A Generalized Model for Indoor Location Estimation Using Environmental Sound from Human Activity Recognition

Abstract

The indoor location of individuals is a key contextual variable for commercial and assisted location-based services and applications. Commercial centers and medical buildings (e.g., hospitals) require location information of their users/patients to offer the services that are needed at the correct moment. Several approaches have been proposed to tackle this problem. In this paper, we present the development of an indoor location system which relies on the human activity recognition approach, using sound as an information source to infer the indoor location based on the contextual information of the activity that is realized at the moment. In this work, we analyze the sound information to estimate the location using the contextual information of the activity. A feature extraction approach to the sound signal is performed to feed a random forest algorithm in order to generate a model to estimate the location of the user. We evaluate the quality of the resulting model in terms of sensitivity and specificity for each location, and we also perform out-of-bag error estimation. Our experiments were carried out in five representative residential homes. Each home had four individual indoor rooms. Eleven activities (brewing coffee, cooking, eggs, taking a shower, etc.) were performed to provide the contextual information. Experimental results show that developing an indoor location system (ILS) that uses contextual information from human activities (identified with data provided from the environmental sound) can achieve an estimation that is 95% correct.

From Offline to Real-Time Distributed Activity Recognition in Wireless Sensor Networks for Healthcare: A Review

“…Du et al [ 27 ] have improved the use of RFID tags on objects with a 3-stage activity recognition framework allowing us to capitalize on the usage state of different objects to infer the activity currently being performed and to predict the next activity in line. Galvan-Tejada et al [ 28 ] have used sound data and a Random Forest for the development of an indoor location system relying on a human activity recognition approach: The activity performed by the user allows us to infer their location in the environment. Chapron et al [ 29 ] have focused on bathroom activity recognition using Infrared Proximity Sensors (IRPS) in a cost-efficient system and have achieved an accuracy of 92.8% and 97.3% for toilet use and showering activities.…”

Human Activity Recognition for Indoor Localization Using Smartphone Inertial Sensors

“…Additionally, Guo et al [ 7 ] developed an IPS that leverages HAR using inertial sensors and a barometer to detect specific activities, such as going up stairs or opening doors, which are then matched using previously annotated landmarks in the floor plan, aiding improvements to the positioning process. Using sound as a source for HAR, Galván-Tejada et al [ 18 ] developed a room-level IPS, leveraging a Random Forest model to identify the specific features of activities traditionally performed in the same location.…”