Real-Time Indoor Localization in Smart Homes Using Semi-Supervised Learning

Abstract


 
 
 Long-term automated monitoring of residential or small in- dustrial properties is an important task within the broader scope of human activity recognition. We present a device- free wifi-based localization system for smart indoor spaces, developed in a collaboration between McGill University and Aerˆıal Technologies. The system relies on existing wifi net- work signals and semi-supervised learning, in order to au- tomatically detect entrance into a residential unit, and track the location of a moving subject within the sensing area. The implemented real-time monitoring platform works by detect- ing changes in the characteristics of the wifi signals collected via existing off-the-shelf wifi-enabled devices in the environ- ment. This platform has been deployed in several apartments in the Montreal area, and the results obtained show the poten- tial of this technology to turn any regular home with an ex- isting wifi network into a smart home equipped with intruder alarm and room-level location detector. The machine learn- ing component has been devised so as to minimize the need for user annotation and overcome temporal instabilities in the input signals. We use a semi-supervised learning framework which works in two phases. First, we build a base learner for mapping wifi signals to different physical locations in the en- vironment from a small amount of labeled data; during its lifetime, the learner automatically re-trains when the uncer- tainty level rises significantly, without the need for further supervision. This paper describes the technical and practical issues arising in the design and implementation of such a sys- tem for real residential units, and illustrates its performance during on-going deployment.
 
 

AI Bedroom

“…(2017). [20] In the table below, we included a couple of our projects, which are the backbone of the current research project. The researcher can acquire knowledge before implementing the AI home project.…”

“…Most researchers have made significant contributions to this field. The researcher Ghourchian et al (2017) [20] described real-time indoor localization in smart homes using semi-supervised learning. The research work by Leong et al (2023).…”

WiFi-Based Human Identification via Convex Tensor Shapelet Learning

“…Furthermore, with the booming development of Internet of Things (IoT), billions of WiFi enabled IoT devices, such as thermostats, sound bar, and smart TV, are en route to being widely deployed in indoor environments. Because the body movements of a human introduce variations in WiFi Received Signal Strength (RSS) measurements, device-free occupancy sensing becomes feasible by analyzing the signals (Ghourchian, Allegue-Martinez, and Precup 2017). Being a coarse measurement, nevertheless, RSS usually fails to capture the multipath effects caused by complicated human motions.…”

Exploiting Wi-Fi Channel State Information for Artificial Intelligence-Based Human Activity Recognition of Similar Dynamic Motions

“…Recent studies leveraged Wi-Fi signals for human identification [32] and localization [33 -36], intruder detection [33,36], vital signs monitoring [37 -39], gesture tracking [40,41], and basic activity recognition [35, 42 -47] © 2019 [1]. WiWho [32] is a system that can identify a person from a small group of two to six people using the CSI of Wi-Fi signals.…”

“…Finally, as part of a project done in collaboration between Aerial Technologies, Inc. and McGill University, Ghourchian et al [33] proposed a Wi-Fi-based indoor localization system, which uses random forests to locate a subject and detect intruders in a residential unit in the context of motion detection in home security applications © 2019 [1]. The state-of-the-art shows that machine learning algorithms can be used to associate the obtained CSI signals with ADLs.…”