Reducing bandwidth requirements and optimizing data flow in distributed data acquisition and processing

With the advent of new more capable technologies and greater availability of sensing technologies the amount of sensor data available for creating situation awareness is also increasing in an exponential rate. Although technology has provided us with means to have access to more data, the processing methodology has not been able to keep up with that pace. Nature has provided humans and other animals with great means to handle this issue - we do not pay attention to the data that has no relevance to us at a given moment in a given situation. However, provided we observe relevant cues (e.g. danger) to indicate that the situation may be changing or that an entity is of interest we immediately start paying attention, collecting additional detailed information of the phenomena and storing that information for future use. Similar approach would be useful also with modern distributed data acquisition systems catering for our situational information needs - there is no need to process all available data from all sources if the data is not relevant at a given moment in a given situation.