Rapid and efficient application design using a signal processing framework for WSN

Abstract

Advances in sensor technology, wireless mesh networking and embedded processors are pushing development of new technologies in Wireless Sensor Networks (WSN). The wide variety of potential applications and platforms makes hard to port and develop new applications to different platforms keeping smart and efficient behavior of the nodes of WSN. For this reason nowadays middlewares are emerging as a valid alternative to the resource-hungry Operating Systems for WSN. In particular, frameworks are valid tools to provide hardware abstraction by creating an intermediate layer that improves interoperability and reduces the development time. Frameworks make available a set of libraries, utilities and functions used to build up distributed systems and algorithms where energy and computational limitations of the individual sensors can be overcome. In this paper we present two interesting real case applications using SPINE2, a framework for signal processing in node environment, ported on Ember EM250 ZigBee platform. We show how to obtain a general behavior on the system that is not achievable without using a suitable framework. © 2010 IEEE.