german vega - Academia.edu (original) (raw)

Papers by german vega

Pervasive and Mobile Computing

Pervasive computing promotes the integration of smart devices in our living spaces to develop ser... more Pervasive computing promotes the integration of smart devices in our living spaces to develop services providing assistance to people. Such smart devices are increasingly relying on cloud-based Machine Learning, which raises questions in terms of security (data privacy), reliance (latency), and communication costs. In this context, Federated Learning (FL) has been introduced as a new machine learning paradigm enhancing the use of local devices. At the server level, FL aggregates models learned locally on distributed clients to obtain a more general model. In this way, no private data is sent over the network, and the communication cost is reduced. Unfortunately, however, the most popular federated learning algorithms have been shown not to be adapted to some highly heterogeneous pervasive computing environments. In this paper, we propose a new FL algorithm, termed FedDist, which can modify models (here, deep neural network) during training by identifying dissimilarities between neurons among the clients. This permits to account for clients' specificity without impairing generalization. FedDist evaluated with three state-of-the-art federated learning algorithms on three large heterogeneous mobile Human Activity Recognition datasets. Results have shown the ability of FedDist to adapt to heterogeneous data and the capability of FL to deal with asynchronous situations.

2022 IEEE International Conference on Smart Computing (SMARTCOMP)

Federated Learning has been introduced as a new machine learning paradigm enhancing the use of lo... more Federated Learning has been introduced as a new machine learning paradigm enhancing the use of local devices. At a server level, FL regularly aggregates models learned locally on distributed clients to obtain a more general model. Current solutions rely on the availability of large amounts of stored data at the client side in order to fine-tune the models sent by the server. Such setting is not realistic in mobile pervasive computing where data storage must be kept low and data characteristic can change dramatically. To account for this variability, a solution is to use the data regularly collected by the client to progressively adapt the received model. But such naive approach exposes clients to the well-known problem of catastrophic forgetting. To address this problem, we have defined a Federated Continual Learning approach which is mainly based on distillation. Our approach allows a better use of resources, eliminating the need to retrain from scratch at the arrival of new data and reducing memory usage by limiting the amount of data to be stored. This proposal has been evaluated in the Human Activity Recognition (HAR) domain and has shown to effectively reduce the catastrophic forgetting effect.

2021 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops)

Pervasive computing promotes the integration of smart electronic devices in our living and workin... more Pervasive computing promotes the integration of smart electronic devices in our living and working spaces in order to provide new, advanced services. Recently many prototype services based on machine learning techniques have been proposed in a number of domains like smart homes, smart buildings or smart plants. However, the number of applications effectively deployed in the real world is still limited. We believe that architectural principles and integrated frameworks are still missing today to successfully and repetitively support application developers and operators. In this paper, we present a novel architecture and a pervasive platform allowing the development of machine learning based applications in smart buildings.

2022 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops)

Federated Learning has been introduced as a new machine learning paradigm enhancing the use of lo... more Federated Learning has been introduced as a new machine learning paradigm enhancing the use of local devices. At a server level, FL regularly aggregates models learned locally on distributed clients to obtain a more general model. In this way, no private data is sent over the network, and the communication cost is reduced. However, current solutions rely on the availability of large amounts of stored data at the client side in order to fine-tune the models sent by the server. Such setting is not realistic in mobile pervasive computing where data storage must be kept low and data characteristic (distribution) can change dramatically. To account for this variability, a solution is to use the data regularly collected by the client to progressively adapt the received model. But such naive approach exposes clients to the well-known problem of catastrophic forgetting. The purpose of this paper is to demonstrate this problem in the mobile human activity recognition context on smartphones.

2021 IEEE International Conference on Pervasive Computing and Communications (PerCom), 2021

Pervasive computing promotes the installation of connected devices in our living spaces in order ... more Pervasive computing promotes the installation of connected devices in our living spaces in order to provide services. Two major developments have gained significant momentum recently: an advanced use of edge resources and the integration of machine learning techniques for engineering applications. This evolution raises major challenges, in particular related to the appropriate distribution of computing elements along an edgeto-cloud continuum. About this, Federated Learning has been recently proposed for distributed model training in the edge. The principle of this approach is to aggregate models learned on distributed clients in order to obtain a new, more general model. The resulting model is then redistributed to clients for further training. To date, the most popular federated learning algorithm uses coordinate-wise averaging of the model parameters for aggregation. However, it has been shown that this method is not adapted in heterogeneous environments where data is not identically and independently distributed (non-iid). This corresponds directly to some pervasive computing scenarios where heterogeneity of devices and users challenges machine learning with the double objective of generalization and personalization. In this paper, we propose a novel aggregation algorithm, termed FedDist, which is able to modify its model architecture (here, deep neural network) by identifying dissimilarities between specific neurons amongst the clients. This permits to account for clients' specificity without impairing generalization. Furthermore, we define a complete method to evaluate federated learning in a realistic way taking generalization and personalization into account. Using this method, FedDist is extensively tested and compared with three state-of-the-art federated learning algorithms on the pervasive domain of Human Activity Recognition with smartphones.

2019 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), 2019

Pervasive computing envisions environments where computers are blended into our environment to pr... more Pervasive computing envisions environments where computers are blended into our environment to provide services. This vision is today very popular and a number of pervasive platforms are already used today in smart homes or smart plants. These platforms however are based on different middleware and networks. Interoperability has recently emerged as a key issue. Interoperability is a difficult problem which requires interactions between platforms that were not designed to do so. It demands to be able to communicate but also to share contextual information on demand. We here present an architecture to federate platforms and a solution to autonomically manage context at the platform level. This is demonstrated in the smart home domain and, more precisely, with the iCasa/iPOJO and Base pervasive platforms.

ArXiv, 2021

Federated Learning, a new machine learning paradigm enhancing the use of edge devices, is receivi... more Federated Learning, a new machine learning paradigm enhancing the use of edge devices, is receiving a lot of attention in the pervasive community to support the development of smart services. Nevertheless, this approach still needs to be adapted to the specificity of the pervasive domain. In particular, issues related to continual learning need to be addressed. In this paper, we present a distillation-based approach dealing with catastrophic forgetting in federated learning scenario. Specifically, Human Activity Recognition tasks are used as a demonstration domain.

2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), 2017

Pervasive applications are today very distributed from devices to cloud facilities, going through... more Pervasive applications are today very distributed from devices to cloud facilities, going through fog-level gateways. Contextual information, based on various models, is used at every level but it needs to be frequently synchronized. In this demo, we present an application guiding users to install, move and replace devices that are used by other pervasive applications. This application has required the development of different contextual models on different platforms and kept synchronized.

2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), 2017

Smart Homes aim to improve the daily lives of the inhabitants by integrating a variety of context... more Smart Homes aim to improve the daily lives of the inhabitants by integrating a variety of context-aware applications, generally pertaining to multiple fields and provided by different actors. These applications share the same context and may have to compete for the access to resources in their surroundings. Sharing resources leads to conflicts, particularly if these applications act in contradictory ways or have interfering effects on the environment. Such conflicts can lead to critical situations by putting the home's inhabitants at risk. In this paper, we present a context-based approach to manage conflicts among pervasive applications in smart home environments. Our approach is optimistic and aims to address conflicts at runtime before their undesired effects will occur. This approach is developed and integrated in the iCASA platform as iPOJO components.

2021 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops), 2021

This document provides the details and guidelines to run the federated learning experiments used ... more This document provides the details and guidelines to run the federated learning experiments used in the study “A Federated Learning Aggregation Algorithm for Pervasive Computing: Evaluation and Comparison” [1]. The artifact is available in a public GitHub repository11PerCom2021-FL - https://github.com/getalp/PerCom2021-FL. Note that The centralized/local training approaches are not included in the repository.

Lecture Notes in Computer Science, 2020

Following the success of image recognition, machine learning approaches have recently been propos... more Following the success of image recognition, machine learning approaches have recently been proposed to improve the efficiency for such systems as industry operation and maintenance, smart buildings, and smart homes. These applications are beginning to be deployed in pervasive environments. This poses greater stress in maintaining the quality of the applications. To date, there is no architecture and tools developed that can automatically support application quality maintenance. Even worse, there is no clear definition on the requirements. In this paper, we present initial experiments that we conducted with real use cases pertaining to Industry 4.0 and discuss a set of requirements that should be met by pervasive platforms to better support AI-based applications running in the edge.

Adjunct Proceedings of the 2020 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2020 ACM International Symposium on Wearable Computers, 2020

Pervasive computing promotes the integration of connected electronic devices in our living spaces... more Pervasive computing promotes the integration of connected electronic devices in our living spaces in order to assist us through appropriate services. Two major developments have gained significant momentum recently: a better use of fog resources and the use of AI techniques. Specifically, interest in machine learning approaches for engineering applications has increased rapidly. This paradigm seems to fit the pervasive environment well. However, federated learning has been applied so far to specific services and remains largely conceptual. It needs to be tested extensively on pervasive services partially located in the fog. In this paper, we present experiments performed in the domain of Human Activity Recognition on smartphones in order to evaluate existing algorithms.

Pervasive and Mobile Computing, 2018

Abstract Middleware support for Pervasive Computing has been extensively researched in the past. ... more Abstract Middleware support for Pervasive Computing has been extensively researched in the past. Different approaches for, e.g., discovery, communication, or interaction, have been explored and successfully tested in different lab environments. Although this works well in lab environments, where one middleware manages all resources, this will not suffice in real-world deployments, where more than one system software is present and services as well as other resources are only accessible by those. In this paper, we present XWARE, an interoperability framework that allows to integrate discovery and interaction of different middleware platforms. The flexible design allows to configure and extend the interoperability framework in order to add new platforms and tailor it for the use in different domains. Furthermore, XWARE instances can communicate with each other enabling interoperability in smart environments as well as in smart peer groups. The feasibility of our approach is discussed and evaluated based on the integration of five different platforms.

Lecture Notes in Computer Science, 2016

Pervasive computing promotes environments where smart, communication-enabled devices cooperate to... more Pervasive computing promotes environments where smart, communication-enabled devices cooperate to provide services to people. Due to their inherent complexity, many pervasive applications are built on top of service-oriented platforms, providing a set of facilities simplifying their development and execution. In this paper, we present such a platform, iCasa, extended with an autonomic, service-oriented context module. This module is programmed with a domain-specific serviceoriented language built on top of iPOJO, the Apache service-oriented component model. It is validated on smart home applications developed with the Orange Labs.

2016 IEEE International Conference on Autonomic Computing (ICAC), 2016

Pervasive computing promotes environments where smart, communication-enabled devices cooperate to... more Pervasive computing promotes environments where smart, communication-enabled devices cooperate to provide services to people. Due to their inherent complexity, many pervasive applications are built on top of service-oriented platforms, providing a set of facilities simplifying their development and execution. In this demo, we present such a platform, iCasa, extended with an autonomic, service-oriented context module. It is validated on diverse applications developed with the Orange Labs in the health domain.

2016 IEEE International Conference on Services Computing (SCC), 2016

Pervasive computing promotes environments where smart, communication-enabled devices cooperate to... more Pervasive computing promotes environments where smart, communication-enabled devices cooperate to provide services to people. Due to their inherent complexity, many pervasive applications are built on top of service-oriented platforms, providing a set of facilities simplifying their development and execution. In this paper, we present such a platform, iCasa, extended with an autonomic, service-oriented context module. This module is programmed with a domainspecific service-oriented language built on top of iPOJO, the Apache service-oriented component model. It is validated on smart home applications developed with the Orange Labs.

2016 IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops), 2016

Pervasive computing envisions environments where computers are blended into everyday objects in o... more Pervasive computing envisions environments where computers are blended into everyday objects in order to provide added-value services to people. This new form of computing gives rise to huge economical and societal expectations. However, pervasive applications raise major challenges in term of software engineering and remain hard to develop, deploy, execute and maintain. Context-awareness, in particular, is a salient and difficult property that must be met by pervasive applications. In this paper, we propose a service-oriented framework facilitating the design and execution of a context management module in pervasive platforms. Our approach is illustrated with a smart home example and implemented on top of iPOJO, the Service-Oriented Component Model of our pervasive platform iCasa.

Lecture Notes in Computer Science, 2007

This paper presents SEEMP, a marketplace to coordinate and integrate public and private employmen... more This paper presents SEEMP, a marketplace to coordinate and integrate public and private employment services (ESs) around the EU Member States. The need for flexible collaboration in the marketplace gives rise to the issue of interoperability in both data exchange and share of services. SEEMP proposes a mixed approach that relies on the concepts of services and semantics. SEEMP approach combines Software Engineering and Semantic Web methodologies/tools in an infrastructure that allows for a meaningful service-based communication among ESs.

Lecture Notes in Computer Science, 2010

The dream of Model Driven Engineering (MDE) is that Software Engineering activities should be per... more The dream of Model Driven Engineering (MDE) is that Software Engineering activities should be performed only on models, but in practice a significant amount of programming is still being performed. There is a clear need to keep code and models strongly synchronized when they represent the same entities at different levels of abstraction. We observe that versioning is ill supported by MDE tools, and that no strong synchronization is ensured between code and model versions. This, among other things, explains why MDE is not widely adopted in industry. This paper presents the solution developed in the CADSE project for providing consistent support for model and code co-evolution. It is shown that it requires to (1) define, what evolution policy is to be applied, (2) closely synchronize both ways, the model entities and the computer artifacts, and (3) enforce consistency constraints and evolution policies during the commit and check-out of both model elements and their corresponding artifacts.

Pervasive and Mobile Computing

Pervasive computing promotes the integration of smart devices in our living spaces to develop ser... more Pervasive computing promotes the integration of smart devices in our living spaces to develop services providing assistance to people. Such smart devices are increasingly relying on cloud-based Machine Learning, which raises questions in terms of security (data privacy), reliance (latency), and communication costs. In this context, Federated Learning (FL) has been introduced as a new machine learning paradigm enhancing the use of local devices. At the server level, FL aggregates models learned locally on distributed clients to obtain a more general model. In this way, no private data is sent over the network, and the communication cost is reduced. Unfortunately, however, the most popular federated learning algorithms have been shown not to be adapted to some highly heterogeneous pervasive computing environments. In this paper, we propose a new FL algorithm, termed FedDist, which can modify models (here, deep neural network) during training by identifying dissimilarities between neurons among the clients. This permits to account for clients' specificity without impairing generalization. FedDist evaluated with three state-of-the-art federated learning algorithms on three large heterogeneous mobile Human Activity Recognition datasets. Results have shown the ability of FedDist to adapt to heterogeneous data and the capability of FL to deal with asynchronous situations.

2022 IEEE International Conference on Smart Computing (SMARTCOMP)

Federated Learning has been introduced as a new machine learning paradigm enhancing the use of lo... more Federated Learning has been introduced as a new machine learning paradigm enhancing the use of local devices. At a server level, FL regularly aggregates models learned locally on distributed clients to obtain a more general model. Current solutions rely on the availability of large amounts of stored data at the client side in order to fine-tune the models sent by the server. Such setting is not realistic in mobile pervasive computing where data storage must be kept low and data characteristic can change dramatically. To account for this variability, a solution is to use the data regularly collected by the client to progressively adapt the received model. But such naive approach exposes clients to the well-known problem of catastrophic forgetting. To address this problem, we have defined a Federated Continual Learning approach which is mainly based on distillation. Our approach allows a better use of resources, eliminating the need to retrain from scratch at the arrival of new data and reducing memory usage by limiting the amount of data to be stored. This proposal has been evaluated in the Human Activity Recognition (HAR) domain and has shown to effectively reduce the catastrophic forgetting effect.

2021 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops)

Pervasive computing promotes the integration of smart electronic devices in our living and workin... more Pervasive computing promotes the integration of smart electronic devices in our living and working spaces in order to provide new, advanced services. Recently many prototype services based on machine learning techniques have been proposed in a number of domains like smart homes, smart buildings or smart plants. However, the number of applications effectively deployed in the real world is still limited. We believe that architectural principles and integrated frameworks are still missing today to successfully and repetitively support application developers and operators. In this paper, we present a novel architecture and a pervasive platform allowing the development of machine learning based applications in smart buildings.

2022 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops)

Federated Learning has been introduced as a new machine learning paradigm enhancing the use of lo... more Federated Learning has been introduced as a new machine learning paradigm enhancing the use of local devices. At a server level, FL regularly aggregates models learned locally on distributed clients to obtain a more general model. In this way, no private data is sent over the network, and the communication cost is reduced. However, current solutions rely on the availability of large amounts of stored data at the client side in order to fine-tune the models sent by the server. Such setting is not realistic in mobile pervasive computing where data storage must be kept low and data characteristic (distribution) can change dramatically. To account for this variability, a solution is to use the data regularly collected by the client to progressively adapt the received model. But such naive approach exposes clients to the well-known problem of catastrophic forgetting. The purpose of this paper is to demonstrate this problem in the mobile human activity recognition context on smartphones.

2021 IEEE International Conference on Pervasive Computing and Communications (PerCom), 2021

Pervasive computing promotes the installation of connected devices in our living spaces in order ... more Pervasive computing promotes the installation of connected devices in our living spaces in order to provide services. Two major developments have gained significant momentum recently: an advanced use of edge resources and the integration of machine learning techniques for engineering applications. This evolution raises major challenges, in particular related to the appropriate distribution of computing elements along an edgeto-cloud continuum. About this, Federated Learning has been recently proposed for distributed model training in the edge. The principle of this approach is to aggregate models learned on distributed clients in order to obtain a new, more general model. The resulting model is then redistributed to clients for further training. To date, the most popular federated learning algorithm uses coordinate-wise averaging of the model parameters for aggregation. However, it has been shown that this method is not adapted in heterogeneous environments where data is not identically and independently distributed (non-iid). This corresponds directly to some pervasive computing scenarios where heterogeneity of devices and users challenges machine learning with the double objective of generalization and personalization. In this paper, we propose a novel aggregation algorithm, termed FedDist, which is able to modify its model architecture (here, deep neural network) by identifying dissimilarities between specific neurons amongst the clients. This permits to account for clients' specificity without impairing generalization. Furthermore, we define a complete method to evaluate federated learning in a realistic way taking generalization and personalization into account. Using this method, FedDist is extensively tested and compared with three state-of-the-art federated learning algorithms on the pervasive domain of Human Activity Recognition with smartphones.

2019 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), 2019

Pervasive computing envisions environments where computers are blended into our environment to pr... more Pervasive computing envisions environments where computers are blended into our environment to provide services. This vision is today very popular and a number of pervasive platforms are already used today in smart homes or smart plants. These platforms however are based on different middleware and networks. Interoperability has recently emerged as a key issue. Interoperability is a difficult problem which requires interactions between platforms that were not designed to do so. It demands to be able to communicate but also to share contextual information on demand. We here present an architecture to federate platforms and a solution to autonomically manage context at the platform level. This is demonstrated in the smart home domain and, more precisely, with the iCasa/iPOJO and Base pervasive platforms.

ArXiv, 2021

Federated Learning, a new machine learning paradigm enhancing the use of edge devices, is receivi... more Federated Learning, a new machine learning paradigm enhancing the use of edge devices, is receiving a lot of attention in the pervasive community to support the development of smart services. Nevertheless, this approach still needs to be adapted to the specificity of the pervasive domain. In particular, issues related to continual learning need to be addressed. In this paper, we present a distillation-based approach dealing with catastrophic forgetting in federated learning scenario. Specifically, Human Activity Recognition tasks are used as a demonstration domain.

2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), 2017

Pervasive applications are today very distributed from devices to cloud facilities, going through... more Pervasive applications are today very distributed from devices to cloud facilities, going through fog-level gateways. Contextual information, based on various models, is used at every level but it needs to be frequently synchronized. In this demo, we present an application guiding users to install, move and replace devices that are used by other pervasive applications. This application has required the development of different contextual models on different platforms and kept synchronized.

2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), 2017

Smart Homes aim to improve the daily lives of the inhabitants by integrating a variety of context... more Smart Homes aim to improve the daily lives of the inhabitants by integrating a variety of context-aware applications, generally pertaining to multiple fields and provided by different actors. These applications share the same context and may have to compete for the access to resources in their surroundings. Sharing resources leads to conflicts, particularly if these applications act in contradictory ways or have interfering effects on the environment. Such conflicts can lead to critical situations by putting the home's inhabitants at risk. In this paper, we present a context-based approach to manage conflicts among pervasive applications in smart home environments. Our approach is optimistic and aims to address conflicts at runtime before their undesired effects will occur. This approach is developed and integrated in the iCASA platform as iPOJO components.

2021 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops), 2021

This document provides the details and guidelines to run the federated learning experiments used ... more This document provides the details and guidelines to run the federated learning experiments used in the study “A Federated Learning Aggregation Algorithm for Pervasive Computing: Evaluation and Comparison” [1]. The artifact is available in a public GitHub repository11PerCom2021-FL - https://github.com/getalp/PerCom2021-FL. Note that The centralized/local training approaches are not included in the repository.

Lecture Notes in Computer Science, 2020

Following the success of image recognition, machine learning approaches have recently been propos... more Following the success of image recognition, machine learning approaches have recently been proposed to improve the efficiency for such systems as industry operation and maintenance, smart buildings, and smart homes. These applications are beginning to be deployed in pervasive environments. This poses greater stress in maintaining the quality of the applications. To date, there is no architecture and tools developed that can automatically support application quality maintenance. Even worse, there is no clear definition on the requirements. In this paper, we present initial experiments that we conducted with real use cases pertaining to Industry 4.0 and discuss a set of requirements that should be met by pervasive platforms to better support AI-based applications running in the edge.

Adjunct Proceedings of the 2020 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2020 ACM International Symposium on Wearable Computers, 2020

Pervasive computing promotes the integration of connected electronic devices in our living spaces... more Pervasive computing promotes the integration of connected electronic devices in our living spaces in order to assist us through appropriate services. Two major developments have gained significant momentum recently: a better use of fog resources and the use of AI techniques. Specifically, interest in machine learning approaches for engineering applications has increased rapidly. This paradigm seems to fit the pervasive environment well. However, federated learning has been applied so far to specific services and remains largely conceptual. It needs to be tested extensively on pervasive services partially located in the fog. In this paper, we present experiments performed in the domain of Human Activity Recognition on smartphones in order to evaluate existing algorithms.

Pervasive and Mobile Computing, 2018

Abstract Middleware support for Pervasive Computing has been extensively researched in the past. ... more Abstract Middleware support for Pervasive Computing has been extensively researched in the past. Different approaches for, e.g., discovery, communication, or interaction, have been explored and successfully tested in different lab environments. Although this works well in lab environments, where one middleware manages all resources, this will not suffice in real-world deployments, where more than one system software is present and services as well as other resources are only accessible by those. In this paper, we present XWARE, an interoperability framework that allows to integrate discovery and interaction of different middleware platforms. The flexible design allows to configure and extend the interoperability framework in order to add new platforms and tailor it for the use in different domains. Furthermore, XWARE instances can communicate with each other enabling interoperability in smart environments as well as in smart peer groups. The feasibility of our approach is discussed and evaluated based on the integration of five different platforms.

Lecture Notes in Computer Science, 2016

Pervasive computing promotes environments where smart, communication-enabled devices cooperate to... more Pervasive computing promotes environments where smart, communication-enabled devices cooperate to provide services to people. Due to their inherent complexity, many pervasive applications are built on top of service-oriented platforms, providing a set of facilities simplifying their development and execution. In this paper, we present such a platform, iCasa, extended with an autonomic, service-oriented context module. This module is programmed with a domain-specific serviceoriented language built on top of iPOJO, the Apache service-oriented component model. It is validated on smart home applications developed with the Orange Labs.

2016 IEEE International Conference on Autonomic Computing (ICAC), 2016

Pervasive computing promotes environments where smart, communication-enabled devices cooperate to... more Pervasive computing promotes environments where smart, communication-enabled devices cooperate to provide services to people. Due to their inherent complexity, many pervasive applications are built on top of service-oriented platforms, providing a set of facilities simplifying their development and execution. In this demo, we present such a platform, iCasa, extended with an autonomic, service-oriented context module. It is validated on diverse applications developed with the Orange Labs in the health domain.

2016 IEEE International Conference on Services Computing (SCC), 2016

Pervasive computing promotes environments where smart, communication-enabled devices cooperate to... more Pervasive computing promotes environments where smart, communication-enabled devices cooperate to provide services to people. Due to their inherent complexity, many pervasive applications are built on top of service-oriented platforms, providing a set of facilities simplifying their development and execution. In this paper, we present such a platform, iCasa, extended with an autonomic, service-oriented context module. This module is programmed with a domainspecific service-oriented language built on top of iPOJO, the Apache service-oriented component model. It is validated on smart home applications developed with the Orange Labs.

2016 IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops), 2016

Pervasive computing envisions environments where computers are blended into everyday objects in o... more Pervasive computing envisions environments where computers are blended into everyday objects in order to provide added-value services to people. This new form of computing gives rise to huge economical and societal expectations. However, pervasive applications raise major challenges in term of software engineering and remain hard to develop, deploy, execute and maintain. Context-awareness, in particular, is a salient and difficult property that must be met by pervasive applications. In this paper, we propose a service-oriented framework facilitating the design and execution of a context management module in pervasive platforms. Our approach is illustrated with a smart home example and implemented on top of iPOJO, the Service-Oriented Component Model of our pervasive platform iCasa.

Lecture Notes in Computer Science, 2007

This paper presents SEEMP, a marketplace to coordinate and integrate public and private employmen... more This paper presents SEEMP, a marketplace to coordinate and integrate public and private employment services (ESs) around the EU Member States. The need for flexible collaboration in the marketplace gives rise to the issue of interoperability in both data exchange and share of services. SEEMP proposes a mixed approach that relies on the concepts of services and semantics. SEEMP approach combines Software Engineering and Semantic Web methodologies/tools in an infrastructure that allows for a meaningful service-based communication among ESs.

Lecture Notes in Computer Science, 2010

The dream of Model Driven Engineering (MDE) is that Software Engineering activities should be per... more The dream of Model Driven Engineering (MDE) is that Software Engineering activities should be performed only on models, but in practice a significant amount of programming is still being performed. There is a clear need to keep code and models strongly synchronized when they represent the same entities at different levels of abstraction. We observe that versioning is ill supported by MDE tools, and that no strong synchronization is ensured between code and model versions. This, among other things, explains why MDE is not widely adopted in industry. This paper presents the solution developed in the CADSE project for providing consistent support for model and code co-evolution. It is shown that it requires to (1) define, what evolution policy is to be applied, (2) closely synchronize both ways, the model entities and the computer artifacts, and (3) enforce consistency constraints and evolution policies during the commit and check-out of both model elements and their corresponding artifacts.