| The rapid advances in a wide range of wireless access technologies along with the efficient use of smart spaces have set up the stage for development of smart homes. Context-awareness is perhaps the most salient feature in such intelligent computing platform. The “location” information of the users plays a vital role in defining this context. To extract the best performance and efficacy of such a smart, computing environment, one needs a scalable, technology-independent location service. This brings forth the debate of choice between the geometric and symbolic forms of location information as a consistent representation. We discuss how symbolic location information can be used to profile personal mobility pattern of users for building a predictive framework which provides supports for a number of location-aware services.; In this thesis we have developed a predictive framework for location aware resource optimization in smart homes. The LeZi update algorithm helps in efficient learning of the movement profiles of the mobile user from symbolic space. The symbol-wise context model in the incremental parse tree preserves the path segments followed by the mobile user in his daily life. Using the concept of typical set for stationary, ergodic sequences, user's future location inside the room as well as most likely path-segments can be predicted with good accuracy. Successful prediction helps in pro-active resource reservations and automated, on-demand operations of devices in the paths and locations, which the user is going to follow. This scheme aids in providing the necessary comfort to the user at near-optimal cost. Simulation results based on real data and synthetic traces provide a clear picture of gain in update costs, prediction costs and prediction success-rates. Sufficient reduction in daily average energy consumption, switching operations and time spent by the user for manual operations provides a good essence of user comfort in smart environment. |
