WLAN FINGERPRINT UNTUK PREDIKSI LOKASI OBJEK DALAM GEDUNG

Tags:

Publication Type:

Journal Article

Source:

Jurnal Teknologi, Volume 5, Number 2 (2012)

Keywords:

Fingerprint, kNN, Naïve Bayes, RSS

Abstract:

Wireless LAN technology has become in public to enterprise networking and widely applied in various places, ranging from the campus, shops, offices and even public areas. Wireless LAN technology by using the RSS (Received Signal Strength) obtained from the access point (AP). RSS can be applied to estimate the location objects in room. This issues are able to estimate the location of objects with fingerprint method.
This research focused on implementation of RSS from 5 Acces point inside and around the Jurusan Teknik Elektro & Teknik Informatika UGM on third floor building. RSS fingerprint collected with different measuring with grid-size 1m x 1m for high accuracy. Location predict of the object is calculated by k-Nearest Neighbor (k-NN) and Naïve Bayes.
Phase off-line, visualization map fingerprint result indicates received signal strength measurements is influence by grid sizet. In on-line phase average error distance of the location estimation algorithm is better than the Naive Bayes algorithm gives highes accuracy of kNN algorithm.

AttachmentSize
143-148_ginting1.pdf889.15 KB

(c) 2009 Jurnal teknologi FTI IST AKPRIND
Developed by UPT PUSKOM IST AKPRIND

Who's online

There are currently 0 users and 58 guests online.