Macroprogramming of Wireless Sensor Networks using Task Graphs and Constraint Solving
2012 (English)Conference paper (Refereed)
Programming a wireless sensor network (WSN) on node level is a tedious, difficult, and error-prone task. A way to address this problem is to use a high-level programming language specifically designed for WSNs. However, we anticipate to go one step further and provide a standard and generic model for macro-level programming of WSNs. We propose a platform for WSNs as a multi-layer abstraction approach for macroprogramming, where, on the highest level of abstraction, the developer uses an interactive graphical interface to specify the features of a sensor network as a data-driven task graph. The task graph expresses the functionality of the entire network as a whole, and it encapsulates the requirements and resource limitations on the network and the sensor nodes, as well as the data flows among the tasks. This platform makes use of several optimization methods, such as constraint programming, to map the tasks to nodes optimally, and to plan the configuration for deployment to maximize the network life time. It makes use of more realistic network abstractions and takes into account the dynamics of WSNs. We present our model for macro-level programming and show that many optimization problems in this context can be solved more efficiently with suitable techniques.
Place, publisher, year, edition, pages
Research subject Computer Science with specialization in Computer Communication; Computer Science with specialization in Embedded Systems
IdentifiersURN: urn:nbn:se:uu:diva-185269OAI: oai:DiVA.org:uu-185269DiVA: diva2:571112
8th Swedish National Computer Networking Workshop (SNCNW), 7-8 June, 2012, Stockholm, Sweden