Return to search

A computational-based methodology for the rapid determination of initial AP location for WLAN deployment

The determination of the optimal location of transceivers is a critical design
factor when deploying a wireless local area network (WLAN). The performance of
the WLAN will improve in a variety of aspects when the transceivers' locations are
adequately determined, including the overall cell coverage to the battery life of the
client units. Currently, the most common method to determine the appropriate
location of transceivers is known as a site survey, which is normally a very time and
energy consuming process.
The main objective of this research was to improve current methodologies for
the optimal or near-optimal placement of APs in a WLAN installation. To achieve
this objective, several improvements and additions were made to an existing
computational tool to reflect the evolution that WLAN equipment has experienced in
recent years. Major additions to the computational tool included the addition of the
capability to handle multiple power levels for the transceivers, the implementation of
a more adequate and precise representation of the passive interference sources for the
path loss calculations, and the definition of a termination criterion to achieve
reasonable computational times without compromising the quality of the solution.
An experiment was designed to assess how the improvements made to the
computational tool provided the desired balance between computational time and the
quality of the solutions obtained. The controlled factors were the level of strictness
of the termination criterion (i.e., high or low), and the number of runs performed
(i.e., 1, 5, 10, 15, and 20 runs). The low level of strictness proved to dramatically
reduce (i.e., from 65 to 70%) the running time required to obtain an acceptable
solution when compared to that obtained at the high level of strictness. The quality
of the solutions found with a single run was considerably lower than that obtained
with the any other number of runs. On the other hand, the quality of the solutions
seemed to stabilize at and after 10 runs, indicating that there is no added value to the
quality of the solution when 15 or 20 runs are performed. In summary, having the
computational tool developed in this research execute 5 runs with the low level of
strictness would generate high quality solutions in a reasonable running time. / Graduation date: 2004

Identiferoai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/30805
Date18 March 2004
CreatorsAltamirano, Esteban
ContributorsPorter, J. David
Source SetsOregon State University
Languageen_US
Detected LanguageEnglish
TypeThesis/Dissertation

Page generated in 0.0015 seconds