Ainul, Rafina D. (2019) Enhancement of weighted centroid algorithm for indoor mobile non-cooperative localization system. In: International Conference on Informatics, Technology and Engineering 2019, 22-23 Agustus 2019, Bali.
Preview |
PDF
Paper D4.pdf Download (744kB) | Preview |
Abstract
Nowadays, indoor wireless localization is being challenged research by providing high accuracy of location information. However lower processing time, resistant to environmental condition, simple network topology are also become main concern. Noncooperative localization based on RSSI allow the anchor nodes as the reference nodes communicate directly to the target node by exchanging the location data. High sensitivity of RSSI to the indoor environment, make difficulties in modelling propagation characteristic called as PLE. Incompatibility the PLE value can influence to the estimated position result. Weighted centroid localization (WCL) is feasible solution for RSSI-based that can obtain the target node location just by RSSI and anchor nodes coordinate without PLE value and estimated distance. While, the centroid determination of WCL give better estimation only to the centralized position of target node between all anchor nodes position. Therefore, we propose enhancement of WCL (eWCL) by replacing the weight based on RSSI with different estimated distance from WCL calculation. The simulation result show that using eWCL can reduce the error estimation around 60.42% compared to the WCL algorithm with 1.85 meters MSE value. Then, compared to the cooperative localization based on trilateration algorithm achieve 12.15% error estimation larger than eWCL at non-cooperative scheme.
Item Type: | Conference or Workshop Item (Paper) |
---|---|
Subjects: | T Technology > T Technology (General) |
Divisions: | Faculty of Engineering > Department of Electrical Engineering |
Depositing User: | RAFINA DESTIARTI AINUL |
Date Deposited: | 14 Feb 2020 03:35 |
Last Modified: | 24 Mar 2021 16:20 |
URI: | http://repository.ubaya.ac.id/id/eprint/37359 |
Actions (login required)
View Item |