MUSIC: an autonomous platform supporting self-adaptive mobile applications

With the continuous improvement of device capabilities, mobile applications are becoming not only context-aware, but also self-adaptive. This new trend of applications are capable of tuning their behavior depending on changes observed in the surrounding environment. However, these adaptations are often restrained by the static features of the supporting middleware, which is not able to exploit the opportunities offered by the environment in order to improve the quality of the adaptations. The MUSIC platform offers a modular support maximizing the user satisfaction by adapting dynamically mobile applications. Beyond this support, the MUSIC platform can itself be adapted towards the evolving environment (e.g., location, network connectivity) in order to satisfy both the user requirements and the device properties. In this paper, we focus on three complementary approaches for adapting the MUSIC platform, namely i) the distribution of the adaptation process among the neighboring devices, ii) the assignment of the devices responsibilities, and iii) the adaptation of the platform components. Thus, MUSIC leverages the adaptation of mobile applications by changing autonomously the middleware characteristics.