An aspect-oriented ambient intelligence middleware platform

Currently, the interest in Ambient Intelligence (or AmI) has increased exponentially due to the widespread use of portable devices. Users demand more and more functionality from these devices, especially in order to perform collaborative tasks and interchange information. As a result, this technology proposes new challenges that must be addressed by both the hardware manufacturers and Software Engineers. The first challenge is to provide a middleware platform providing specific AmI services like communication or device discovery and able to cope with several challenges posed by AmI applications. One of these challenges is to manage heterogeneity of devices that are present in AmI environments in a transparent way, and in particular to manage the evolution of such devices without breaking the code of already developed applications. Moreover, an AmI middleware platform has to support the evolution of the software architecture of AmI applications over time, making it possible to add new functionalities, to adapt applications to any technological changes, and to produce a product line of AmI platforms to be executed in different devices. A final challenge is that the middleware platform has to support the adaptation of AmI applications to unexpected and dynamic changes in AmI environments. Aspect Oriented Programming (AOP) provides good and advanced solutions to the evolution management problem at different levels, so our work focuses on applying AOP to develop an AmI platform. In this paper, we will discuss these problems, propose solutions and present how these problems are handled in our AmI aspect-oriented platform named AOPAmI. This paper describes the internal platform structure and how it provides support to AmI devices. We will focus on the dynamic nature of AmI applications and on how to solve the above mentioned problems.