Lakon: a middle-ground approach to high-frequency data acquisition and in-network processing in sensor networks

  • Authors:

  • Prashanth G. Reddy;Nigamanth Sridhar

  • Affiliations:

  • Cleveland State University, Cleveland, OH;Cleveland State University, Cleveland, OH

  • Venue:
  • Proceedings of the 9th ACM/IEEE International Conference on Information Processing in Sensor Networks
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

The need for high-frequency signal acquisition and processing is becoming increasingly prevalent in sensor networks. Applications that require high-frequency data sampling are presently at a disadvantage; applications that only sample at high data rates (and not process any of it locally) end up transmitting large quantities of data, greatly reducing network lifetime. Other applications that do use in-network signal processing rely on power-hungry motes. We present Lakon, a mote architecture capable of onboard signal processing of high-frequency data that provides a middle ground for more general classes of applications that require signal processing. Our design takes advantage of an energy-efficient on-board digital signal processor (DSP) that can be intelligently enabled on demand. Our contribution here is threefold. First, we present a general mote architecture that is more appropriate for applications such as body sensor networks and habitat monitoring. Second, we present a switching scheme for processor scheduling that determines the co-processor's usage. Finally, we demonstrate the use and potential of Lakon in the context of a text-independent speaker recognition system that takes an audio signal as input and performs classification on that signal to identify the owner of the voice.