Teresa Garcia-Valverde

A Fuzzy Logic-Based System for Indoor Localization Using WiFi in Ambient Intelligent Environments

Ambient intelligence is a new information paradigm, where people are empowered through a digital environment that is “aware” of their presence and context and is sensitive, adaptive, and responsive to their needs. Hence, one of the important requirements for ambient intelligent environments (AIEs) is the ability to localize the whereabouts of the user in the AIE to address her/his needs. In order to protect user privacy, the use of cameras is not desirable in AIEs, and hence, there is a need to rely on nonintrusive sensors. There are various localization means that are available for outdoor spaces such as those which rely on satellite signals triangulation. However, these outdoor localization means cannot be used in indoor environments. The majority of nonintrusive and noncamera-based indoor localization systems require the installation of extra hardware such as ultrasound emitters/antennas, radio-frequency identification (RFID) antennas, etc. In this paper, we propose a novel indoor localization system that is based on WiFi signals which are free to receive, and they are available in abundance in the majority of domestic spaces. However, free WiFi signals are noisy and uncertain, and their strengths and availability are continuously changing. Hence, we present a fuzzy logic-based system which employs free available WiFi signals to localize a given user in AIEs. The proposed system receives WiFi signals from a large number of existing WiFi access points (up to 170 access points), where no prior knowledge of the access points locations and the environment is required. The system employs an incremental lifelong learning approach to adjust its behavior to the varying and changing WiFi signals to provide a zero-cost localization system which can provide high accuracy in real-world living spaces. We have compared our system in both simulated and real environments with other relevant techniques in the literature, and we have found that our system outperforms the other systems in the offline learning process, whereas our system was the only system which is capable of performing online learning and adaptation. The proposed system was tested in real-world spaces from a living lab intelligent apartment (iSpace) to a town center apartment to a block of offices. In all these experiments, our system has been highly accurate in detecting the user in the given AIEs, and the system was able to adapt its behavior to changes in the AIE or the WiFi signals. We envisage that the proposed system will play an important role in AIEs, especially for privacy concerned situations like elderly care scenarios.

Flexible Simulation of Ubiquitous Computing Environments

Ubiquitous computing software must be reliable. As it occurs in conventional software, one of hardest tasks nowadays is testing. Moreover, if testing is focused on context-aware and adaptive services, the task is even harder. In this case, it is not sufficient testing the software in order to find bugs in the code and repair them (i.e. debugging) and, at the same time, checking that procedures and functions responses to specific interesting inputs are correct. It is also needed checking that the responses of services (i.e. the system under test) to changes in the environment are correct. And the environment includes also human users. This paper proposes UbikSim. An ubiquitous computing environments simulator which tries to alleviate the particularities of testing services and applications whose behaviour depends on both physical environment and users.

Automatic Design of an Indoor User Location Infrastructure Using a Memetic Multiobjective Approach

Services in ambient intelligence (AmI) environments should adapt to contextual information (context-aware) of users and environment in a nonintrusive and natural way. Location-aware, i.e., the user location, is one of the most important pieces in context-aware. According to this premise, a location-based service (LBS) using radio frequency identification technology is presented. The service is based on hidden Markov models for location within an intelligent building. This problem leads to a multiobjective optimization problem, in which, the best configuration of antennas that minimizes the set of antennas but maximizes the precision of the prediction should be found. Specifically, this study presents a memetic approach for multiobjective improvement of LBS in AmI environments. The memetic algorithm provides, in this problem, the exploitation of domain knowledge and the combination of metaheuristics. Experimental results show that the approach obtains a configuration of antennas which optimally configures the number and position of the antennas while keeping a high quality of the precision of the location prediction.

Combining the real world with simulations for a robust testing of Ambient Intelligence services

This paper proposes a general architecture for testing, validating and verifying Ambient Intelligence (AmI) environments: AmISim. The development of AmI is a very complex task because this technology must often adapt to contextual information as well as unpredictable behaviours and environmental features. The architecture presented deals with AmI applications in order to cover the different components of these kinds of systems: environment, users, context and adaptation. This architecture is the first one that is able to cover all these features, which are needed in a full AmI system. The paper shows that AmISim is able to cover a complete AmI system and to provide a framework which can test scenarios that would be impossible to test in real environments or even with previous simulation approaches. Simulated and real elements coexist in AmISim for a robust testing, validation and verification of the AmI systems, which provide an easier and less costly deployment.

An adaptive learning fuzzy logic system for indoor localisation using Wi-Fi in Ambient Intelligent Environments

One of the important requirements for Ambient Intelligent Environments (AIEs) is the ability to localise the whereabouts of the user in the AIE to address her/his needs. The outdoor localisation means (like GPS systems) cannot be used in indoor environments. The majority of non intrusive and non camera based indoor localisation systems require the installation of extra hardware such as ultra sound emitters/antennas, RFID antennas, etc. In this paper, we will propose a novel fuzzy logic based indoor localisation system which is based on the WiFi signals which are free to receive and they are available in abundance in the majority of domestic spaces. The proposed system receives WiFi signals from a big number of existing WiFi Access Points (up to 170 Access Points) with no prior knowledge of the access points locations and the environment. The proposed system is able to adapt online incrementally in a lifelong learning mode to deal with the uncertainties and changing conditions facing unknown indoor structures with a few days of calibration at zero-cost deployment with high accuracy. The proposed system was tested in simulated and real environments where the system has given high accuracy (that outperformed the existing techniques) to detect the user in the given AIE and the system was able also to adapt its behaviour to changes in the AIE or the WiFi signals.

Generation of human computational models with machine learning

Abstract Services in smart environments pursue to increase the quality of people’s lives. The most important issues when developing this kind of environments is testing and validating such services. These tasks usually imply high costs and annoying or unfeasible real-world testing. In such cases, artificial societies may be used to simulate the smart environment (i.e. physical environment, equipment and humans). With this aim, the CHROMUBE methodology guides test engineers when modeling human beings. Such models reproduce behaviors which are highly similar to the real ones. Originally, these models are based on automata whose transitions are governed by random variables. Automaton’s structure and the probability distribution functions of each random variable are determined by a manual test and error process. In this paper, it is presented an alternative extension of this methodology which avoids the said manual process. It is based on learning human behavior patterns automatically from sensor data by using machine learning techniques. The presented approach has been tested on a real scenario, where this extension has given highly accurate human behavior models.

Engineering Ambient Intelligence Services by Means of MABS

In this work, the methodology AmISim to test and to deployment of Ambient Intelligence (AmI) system is presented. The development of AmI systems is a complex task because this technology must adapt to users and contextual information as well as unpredictable and changeable behaviours. So, we focused in how the methodology AmISim can help to the engineering of adaptative services for users. In this case, we propose a predictor of location based on Hidden Markov Models (HMMs). So, the system can offer Location-Based Services(LBS) that adapt to the users. To this end, we propose a methodology based on a previous social multi-agent based simulation (MABS) and a following deployment of the service in a real environment.

Ubiquitous deployment configuration of indoor location services

The development of services in Ubiquitous Computing is a hard task. Services must adapt to context information about users. One of the most important pieces of context is user location, which allows Location Based Services (LBS) to adapt their functionality regarding the users nearest features of interest. In this paper, we will propose a hybrid system to solve the problem of finding the best configuration of antennas within an intelligent environment that minimizes cost and intrusion but maximizes the accuracy of the LBS in the prediction task. The approach combines Hidden Markov Models (HMM) for user location prediction with a multiobjective genetic algorithm which is able to get suboptimal configurations of the number and position of the antennas in the intelligent building. In the experiments, our system has given configurations of antennas which provide high accuracy to predict the location (based on Radio Frequency Identification, RFID) of the user while a minimal deployment of antennas in the building is needed.

Reasoning on a Semantic Web Based Context-Awareness Middleware

We present in this paper the OCP middleware for context information management. More specifically, we focus on the reasoning aspects of the middleware. In order to provide the OCP middleware with reasoning capabilities, we have integrated a semantic web ontology management API based on Jena and Pellet, developed in our lab, i.e. the ORE API. We present some projects in which the reasoning capabilities have been applied and some examples trying to show the power of this open source implementation.

Semantic description of multimedia contents for the optimization of the advertising impact on TV program grids

The problem of advertising impact optimization in program grids consists to find a fully design of advertising contents in the program grid maximizing the satisfaction of advertisers and viewers. In this work, the problem of advertising impact optimization of program grids is approached. Standards for semantic description of multimedia contents are used for expressing contents in a television grid and the optimization process is based on semantic similarity measures between the descriptions of the TV contents. The overall optimization of the advertising impact is guaranteed using an evolutive approach.