SenseIO: Realistic Ubiquitous Indoor Outdoor Detection System Using Smartphones

Author's Department

Computer Science & Engineering Department

Find in your Library

https://ieeexplore.ieee.org/abstract/document/8303687

All Authors

Mohsen Ali; Tamer ElBatt; Moustafa Youssef

Document Type

Research Article

Publication Title

IEEE Sensors Journal

Publication Date

5-1-2018

doi

10.1109/JSEN.2018.2810193

Abstract

Indoor/outdoor localization, tracking, and positioning applications are developed using the Global Positioning System receivers, ultrasound, infrared, and radio frequency (Wi-Fi and cellular) signals. The key point of such upper layer applications is to detect precisely whether a user is indoor or outdoor. This detection is crucial to improve the performance drastically through making a clever decision whether it is suitable to turn ON/OFF the sensors. Due to this, an unrealistic assumption is posed by the applications that the testbed environment type (indoor or outdoor) must be pre-known. In this paper, we present a realistic and ubiquitous (SenseIO) system which provides not only binary indoor/outdoor, but also a fine-grained detection (i.e., Rural, Urban, Indoor and Complex places). Without any prior knowledge, SenseIO leverages the measurements of sensor-rich smartphones (e.g., cellular, Wi-Fi, accelerometer, proximity, light and time-clock) to infer automatically the ambient environment type. A novel SenseIO multi-model system consists of four modules: 1) single serving cell tower; 2) Wi-Fi based; 3) activity recognition; and 4) light intensity. In addition, to achieve realism and ubiquity goals, we develop a SenseIO framework which includes three scenarios (A, B, C). We implement SenseIO on android-based smartphones and test it through multi-path tracing in real I/O environments. Our experiments for each individual module and all framework scenarios show that the SenseIO provides promising detection accuracy (above 92%) and outperforms existing indoor-outdoor techniques in terms of both accuracy and fine-grained detection.

First Page

3684

Last Page

3693

This document is currently not available here.

Share

COinS