System on Fabrics Architecture using Distributed Computing.pdf (3.74 MB)
System on fabrics architecture using distributed computing
journal contribution
posted on 2018-06-12, 10:18 authored by Partheepan Kandaswamy, James FlintJames Flint, Vassilios ChouliarasThis paper describes a novel, distributed sensor
network with parallel processing capability based on the Instruction Systolic Array (ISA). A new computing paradigm is introduced where spatially distributed sensors are closely coupled to processing elements and the whole array forms a parallel computer. This may find applications in wearable devices for sensing the position and other metrics of a human body and
rapidly processing that data. A new programming model to implement the distributed computer on fabrics is described. The fabric-based distributed computing concept has been validated using a number of parallel applications including a real-time shape sensing and reconstruction application. The exemplar wearable system based on the ISA concept has been realized using
off-the-shelf microcontrollers and sensors. Results show that the application executes on the prototype ISA implementation in real time thus confirming the viability of the proposed architecture for fabric-resident computing devices.
History
School
- Mechanical, Electrical and Manufacturing Engineering
Published in
IEEE Sensors JournalCitation
KANDASWAMY, P., FLINT, J.A. and CHOULIARAS, V.A., 2018. System on fabrics architecture using distributed computing. IEEE Sensors Journal, 18(14), pp.5929-5936.Publisher
Institute of Electrical and Electronics EngineersVersion
- AM (Accepted Manuscript)
Acceptance date
2018-05-02Publication date
2018Notes
© 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.ISSN
1530-437XPublisher version
Language
- en