Abstract
Wireless networks are ubiquitous nowadays and hence provide a promising approach for indoor localization. Many algorithms have been proposed for exploiting wireless signals for localization purposes. Among the methods, ANNbased methods have attracted particular attention due to their robustness in complex signal environments. However, their accuracy is still degraded by multi-path effects, signal fluctuations, and so on. Accordingly, this study commences by examining the effects of fluctuations in the received signal strength indicator (RSSI) measurement on the accuracy of an ANN-based localization algorithm. This study list some strategies and illustrate by simulation experiment. Based on the investigation results, a systematic methodology is proposed for improving the localization performance by increasing the number of APs. The feasibility of the proposed method is demonstrated by means of numerical simulations.
Highlights
Location-based service (LBS) technology has entered the mainstream as the location accuracy of the global positioning system (GPS) has improved to within 10 m
[4] The models perform well in outdoor, uncluttered environments, and are widely deployed in traditional wireless sensor networks (WSNs).the authors in [2] showed that the received signal strength indication (RSSI) signal fluctuations produced in general indoor environments by scattering and other physical effects severely degrade the accuracy of the distance estimates in WSNs
The authors in [7] proposed an RSSI real-time correction method based on the particle swarm optimization – back propagation neural network (PSO-BPNN) RSSI-distance model presented in [8].In the proposed approach, a terminal was established to collect the RSSIs of the surrounding access points (APs) and the RSSI measurements were adjusted intelligently in realtime using RSSI fluctuation data stored on a local server
Summary
Location-based service (LBS) technology has entered the mainstream as the location accuracy of the global positioning system (GPS) has improved to within 10 m. Wi-Fi, Bluetooth, Infrared/Laser, Radio Frequency Identification (RFID), ZigBee and Ultra-wideband (UWB) have all attracted significant attention as enabling technologies for indoor localization [1]. Among these techniques, Wi-Fi is regarded as a promising solution due to the proliferation of wireless networks in both outdoor and indoor environments nowadays and the Wi-Fi capability implemented as standard on most handheld and portable devices (e.g., smartphones, tablets, watches, and so on). It has been shown that ANNs enable a localization error of as little as 2 m in ideal scenarios.[2] As a result, they vastly outperform conventional GPS methods.their accuracy is still adversely affected by RSSI signal fluctuations. Based localization methods, and that of conventional triangulation schemes
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call