Abstract
Green WLAN is a promising technique for accessing future indoor Internet services. It is designed not only for high-speed data communication purposes but also for energy efficiency. The basic strategy of green WLAN is that all the access points are not always powered on, but rather work on-demand. Though powering off idle access points does not affect data communication, a serious asymmetric matching problem will arise in a WLAN indoor positioning system due to the fact the received signal strength (RSS) readings from the available access points are different in their offline and online phases. This asymmetry problem will no doubt invalidate the fingerprint algorithm used to estimate the mobile device location. Therefore, in this paper we propose a green WLAN indoor positioning system, which can recover RSS readings and achieve good localization performance based on singular value thresholding (SVT) theory. By solving the nuclear norm minimization problem, SVT recovers not only the radio map, but also online RSS readings from a sparse matrix by sensing only a fraction of the RSS readings. We have implemented the method in our lab and evaluated its performances. The experimental results indicate the proposed system could recover the RSS readings and achieve good localization performance.
Highlights
Wireless local area networks (WLANs) based on IEEE 802.11 are widely deployed in the indoor environment for mobile devices to access the internet
Due to its ubiquitous network architecture and no additional hardware requirements, the wireless local area networks (WLANs) indoor positioning system based on received signal strength (RSS) has become the most popular option for indoor localization and navigation, as it offers the merits of relative measurement simplicity and minimal hardware requirements to provide a very beneficial supplement to the WLAN application
We have proposed an RSS-based indoor positioning system in a green WLAN using
Summary
Wireless local area networks (WLANs) based on IEEE 802.11 are widely deployed in the indoor environment for mobile devices to access the internet. When the access points in a green WLAN are powered off randomly according to the data communication demands, the KNN algorithm is seriously challenged in both the offline phase and online phase by the working on-demand strategy. Readings both in the offline phase and online phase, which could help validate the KNN algorithm to provide localization estimations in green WLANs. SVT theory is derived from the matrix completion problem, which aims to recover an unknown matrix when only a fraction of its entries are known.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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
