1 / 5

Optimizing QR Decomposition for Resource-Constrained Wireless Sensor Networks

This demonstration showcases a distributed computing approach to QR decomposition within wireless sensor networks (WSNs), emphasizing algorithms designed for energy and resource-constrained embedded platforms. The proposed methodology blends Householder reflections, Givens rotations, and an AllReduce technique to develop an efficient distributed QR algorithm. Achieving a balance between parallelism and communication costs, our approach delivers a significant 2x speedup and energy savings per node compared to traditional centralized methods. This research contributes valuable insights for enhancing adaptive filtering and beamforming in WSNs.

afi
Télécharger la présentation

Optimizing QR Decomposition for Resource-Constrained Wireless Sensor Networks

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. QR Decomposition: Demonstration of Distributed Computing on Wireless Sensor Networks CACS, University of Louisiana at Lafayette Lafayette, LA 70504 {spa9242, sxg5317, rxc2763, mab} @ cacs.louisiana.edu By Sherine Abdelhak, Soumik Ghosh, Rabi Chaudhuri, Magdy Bayoumi High Performance Embedded Computing (HPEC) Workshop 22−23 September 2009 (A) Approved for public release; distribution is unlimited.

  2. Scope • Kernel for distributed numerical algorithms on Energy and Resource-constrained Wireless Embedded Platforms • Wireless Sensor Network is an example of a severely resource constrained platform QR in WSN • QR-based least mean square (LMS) • QR-based recursive least mean square (RLS) • RLS and LMS in adaptive filtering and beamforming Proposed QR Blend of Householder reflections, Givens rotations, and AllReduce Introduction

  3. Develop a new distributed QR algorithm • Achieve a balanced tradeoff between degree of parallelism and communication cost • Propose a resource-aware distribution • Achieve 2x speedup and 2x energy savings per node vs. centralized (single node) approach Contributions Introduction

  4. Givens Phase Proposed Algorithm Introduction Proposed Scheme

  5. Proposed Distribution on WSN Introduction Proposed Scheme

More Related