| Application of wireless technologies in the smart home is dealt with by pointing out advantages and limitations of available approaches for the solution of heterogeneous and coexisting problems related to the distributed monitoring of the home and the inhabitants. Some hot challenges facing the exploitation of noninvasive wireless devices for user behavior monitoring are then addressed and the application fields of smart power management and elderly people monitoring are chosen as representative cases where the estimation of user activities improves the potential of location-aware services in the smart home. The problem of user localization is considered with great care to minimize the invasiveness of the monitoring system. Wireless architectures are reviewed and discussed as flexible and transparent tools toward the paradigm of a totally automatic/autonomic environment. With respect to available state-of-the-art solutions, our proposed architecture is based also on existing wireless devices and exploits, in an opportunistic way, the characteristics of wireless signals to estimate the presence, the movements, and the behaviors of inhabitants, reducing the system complexity and costs. Selected and representative examples from real implementations are pre-sented to give some insight on state-of-the-art solutions also envisaging possible future trends.

Sparsening conformal arrangements is carried out through a versatile Multi-Task Bayesian Compressive Sensing strategy. The problem, formulated in a probabilistic fashion as a pattern-matching synthesis, is that of determining the sparsest excitation set (locations and weights) fitting a reference pattern subject to user-defined geometrical constraints. Results from a set of representative numerical experiments are presented to illustrate the key-features of the proposed approach as well as to assess, also through comparisons, its potentials in terms of matching accuracy, element saving, and computational costs.

No abstract

No abstract

International audienc