Beta
ManageSearchChatDiscover
63480

Comparison between On-line Supervised Learning and Unsupervised Learning of Link Quality estimates in wireless networks Performance

Article

Last updated: 25 Dec 2024

OverviewSimilar Items

Subjects

-

Tags

-

Abstract

In this paper, Optimized Link State routing protocol (OLSR) and Link Quality Learning protocol (LQL) are employed to show how to control the rerouting time process of a local area network. At every node and for every wireless link, OLSR passively gathers measurements to quantify the network features current value and to assess the related link quality value according to many metrics. The measurements, of These metrics, are used as training samples for the incremental, regression mapping, and supervised learning model which can be used by network protocol LQL to derive link qualities in real-time and robust manner. According to OLSR and LQL, Nodes can also cooperate by exchanging training samples, speeding up the overall learning process. This cooperative feature is useful when the local network configurations are continually changing because of mobility and/or varying traffic pattern. We propose a distributed online protocol based on organized maps learning technique called Self Organized Mapping (SOM). Three performance methods are applied to the wireless network used within this paper: OLSR_SOM that cares making modification, OLSR protocol code, OLSR_LQL that depends on the introducing LQL as a supervised learning protocol and OLSR_SOM that depends on the introducing SOM as an unsupervised learning protocol.

DOI

10.21608/mjeer.2017.63480

Authors

First Name

M. F.

Last Name

ZAGHLOUL

MiddleName

-

Affiliation

Dept. of Computers and Systems Eng., Faculty of Eng., Al-azhar University

Email

-

City

-

Orcid

-

First Name

R. H.

Last Name

ABO EL-EZZ

MiddleName

-

Affiliation

Dept. of Computers and Systems Eng., Faculty of Eng., Al-azhar University

Email

-

City

-

Orcid

-

First Name

M. M.

Last Name

ABO-GHAZALA

MiddleName

-

Affiliation

Dept. of Computers and Systems Eng., Faculty of Eng., Al-azhar University

Email

-

City

-

Orcid

-

Volume

26

Article Issue

1

Related Issue

9502

Issue Date

2017-01-01

Receive Date

2019-12-03

Publish Date

2017-01-01

Page Start

226

Page End

246

Print ISSN

1687-1189

Online ISSN

2682-3535

Link

https://mjeer.journals.ekb.eg/article_63480.html

Detail API

https://mjeer.journals.ekb.eg/service?article_code=63480

Order

14

Type

Original Article

Type Code

1,088

Publication Type

Journal

Publication Title

Menoufia Journal of Electronic Engineering Research

Publication Link

https://mjeer.journals.ekb.eg/

MainTitle

Comparison between On-line Supervised Learning and Unsupervised Learning of Link Quality estimates in wireless networks Performance

Details

Type

Article

Created At

22 Jan 2023