PhyForm - A cloud SDR framework for security research supporting machine learning of wireless IoT signal data sets - Research Portal

Computing and Communications

PhyForm - A cloud SDR framework for security research supporting machine learning of wireless IoT signal data sets

Research output: Contribution to conference - Without ISBN/ISSN › Poster › peer-review

Published

Antony Chung

Publication date	2020
<mark>Original language</mark>	English
Event	Seventeenth International Conference on Embedded Wireless Systems and Networks (EWSN 2020) - Lyon, France Duration: 17/02/2020 → 19/02/2020 Conference number: 17 https://ewsn2020.conf.citi-lab.fr/

Conference

Conference	Seventeenth International Conference on Embedded Wireless Systems and Networks (EWSN 2020)
Abbreviated title	EWSN
Country/Territory	France
City	Lyon
Period	17/02/20 → 19/02/20
Internet address	https://ewsn2020.conf.citi-lab.fr/

Abstract

Software defined radio (SDR) enables the use of digital signal processing (DSP) to identify IoT security issues based on waveform analysis. Such research requires the handling, processing and interaction with large data sets of digitised RF. Those supporting activities are a high overhead.

An extensible framework is introduced for the curation, filtering, pre-processing, and analysis tasks associated with RF data sets in machine learning and IoT research. It provides a web interface, API, SigMF data sharing and integration with GNU Radio. The aim is improved data set and algorithm collaboration. A LoRa example provides context.

Research

Links