Practical Resource-Aware Optimization of Cascade Detection Systems with Multiple Applications
Long Le, Douglas L. Jones

Citation
Long Le, Douglas L. Jones. "Practical Resource-Aware Optimization of Cascade Detection Systems with Multiple Applications". Talk or presentation, 26, October, 2016; Poster presented at the 2016 TerraSwarm Annual Meeting.

Abstract
Traditional distributed detection systems are often designed for a single target application. However, with the emergence of the Internet of Things (IoT) paradigm, next-generation systems are expected to be a shared infrastructure for multiple applications. To achieve this goal, a detection system needs to be modularly (re-)designed. To this end, we propose a modular, cascade design for resource-efficient, multi-task detection systems. Two (classes of) applications are considered in the system, a primary and a secondary one. The primary application has universal features that can be shared with other applications, to reduce the overall feature extraction cost, while the secondary application does not. In this setting, the two applications can collaborate via feature sharing. We provide a method to optimize the operation of the multi-application cascade system based on a practical resource consumption model. Our system simulation shows that by using the optimal feature sharing strategy, a system can achieve 9$\times$ resource saving and 2$\times$ improvement in detection performance.

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

  Long Le, Douglas L. Jones. <a
  href="http://www.terraswarm.org/pubs/847.html"><i>Practical
  Resource-Aware Optimization of Cascade Detection Systems
  with Multiple Applications</i></a>, Talk or
  presentation,  26, October, 2016; Poster presented at the
  <a
  href="http://terraswarm.org/conferences/16/annual"
  >2016 TerraSwarm Annual Meeting</a>.

• Plain text

  Long Le, Douglas L. Jones. "Practical Resource-Aware
  Optimization of Cascade Detection Systems with Multiple
  Applications". Talk or presentation,  26, October,
  2016; Poster presented at the <a
  href="http://terraswarm.org/conferences/16/annual"
  >2016 TerraSwarm Annual Meeting</a>.

• BibTeX

  @presentation{LeJones16_PracticalResourceAwareOptimizationOfCascadeDetection,
      author = {Long Le and Douglas L. Jones},
      title = {Practical Resource-Aware Optimization of Cascade
                Detection Systems with Multiple Applications},
      day = {26},
      month = {October},
      year = {2016},
      note = {Poster presented at the <a
                href="http://terraswarm.org/conferences/16/annual"
                >2016 TerraSwarm Annual Meeting</a>.},
      abstract = {Traditional distributed detection systems are
                often designed for a single target application.
                However, with the emergence of the Internet of
                Things (IoT) paradigm, next-generation systems are
                expected to be a shared infrastructure for
                multiple applications. To achieve this goal, a
                detection system needs to be modularly
                (re-)designed. To this end, we propose a modular,
                cascade design for resource-efficient, multi-task
                detection systems. Two (classes of) applications
                are considered in the system, a primary and a
                secondary one. The primary application has
                universal features that can be shared with other
                applications, to reduce the overall feature
                extraction cost, while the secondary application
                does not. In this setting, the two applications
                can collaborate via feature sharing. We provide a
                method to optimize the operation of the
                multi-application cascade system based on a
                practical resource consumption model. Our system
                simulation shows that by using the optimal feature
                sharing strategy, a system can achieve 9$\times$
                resource saving and 2$\times$ improvement in
                detection performance.},
      URL = {http://terraswarm.org/pubs/847.html}
  }
  

Posted by Long Le on 24 Oct 2016.
Groups: pw