Peter Lorkowski Knihy

Monitoring continuous phenomena using stationary and mobile sensors has become prevalent due to advancements in hardware, communication infrastructure, and cost reductions. Sensor data is now accessible in near real-time through web interfaces and machine-readable formats, thanks to the Internet of Things (IoT). However, challenges remain in data usability, particularly when the positions of observations and points of interest do not align. Interpolation serves as a method to bridge these gaps, with various techniques available. Operating a monitoring system involves addressing issues such as selecting appropriate interpolation methods, storing observations, retrieving interpolated data, updating models for real-time monitoring, compressing observational data, and defining critical states through value aggregation. This work proposes a comprehensive system architecture to tackle these challenges, emphasizing a holistic approach rather than focusing solely on specific interpolation methods. It introduces state-of-the-art technologies like geostatistics and sensor web enablement, offering a robust toolset for the domain. The emphasis is on the overall organization of monitoring systems and the architectural design of the software, alongside a simulation framework for evaluating different monitoring strategies. The entire monitoring cycle—observation, interpolation, discretization, storage, retrieval, and notification—is addresse