Corona: a Sun SPOT Distributed Query Processor

Wireless Sensor Networks (WSNs) are an emerging technology for automatic and continuous monitoring of physical phenomena in a way that has not been possible before. Some of the many applications include habitat monitoring, structural monitoring, and traffic control. The Sun(TM) Small Programmable Object Technology (SPOT) is a new sensor network hardware with full Java support, that has been recently (early 2007) introduced by Sun Microsystems(TM). It is a state-of-the-art WSN platform that provides much more computational power and memory than the previous generations of very limited sensor nodes. These recent hardware developments make complex distributed data processing tasks feasible which go beyond the previous 'sensing - collecting - centrally processing' approaches.

SSDQP GUI Snapshot As part of the WSN Initiative of our School, we are interested in developing novel adaptive methods for true in-network data processing within a WSN. As research platform, we are currently developing our own prototype system for in-network querying and data processing for wireless sensor networks, called the Sun SPOT Distributed Query Processor (SSDQP). We have currently about 20 Sun SPOT devices available here in the School of IT for our research in wireless sensor networks. Our SSDQP system consists of two components: The distributed query engine that is executed directly on the Sun SPOT devices, and the control system on the user's workstation that is connected to the WSN base station (see GUI snapshot on the right).

SSDQP is built around the following core features:

  • Multi-Tasking
    The whole functionality of the query engine is implemented as a set of tasks that are executed by a scheduler running locally on each node. Tasks can be periodically executed with a fixed time period and a certain lifetime. In particular, SSDQP can schedule several query tasks in parallel on each node.
  • Time Synchronization
    The query engine is time triggered, i.e. each task has a specific start time when it starts executing. The query engine has a specific time synchronisation module that is responsible for establishing a global time in the ad-hoc network.
  • In-network Querying Processing
    The query engine supports all the fundamental query operators such as sensing, selection, projection, join, and in-network aggregation. Users can enter declarative queries at the control system using an SQL-like query language. Queries are translated into a physical query plan which is then distributed and executed in the WSN.
  • Optimized Data Communication
    To minimise the size of messages (and therefore the energy consumption), a basic data compression method is used for encoding the query plans and results send in the network. The data compression achieves compression rates of about 62% in practice.
  • Resource-Awareness
    We have developed a resource monitoring framework that allows to monitor and query the internal resource consumption (battery, memory, CPU) at each node. We are currently integrating this framework into our SSDQP system so that node resource levels can be queried centrally and used to adapt in-network data processing.
  • In-network Data Stream Processing
    We have integrated an adaptive online clustering algorithm within SSDQP. The primary goal of this research is to enable resource-awareness for data processing in wireless sensor networks by using resource monitoring components for the online Cluster algorithm.
  • In-network Event Detection
    This is work in progress.
  • Graphical User Interface
    The control system on the workstation features a graphical user interface to enter and manage queries in the system and to visualise results. It also provides a client/server interface so that several client computers can connect and interact with the same WSN infrastructure.

Sun SPOT We are always looking for interested students to extend and improve our current system.
Currently, we are working on the following topics:

For further information see also our dedicated WSN @ USYD website.

The SSDQP Team

This project would not have been possible without the impressive contributions of our many project students:
Tim Dawborn, Raymes Khoury, Edmund Tse, Quincy Tse, Duc Nhan Phung, Karel Herink, and Yu Hsiang 'Dion' Tsai.

Academic Staff: Dr. Uwe Roehm, Dr. Bernhard Scholz, Dr. Selvakennedy Selvadurai, and Dr. Mohamed M. Gaber


This work is funded by Sun Microsystems and the Australian Research Council as part of the ARC Discovery project DP0664782.


Bernhard Scholz, Mohamed M. Gaber, Tim Dawborn, Raymes Khoury, and Edmund Tse. Efficient time triggered query processing in wireless sensor networks. In: Proceedings of the 2007 International Conference on Embedded Systems and Software (ICESS-07), Daegu, Korea, May 14-16. Springer Press, 2007.

Nhan Duc Phung, Mohamed M. Gaber, and Uwe Röhm. Resource-aware Online Data Mining in Wireless Sensor Networks. In: Proceedings of the IEEE Symposium on Computational Intelligence and Data Mining (CIDM 2007), April 1-5, Honolulu, HI, USA, 2007.

S. Selvakennedy, S. Sinnappan, and Yi Shang. A Biologically-Inspired Clustering Protocol for Wireless Sensor Networks. Elsevier Computer Communications: SI on Network Coverage and Routing Schemes for Wireless Sensor Networks, September 2006.

Uwe Röhm, Bernhard Scholz, and Mohamed M. Gaber. Integration of Data Stream Clustering into a Query Processor for Wireless Sensor Networks. In: Proceedings of the International Workshop on Data Intensive Sensor Networks (DISN'07), in conjunction with MDM'07, May 7-11, Mannheim, Germany, 2007.

Selvakennedy Selvadurai, Uwe Röhm, and Bernhard Scholz. Event Processing Middleware for Wireless Sensor Networks. In: Proceedings of the First International Workshop on Ubiquitous Computing for Parallel and Distributed Systems (uPADS07), in conjunction with ICPP 2007, September 10-14, Xian China, 2007.

Nhan Duc Phung, Mohamed M. Gaber, and Uwe Röhm. Resource-aware Distributed Online Data Mining for Wireless Sensor Networks. In: Proceedings of the International Workshop on Knowledge Discovery from Ubiquitous Data Streams (IWKDUDS07), in conjunction with ECML and PKDD 2007, September 17, Warsaw, Poland, 2007.

In the pipe:

Uwe Röhm, Mohamed M. Gaber, and Quincy Tse. "Enabling Resource-Awareness for In-network Data Processing in Wireless Sensor Networks". 2007.