Radio Frequency technologies are getting more and more attention as they are attractive solutions to many application domains. Accordingly, RF-based indoor localization has become a very hot research topic currently. As we know, there are several major challenges in RF-based indoor localization domain, such as radio signal multi-path problem, absorption effect of radio signal, etc. A few traditional indoor positioning approaches (for example, the classic LANDMRAC approach) have obtained achievements on this domain. In particular, they have enlightened later researchers very much, and provided many promising research directions. However, the traditional RF-based indoor positioning approaches suffer from several major drawbacks. In this thesis, we propose two novel RF-based indoor localiz...[ Read more ]

Radio Frequency technologies are getting more and more attention as they are attractive solutions to many application domains. Accordingly, RF-based indoor localization has become a very hot research topic currently. As we know, there are several major challenges in RF-based indoor localization domain, such as radio signal multi-path problem, absorption effect of radio signal, etc. A few traditional indoor positioning approaches (for example, the classic LANDMRAC approach) have obtained achievements on this domain. In particular, they have enlightened later researchers very much, and provided many promising research directions. However, the traditional RF-based indoor positioning approaches suffer from several major drawbacks. In this thesis, we propose two novel RF-based indoor localization schemes to attempt to overcome the existing drawbacks. The proposed SVR-LANDMRC approach is our first interesting attempt, which can guarantee more stable location sensing results based on a same infrastructure to LANDMARC. Despite the stableness of the SVR-LANDMRC approach is good, its degree of prediction accuracy is similar to existing LANDMARC approach. The key idea of this thesis is the proposed SA-LANDMARC approach, which is an RF-based hybrid indoor localization mechanism. The SA-LANDMARC approach is based on active RFID equipments and Wireless Sensor Networks equipments jointly. This scheme uses wireless sensor nodes to implement effective target-clustering process to appoint a subset of all reference tags as candidates. Then the final location is estimated by radio signal relationship between the target tag and candidate reference tags. With a similar hardware cost to LANDMARC, the SA-LANDMARC approach can provide a remarkable high degree of localization accuracy. Our experimental results show that, under our system setup, the accuracy of SA-LANDMARC approach can reach to about 0.7m.