Trimming the tree: tailoring adaptive huffman coding to wireless sensor networks

  • Authors:

  • Andreas Reinhardt;Delphine Christin;Matthias Hollick;Johannes Schmitt;Parag S. Mogre;Ralf Steinmetz

  • Affiliations:

  • Multimedia Communications Lab, Technische Universität Darmstadt, Darmstadt, Germany;Secure Mobile Networking Lab, Technische Universität Darmstadt, Darmstadt, Germany;Secure Mobile Networking Lab, Technische Universität Darmstadt, Darmstadt, Germany;Multimedia Communications Lab, Technische Universität Darmstadt, Darmstadt, Germany;Multimedia Communications Lab, Technische Universität Darmstadt, Darmstadt, Germany;Multimedia Communications Lab, Technische Universität Darmstadt, Darmstadt, Germany

  • Venue:
  • EWSN'10 Proceedings of the 7th European conference on Wireless Sensor Networks
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Nodes in wireless sensor networks are generally designed to operate on a limited energy budget, and must consciously use the available charge to allow for long lifetimes. As the radio transceiver is the predominant power consumer on current node platforms, the minimization of its activity periods and efficient use of the radio channel are major targets for optimization. Data compression is a viable option to increase the packet information density, resulting in reduced transmission durations and thus allowing for an optimized channel utilization. The computational and memory demands of many current compression algorithms however hamper their applicability on sensor nodes. In this paper, we present a novel variant of the adaptive Huffman coding algorithm, operating on reduced code table sizes and thus significantly alleviating the resource demands for storing and updating the code table during runtime. An implementation for tmote sky hardware proves its adequacy to the capabilities of sensor nodes, and we present its achievable compression gains and energy requirements in both simulation and real world experiments. Results anticipate that overall energy savings can be achieved when transferring packets of reduced sizes, even when increased CPU utilization is incurred.