An Integrated Signal Processing Architecture for Human Activity Recognition
Jingxiang Tian, Carl Sechen, Roozbeh Jafari

Citation
Jingxiang Tian, Carl Sechen, Roozbeh Jafari. "An Integrated Signal Processing Architecture for Human Activity Recognition". Talk or presentation, 5, November, 2013; Poster presented at the 2013 TerraSwarm Annual Meeting.

Abstract
We are developing a low-power, highly reliable wearable computer for real-time human activity recognition based on sub-threshold signal processing which utilizes tri-axial accelerometer data, the dynamic time warping algorithm, and a granular decision making architecture.

• Granular decision making architecture
• Parallel processing with multiple templates
• Priority based energy-aware decision path New maximal reliability 11T sub-threshold 45nm SRAM

Electronic downloads

Internal. This publication has been marked by the author for TerraSwarm-only distribution, so electronic downloads are not available without logging in.

Citation formats

• HTML

  Jingxiang Tian, Carl Sechen, Roozbeh Jafari. <a
  href="http://www.terraswarm.org/pubs/167.html"
  ><i>An Integrated Signal Processing Architecture
  for Human Activity Recognition</i></a>, Talk or
  presentation,  5, November, 2013; Poster presented at the
  <a 
  href="http://www.terraswarm.org/conferences/13/annual">2013
  TerraSwarm 
  Annual Meeting</a>.

• Plain text

  Jingxiang Tian, Carl Sechen, Roozbeh Jafari. "An
  Integrated Signal Processing Architecture for Human Activity
  Recognition". Talk or presentation,  5, November, 2013;
  Poster presented at the <a 
  href="http://www.terraswarm.org/conferences/13/annual">2013
  TerraSwarm 
  Annual Meeting</a>.

• BibTeX

  @presentation{TianSechenJafari13_IntegratedSignalProcessingArchitectureForHumanActivity,
      author = {Jingxiang Tian and Carl Sechen and Roozbeh Jafari},
      title = {An Integrated Signal Processing Architecture for
                Human Activity Recognition},
      day = {5},
      month = {November},
      year = {2013},
      note = {Poster presented at the <a 
                href="http://www.terraswarm.org/conferences/13/annual">2013
                TerraSwarm 
  Annual Meeting</a>.},
      abstract = {We are developing a low-power, highly reliable
                wearable computer for real-time human activity
                recognition based on sub-threshold signal
                processing which utilizes tri-axial accelerometer
                data, the dynamic time warping algorithm, and a
                granular decision making architecture. <ul>
                <li>Granular decision making architecture
                <li>Parallel processing with multiple templates
                <li>Priority based energy-aware decision path
                <li:New sub-threshold, low-overhead (10%)
                asynchronous logic <li>New maximal reliability 11T
                sub-threshold 45nm SRAM </ul> },
      URL = {http://terraswarm.org/pubs/167.html}
  }
  

Posted by Jingxiang Tian on 3 Nov 2013.
Groups: services