PENENTUAN LOKASI OBJEK DI DALAM GEDUNG BERBASIS WLAN FINGERPRIN

• Main Authors: , TAMAN GINTING, , Widyawan, ST, M.Sc, Ph.D
• Format: Theses and Dissertations NonPeerReviewed
• Published: [Yogyakarta] : Universitas Gadjah Mada 2013
• Subjects: ETD
• Online Access: https://repository.ugm.ac.id/119467/; http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=59469
• Institution: Universitas Gadjah Mada

WLAN has become very popular in public and enterprise networking during the last few years. IEEE 802.11 is currently the dominant local wireless networking standard. It is appealling to use an existing WLAN infrastructure for indoor location based WLAN positioning system using RSS from APs that have been available. This research focused on implementation of RSS from APs inside and around the JTETI UGM building without placing additional APs. RSS fingerprint are collected with four different measuring, with grid-size 1m x 1m and 4m x 4m. RSS fingerprint from third floor are collected. Location estimation of the object is calculated by k-Nearest Neighbor (k-NN), Naïve Bayes and Decision Tree algorithm as comparator. From the survey results revealed that location estimation results are influenced by several factors including the size of the grid fingerprint, algorithms and data from the wide amount of data measuring location the fingerprint. The best results on three phases testing on-line with amount data used is 5760 and data test data by 40 real-time data. Average Distance Estimation Error in phase online using k-NN algorithm with k = 1 is 0:11 Meter, Naive Bayes Meter 4:01, 1:38 Desicion Tree Meter and standard deviation k-NN algorithm with k = 1 is 0:43 Meter, Naive Bayes 1.95 Meters 2.82 Meters and Decision Tree. k-NN algorithms results a better accuracy than the algorithm.