Design, Implementation and Optimisation of a ZigBee Wireless Sensor


Eryk Dutkiewicz.

Student: Nandika Thapar.
Introduction: ZigBee is a promising new international standard being promoted by the ZigBee Alliance. It is designed for home automation and industrial control networks, among other applications.
Significance: This evolving standard is aimed to be a low power, low cost wireless solution that can support many different applications that consist of unsupervised devices, in an interoperable and scalable way. This thesis project aims to investigate these claims with a focus on ZigBees' application in industrial control. For Wireless Sensor Networks (WSN), energy conservation, self con_guration and reliability are the primary considerations while designing a protocol.
Applications: This thesis lays grounds for future development and testing of ZigBee WSN with the latest version of Freescale's Beekit GUI.
Challenges: In this project, channel access and routing protocols are investigated to identify the design challenges and optimisation parameters for ZigBee WSN. The target application for the ZigBee network is temperature sensing in the steel manufacturing process at the BlueScope Steel plant at Port Kembla, NSW, Australia. The requirements of this application, with a focus on the BlueScope environment, have been investigated. This thesis is also a testbed of Freescale Semiconductors ZigBee development kits. Initial assessment of Freescale's development kit capabilities has been made by implementing and evaluating a test ZigBee network. A prevailing mathematical model was corrected and then modified to achieve an optimal network configuration that is power efficient, while guaranteeing reliable sensor data delivery to the control centre. This optimisation was realised within the constraints of the ZigBee Specifications and the BlueScope Steel environment and requirements. Comparisons have been drawn between the Freescale and Chipcon proprietary hardware to establish which device gives more efficient performance in terms of power consumption under similar network conditions. The battery lifetime of Freescale devices has also been estimated using the modified mathematical model. This has been compared with the battery life estimate drawn from the data sheet values and also through practical measurements.
top of page