A survey of middleware for sensor networks: state-of-the-art and future directions

In future computing environments, networked sensors will play an increasingly important role in mediating between the physical and virtual worlds. However, programming sensor networks, and the applications that depend on the data they produce, is extremely challenging. The need for suitable middleware to address this problem is evident. In the last few years, various middleware solutions for sensor networks have emerged. These differ in terms of their models for querying and data aggregation, and their assumptions about the topology and other characteristics of the network. Naturally, the assumptions made for each particular middleware limit its potential applicability. Most of the current solutions provide relatively simple query abstractions, and therefore are not suitable for applications that have sophisticated requirements for processing of sensor data in the network. This paper presents a survey and analysis of the current state-of-the art in the field, highlighting the open research challenges. It also draws on the authors' experience with developing middleware for context-aware systems - that is, systems that rely on sensor-derived data to intelligently adapt their behaviour - to propose some future directions for the development of middleware for sensor networks.