Learning-based admission control adjustment in data networks
US6791941B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Sep 28, 1999 |
| Grant date | Sep 14, 2004 |
| Priority date | — |
| Expiry date | Sep 28, 2019 |
Classification
- Technology area (CPC H)Electricity
- CPC primaryH04L47/83
- WIPO fieldDigital communication
- WIPO sectorElectrical engineering
Abstract
Tuning for connection admission control (CAC) algorithms in broadband ATM networks is accomplished using an overbooking technique based on aggregate effective bandwidth, AEBW, as an approximation to required bandwidth for given levels and classes of network traffic. An overbooking gain factor, &agr;t is developed—typically in a network operations system—based on long-term measurements supplied periodically by network switches. In operation, overbooking is introduced in small increments until a threshold (such as a cell loss threshold) is reached, at which point overbooking is reduced in a large step. A variety of network measurements can be used to apply the present invention to a range of traffic and service contexts.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.