Ethernet Ring Node with Duplicated Rings for Failure Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ethernet-based ring communication systems face challenges in configuring a ring topology without repeaters, managing failures, and ensuring punctuality due to collision and congestion issues, and lack a method for easily checking station information delivery across all nodes.
Innovation Solution
A communication node design with duplicated normal-system and standby-system rings, where each node has specific ports for input/output operations and a processor for frame management, enabling loop-back processes to isolate failures and ensure continuous communication without repeaters, and a method for checking ring configuration and station information delivery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If Ethernet-based ring topology is configured without repeaters, then cost is reduced and device complexity is simplified, but collision and congestion occur making it difficult to ensure punctuality
Solution Approach 1:
The patent segments the ring network into multiple communication channels by introducing virtual channel identifiers (VCIs). Each channel operates independently with its own token circulation, dividing the shared medium into dedicated logical paths. This segmentation prevents collisions between different data streams while maintaining the simple ring topology structure without repeaters.
Solution Approach 2:
The patent implements dynamic channel allocation where communication channels are created, modified, and deleted based on real-time network requirements. Virtual channels are dynamically established between node pairs needing communication, and the system adapts channel configurations as nodes are added or removed from the ring, providing flexible resource utilization without fixed assignments.
2Ease of operation
If star topology is used with a repeater (hub), then communication is simplified, but the repeater becomes a bottleneck and communication performance deteriorates
Solution Approach 1:
The patent extracts the centralization function from a physical repeater and distributes it across multiple nodes in the ring. Each node independently manages its own communication channels and performs local frame forwarding decisions, eliminating the single-point bottleneck while maintaining simplified operation through standardized node behavior.
Solution Approach 2:
Each node in the ring is designed with multi-functionality, serving both as an endpoint for communication and as a forwarding node for other nodes' traffic. Nodes can simultaneously participate in multiple virtual channels, acting as sources, destinations, and intermediate relay points, thereby distributing the workload and eliminating communication bottlenecks.
3Reliability
If FDDI token passing system is implemented, then collision control is achieved, but the system cannot be applied to Ethernet-based networks
Solution Approach 1:
The patent adapts the token passing mechanism by changing key parameters: instead of FDDI's physical layer tokens, it uses Ethernet-layer frames with special token identifiers. The token is implemented as a standard Ethernet frame with a specific VCI value, allowing it to traverse Ethernet switches and bridges while maintaining collision-free access through logical token control rather than physical carrier sensing.
Solution Approach 2:
The patent creates an Ethernet-compatible version of the token passing system by copying the essential control logic of FDDI token management into the Ethernet protocol stack. Virtual channel identifiers and token frame structures are designed to be Ethernet-compatible, allowing the system to replicate FDDI's collision-free performance while using standard Ethernet hardware and cabling.
4Loss of information
If station information transfer is performed from one source to all nodes, then network configuration is established, but the process becomes complicated requiring response checking
Solution Approach 1:
The patent merges the station information distribution process with the virtual channel establishment procedure. When virtual channels are created, station information is automatically carried along with the channel setup frames. Nodes receive station information as a byproduct of channel configuration, eliminating the need for separate information transfer phases and complex acknowledgment protocols.
Solution Approach 2:
The patent performs station information distribution in advance during network initialization and node joining processes. Before normal data communication begins, all nodes are pre-configured with the station information of other network members through control frames circulated during the setup phase, so that no additional information exchange is needed during operational communication.
Data Source
AI summary
A plurality of communication nodes are one-to-one connected to each other by Ethernet (registered trademark) which is duplicated with a normal-system ring and with a standby-system ring. The communication node includes an A port, a B port, and a communication processor that performs a process of receiving a frame output from the B port and a process of transmitting the frame from the normal-system ring from the B port when the communication is normal. When other communication node performs a loop-back process due to communication abnormality, the communication node performs only a process of forwarding a frame input from the standby-system ring in the B port without receiving the frame and performs a process of transmitting the frame from the A port to the standby-system ring.