Sensing and Systems in Pervasive Computing: Engineering Context Aware Systems (Undergraduate Topics in Computer Science)
Format: PDF / Kindle (mobi) / ePub
Focus on issues and principles in context awareness, sensor processing and software design (rather than sensor networks or HCI or particular commercial systems).
Designed as a textbook, with readings and lab problems in most chapters.
Focus on concepts, algorithms and ideas rather than particular technologies.
action is generally described as a code fragment, which might cause a reconfiguration, or make some application call. In the example in Listing 5.1 a location change event is caught. The scale of the event is limited to at least room-level changes. The limit is written as a delta filter on the event, but may in fact be used to configure the event system. Next conditions are checked, confirming that the user is at work (and so wants phone calls re-routed) but not in a meeting which should not be
This basic model is found in many spatial data definitions, although often with extensions, such as: Collections, defined as multiples of one of the other types; relationships, e.g. road lines connected by junction points; and additional attributes, including directionality for a line. OpenGIS1 is a popular current GIS description standard which embodies many of these characteristics. Google Maps2 provides an easily accessed view of combined raster (satellite) and vector drawings (e.g. roads,
indoor and outdoor, scalable, reliable and cheap coverage has not yet been achieved. To obtain even reliable, global coverage requires a system which can combine location data from different sources—converting between reference and coordinate systems and adapting to varying resolution, error characteristics. Where the object to be located is covered by multiple location systems at once, the data from these systems can still usefully be combined. Depending on the reference and coordinate systems
model is adjusting to change or suffering from sensor error. Fig. 8.1An illustration of a noisy numeric model converting into a hierarchical state model In our notation of context aspects from Chap. 5 we should extend the overall context for an object to include time (t) and other meta data (): (8.1) Here is meta data associated with the context of an object at time t′. Meta data may include models of noise, identification of data sources. The time associated with a context reading may be
formation of queries over sensor networks The use location in these networks The coverage is necessarily incomplete in a single chapter, but the issues raised are particularly pertinent to the rest of this book and re-illustrate the various techniques considered. The collection and processing of sensor data in an efficient, adaptive manner which reflects the context of nodes, e.g. energy level, location, is an important part of the design of ubiquitous computing infrastructure. Not only do