![Institutional Repository [SANDBOX]](images/dias.logo.png "Institutional Repository [SANDBOX]"){kind=logo}
Institutional Repository [SANDBOX]
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/BDE54956-481D-4F14-9C35-98F9ECB0EBA6 | ||
| Year | 2024 | ||
| Type of Item | Diploma Work | ||
| License |
|
||
| Bibliographic Citation | Anastasios Kleniatis, "Device-free indoor localization of people with radio frequency", Diploma Work, School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece, 2024 https://doi.org/10.26233/heallink.tuc.100544 | ||
| Appears in Collections |
Device-free localization (DFL) is an emerging technology with diverse applications, including security, customer behavior analysis, and smart buildings. DFL focuses on locating individuals without requiring them to carry devices or actively participate in the localization process. Recent research has leveraged wireless technologies for DFL due to their cost-effectiveness and non-intrusive nature. However, accurately localizing multiple people in an area remains challenging due to severe multipath effects.This work addresses these challenges by modeling multipath reflections and proposing two DFL methods that utilize additional signal characteristics, such as phase and read rate, unlike existing methods that rely solely on received signal strength. Additionally, it introduces a practical fusion technique to combine different DFL methods, enhancing overall performance. Another key component is the deployment of multiple RFID tags within the area of interest, which function as additional wireless antennas, significantly reducing the overall cost compared to the wireless sensor networks commonly used in the literature.Real-world experiments demonstrated that the proposed methods can achieve accuracy comparable to state-of-the-art techniques and can even surpass them in certain scenarios. Notably, one of the proposed methods achieves 25% better localization accuracy than an existing method, when localizing three targets. Additionally, combining one of the proposed methods with an existing technique enhances the existing method’s performance by 40%, 13%, and 28% in one, two, and three target localization scenarios, respectively. Specifically, using 84 passive RFID tags, the localization error is reduced to less than 36 cm for one target, less than 41 cm for two targets, and less than 96 cm for three targets, 90% of the time.
Copyright © DIAS 2013
