Frédéric Weis

Multi-sensor data fusion within the belief functions framework: application to smart home services

In Smart Home, understanding the environment and what is going on is the basis of all adapted services. Unfortunately, inferring situations and activity recognition directly from raw data is way too complex to be applied. Firstly, we present a layered architecture we are building to process raw data into abstract situations and activities. Secondly, data fusion tools using the belief functions theory are introduced as a general framework to provide a first level of abstraction from raw data given by sensors to a more complex context model. Then a methodology to apply the model to our Smart Home within the belief functions framework, a first implementation and the encountered issues in modeling are discussed.

S IDE surfer: enriching casual meetings with spontaneous information gathering

Development of wireless communications enables the rise of networking applications in mobile personal systems. Thus, Web access from wireless PDAs is now available. More recently, emergence of proximity and wireless communication technologies has allowed to envision new kinds of applications taking part from their physical neighbourhood. In the field of personal applications, physical neighbourhood awareness can contribute to manage direct interactions between people during opportunistic encounters. Such encounters can be exploited to perform spontaneous and direct information exchanges. In this context, the SIDE Surfer system proposes to enrich casual meetings with proximate and spontaneous Web interactions.

Data Exchanging in Hierarchical Infostation Systems

Infostation systems, based on pico-cells, provide intermittent but very high speed rates. Many problems appear because of discontinuous coverage and the mobile environment (mainly service disruptions). These problems directly affect applications performance. In this paper, we propose a hierarchical architecture that enables the network to resolve these problems. A new entity called access controller has been introduced and validated. Then, we focus on applications made possible using this architecture. Mainly, a distributed data storage model is described that proposes many classes of service to give more flexibility to users. Finally, we present a middleware adapted to the architecture that implements the storage model and class of services.

Experiences in managing uncertainty and ignorance in a lightly instrumented smart home

Purpose – The purpose of this paper is to report an inter‐disciplinary experience in building a context‐aware system that provides adapted functionalities to inhabitants of a smart home. The paper focuses on the management of uncertainty that is intrinsic to pervasive computing systems.Design/methodology/approach – The paper presents the principles that characterize the context‐aware architecture: the acceptability‐driven design, where privacy and acceptability are favored; the awareness of the gap between the reality of human activity and the capabilities of the capture process; the step‐by‐step abstraction of contextual information; the management of uncertainty imprecision and ignorance at individual‐ and cross‐layer levels. The paper presents the principles and describes the system architecture, focusing on the management of uncertainty.Findings – The authors built a layered architecture that manages and propagates uncertainty, imprecision and ignorance, allowing the recognition of ambiguous contexts ...

Towards a feasibility-driven uncertainty-aware layered architecture for recognizing complex domestic activity

This paper illustrates some recommended design principles for activity recognition techniques in the smart home domain. These include considerations about acceptability and feasibility and the adoption of realistic human activity models. The paper also presents the architecture of a smart home prototype that is currently under development. In this system, contextual information is obtained through successive abstractions. Mechanisms for assessing and propagating the uncertainty of the recognition process are also discussed.

Sensing and Filtering Surrounding Data: The PERSEND Approach

In the mobile computing area, short-range wireless communication technologies make it possible to envision direct interactions between mobile devices. In the scope of data access, devices can now be considered as both data providers and data consumers. Thus, each device can be provided with a remote access to data its neighbours agree to share. Such a service enables applications to consult a set of data providers which dynamically evolves according to the mobility of the neighbouring devices. The set of data sources an application may access by this way is therefore representative of its physical neighbourhood. In this context, we propose to design a tool making possible the continuous consultation of neighbouring shared data. We present, in this paper, the PERSEND system we develop in this scope. Based on relational databases systems, PERSEND enables applications to define continuous queries over neighbouring data.

Proximate interactions of wireless appliances

Development and progress of appliances enable to envision the design of new application classes that exploit direct communication with a limited range. Mobile users will be able to exchange information dynamically and spontaneously when they are physically close together. The purpose of this position paper is to discuss about some research problems related to these new usage scenarios.

An improved extreme learning machine model for the prediction of human scenarios in smart homes

One of the main objectives of smart homes is healthcare monitoring and assistance, especially for elderly and disabled people. Therefore, an accurate prediction of the inhabitant behavior is very helpful to provide the required assistance. This work aims to propose a prediction model that satisfies the accuracy as well as the rapidity of the learning phase. To do so, we propose to improve the existing extreme learning machine (ELM) model by defining a recurrent form. This form ensures a temporal relationship of inputs between observations at different time steps. The new model uses feedback connections to the input layer from the output layer which allows the output to be included in the long-term prediction. A recurrent dynamic network, with feedback connections of the output of the network, is proposed to predict the future series representing future activities of the inhabitant. The resulting model, called Recurrent Extreme Learning Machine (RELM), provides the ability to learn the human behavior and ensures a good balance between the learning time and the prediction accuracy. The input data is based on the real data representing the activities of persons belonging to the profile of first level (i.e. P1) as measured by the dependency model called Functional Autonomy Measurement System (SMAF) used in the geriatric domain. The experimental results reveal that the proposed RELM model requires a minimum time during the learning phase with a better performance compared to existing models.

A Genetic Neural Network Approach for Unusual Behavior Prediction in Smart Home

Detect efficiently the activities of daily living of elderly people at home in order to provide a secure life and to intervene in the necessary time is an important problem we propose here an improved artificial neural network model. As we need an efficient prediction model, we propose a recurrent output neural network model (RO-NN) combined with a genetic algorithm (GA) which surely monitors and predicts the state of the concerned elderly person. Furthermore, we propose a prediction algorithm “Unusual Behavior Algorithm (UBA)” dedicated to detect the unusual activities and hold us account in the dangerous state.

Design of a Smart Information Diffusion Service for Museums Using RFID-Based Location System

In this paper, we present the design and the imple- mentation of a contextual and multimodal information diffusion system for museums. We propose a proactive service (1) with the ability to sense the profile of nearby visitors, using RFID and WLAN connectivity, (2) that adapts the information content to the target user, and finally (3) that distributes automatically the information to several devices, collective devices like large wall panels disseminated in museum or directly to digital guide carried by visitors.