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

Long-term automated monitoring of residential or small industrial 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 Aerial Technologies. The system relies on existing wifi network signals and semi-supervised learning, in order to automatically 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 detecting changes in the characteristics of the wifi signals collected via existing off-the-shelf wifi-enabled devices in the environment. This platform has been deployed in several apartments in the Montreal area, and the results obtained show the potential of this technology to turn any regular home with an existing wifi network into a smart home equipped with intruder alarm and room-level location detector. The machine learning 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 environment from a small amount of labeled data; during its lifetime, the learner automatically re-trains when the uncertainty 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 system for real residential units, and illustrates its performance during on-going deployment.