À la Une

Soutenance de thèse Abbass Hammoud


M. Abbass Hammoud soutiendra en anglais, en vue de l'obtention du grade de docteur en systèmes d'information de la Faculté d'économie et de management (GSEM), sa thèse intitulée:

Indoor Occupancy Sensing with Ultrasounds

Date: Vendredi 23 novembre 2018 à 15h30

Lieu: CUI / Battelle bâtiment A, Auditoire rez-de-chaussée



  •  Dr. Michel Deriaz (co-directeur)
  •  Prof. Dimitri Kontantas (co-directeur)
  •  Prof. Gilles Falquet (président de jury)
  •  Prof. Jean-Henry Morin (SdS, interne)
  •  Dr. Pedram Pad (CSEM, externe)
  •  M. Serge Grisard (CEO DomoSafety, externe)


As human beings, we rely on audible sounds as one way to communicate between each other and to infer information about our surrounding environment. Similarly, ultrasounds are used by some species in the animal kingdom to sense objects around them and get relevant information about their environment. In this thesis, we build on the inherent characteristics of ultrasounds and explore their application in occupancy sensing of indoor spaces, as ultrasounds exhibit interesting advantages compared to other technologies. Specifically, we design methods and algorithms to generate and process ultrasonic signals and infer the room occupancy, and we develop systems to evaluate their performance. Throughout the work, we address the implementation of our methods using commodity hardware, we pay attention to design algorithms that are computationally efficient, and we evaluate their time and space complexity. We focus on the reusability aspects in our designs, with the aim of bringing the technology to a wide range of existing and potential commercial devices, that would be able to implement our methods and algorithms seamlessly, and offer insights for new applications (like improving users' experience, enhancing home automation, etc.).
This thesis brings four main contributions. We start off by presenting our solution for a device-based occupancy detection system, in which the room occupancy is determined using people's smartphones. The system wouldn't be robust, unless the problem of signal interference and packet collision is mitigated. Therefore, we show how collisions could be detected, and propose a solution to reduce their occurrence probability.
Then, we move on to address device-free occupancy sensing, where we sense the presence of persons without requiring them to carry or wear any devices. In this regard, our contribution is a self-calibrating motion sensing system that is based on the Doppler effect. We show how unsupervised learning can be used to auto-calibrate the parameters of the system without prior information of the installation environment.
While active ultrasonic motion sensing offers a higher accuracy, it generally consumes more energy than traditional passive sensing technologies (like passive infrared sensors). To alleviate this limitation, our third contribution is a novel automatic power switching method that can reduce the energy consumption of the sensors. The method, which we call "power hopping", allows a motion sensor to optimize its transmit power in function of the surrounding environment's conditions, and is automatically triggered every time the layout of the environment is detected to have changed.
In our last contribution, we address the sensing of still persons. For this, we explore the use of reflection patterns of ultrasonic signals. We show how we can process the signals and make use of supervised learning techniques, to accurately detect the presence of still persons, even in low signal-to-noise ratio conditions. All of the presented methods and algorithms were experimentally evaluated using working prototypes. To summarize this dissertation, we discuss how our proposed methods and algorithms can be applied to make devices and appliances smarter, more aware and responsive to their users. These include smartphones, digital speaker assistants, PCs, smart TVs, and virtually any devices equipped with sound speakers and microphones.