CoughLoc: Location-Aware Indoor Acoustic Sensing for Non-Intrusive Cough Detection

Zheng Sun, Carnegie Mellon University
Aveek Purohit, Carnegie Mellon University
Kathleen Yang, Samsung
Neha Pattan, Carnegie Mellon University
Dan Siewiorek, Carnegie Mellon University
Asim Smailagic, Carnegie Mellon University
Ian Lane, Carnegie Mellon University
Pei Zhang, Carnegie Mellon University

Abstract

Pervasive medical monitoring has become an ideal alter- native to nursing care for elderly people and patients in hospitals. Existing systems using single body-worn sensors are often intrusive and less reliable. By contrast, ubiqui- tous acoustic sensing techniques can support non-intrusive and robust medical monitoring. In this paper, we describe CoughLoc, a ubiquitous acoustic sensing system for con- tinuous cough detection using a wireless sensor network. We show how knowledge of sound source locations can be leveraged to improve the detection accuracy of sound events caused by mobile users. Experiments in indoor environ- ments show our system achieves over 90% cough detection performance under quiet backgrounds, and 1.6 times higher performance compared to a baseline approach with no loca- tion information.