Abstract
Ambient radio signals, such as commercial FM radio broadcasts, provide an attractive opportunity of infrastructure-free indoor positioning with global coverage. While feasibility of FM-based indoor positioning has already been demonstrated in several studies, they provided little insight into the long-term performance of the ambient approach and its sensitivity to environment changes.This paper presents a year-long experimental evaluation of FM indoor positioning, using a multi-testbed dataset collected bi-weekly over the course of 12 months. Moreover, we improve the accuracy of classical receiver signal strength (RSS) fingerprinting by introducing a number of novel multipath-sensitive signal features. Finally, we analyze the robustness of FM signal features to human presence and weather conditions.
Sign-in/Register to access full text options 
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
