Ubiquitous computing and ambient intelligence for smart homes applications

The increasing presence of ubiquitous computing and ambient intelligence in products and services is presently creating huge opportunities in a variety of different areas. To create a robust and reliable platform for complex distributed embedded systems like a smart home, we address two different challenges which are not often met in existing complex distributed embedded systems. The first focus of this study is on hypervision (virtualization). In this work we posit that by completely isolating the subsystem, users can run several independent operating systems or applications directly over the hypervisor. To support safety critical real-time applications, the hypervisor must execute the different sub-domains with real-time characteristics. Our second challenge is to implement a micro middleware called μDDS, which enables wireless embedded systems and access networks to operate seamlessly by providing a programming language and protocol interoperability based on the standard Data Distribution Service (DDS) specification. In this study, we deployed the PANDORA algorithm for wireless hybrid networks as part of the μDDS. As a proof of concept, a small Wireless Hybrid Network (WHN) is evaluated with three embedded platforms and two laptops computers (backbone network), one of the embedded platforms is employed as the IROOT (Internet Gateway). Furthermore, a wireless sensor network with ten sensor nodes was also evaluated as part of the WHN.