Wireless sensor network-based distributed GNSS receiver architecture for infrastructure monitoring

Wireless Sensor Networks (WSN) have received a marked interest in the recent years. They are constituted of low cost and completely autonomous nodes with extremely limited power and processing capabilities that wirelessly communicate between each other. The majority of field deployments is executed in an ad-hoc manner in an area with no power or communication infrastructure, and in unpredictable locations. Their essential use is to provide spatio-temporal information about their environment. They can collect physical information such as temperature, or boundaries of a diffuse phenomenon over time and space such as expanding chemical plumes. They can be also used in collaborative data collection such as seismologic monitoring, or construction monitoring. Their optimal use presupposes knowledge of their relative location, either in an approximate manner, or in a very accurate manner. Earlier solutions to provide the position or location function were not using Global Navigation Satellite Systems, usually resulting in low accuracy. Later, existing GPS receivers were attached to a subset of the nodes, without trying to rethink their architecture in the context of WSN. In order to achieve optimality, the GPS function should become an integral part of the node and the network architectures in such a way that the redundancy is eliminated across nodes, and the energy consumption is reduced as much as possible and distributed among nodes as evenly as possible.The main goal of this work is to propose a practical solution for accurate relative positioning between nodes of a Wireless Sensor Network, outlining their fundamental limitations, and developing a GPS distributed architecture aligned with these limitations. In particular a new concept, namely the WA-GPS or Wireless Sensor Network A-GPS concept, will be introduced, where the per node support to the location function is extremely limited, and the effort to collect GPS necessary information is distributed among all nodes. The problem of synchronization between nodes and its unacceptable impact on carrier phase relative positioning accuracy will be exposed and a novel technique, the Ambiguity Resolution of Time Integer (ARTI), will be proposed as mitigation. Three WSN architectures apt to support relative positioning will be proposed and assessed in terms of energy consumption and robustness.