No Detail Found!

Detecting anomalies in AOA data measured for indoor positioning to improve accuracy

Author(s)
HYEONGSEOL SHIN Division of Marine System Engineering, Korea Maritime and Ocean University, Busan, Korea
QIANFENG LIN Department of Computer Engineering, Korea Maritime and Ocean University, Busan, Korea
JOOYOUNG SON Department of Computer Engineering, Korea Maritime and Ocean University, Busan, Korea
Abstract
Ships are made of steel. The space of ship's environments is generally narrow. In the ship's environments noises are included in the signal. Accurate indoor positioning technology can contribute to preventing infectious diseases by identifying and locating infected people at an early stage. As a method, this paper makes the Angle of Arrival using Bluetooth 5.1 to locate infected people. The function of Anomaly Detection methods as filters may be used to eliminate outliers in elevation and azimuth angles when measuring the Angle of Arrival. Three different Anomaly Detection methods are evaluated in this paper, they are Generalized Extreme Studentized Deviation, LevelShift, and Persist. As a result, LevelShift shows an improvement by 14% in Root Mean Square Error (RMSE) and Generalized Extreme Studentized Deviation shows by 5%. Persist, however, gets worse by 2%.
Publication Details
Page(s) 5895-5905
DOI DOI not available
Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 100, Issue: 20, Year: 2022

Request the full text from PASTIC

Keywords
Anomaly Detection, Bluetooth 5.1, Angle of Arrival, Indoor Positioning
0

Citations

0

Downloads

46

Views