Smart Homes - Academia.edu (original) (raw)
Papers by Smart Homes
As the number of elderly people has increased worldwide, there has been a surge of research into ... more As the number of elderly people has increased worldwide, there has been a surge of research into assistive technologies to provide them with better care by recognizing their normal and abnormal activities. However, existing abnormal activity recognition (AAR) algorithms rarely consider sub-activity relations when recognizing abnormal activities. This paper presents an application of the Hidden State Conditional Random Field (HCRF) method to detect and assess abnormal activities that often occur in elderly persons' homes. Based on HCRF, this paper designs two AAR algorithms, and validates them by comparing them with a feature vector distance based algorithm in two experiments. The results demonstrate that the proposed algorithms favorably outperform the competitor, especially when abnormal activities have same sensor type and sensor number as normal activities.
Smart homes are considered a vital technology in an aging society as they compensate for a shorta... more Smart homes are considered a vital technology in an aging society as they compensate for a shortage in care workers. However, often smart homes do not perform well. Performance management is well known in the manufacturing industry but not common in the building industry. The performance approach is the practice of thinking and working in terms of ends rather than means. It is concerned with what a building or a building product is required to do, and not with prescribing how it is to be constructed. Performance based building (PBB) includes amongst others functional briefing and performance assessments. However, an overview of tasks in PBB is lacking and the principle has not yet been applied to smart homes. Aim of this research is to contribute to performance management in the construction of health smart homes by identifying tasks and proposing a task assignment. Tasks are identified by studying the extended reporting of the Performance Based Building Network (PeBBu). In conclusi...
Une approche orientée événements discrets pour la localisation des habitants dans des habitats in... more Une approche orientée événements discrets pour la localisation des habitants dans des habitats intelligents basée sur le modèle Ein ereignisdiskreter Ansatz zur modellbasierten Lokalisierung der Bewohner in intelligenten Wohnungen
Competitive Building. Without the unfaltering support and encouragement from a number of people, ... more Competitive Building. Without the unfaltering support and encouragement from a number of people, this work would not have been possible, and definitely not as enjoyable. I would like to thank all of them, albeit only a few can be mentioned here. First and foremost, I would like to thank my principal supervisor Professor Ulf Keijer for giving me the opportunity to embark this scientific journey, for having faith in me, and for always being supportive and open minded. Thank you Ulf – all the discussions with you over the years were a true pleasure. I would also like to thank my second supervisor Associate Professor Inga Britt Werner, who at different occasions contributed to the advancement of the work with creative criticism and, in addition, provided valuable advice about the statistical analyses.
Abstract—The research of adaptive systems has a long tradition in different fields of computer sc... more Abstract—The research of adaptive systems has a long tradition in different fields of computer science. Especially in the area of ambient intelligence context aware systems play a key role towards smart environments. Position data are in many approaches the basis for context interpretation. The interpretation of various combined sensor data however lead to more sophisticated models. Experiments in smart home settings allow the validation of the combination of context theories. A flexible system architecture with the possibility for easy integration of new sensors in combination with a rule-based semantic interpretation engine for fast changes of a situational guess allows complex experiments in adaptive environments. This paper describes first steps towards an adaptive living environment setting. The overall architecture is given as well as the interpretation processes for context estimations. The role of location information and the semantic building model as basis for the detectio...
A “smart house ” that responds to the dweller’s needs and desires by adjusting lighting, temperat... more A “smart house ” that responds to the dweller’s needs and desires by adjusting lighting, temperature, even ambient music, has reached the millions of homes in the present century. The basic idea of home automation is to employ sensors and control systems to monitor a dwelling, and accordingly adjust the various mechanisms that provide heat, ventilation, lighting, and other services. The work is concerned with the development of a smart home architecture allowing to integrate information from a wide variety of sensors and actuators: information recruited for these elements is processed into microprocessors implementing computational intelligence techniques; cooperative communication between units is implemented through a wireless net into the home; and internet resources allow to link the home with external services. The paper presents an agent-based
Abstract. While people have many ideas about how a smart home should react to particular behaviou... more Abstract. While people have many ideas about how a smart home should react to particular behaviours from their inhabitant, there seems to have been relatively little attempt to organise this systematically. In this paper, we attempt to rectify this in consideration of context awareness and novelty detection for a smart home that monitors its inhabitant for illness and unexpected behaviour. We do this through the concept of the Use Case, which is used in software engineering to specify the behaviour of a system. We describe a set of scenarios and the possible outputs that the smart home could give and introduce the SHMUC Repository of Smart Home Use Cases. Based on this, we can consider how probabilistic and logic-based reasoning systems would produce different capabilities.
Distributed network of wireless nodes that monitor the environment or other systems. 3 Deployed i... more Distributed network of wireless nodes that monitor the environment or other systems. 3 Deployed in large numbers Nodes have limited battery life Nodes have low computational power Nodes have small data storage
Abstract. In the healthcare domain, the evaluation of elders ’ daily activities is important to d... more Abstract. In the healthcare domain, the evaluation of elders ’ daily activities is important to define their dependencies; professionals define manually this information by using the AGGIR grid. With the aging of the word population, it is interesting to automate the dependency’s evaluation. In our work, we use the data provided by the different sensors in iCASA simulator and give them semantic. We use the RDF to add semantic to our data. We extend our system by using the ontology and use the SWRL to get more useful information. Our system is based on 3 levels, each level provides information to help us to evaluate the dependency. The first level collects the raw data and represents them in RDF, the second level represents the ambient services, those services use the aggregation of data to get information that help to evaluate the AGGIR services that represents the daily activity of the elder person, those services are represented in the third layer. Having the dependency’s value, w...
Application of sensor-based technology within activity monitoring systems is becoming a popular t... more Application of sensor-based technology within activity monitoring systems is becoming a popular technique within the smart environment paradigm. Nevertheless, the use of such an approach generates complex constructs of data, which subsequently requires the use of intricate activity recognition techniques to automatically infer the underlying activity. This paper explores a cluster-based ensemble method as a new solution for the purposes of activity recognition within smart environments. With this approach activities are modelled as collections of clusters built on different subsets of features. A classification process is performed by assigning a new instance to its closest cluster from each collection. Two different sensor data representations have been investigated, namely numeric and binary. Following the evaluation of the proposed methodology it has been demonstrated that the cluster-based ensemble method can be successfully applied as a viable option for activity recognition. Results following exposure to data collected from a range of activities indicated that the ensemble method had the ability to perform with accuracies of 94.2% and 97.5% for numeric and binary data, respectively. These results outperformed a range of single classifiers considered as benchmarks.
Abstract-We present an ontology-based reasoning approach for saving energy in a smart home settin... more Abstract-We present an ontology-based reasoning approach for saving energy in a smart home setting where a mobile phone can serve as a generic sensor which can collect the inhabitant’s contextual data. The paper details an ontology that describes the smart home domain and a prototype to test the system. Finally, we conclude with lessons learned from our work in developing an energy-aware smart home prototype and suggestions for future work. I.
Inexpensive wireless sensing products are dramatically reducing the cost of in-home sensing. Howe... more Inexpensive wireless sensing products are dramatically reducing the cost of in-home sensing. However, these sensors have been found to fail often and prohibitive maintenance costs may negate the cost benefits of inexpensive hardware and do-it-yourself installation. In this paper, we describe a new technique called SMART that uses applicationlevel semantics to detect, assess, and adapt to sensor failures. SMART detects node failures at run-time by analyzing the relative behavior of multiple classifier instances trained to recognize the same set of activities based on different subsets of sensors. Once a failure is detected, SMART assesses its importance and adapts the classifier ensemble in attempt to avoid maintenance dispatch. Evaluation on two public datasets shows that SMART decreases the number of maintenance dispatches by 55 % on average, identifies non-failstop failures at run-time with more than 85 % accuracy, and improves the activity recognition accuracy under node failures...
Devices and applications communicate invisibly behind the back of inhabitants “Poor strategies fo... more Devices and applications communicate invisibly behind the back of inhabitants “Poor strategies for diagnosis and troubleshooting ” in home networks [4], debugging tools are hard to operate and almost never used by ordinary end users [3] → Privacy-sensitive data flows out of home network (e.g., from smart thermostats) and “fear of pirating ” by remote services [2, 5]
partial fulfillment of the requirements for the degree of Master of Science in Software Engineeri... more partial fulfillment of the requirements for the degree of Master of Science in Software Engineering. The thesis is equivalent to 10 weeks of full time studies. Contact Information: Author(s):
UPnP is a distributed computing framework based on Web technologies that is intended for use in l... more UPnP is a distributed computing framework based on Web technologies that is intended for use in local area networksespecially home networks. Over the years, valuable lessons have been learned from the efforts of various companies to develop and deploy products based on this technology.
Journal of Biological Education, 2018
Mulder (2019) The effects of an online learning environment with worked examples and peer feedbac... more Mulder (2019) The effects of an online learning environment with worked examples and peer feedback on students' argumentative essay writing and domain-specific knowledge acquisition in the field of biotechnology,
SSRN Electronic Journal, 2016
In the following article, the theory of knowledge management (TKM) is described as the systematis... more In the following article, the theory of knowledge management (TKM) is described as the systematisation of an organisation’s practices to facilitate consistent service and product delivery across divisions and regions. Organisation theory is approached from the postmodern perspective and expressed as comprising five separate elements: organisational activism, workplace humanism, societal context, multiple disciplines, and the value given to historical knowledge. Corporate social responsibility (CSR) is addressed in its importance as a popular project for organisations, however, studies show that CSR does not present long-term value for most firms. Chia’s (1996) consideration of being-realism in organisational theory is evaluated, questioning the rationality of depicting strategies and goals with physicality in order to compare them. Mills, Boylstein, & Lorean (2001) provide an argument for a society heavily influenced by technology which causes individuals to feel a deep sense of impending doom, which is examined against related studies. Finally, Newton’s (1996) comparison of postmodernist theory, organisational theory, and feminist study within the context of actions in the postmodernist movement, is reviewed. The article concludes by critiquing postmodernist applications of organisational theory and their applicability when faced with real-world constraining elements, such as physical risks and time factors. It is suggested that this uneven transition outside of academics makes creating a functional organisational CSR model for ecommerce corporations impractical in long-range planning.
Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2016
Understanding the argumentative structure of a persuasive essay involves addressing two challengi... more Understanding the argumentative structure of a persuasive essay involves addressing two challenging tasks: identifying the components of the essay's argument and identifying the relations that occur between them. We examine the under-investigated task of end-toend argument mining in persuasive student essays, where we (1) present the first results on end-to-end argument mining in student essays using a pipeline approach; (2) address error propagation inherent in the pipeline approach by performing joint inference over the outputs of the tasks in an Integer Linear Programming (ILP) framework; and (3) propose a novel objective function that enables F-score to be maximized directly by an ILP solver. We evaluate our joint-inference approach with our novel objective function on a publiclyavailable corpus of 90 essays, where it yields an 18.5% relative error reduction in F-score over the pipeline system.
21st International Conference on Advanced Information Networking and Applications Workshops (AINAW'07), 2007
We present a framework for a wireless health monitoring system within a smart home using ZigBee t... more We present a framework for a wireless health monitoring system within a smart home using ZigBee technology. Vital signals are collected and processed using a 3-tiered architecture. The first stage is the mobile device carried on the body that runs a number of wired and wireless probes. This device is also designed to perform some basic processing such as the heart rate and fatal failure detection. At the second stage, further processing is performed by a local server using the raw data transmitted by the mobile device continuously. The raw data is also stored at this server. The processed data as well as the outcomes of the analysis are then transmitted to the service provider center for expertsŕeview for diagnosis as well as storage. The main advantages of the proposed framework are (1) The ability to detect signals wirelessly within a Body Area Network (BAN) (2)Low-power and reliable data sensing through ZigBee network nodes and (3) Optimized analysis of data through an adaptive tiered architecture that maximizes the utility of processing and computational capacity at each of three stages. We are currently building a prototype of the proposed system using in-house ECG probes and ZigBee radio modules.
2011 IEEE GCC Conference and Exhibition (GCC), 2011
The future home will be pervaded by mobile and stationary devices which need to dynamically conne... more The future home will be pervaded by mobile and stationary devices which need to dynamically connect and coordinate in order to seamlessly help people in accomplishing their tasks. However, for this vision to become a reality, it is important for researcher and developer to build applications that constantly adapt and integrate heterogeneous devices supported by different communication technologies. Bluetooth efficiently fits in the home environment: the range of communication it provides sufficient for Smart Homes applications consisting of a large number of resource-constraint devices. In this paper, we present a Pervasive Discovery Framework for Smart Homes, and an implementation currently supporting the Bluetooth and UPnP device discovery. The framework is based on robust technologies such as the UPnP protocol and the Java based OSGi bundles.
As the number of elderly people has increased worldwide, there has been a surge of research into ... more As the number of elderly people has increased worldwide, there has been a surge of research into assistive technologies to provide them with better care by recognizing their normal and abnormal activities. However, existing abnormal activity recognition (AAR) algorithms rarely consider sub-activity relations when recognizing abnormal activities. This paper presents an application of the Hidden State Conditional Random Field (HCRF) method to detect and assess abnormal activities that often occur in elderly persons' homes. Based on HCRF, this paper designs two AAR algorithms, and validates them by comparing them with a feature vector distance based algorithm in two experiments. The results demonstrate that the proposed algorithms favorably outperform the competitor, especially when abnormal activities have same sensor type and sensor number as normal activities.
Smart homes are considered a vital technology in an aging society as they compensate for a shorta... more Smart homes are considered a vital technology in an aging society as they compensate for a shortage in care workers. However, often smart homes do not perform well. Performance management is well known in the manufacturing industry but not common in the building industry. The performance approach is the practice of thinking and working in terms of ends rather than means. It is concerned with what a building or a building product is required to do, and not with prescribing how it is to be constructed. Performance based building (PBB) includes amongst others functional briefing and performance assessments. However, an overview of tasks in PBB is lacking and the principle has not yet been applied to smart homes. Aim of this research is to contribute to performance management in the construction of health smart homes by identifying tasks and proposing a task assignment. Tasks are identified by studying the extended reporting of the Performance Based Building Network (PeBBu). In conclusi...
Une approche orientée événements discrets pour la localisation des habitants dans des habitats in... more Une approche orientée événements discrets pour la localisation des habitants dans des habitats intelligents basée sur le modèle Ein ereignisdiskreter Ansatz zur modellbasierten Lokalisierung der Bewohner in intelligenten Wohnungen
Competitive Building. Without the unfaltering support and encouragement from a number of people, ... more Competitive Building. Without the unfaltering support and encouragement from a number of people, this work would not have been possible, and definitely not as enjoyable. I would like to thank all of them, albeit only a few can be mentioned here. First and foremost, I would like to thank my principal supervisor Professor Ulf Keijer for giving me the opportunity to embark this scientific journey, for having faith in me, and for always being supportive and open minded. Thank you Ulf – all the discussions with you over the years were a true pleasure. I would also like to thank my second supervisor Associate Professor Inga Britt Werner, who at different occasions contributed to the advancement of the work with creative criticism and, in addition, provided valuable advice about the statistical analyses.
Abstract—The research of adaptive systems has a long tradition in different fields of computer sc... more Abstract—The research of adaptive systems has a long tradition in different fields of computer science. Especially in the area of ambient intelligence context aware systems play a key role towards smart environments. Position data are in many approaches the basis for context interpretation. The interpretation of various combined sensor data however lead to more sophisticated models. Experiments in smart home settings allow the validation of the combination of context theories. A flexible system architecture with the possibility for easy integration of new sensors in combination with a rule-based semantic interpretation engine for fast changes of a situational guess allows complex experiments in adaptive environments. This paper describes first steps towards an adaptive living environment setting. The overall architecture is given as well as the interpretation processes for context estimations. The role of location information and the semantic building model as basis for the detectio...
A “smart house ” that responds to the dweller’s needs and desires by adjusting lighting, temperat... more A “smart house ” that responds to the dweller’s needs and desires by adjusting lighting, temperature, even ambient music, has reached the millions of homes in the present century. The basic idea of home automation is to employ sensors and control systems to monitor a dwelling, and accordingly adjust the various mechanisms that provide heat, ventilation, lighting, and other services. The work is concerned with the development of a smart home architecture allowing to integrate information from a wide variety of sensors and actuators: information recruited for these elements is processed into microprocessors implementing computational intelligence techniques; cooperative communication between units is implemented through a wireless net into the home; and internet resources allow to link the home with external services. The paper presents an agent-based
Abstract. While people have many ideas about how a smart home should react to particular behaviou... more Abstract. While people have many ideas about how a smart home should react to particular behaviours from their inhabitant, there seems to have been relatively little attempt to organise this systematically. In this paper, we attempt to rectify this in consideration of context awareness and novelty detection for a smart home that monitors its inhabitant for illness and unexpected behaviour. We do this through the concept of the Use Case, which is used in software engineering to specify the behaviour of a system. We describe a set of scenarios and the possible outputs that the smart home could give and introduce the SHMUC Repository of Smart Home Use Cases. Based on this, we can consider how probabilistic and logic-based reasoning systems would produce different capabilities.
Distributed network of wireless nodes that monitor the environment or other systems. 3 Deployed i... more Distributed network of wireless nodes that monitor the environment or other systems. 3 Deployed in large numbers Nodes have limited battery life Nodes have low computational power Nodes have small data storage
Abstract. In the healthcare domain, the evaluation of elders ’ daily activities is important to d... more Abstract. In the healthcare domain, the evaluation of elders ’ daily activities is important to define their dependencies; professionals define manually this information by using the AGGIR grid. With the aging of the word population, it is interesting to automate the dependency’s evaluation. In our work, we use the data provided by the different sensors in iCASA simulator and give them semantic. We use the RDF to add semantic to our data. We extend our system by using the ontology and use the SWRL to get more useful information. Our system is based on 3 levels, each level provides information to help us to evaluate the dependency. The first level collects the raw data and represents them in RDF, the second level represents the ambient services, those services use the aggregation of data to get information that help to evaluate the AGGIR services that represents the daily activity of the elder person, those services are represented in the third layer. Having the dependency’s value, w...
Application of sensor-based technology within activity monitoring systems is becoming a popular t... more Application of sensor-based technology within activity monitoring systems is becoming a popular technique within the smart environment paradigm. Nevertheless, the use of such an approach generates complex constructs of data, which subsequently requires the use of intricate activity recognition techniques to automatically infer the underlying activity. This paper explores a cluster-based ensemble method as a new solution for the purposes of activity recognition within smart environments. With this approach activities are modelled as collections of clusters built on different subsets of features. A classification process is performed by assigning a new instance to its closest cluster from each collection. Two different sensor data representations have been investigated, namely numeric and binary. Following the evaluation of the proposed methodology it has been demonstrated that the cluster-based ensemble method can be successfully applied as a viable option for activity recognition. Results following exposure to data collected from a range of activities indicated that the ensemble method had the ability to perform with accuracies of 94.2% and 97.5% for numeric and binary data, respectively. These results outperformed a range of single classifiers considered as benchmarks.
Abstract-We present an ontology-based reasoning approach for saving energy in a smart home settin... more Abstract-We present an ontology-based reasoning approach for saving energy in a smart home setting where a mobile phone can serve as a generic sensor which can collect the inhabitant’s contextual data. The paper details an ontology that describes the smart home domain and a prototype to test the system. Finally, we conclude with lessons learned from our work in developing an energy-aware smart home prototype and suggestions for future work. I.
Inexpensive wireless sensing products are dramatically reducing the cost of in-home sensing. Howe... more Inexpensive wireless sensing products are dramatically reducing the cost of in-home sensing. However, these sensors have been found to fail often and prohibitive maintenance costs may negate the cost benefits of inexpensive hardware and do-it-yourself installation. In this paper, we describe a new technique called SMART that uses applicationlevel semantics to detect, assess, and adapt to sensor failures. SMART detects node failures at run-time by analyzing the relative behavior of multiple classifier instances trained to recognize the same set of activities based on different subsets of sensors. Once a failure is detected, SMART assesses its importance and adapts the classifier ensemble in attempt to avoid maintenance dispatch. Evaluation on two public datasets shows that SMART decreases the number of maintenance dispatches by 55 % on average, identifies non-failstop failures at run-time with more than 85 % accuracy, and improves the activity recognition accuracy under node failures...
Devices and applications communicate invisibly behind the back of inhabitants “Poor strategies fo... more Devices and applications communicate invisibly behind the back of inhabitants “Poor strategies for diagnosis and troubleshooting ” in home networks [4], debugging tools are hard to operate and almost never used by ordinary end users [3] → Privacy-sensitive data flows out of home network (e.g., from smart thermostats) and “fear of pirating ” by remote services [2, 5]
partial fulfillment of the requirements for the degree of Master of Science in Software Engineeri... more partial fulfillment of the requirements for the degree of Master of Science in Software Engineering. The thesis is equivalent to 10 weeks of full time studies. Contact Information: Author(s):
UPnP is a distributed computing framework based on Web technologies that is intended for use in l... more UPnP is a distributed computing framework based on Web technologies that is intended for use in local area networksespecially home networks. Over the years, valuable lessons have been learned from the efforts of various companies to develop and deploy products based on this technology.
Journal of Biological Education, 2018
Mulder (2019) The effects of an online learning environment with worked examples and peer feedbac... more Mulder (2019) The effects of an online learning environment with worked examples and peer feedback on students' argumentative essay writing and domain-specific knowledge acquisition in the field of biotechnology,
SSRN Electronic Journal, 2016
In the following article, the theory of knowledge management (TKM) is described as the systematis... more In the following article, the theory of knowledge management (TKM) is described as the systematisation of an organisation’s practices to facilitate consistent service and product delivery across divisions and regions. Organisation theory is approached from the postmodern perspective and expressed as comprising five separate elements: organisational activism, workplace humanism, societal context, multiple disciplines, and the value given to historical knowledge. Corporate social responsibility (CSR) is addressed in its importance as a popular project for organisations, however, studies show that CSR does not present long-term value for most firms. Chia’s (1996) consideration of being-realism in organisational theory is evaluated, questioning the rationality of depicting strategies and goals with physicality in order to compare them. Mills, Boylstein, & Lorean (2001) provide an argument for a society heavily influenced by technology which causes individuals to feel a deep sense of impending doom, which is examined against related studies. Finally, Newton’s (1996) comparison of postmodernist theory, organisational theory, and feminist study within the context of actions in the postmodernist movement, is reviewed. The article concludes by critiquing postmodernist applications of organisational theory and their applicability when faced with real-world constraining elements, such as physical risks and time factors. It is suggested that this uneven transition outside of academics makes creating a functional organisational CSR model for ecommerce corporations impractical in long-range planning.
Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2016
Understanding the argumentative structure of a persuasive essay involves addressing two challengi... more Understanding the argumentative structure of a persuasive essay involves addressing two challenging tasks: identifying the components of the essay's argument and identifying the relations that occur between them. We examine the under-investigated task of end-toend argument mining in persuasive student essays, where we (1) present the first results on end-to-end argument mining in student essays using a pipeline approach; (2) address error propagation inherent in the pipeline approach by performing joint inference over the outputs of the tasks in an Integer Linear Programming (ILP) framework; and (3) propose a novel objective function that enables F-score to be maximized directly by an ILP solver. We evaluate our joint-inference approach with our novel objective function on a publiclyavailable corpus of 90 essays, where it yields an 18.5% relative error reduction in F-score over the pipeline system.
21st International Conference on Advanced Information Networking and Applications Workshops (AINAW'07), 2007
We present a framework for a wireless health monitoring system within a smart home using ZigBee t... more We present a framework for a wireless health monitoring system within a smart home using ZigBee technology. Vital signals are collected and processed using a 3-tiered architecture. The first stage is the mobile device carried on the body that runs a number of wired and wireless probes. This device is also designed to perform some basic processing such as the heart rate and fatal failure detection. At the second stage, further processing is performed by a local server using the raw data transmitted by the mobile device continuously. The raw data is also stored at this server. The processed data as well as the outcomes of the analysis are then transmitted to the service provider center for expertsŕeview for diagnosis as well as storage. The main advantages of the proposed framework are (1) The ability to detect signals wirelessly within a Body Area Network (BAN) (2)Low-power and reliable data sensing through ZigBee network nodes and (3) Optimized analysis of data through an adaptive tiered architecture that maximizes the utility of processing and computational capacity at each of three stages. We are currently building a prototype of the proposed system using in-house ECG probes and ZigBee radio modules.
2011 IEEE GCC Conference and Exhibition (GCC), 2011
The future home will be pervaded by mobile and stationary devices which need to dynamically conne... more The future home will be pervaded by mobile and stationary devices which need to dynamically connect and coordinate in order to seamlessly help people in accomplishing their tasks. However, for this vision to become a reality, it is important for researcher and developer to build applications that constantly adapt and integrate heterogeneous devices supported by different communication technologies. Bluetooth efficiently fits in the home environment: the range of communication it provides sufficient for Smart Homes applications consisting of a large number of resource-constraint devices. In this paper, we present a Pervasive Discovery Framework for Smart Homes, and an implementation currently supporting the Bluetooth and UPnP device discovery. The framework is based on robust technologies such as the UPnP protocol and the Java based OSGi bundles.