Reception region characterisation using a 2.4GHz direct sequence spread spectrum radio

  • Authors:
  • Damien O'Rourke;Szymon Fedor;Conor Brennan;Martin Collier

  • Affiliations:
  • Dublin City University, Dublin, Ireland;Dublin City University, Dublin, Ireland;Dublin City University, Dublin, Ireland;Dublin City University, Dublin, Ireland

  • Venue:
  • Proceedings of the 4th workshop on Embedded networked sensors
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Experimental studies on Wireless Sensor Networks (WSNs) have revealed the existence of three distinct reception regions. These regions can be classified as connected, transitional, and disconnected. Their location and size may have a significant impact on the performance of communication protocols. The underlying causes for the transitional (or so-called "grey") region have been determined in previous work using both analytical and empirical techniques. Conclusions drawn from these experiments suggest that the transitional region is caused in part by multi-path fading. A natural assumption therefore is that any system used to combat multipath effects will also reduce the extent of the transitional region. Measurements from previous works were mainly carried out on radio architectures using (for example) NRZ and noncoherent FSK and therefore do not validate this assumption. This work shows that the above hypothesis is invalid due to a smaller decay factor and increased standard deviation of the shadowing component for high mulit-path environments. Along with the usual Packet Reception Rate (PRR) tests, a BER test was also undertaken to help identify and characterise the transitional region. A side result of this is that the measured BER shows some deviation from the theoretical value given in the IEEE 802.15.4 standard and provides evidence why RSSI may not be a good link quality estimator in certain circumstances. In addition a number of practical issues encountered and resolved are reported in the paper.