Autonomous Smart Routing for Network QoS

Erol Gelenbe; Michael Gellman; Ricardo Lent; Peixiang Liu; Pu Su

doi:10.1109/ICAC.2004.1301368

Autonomous Smart Routing for Network QoS

Erol Gelenbe
, Michael Gellman
, Ricardo Lent
, Peixiang Liu
, Pu Su

Research output: Contribution to conference › Presentation

Abstract

We present an autonomous adaptive quality of service (QoS) driven network system called a "Cognitive Packet Network" (CPN), which adoptively selects paths so as to offer best effort QoS to the end users based on user defined QoS. CPN uses neural network based reinforcement learning to make routing decisions separately at each node. Measurements on an experimental test-bed are provided to show how the system responds to the choice of QoS goals. We also discuss and evaluate an extension of CPN that uses a genetic algorithm to generate and maintain paths from previously discovered information by matching their "fitness" with respect to the desired QoS.

Original language	American English
Pages	232-239
Number of pages	8
DOIs	https://doi.org/10.1109/ICAC.2004.1301368
State	Published - May 17 2004
Event	Proceedings from International Conference On Autonomic Computing - Duration: May 17 2004 → …

Conference

Conference	Proceedings from International Conference On Autonomic Computing
Period	5/17/04 → …

ASJC Scopus Subject Areas

General Engineering

Keywords

transport protocols
computer networks
telecommunication network routing
AI
learning
genetic algorithms
quality of service
neural nets

Disciplines

Computer Sciences

Access to Document

10.1109/ICAC.2004.1301368

http://dx.doi.org/10.1109/ICAC.2004.1301368

Cite this

@conference{f575a5b4dace4f61952663507d978e07,

title = "Autonomous Smart Routing for Network QoS",

abstract = "We present an autonomous adaptive quality of service (QoS) driven network system called a {"}Cognitive Packet Network{"} (CPN), which adoptively selects paths so as to offer best effort QoS to the end users based on user defined QoS. CPN uses neural network based reinforcement learning to make routing decisions separately at each node. Measurements on an experimental test-bed are provided to show how the system responds to the choice of QoS goals. We also discuss and evaluate an extension of CPN that uses a genetic algorithm to generate and maintain paths from previously discovered information by matching their {"}fitness{"} with respect to the desired QoS.",

keywords = "transport protocols, computer networks, telecommunication network routing, AI, learning, genetic algorithms, quality of service, neural nets",

author = "Erol Gelenbe and Michael Gellman and Ricardo Lent and Peixiang Liu and Pu Su",

year = "2004",

month = may,

day = "17",

doi = "10.1109/ICAC.2004.1301368",

language = "American English",

pages = "232--239",

note = "Proceedings from International Conference On Autonomic Computing ; Conference date: 17-05-2004",

}

TY - CONF

T1 - Autonomous Smart Routing for Network QoS

AU - Gelenbe, Erol

AU - Gellman, Michael

AU - Lent, Ricardo

AU - Liu, Peixiang

AU - Su, Pu

PY - 2004/5/17

Y1 - 2004/5/17

N2 - We present an autonomous adaptive quality of service (QoS) driven network system called a "Cognitive Packet Network" (CPN), which adoptively selects paths so as to offer best effort QoS to the end users based on user defined QoS. CPN uses neural network based reinforcement learning to make routing decisions separately at each node. Measurements on an experimental test-bed are provided to show how the system responds to the choice of QoS goals. We also discuss and evaluate an extension of CPN that uses a genetic algorithm to generate and maintain paths from previously discovered information by matching their "fitness" with respect to the desired QoS.

AB - We present an autonomous adaptive quality of service (QoS) driven network system called a "Cognitive Packet Network" (CPN), which adoptively selects paths so as to offer best effort QoS to the end users based on user defined QoS. CPN uses neural network based reinforcement learning to make routing decisions separately at each node. Measurements on an experimental test-bed are provided to show how the system responds to the choice of QoS goals. We also discuss and evaluate an extension of CPN that uses a genetic algorithm to generate and maintain paths from previously discovered information by matching their "fitness" with respect to the desired QoS.

KW - transport protocols

KW - computer networks

KW - telecommunication network routing

KW - AI

KW - learning

KW - genetic algorithms

KW - quality of service

KW - neural nets

UR - https://nsuworks.nova.edu/gscis_facpres/76

UR - http://dx.doi.org/10.1109/ICAC.2004.1301368

U2 - 10.1109/ICAC.2004.1301368

DO - 10.1109/ICAC.2004.1301368

M3 - Presentation

SP - 232

EP - 239

T2 - Proceedings from International Conference On Autonomic Computing

Y2 - 17 May 2004

ER -

Autonomous Smart Routing for Network QoS

Abstract

Conference

ASJC Scopus Subject Areas

Keywords

Disciplines

Access to Document

Other files and links

Fingerprint

Cite this