0010] The first and second storage modules are identical: the data collected or calculated by each module is pushed to the first and second storage modules for storage. In addition, the first and second storage modules will save the latest data and remove the corresponding invalid data according to the data's age.

[0011] Routing calculation module: Reads the link status in the network area from the first storage module, uses Dijkstra's algorithm, and uses link cost as the key indicator to calculate the route with the lowest total link cost between any two nodes;

[0012] The first and second control modules are identical: the first and second control modules identify the status of their respective nodes and control the execution of each module.

[0013] Interface detection module: Link cost refers to the bandwidth utilization efficiency of the interface. The function of this module is to detect the bandwidth utilization of the interface in real time and promptly report it to the second control module.

[0014] Photoelectric conversion module: Converts the data in the second storage module from electrical signals into optical signals for transmission;

[0015] Wavelength conversion module: Based on the control signals sent by the master node, the slave node adjusts the wavelength transmitted by this node;

[0016] The master node control layer coordinates global optical network resources and makes routing decisions. The specific operation steps are as follows:

[0017] Step 1.1: Periodically send Hello messages to establish and maintain adjacency with directly connected nodes;

[0018] Step 1.2: After the adjacency state is established, broadcast an LSR link state request message to the slave nodes in the area, requesting the slave nodes to send their own LSA link state advertisement messages;

[0019] Step 1.3: Upon receiving the LSA link status advertisement message from the slave node, the master node sends an LSAck receive response message back to the slave node, indicating that the master node has received the LSA link status advertisement message.

[0020] Step 1.4: The first storage module saves the LSA link status announcement messages of all slave nodes in the area and establishes an area database, which includes each node ID number, interface address, link cost, interface wavelength, ID number of directly connected nodes and address of directly connected interfaces. The topology structure is established through the above data.

[0021] Step 1.5: The routing calculation module accesses the database in the first storage module and uses the Dijkstra algorithm to calculate a route with the minimum total cost of the link. Therefore, the link cost, as the most important parameter, will directly determine the route between nodes. The link cost refers to the efficiency of interface bandwidth utilization, that is, Cost = business data volume / interface rate.

[0022] The node link cost is detected and uploaded in real time at the node layer, and data is transmitted according to the routing plan published by the master node control layer. The specific operation steps are as follows:

[0023] Step 2.1: Periodically send Hello messages to establish and maintain adjacency with directly connected nodes;

[0024] Step 2.2: Receive the LSR link status request message sent by the master node;

[0025] Step 2.3: Respond to the master node's request, encapsulate the local node's LSA link status announcement message. The message content includes the local node ID number, interface address, link cost, interface wavelength, ID number of the directly connected node, and address of the directly connected node. After sending the LSA link status announcement message, receive the master node's LSAck receive response message.

[0026] Step 2.4: Receive the routing table calculated by the master node and transmit data according to the routing table;

[0027] Step 2.5: Determine whether the connection with the directly connected node is broken, and then exit the adjacency state; if exiting the adjacency state, update the LSA link state advertisement message of this node and immediately send the LSA link state advertisement message to the master node.

[0028] Step 2.6: Determine whether the link cost (Cost) of this node has changed, i.e., whether the interface bandwidth utilization efficiency has changed; divide the interface bandwidth utilization efficiency into three states: A, B, and C. The bandwidth utilization efficiency corresponding to state A is [0, 0.5], the bandwidth utilization efficiency corresponding to state B is [0.5, 0.8], and the bandwidth utilization efficiency corresponding to state C is [0.8, 1]. If the Cost state of any interface in the node changes, update the LSA link status announcement message of this node and immediately send this message to the master node.

[0029] Step 2.7: Receive the latest routing table sent by the master node and transmit data according to the routing table.

[0030] The Hello message and the LSR link state request message have the same message header. The message header consists of message type (Type), total message length (Packet length), local node identifier (Node ID), and checksum. Message type (Type) = 1 and 2 indicate that the message is a Hello message and an LSR link state request message, respectively. Message length (Packet length) represents the total message length, including the message header. Local node identifier (Node ID) indicates the identifier of the node sending this message. Checksum is the checksum of the entire message, including the authentication field.

[0031] The Hello message consists of a header and a content. Its function is to establish and maintain adjacency relationships. It is sent periodically on the enabled interface. The content includes the Link ID of the interface that sent the message, the Hello Interval of sending the message, the Dead Interval used to determine whether the directly connected node has lost connection, and the Neighbor ID of the known neighbor node.

[0032] The LSR link state request message consists of a message header and a message content. Its function is for the master node to send the required LSA link state advertisement message to the slave node. The message content includes a sequence number. The master node periodically sends LSR link state request messages, and the sequence number is incremented by 1 at this time. The node ID identifier that generates this message is the Advertising ID.

[0033] The LSA link-state advertisement message and the LSAck receive response message have the same header. This header consists of message type (Type), total message length (Packet length), node ID (Node ID that sent the message), checksum, LSA sequence number (LS sequence number), advertisingID (Node ID that generated the message), and message generation time (LS age). Specifically, message type (Type = 3 and 4) indicates that this message is both an LSA link-state advertisement message and an LSAck receive response message; packet length indicates the total message length including the header; LSA sequence number (LS sequence number) is used by the master node to determine if it is the latest LSA link-state advertisement message; and LS age is the time the message was generated.

[0034] The LSA Link State Advertisement message consists of a header and a content. This message is generated by the slave node and describes the slave node's own link status and directly connected nodes. The header content of the LSA Link State Advertisement message includes the number of interfaces of this node (links), the number of interfaces directly connected to this node (#Links), the port identifier (Link ID) of this node, the link cost (Cost), the output wavelength (Wave length) of the interface, the neighbor node identifier (Neighbor ID), the corresponding directly connected port identifier (#Link ID), and other known neighbors and ports.

[0035] The LSAck receive response message consists of a message header and a message content. This message is generated by the master node and is used to confirm receipt of the LSA link status announcement message from the slave node, so as to prevent the slave node from repeatedly sending LSA link status announcement messages.

[0036] This method primarily consists of a master node control layer and a slave node data transmission layer. The master node control layer collects LSA link-state advertisement messages from all nodes within the region, viewing the link status of each slave node and its neighboring nodes from a global perspective. Based on changes in link cost, it adaptively adjusts optical network routing, thereby reducing local link congestion and improving overall resource utilization. Attached Figure Description

[0037] Figure 1 is a functional block diagram of the embodiment;

[0038] Figure 2 is a schematic diagram of the main node process in the embodiment;

[0039] Figure 3 is a schematic diagram of the node process in the embodiment;

[0040] Figure 4 is a schematic diagram of the structure of the Hello message and LSR link status request message in the embodiment;

[0041] Figure 5 is a schematic diagram of the structure of the LSA link status announcement message and the LSAck receive response message in the embodiment.

[0042] Figure 6 is a schematic diagram of the optical network topology of the embodiment;

[0043] Figure 7 is a schematic diagram of the adjacency relationship establishment in the embodiment;

[0044] Figure 8 is a schematic diagram of the master-slave node interaction in an embodiment. Detailed Implementation

[0045] The invention will now be described in further detail with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the invention. Embodiments

[0046] A centralized dynamic routing method for optical networks is provided, which is based on the interaction between the master node control layer and the slave node data transmission layer, as shown in Figure 1, wherein the slave node data transmission layer obtains services from the terminal.

[0047] The master node control layer includes a first message generation and processing module, a route calculation module, a first storage module, and a first control module. The master node control layer collects the link status of all nodes in the storage area and calculates the routes between each pair of nodes, controls the data transmission routes between nodes in the area, reduces local link congestion, and improves the overall resource utilization.

[0048] The slave node data transmission layer includes a second message generation and processing module, an interface detection module, a second storage module, a wavelength conversion module, a photoelectric conversion module, and a second control module. It transmits data uploaded by the terminal physical layer according to the route calculated by the master node control layer. The terminal physical layer is the user terminal, which uploads business data between users.

[0049] The first and second message generation and processing modules are identical: the messages in this module include Hello messages, LSR link state request messages, LSA link state announcement messages, and LSAck receive response messages. This module is mainly responsible for two aspects: first, generating and sending corresponding messages based on the state changes of this node; second, receiving messages sent by other nodes, identifying the message type (Type), extracting relevant information from the message, and pushing this information to the first and second storage modules for storage. Both master and slave nodes have this functional module, but they differ in function: the messages generated by the master node include Hello messages, LSR link state request messages, and LSAck receive response messages; the messages generated by the slave node include Hello messages, LSA link state announcement messages, and LSR link state request messages.

[0050] The first and second storage modules are identical: the data collected or calculated by each module is pushed to the first and second storage modules for storage. In addition, the first and second storage modules will save the latest data and remove the corresponding invalid data according to the data's age.

[0051] Routing calculation module: Reads the link status in the network area from the first storage module, uses Dijkstra's algorithm, and uses link cost as the key indicator to calculate the route with the lowest total link cost between any two nodes;

[0052] The first and second control modules are identical: the first and second control modules identify the status of their respective nodes and control the execution of each module.

[0053] Interface detection module: Link cost refers to the bandwidth utilization efficiency of the interface. The function of this module is to detect the bandwidth utilization of the interface in real time and promptly report it to the second control module.

[0054] Photoelectric conversion module: Converts the data in the second storage module from electrical signals into optical signals for transmission;

[0055] Wavelength conversion module: Based on the control signals sent by the master node, the slave node adjusts the wavelength transmitted by this node;

[0056] The master node control layer coordinates global optical network resources and makes routing decisions. The specific operation steps are shown in Figure 2.

[0057] Step 1.1: Periodically send Hello messages to establish and maintain adjacency with directly connected nodes;

[0058] Step 1.2: After the adjacency state is established, broadcast an LSR link state request message to the slave nodes in the area, requesting the slave nodes to send their own LSA link state advertisement messages;

[0059] Step 1.3: Upon receiving the LSA link status advertisement message from the slave node, the master node sends an LSAck receive response message back to the slave node, indicating that the master node has received the LSA link status advertisement message.

[0060] Step 1.4: The first storage module saves the LSA link status announcement messages of all slave nodes in the area and establishes an area database, which includes each node ID number, interface address, link cost, interface wavelength, ID number of directly connected nodes and address of directly connected interfaces. The topology structure is established through the above data.

[0061] Step 1.5: The routing calculation module accesses the database in the first storage module and uses the Dijkstra algorithm to calculate a route with the minimum total cost of the link. Therefore, the link cost, as the most important parameter, will directly determine the route between nodes. The link cost refers to the efficiency of interface bandwidth utilization, that is, Cost = business data volume / interface rate.

[0062] The node link cost (Cost) is detected and uploaded in real time at the node layer, and data is transmitted according to the routing plan published by the master node control layer. The specific operation steps are shown in Figure 3.

[0063] Step 2.1: Periodically send Hello messages to establish and maintain adjacency with directly connected nodes;

[0064] Step 2.2: Receive the LSR link status request message sent by the master node;

[0065] Step 2.3: Respond to the master node's request, encapsulate the local node's LSA link status announcement message. The message content includes the local node ID number, interface address, link cost, interface wavelength, ID number of the directly connected node, and address of the directly connected node. After sending the LSA link status announcement message, receive the master node's LSAck receive response message.

[0066] Step 2.4: Receive the routing table calculated by the master node and transmit data according to the routing table;

[0067] Step 2.5: Determine whether the connection with the directly connected node is broken, and then exit the adjacency state; if exiting the adjacency state, update the LSA link state advertisement message of this node and immediately send the LSA link state advertisement message to the master node.

[0068] Step 2.6: Determine whether the link cost (Cost) of this node has changed, i.e., whether the interface bandwidth utilization efficiency has changed; divide the interface bandwidth utilization efficiency into three states: A, B, and C. The bandwidth utilization efficiency corresponding to state A is [0, 0.5], the bandwidth utilization efficiency corresponding to state B is [0.5, 0.8], and the bandwidth utilization efficiency corresponding to state C is [0.8, 1]. If the Cost state of any interface in the node changes, update the LSA link status announcement message of this node and immediately send this message to the master node.

[0069] Step 2.7: Receive the latest routing table sent by the master node and transmit data according to the routing table.

[0070] The structures of Hello messages and LSR link state request messages are shown in Figure 4. Both messages have the same message header, which consists of message type (Type), total message length (Packet length), local node identifier (Node ID), and checksum. Message type (Type) = 1 and 2, indicating that the message is a Hello message and an LSR link state request message, respectively. The message length (Packet length) represents the total length of the message, including the header. The local node identifier (Node ID) indicates the identifier of the node sending this message. The checksum is the verification sum of the entire message, including the authentication field.

[0071] The Hello message structure is shown in Figure 4(a). It consists of a message header and a message content. Its function is to establish and maintain adjacency relationships. It is sent periodically on the enabled interface. The message content includes the Link ID of the interface that sent the message, the Hello Interval of the time interval for sending the message, the DeadInterval used to determine whether the directly connected node is disconnected, and the Neighbor ID of the known neighbor node.

[0072] The structure of the LSR (Link State Request) message is shown in Figure 4(b). It consists of a message header and a message content. Its function is for the master node to send the required LSA (Link State Advertisement) message to the slave node. The message content includes the sequence number (Sequence Number). The master node periodically sends LSR messages, and the sequence number is incremented by 1 at this time. The node ID identifier (Advertising ID) that generates this message is also included.

[0073] The structure of the LSA link-state advertisement message and the LSAck receive response message is shown in Figure 5. They have the same message header, as shown in Figure 5(a). The message header consists of message type (Type), total message length (Packet length), node ID (Node ID that sent the message), checksum, LSA sequence number (LS sequence number), advertising ID (Node ID that generated the message), and message generation time (LS age). Among them, message type (Type) = 3 and 4, indicating that this message is an LSA link-state advertisement message and an LSAck receive response message; message length (Packet length) indicates the total length of the message including the message header; LSA sequence number (LS sequence number) is used by the master node to determine whether it is the latest LSA link-state advertisement message; and LS age is the time when the message was generated.

[0074] The structure of the LSA Link State Advertisement message is shown in Figure 5(b). It consists of a message header and a message content. This message is generated by the slave node and describes the slave node's own link status and directly connected nodes. The LSA Link State Advertisement message header and message content include the number of interfaces of this node (links), the number of interfaces directly connected to this node (#Links), the port identifier (Link ID) of this node, the link cost (Cost), the interface output wavelength (Wave length), the neighbor node identifier (Neighbor ID), the corresponding directly connected port identifier (#Link ID), and other known neighbors and ports.

[0075] The LSAck receive response message structure is shown in Figure 5(c). It consists of a message header and a message content. This message is generated by the master node to confirm receipt of the slave node's LSA link state announcement message, so as to prevent the slave node from repeatedly sending LSA link state announcement messages.

[0076] The following specific examples illustrate this:

[0077] An optical network based on a multi-node mesh topology is divided into a control layer and a transport layer. The control layer is mainly responsible for routing planning, while the transport layer is responsible for transmitting service data. As shown in Figure 6, four slave nodes act as the transport layer, interconnected by two optical fibers of different wavelengths. The master node, acting as the control layer, obtains the link status and link cost within the region and makes routing decisions.

[0078] The node and the link status between nodes are as follows:

[0079] (1) Slave node 1 and slave node 3: Slave node 1 is identified by ID_1 and slave node 3 is identified by ID_3. The two nodes are interconnected by optical fibers with wavelengths of λ1 and λ2, respectively connecting the interface of slot 2 (1 / 2) and the interface of slot 1 (1 / 1) of the two nodes. The cost of the λ1 optical fiber link is 0.7 and the cost of the λ2 optical fiber link is 0.7.

[0080] (2) Slave node 1 and slave node 2: Slave node 1 is identified by ID_1 and slave node 2 is identified by ID_2. The two nodes are interconnected by optical fibers with wavelengths of λ3 and λ4, respectively connecting the 0 slot 2 interface (0 / 2) and the 0 slot 1 interface (0 / 1) of the two nodes. The cost of the λ3 optical fiber link is 0.5 and the cost of the λ4 optical fiber link is 0.6.

[0081] (3) Slave node 2 and slave node 4: Slave node 2 is identified by ID_2 and slave node 4 is identified by ID_4. The two nodes are interconnected by optical fibers with wavelengths of λ5 and λ6, respectively connecting the interface of slot 2 (1 / 2) and the interface of slot 1 (1 / 1) of the two nodes. The cost of the λ5 optical fiber link is 0.6 and the cost of the λ6 optical fiber link is 0.5.

[0082] (4) Slave node 3 and slave node 4: Slave node 3 is identified by ID_3 and slave node 4 is identified by ID_4. The two nodes are interconnected by optical fibers with wavelengths of λ7 and λ8, respectively connecting the 0 slot 2 interface (0 / 2) and the 0 slot 1 interface (0 / 1) of the two nodes. The cost of the λ7 optical fiber link is 0.7 and the cost of the λ8 optical fiber link is 0.7.

[0083] (5) Slave node 2 and slave node 3: Slave node 2 is identified by ID_2 and slave node 3 is identified by ID_3. The two nodes are interconnected by optical fibers with wavelengths of λ9 and λ10, respectively connecting the interface of slot 1 (2 / 1) and the interface of slot 2 (2 / 2) of the two nodes. The cost of the λ7 optical fiber link is 0.5 and the cost of the λ8 optical fiber link is 0.5.

[0084] The specific steps of this method are as follows:

[0085] Step 1: As shown in Figure 7, the node periodically sends Hello messages at 10-second intervals to establish and maintain the adjacency state. If it does not receive a Hello message from the other party within 40 seconds, it is considered that the other party has lost connection.

[0086] Step 1.1: The 0 / 1 ports of slave node 1 and slave node 2 are interconnected via optical fiber. Slave node 1 identifies its own node ID and enabled port. The second message generation and processing module creates the initial Hello message for this node: Message type Type=1, Node ID = ID_1, Link ID = 0 / 1, Hello Interval = 10, Dead Interval = 40. At this time, Neighbor ID is empty.

[0087] Step 1.2: After receiving the Hello message from slave node 1, the second message generation and processing module updates its own Hello message: Message type Type=1, Node ID = ID_2, Link ID = 0 / 1, Hello Interval =10, Dead Interval = 40, Neighbor ID = ID_1. After updating the message, slave node 2 sends the message to slave node 1.

[0088] Step 1.3: Node 1 receives a Hello message from Node 2 and identifies that the Neighbor ID in the message contains Node 1's own node identifier. At this point, Node 1 enters the adjacency state.

[0089] Step 1.4: Update the Hello message from node 1 with Type=1, Node ID=ID_1, Link ID=0 / 1, Hello Interval=10s, Dead Interval=40s, Neighbor ID=ID_2. After updating the message, node 1 sends the message to node 2.

[0090] Step 1.5: Node 2 receives the Hello message from Node 1 and identifies that the Neighbor ID in the message contains Node 2's own node identifier. At this point, Node 2 enters the adjacency state.

[0091] Step 1.6: Every 10 seconds, slave node 1 and slave node 2 send Hello messages to each other to maintain their adjacency.

[0092] Step 1.7: If no Hello message is received within 40 seconds, it is considered a disconnection and the device exits the adjacency state;

[0093] Step 1.8: The remaining nodes establish their adjacency states in the same way as in steps 1.1 to 1.7 above;

[0094] After the adjacency state is established, the master node needs to grasp the global link state to facilitate route planning. In this example, the link cost is the main indicator for route planning. Changes in the cost will directly affect the nodes and wavelengths of the data transmission path.

[0095] Step 2: The master node broadcasts an LSR link state request and receives an LSA link state advertisement message sent by the slave node. The interaction process is shown in Figure 8.

[0096] Step 2.1: After the nodes in the region are in an adjacency state, the master node creates an LSR link state request message;

[0097] Step 2.2: The master node periodically broadcasts LSR link state request messages to all slave nodes in the region and requests the required LSA link state advertisement messages from the slave nodes;

[0098] Step 2.3: The slave node receives the LSR (Link State Request) message from the master node. At time T1=20s, it sends an LSA (Link State Advertisement) message to the master node. As shown in Figure 6, the LSA message sent by slave node 1 contains the following information: Type = 3, Node ID = ID_1, LS sequence number = 1, Advertising ID = ID_1, LS age = 20, links = 4, #links = 4, Link ID = 1 / 2, cost = 0.7, Wave length = λ1, Neighbor ID = ID_3, #Link ID = 1 / 2; Link ID = 1 / 1, cost = 0.7, Wave length = λ2, Neighbor ID = ID_3, #Link ID = 1 / 1; Link ID = 0 / 1, cost = 0.6, Wave length = λ4, Neighbor ID = ID_2, #Link ID = 0 / 1; Link ID = 0 / 2, cost = 0.5, Wave length = λ3, Neighbor ID = ID_3, #Link ID=0 / 2; The message describes the number of interfaces on node 1, the interface identifiers, and the interface information of its directly connected nodes;

[0099] Step 2.4: After receiving the LSA link status advertisement message, the master node sends back an LSAck reception response message;

[0100] Step 2.5: The remaining slave nodes also send LSA link status announcement messages to the master node through steps 2.1 to 2.4;

[0101] Step 2.6: Based on the obtained LSA link state advertisement messages from the slave nodes, the master node infers the area topology. The routing calculation module uses Dijkstra's algorithm to calculate a route with the minimum total link cost. As shown in Figure 6, the minimum total route cost between slave node 1 and slave node 4 is Cost. （λ3, λ6） =cost λ3 + cost λ6 =1, where (λ3, λ6) represents λ3 between slave node 1 and slave node 2 and λ6 between slave node 2 and slave node 4. Then the route between slave node 1 and slave node 4 is planned as (slave node 1, slave node 2, slave node 4) with wavelength (λ3, λ6).

[0102] Step 2.7: Receive the routing plan calculated by the master node from the slave node, store it in the second storage module, and send data according to the routing plan;

[0103] Step 2.8: The node interface detection module detects its own port cost in real time. If it exceeds the threshold range, it updates the LSA link status announcement message of this node and sends it to the master node. As shown in Figure 6, the cost of port 2 (0 / 2) of frame 0 of node 1 changes from 0.5 to 0.8, that is, the state changes from B to C. This triggers the mechanism of sending LSA link status announcement messages, updates the cost value in the LSA link status announcement message, and immediately sends this message to the master node.

[0104] Step 2.9: The master node receives the LSA link status advertisement message and re-plans the route;

[0105] Step 2.10: If the Hello message from a slave node does not respond for more than the Dead Interval, it exits the adjacency state. Then, it updates the LSA link state advertisement message of this node and sends it to the master node.

[0106] Step 2.11: The master node receives the LSA link state advertisement message and re-plans the route.

Claims

1. A centralized dynamic routing method for optical networks, characterized in that, The method is based on the interaction between the master node control layer and the slave node data transmission layer. The slave node data transmission layer obtains services from the terminal. The master node control layer includes a first message generation and processing module, a route calculation module, a first storage module, and a first control module. The master node control layer collects the link status of all nodes within the storage area and calculates the routes between any two nodes, controlling the data transmission routes between nodes within the area to reduce local link congestion and improve overall resource utilization. The slave node data transmission layer includes a second message generation and processing module, an interface detection module, a second storage module, a wavelength conversion module, a photoelectric conversion module, and a second control module. It transmits data uploaded by the terminal physical layer according to the routes calculated by the master node control layer. The terminal physical layer is the user terminal, which uploads service data between users. The second message generation and processing module is the same: the messages in this module include Hello messages, LSR link state request messages, LSA link state announcement messages, and LSAck receive response messages. This module is mainly responsible for two aspects: first, generating and sending corresponding messages based on the state changes of this node; second, receiving messages sent by other nodes, identifying the message type (Type), extracting relevant information from the message, and pushing this information to the storage module for storage. Both master and slave nodes have this function module, but there are some differences in functionality: the messages generated by the master node include Hello messages, LSR link state request messages, and LSAck receive response messages; the messages generated by the slave node include Hello messages, LSA link state announcement messages, and LSR link state request messages. The first and second storage modules are identical: the data collected or calculated by each module is pushed to the first and second storage modules for storage. In addition, the first and second storage modules will save the latest data and remove the corresponding invalid data according to the data's age. The routing calculation module reads the link status in the network area from the first storage module, uses Dijkstra's algorithm, and uses the link cost as the key indicator to calculate the route with the lowest total link cost between any two nodes. The first and second control modules are identical: they identify the status of their respective nodes and control the execution of each module; the interface detection module (link cost refers to the bandwidth utilization efficiency of the interface) is responsible for real-time detection of interface bandwidth usage and timely feedback to the second control module; the photoelectric conversion module converts the data in the second storage module from electrical signals to optical signals for transmission; the wavelength conversion module adjusts the transmission wavelength of the slave node according to the control signals sent by the master node; the master node control layer coordinates global optical network resources and makes routing decisions. The specific operation steps are as follows: Step 1.1: Periodically send Hello messages to establish and maintain adjacency with directly connected nodes; Step 1.2: After the adjacency is established, broadcast LSR link status request messages to slave nodes in the area, requesting slave nodes to send their own LSA link status announcement messages; Step 1.3: ... 1.3: Upon receiving an LSA link status advertisement message from a slave node, the master node sends an LSAck response message back to the slave node, indicating that the master node has received the LSA link status advertisement message; Step 1.4: The first storage module saves the LSA link status advertisement messages of all slave nodes in the area, establishes an area database, including each node ID, interface address, link cost, interface wavelength, ID of directly connected nodes, and address of directly connected interfaces, and establishes the topology structure based on the above data; Step 1.5: The routing calculation module accesses the database in the first storage module and uses the Dijkstra algorithm to calculate a route with the minimum total link cost. Therefore, the link cost, as the most important parameter, will directly determine the routing between nodes, and the link cost refers to the efficiency of interface bandwidth utilization, i.e., Cost = service data volume / interface rate; The node layer detects and uploads the node link cost in real time, and transmits data according to the routing plan published by the master node control layer. The specific operation steps are as follows: Step 2.1: Periodically send Hello messages to establish and maintain adjacency with directly connected nodes; Step 2.2: Receive LSR link state request messages from the master node; Step 2.3: Respond to the master node's request by encapsulating the node's LSA link state advertisement message. The message content includes the node ID, interface address, link cost, interface wavelength, ID and address of directly connected nodes. After sending the LSA link state advertisement message, receive the master node's LSAck reception response message; Step 2.4: Receive the routing table calculated by the master node and transmit data according to the routing table; Step 2.5: Determine whether the connection with directly connected nodes is broken, and then exit the adjacency state. If the node exits the adjacency state, update the local node's LSA link state advertisement message and immediately send an LSA link state advertisement message to the master node; Step 2.6: Determine whether the local node's link cost (Cost) has changed, i.e., whether the interface bandwidth utilization efficiency has changed; Divide the interface bandwidth utilization efficiency into three states: A, B, and C. The bandwidth utilization efficiency corresponding to state A is [0, 0.5], the bandwidth utilization efficiency corresponding to state B is [0.5, 0.8], and the bandwidth utilization efficiency corresponding to state C is [0.8, 1]. If the Cost state of any interface in the node changes, update the local node's LSA link state advertisement message and immediately send this message to the master node; Step 2.7: Receive the latest routing table sent by the master node and transmit data according to the routing table.

2. The centralized dynamic routing method for optical networks according to claim 1, characterized in that, The Hello message and the LSR Link State Request message have the same message header. This message header consists of message type (Type), total message length (Packet length), local node identifier (Node ID), and checksum. Message type (Type) = 1 and 2 indicate that the message is a Hello message and an LSR Link State Request message, respectively. Message length (Packet length) represents the total length of the message, including the message header. Local node identifier (Node ID) indicates the identifier of the node sending this message. Checksum is the checksum of the entire message, including the authentication field.

3. The centralized dynamic routing method for optical networks according to claim 1, characterized in that, The Hello message consists of a header and a content. Its function is to establish and maintain adjacency relationships. It is sent periodically on the enabled interface. The content includes the Link ID of the interface that sent the message, the HelloInterval of the time interval for sending the message, the Dead Interval used to determine whether the directly connected node has lost connection, and the Neighbor ID of the known neighbor node.

4. The centralized dynamic routing method for optical networks according to claim 1, characterized in that, The LSR link state request message consists of a message header and a message content. Its function is for the master node to send the required LSA link state advertisement message to the slave node. The message content includes a sequence number. The master node periodically sends LSR link state request messages, and the sequence number is incremented by 1 at this time. The node ID identifier that generates this message is the Advertising ID.

5. The centralized dynamic routing method for optical networks according to claim 1, characterized in that, The LSA link-state advertisement message and the LSAck receive response message have the same header. This header consists of message type (Type), total message length (Packet length), node ID (Identifier of the node sending this message), checksum, LSA sequence number (LS sequence number), advertising ID (Identifier of the node that generated this message), and message generation time (LS age). Among them, message type (Type=3 and 4) indicates that this message is an LSA link-state advertisement message and an LSAck receive response message; message length (Packet length) indicates the total length of the message including the header; LSA sequence number (LS sequence number) is used by the master node to determine whether it is the latest LSA link-state advertisement message; and LS age is the time when the message was generated.

6. The centralized dynamic routing method for optical networks according to claim 1, characterized in that, The LSA Link State Advertisement message consists of a header and a content. This message is generated by the slave node and describes the slave node's own link status and directly connected nodes. The header content of the LSA Link State Advertisement message includes the number of interfaces of this node (links), the number of interfaces directly connected to this node (#Links), the port identifier (Link ID) of this node, the link cost (Cost), the output wavelength (Wave length) of the interface, the neighbor node identifier (Neighbor ID), the corresponding directly connected port identifier (#Link ID), and other known neighbors and ports.

7. The centralized dynamic routing method for an optical network according to claim 1, characterized in that, The LSAck receive response message consists of a message header and a message content. This message is generated by the master node and is used to confirm receipt of the LSA link status announcement message from the slave node, so as to prevent the slave node from repeatedly sending LSA link status announcement messages.

Images