The recent advances in digital circuitry, wireless communications and MicroElectroMechanical Systems (MEMS) have allowed the emergence of devices (sensor motes) incorporating wireless communication, processing and storage capabilities, very diverse sensing functionalities as well as (potentially) actuators, in a single unit that is compact, cheap, autonomous and destined to become ubiquitous. This revolution is appearing in the form of dense and distributed Wireless Sensor Networks.

The potential of self-organized wireless sensor networks is enormous for several important applications (e.g. environmental monitoring, structural health monitoring and seismic prediction, monitoring of urban areas and cities, etc) because its deployment may be the only way to achieve the large scales, full decentralization and robustness demanded by this type of applications. Despite recent noteworthy efforts to build infrastructure and clever protocols to network these power-limited devices, a fundamental understanding of both system performance theoretical limits and the optimal design of practical in-network algorithms for the various tasks, remains far from nature. This proposal is motivated by the grand challenge of providing both: a) a fundamental understanding of the performance bounds of large-scale robust wireless sensor networks, b) concrete design guidelines, algorithmic prescriptions and software architectures to provide the required robustness, power efficiency, autonomy and adaptation. The research in this project covers the distributed acquisition of physical fields, In-network distributed processing with communication constraints and partial information in order to achieve various tasks of interest (inference, detection, estimation, classification, decision-making, control), the various problems involved to be able to achieve a truly self-organized communication, and we also analyze and optimize the cooperation in heterogeneous networks. On the more practical side, we also implement a wireless sensor network testbed that will be demonstrated and where the various algorithms will be tested.

Ministerio de Educación y Ciencia