ISSNIP

BigNet Testbed

The development of a world class national research and industry capability in sensor network technology is of great national importance. Although Wireless Sensor Networks (WSNs) are being studied on a global scale, current research is still focusing on simulations and small-scale experiments. In particular for sensor networks, which have to deal with very stringent resource limitations and that are exposed to severe physical conditions, real experiments with real applications are essential. The effectiveness of simulation studies is severely limited in terms of the lack of scalability of network simulators for large networks, as well as the difficulty in modelling the complexities of the radio environment, power consumption on sensors, and the interactions between the physical, network and application layers. Therefore, we propose to face the real world, extending the simulations and laboratory experiments to large scale, real applications. Only by doing so will WSN research: (1) get an insight into the real needs of the users of sensor network; (2) face the world of errors, incomplete information, dynamics, etc. that is very hard to model; (3) perform large-scale experiments of several hundreds of sensor nodes. In the BigNet project, we intend to use this experimental facility as a testbed to demonstrate the algorithms developed on a large scale, facing real world problems, and developing solutions for a variety of applications. This is the first of its kind in Australia and the first truly open, vendor independent large scale testbed in the world. In a National Science Foundation (NSF) Workshop on Sensor Network Testbeds, the importance of Testbeds is clearly argued, further strengthening our case for the testbed. Sensor networks provide the ability to gather information cheaply, accurately and reliably over both small and vast physical regions. Unlike other large computer network forms, where the ultimate I/O interface is a human being, WSNs are about collecting data from the physical. The applications context therefore has significant bearing on the ultimate realisation of WSNs and therefore an effective testbed. The testbed being created needs to mirror such practical requirements. It needs to be large enough to be partitioned into meaningful sub-networks (allowing testing of sub-networks with adjacent nodes).

 

ARC LIEF BigNet – Sensor Network Testbed

ISSNIP CI’s have been successful in obtaining ARC LIEF funding for BigNet - A Distributed Wireless Sensor Network Testbed. The testbed is to be configured into subnets distributed between The University of Melbourne (Electrical and Electronic Engineering; Computer Science and Software Engineering; Geomatics Engineering; and Civil and Environmental Engineering), James Cook University and Deakin University. BigNet consolidates the national/international research and industry collaborations of the ARC Research Network on Intelligent Sensors, Sensor Networks and Information Processing. The infrastructure developed will be of national/international significance, given the rapid emergence of wireless sensor networks. This integrated facility will ensure that Australia is a world leading player in the research and technology
development as well as the socially responsible deployment of sensor networks. The facility has the explicit aim to ensure that Australia is a technology leader rather than mere technology user in sensor networks. The test facility will mirror practical requirements for WSN implementation in the Great Barrier Reef and in a timber plantation, which would offer substantial economic benefits to Australia. Although Wireless Sensor Networks (WSNs) are being studied on a global scale, current research is still largely based on simulations and small-scale experiments. In particular for sensor networks, that have to deal with very stringent resource limitations and that are exposed to severe physical conditions, real experiments with real applications are essential. The effectiveness of simulation studies is severely limited in terms of the lack of scalability of network simulators for large networks, as well as the difficulty in modelling the complexities of the radio environment, power consumption on
sensors, and the interactions between the physical, network and application layers. This infrastructure extends the simulations and laboratory experiments to large scale, real applications. Through the BigNet project, this experimental facility will demonstrate the algorithms developed on a large scale, facing real world problems and developing solutions for a variety of applications.

Testbed Deployment Projects:

  •  Generic Sensor Network Testbed for Anomaly Detection
  •  Wireless marine sensor networks
  •  WSN Security
  •  Wireless Multimedia Sensor Network – A case study
  •  Wireless Body Area Networks
  •  Wireless Sensing and Monitoring for Aged Healthcare
  •  Spatial Computation in WSNs
  •  Data-centric routing and collection
  •  Web and grid enablement
  •  Laboratory for Integrated sensing and networking for real time data acquisition and visualization (Deakin)
  •  High Data Rate Marine Information Stream
Equipment (Melbourne node):
  •  75 Sun Microsystems SunSPOT motes
  •  210 Crossbow motes (Imote2/IRIS/MICA2/MICAz/TelosB)
  •  Crossbow development kits (Imote2/IRIS/MICA2/MICAz)
  •  Bluetooth development kits
  •  100 Sensors/sensor boards
    •  Light
    •  Temperature
    •  Humidity
    •  Soil moisture
    •  Soil temperature
    •  Multi-media (audio/video)
    •  Sonar
    •  Inertia
    •  Blood pressure
    •  ECG/EEG

 

BigNET integrated tracking and sensing

Networked Sensing and Control Laboratory at Deakin University has developed the hardware and necessary software to integrate sensory information from diverse sensing mechanisms including static to mobile sensors. This video demonstrates the wireless sensor network testbed developed by the group.

http://www.deakin.edu.au/scitech/eng/nsc/BigNET_Video.php

Previous | Next 1
top of page