THESIS
2019
x, 47 pages : color illustrations ; 30 cm
Abstract
One key feature of Bluetooth low energy (BLE) beacons is the received signal strength
(RSS), which can be used to estimate the distance between any Bluetooth-compatible
receiver (e.g., smartphone, tablet, etc.) and fixed deployed beacon. Although RSS
can be measured easily with commonly available smart devices, the measurements are
unreliable due to its fluctuation. While distance estimation models are widely available
in multiple development libraries, the lack of consideration for object mobility in these
models undermines its practicality. Furthermore, general estimation models are not
robust to different hardware and settings for real deployed beacon networks. Motivated
by the above limitations, this thesis proposes a novel distance classifier, d-Classifier, to
classify the...[
Read more ]
One key feature of Bluetooth low energy (BLE) beacons is the received signal strength
(RSS), which can be used to estimate the distance between any Bluetooth-compatible
receiver (e.g., smartphone, tablet, etc.) and fixed deployed beacon. Although RSS
can be measured easily with commonly available smart devices, the measurements are
unreliable due to its fluctuation. While distance estimation models are widely available
in multiple development libraries, the lack of consideration for object mobility in these
models undermines its practicality. Furthermore, general estimation models are not
robust to different hardware and settings for real deployed beacon networks. Motivated
by the above limitations, this thesis proposes a novel distance classifier, d-Classifier, to
classify the distance on both stationary and mobile objects. Comprehensive experiments
related to mobility are conducted to study the relationship between packet receiving
rate and estimation accuracy. Improved performance can be achieved by providing
extra mobility information with a list of RSS values during estimation. The robustness
is improved by constructing a feature vector with factors such as hardware type and
deployment environment. The proposed classifier is validated with an extensive dataset
that includes over 200k data collected from real beacon networks. Overall, our proposed
d-Classifier achieves a significant performance gain, > 25% accuracy improvement, over
its prior arts.
Post a Comment