Particle Filters for Robust Estimation from Physiological Sensors in the Swarm
Viswam Nathan

Citation
Viswam Nathan. "Particle Filters for Robust Estimation from Physiological Sensors in the Swarm". Tutorial, 12, May, 2016.

Abstract
With the increased adoption of wearable sensors, we expect personal physiological monitoring to be an important application of the future. Pervasive and continuous monitoring of vitals offers several advantages over traditional, sporadic clinical visits. However, wearable sensors are typically much noisier than their medical grade counterparts. This talk explains how the particle filter, a probabilistic state estimation technique, can be applied to this problem to robustly estimate underlying physiological parameters in noisy environments. This will include real experimental results on heart rate monitoring under the condition of motion artifacts. Furthermore, we will explore the possibility of using the particle filter to effectively fuse information from multiple simultaneous sensors.

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

  Viswam Nathan. <a
  href="http://www.terraswarm.org/pubs/765.html"
  ><i>Particle Filters for Robust Estimation from
  Physiological Sensors in the Swarm</i></a>,
  Tutorial,  12, May, 2016.

• Plain text

  Viswam Nathan. "Particle Filters for Robust Estimation
  from Physiological Sensors in the Swarm". Tutorial, 
  12, May, 2016.

• BibTeX

  @tutorial{Nathan16_ParticleFiltersForRobustEstimationFromPhysiologicalSensors,
      author = {Viswam Nathan},
      title = {Particle Filters for Robust Estimation from
                Physiological Sensors in the Swarm},
      day = {12},
      month = {May},
      year = {2016},
      abstract = {With the increased adoption of wearable sensors,
                we expect personal physiological monitoring to be
                an important application of the future. Pervasive
                and continuous monitoring of vitals offers several
                advantages over traditional, sporadic clinical
                visits. However, wearable sensors are typically
                much noisier than their medical grade
                counterparts. This talk explains how the particle
                filter, a probabilistic state estimation
                technique, can be applied to this problem to
                robustly estimate underlying physiological
                parameters in noisy environments. This will
                include real experimental results on heart rate
                monitoring under the condition of motion
                artifacts. Furthermore, we will explore the
                possibility of using the particle filter to
                effectively fuse information from multiple
                simultaneous sensors.},
      URL = {http://terraswarm.org/pubs/765.html}
  }
  

Posted by Elizabeth Coyne on 17 Mar 2016.
Groups: services