Method of self-learning for the switching nodes of a data transmission network
US6628655B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Jan 24, 2000 |
| Grant date | Sep 30, 2003 |
| Priority date | — |
| Expiry date | Jan 24, 2020 |
Classification
- Technology area (CPC H)Electricity
- CPC primaryH04L2012/5667
- WIPO fieldDigital communication
- WIPO sectorElectrical engineering
Abstract
Method of self-learning for a switching node in a data transmission network (10) wherein Internet Protocol (IP) data frames are transmitted between a sending unit (20 or 24) and a receiving unit (22 or 26) via an ingress switching node (12) linked to an egress switching node (16 or 18) by a plurality of intermediate switching nodes (14) interconnected by trunks. The method consists principally in generating, in the input port/trunk interface of the switching node, a switch header including a source identification field, a destination identification field and a temporary label identifying the flow of data in response to its first data frame, this first switch header being concatenated to said data frame before being transmitted to the router interface of the switching node via its switch engine when the data frame does not belong to a known flow of data and the router has to define the routing of the flow of data. An update message is sent back from the output trunk/port interface to the input port/trunk interface to update the latter with the destination identification and with a switching label replacing the temporary label associated with the flow of data, in order to transmit th…
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.