SurroundSense: mobile phone localization via ambience fingerprinting
Proceedings of the 15th annual international conference on Mobile computing and networking
Indoor localization without infrastructure using the acoustic background spectrum
MobiSys '11 Proceedings of the 9th international conference on Mobile systems, applications, and services
Hi-index | 0.00 |
This paper presents a system design and experimental evaluations for ambient sound-based proximity detection with smartphones. Ambient sound is useful as spatiotemporal identifier. This is because ambient sound contains abundant information which depends on time and space. To detect proximity at high accuracy, we calculate cross correlation in frequency domain for the similarity measure of ambient sound. The results of experiments show that our system discriminates 5 rooms with the accuracy of true positive 94.9% and false positive 0.1%.