Deakin University
Browse

File(s) not publicly available

Intelligent processing of K-nearest neighbors queries using mobile data collectors in a location aware 3D wireless sensor network

conference contribution
posted on 2010-12-01, 00:00 authored by P P Jayaraman, Arkady ZaslavskyArkady Zaslavsky, J Delsing
The increased acceptance of sensor networks into everyday pervasive environments has lead to the creation of abundant distributed resource constrained data sources. In this paper, we propose an intelligent mobile data collector-based K-Nearest Neighbor query processing algorithm namely 3D-KNN. The K-Nearest Neighbor query is an important class of query processing approach in sensor networks. The proposed algorithm is employed over a sensor network that is situated within a 3 dimensional space. We propose a novel boundary estimation algorithm which computes an energy efficient sensor boundary that encloses at least k nearest nodes. We then propose a 3D plane rotation algorithm that maps selected sensor nodes on different planes onto a reference plane and a novel k nearest neighbor selection algorithm based on node distance and signal-to-noise ratio parameters. We have implemented the 3D-KNN algorithm in GlomoSim and validate the proposed algorithm's cost efficiency by extensive performance evaluation over well defined system criteria. © 2010 Springer-Verlag.

History

Volume

6098 LNAI

Issue

PART 3

Pagination

260 - 270

ISSN

0302-9743

eISSN

1611-3349

ISBN-13

9783642130328

ISBN-10

3642130321

Publication classification

E1.1 Full written paper - refereed

Title of proceedings

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Usage metrics

    Research Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC