Ethernet Neighbor Discovery via Link Configuration Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current transport network devices are unable to enable neighbor discovery between a client layer network device and an OTN cross-connect when connected via Ethernet, as they cannot decode or generate messages containing IP addresses and port numbers due to transparent mapping of Ethernet data packets to optical multiplex units.
Innovation Solution
A method where a transport network device sends a link configuration pattern, such as a predetermined bit sequence, invalid, or valid Ethernet packet, containing its ID and port number over the Ethernet connection to a client layer device, allowing both devices to determine their connection and exchange information, enabling auto-discovery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If transparent mapping of Ethernet packets to multiplex units is used in transport network devices, then data transmission efficiency is improved, but neighbor discovery capability deteriorates
Solution Approach 1:
The patent segments the Ethernet packet processing function into two parts: standard Ethernet packets are transparently mapped for efficient transmission, while special discovery packets are extracted and processed separately by the CPU to enable neighbor discovery. This segmentation allows both high-speed transparent mapping and intelligent discovery operations to coexist without interfering with each other.
Solution Approach 2:
The patent introduces a mediator mechanism where the CPU acts as an intermediary between the transparent mapping function and neighbor discovery requirement. The CPU receives special discovery packets, processes them to extract neighbor information, and generates appropriate responses, thereby enabling discovery capability while maintaining the efficiency of transparent mapping for regular traffic.
2Device complexity
If standard Ethernet packet processing is used in transport network devices, then device complexity is reduced, but neighbor discovery function is lost
Solution Approach 1:
The patent applies local quality by making the Ethernet interface selectively process different types of packets differently. Standard Ethernet packets are handled by the simple transparent mapping logic, while special discovery packets are identified and routed to the CPU for intelligent processing. This localized differentiation adds discovery functionality only where needed without complicating the entire device.
Solution Approach 2:
The patent uses partial action by implementing only the minimal necessary processing capability in the transport network device - specifically, the ability to identify and process special discovery packets. The device does not need full Ethernet processing capability, only the partial function required for discovery, thereby maintaining low complexity while gaining the essential discovery function.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention relates to network management in an optical network for performing automatic neighbor discovery. A client layer device (10) is connected over an Ethernet connection (12) to an transport network device (11), which transparently maps Ethernet packets to multiplex units and vice versa. The transport network device may be an OTN or an SDH device. The method comprises the steps of: determining (21), by a first of the devices (10,11), that a port (14) of the transport network device (11) is connected via the Ethernet connection (12) to a port (13) of the client layer device (10);' and sending (23), by the first device; a link configuration pattern over the Ethernet connection (12), the link configuration pattern being one of: a bit sequence, an invalid Ethernet packet and a valid Ethernet packet, the link configuration pattern containing an identifier identifying the first device and a port identifier identifying the port of the first device.