Easy Programming of Integrated Wireless Sensor Networks

Welcome to the website of the makeSense project. This project aims at simplifying both the programming of wireless sensor networks and their integration with business processes.

Wireless Sensor Networks (WSN) are a key component of the upcoming Internet of Things revolution that will revolve around sm art objects cooperating in an autonomous fashion to achieve a higher level goal(s). However, sensor networks currently are exceedingly difficult to program. Programming is focused on individual sensor nodes, and programmers focus on low-level details rather than high-level objectives.

Please watch our video at for an intro to makeSense http://www.youtube.com/embed/n4ospuvPrWA

makeSense intends to drastically improve the ease of wireless sensor network programming by allowing programmers to express high-level objectives, and leave the low-level details to the compiler and run-time system. The goal is also to enable easy integration with other systems, such as business systems, mainly via the usage of business process modeling.

makeSense aims at lowering the barrier for adoption of WSNs in real-world applications, notably business processes. In making this vision a reality, the makeSense approach must be sufficiently expressive to specify a range of different business applications, and be easy to use with no need for specialized training for domain-experts.

Today it is notoriously difficult to realize WSN applications. Current programming platforms are lowlevel, typically at the operating system level. The programmer therefore must concentrate on details quite far from the application logic, such as communication protocols, or power management. The programming task is particularly hard in WSNs because of the inherent antagonism between the desired application requirements and the constraints of WSNs: requirements such as high reliability or long lifetime are difficult to realize in the highly dynamic, unreliable, and resource-constrained setting characterizing WSNs. This situation generates a lack of visibility: it is hard to investigate the performance or correctness of WSN processes, since they are typically hard-coded from scratch every time. The research community has tackled these problems, but has not yet solved two fundamental issues: the problems of unification and integration, schematically shown in the figure below

makeSense solves the two main problems in integration of WSNs and business processes: the back-end
must be integrated with the WSNs and a comprehensive programming framework is needed.