Accession Number:

ADA604052

Title:

A Comparison of RF-DNA Fingerprinting Using High/Low Value Receivers with ZigBee Devices

Descriptive Note:

Master's thesis

Corporate Author:

AIR FORCE INSTITUTE OF TECHNOLOGY WRIGHT-PATTERSON AFB OH GRADUATE SCHOOL OF ENGINEERING AND MANAGEMENT

Personal Author(s):

Report Date:

2014-03-27

Pagination or Media Count:

79.0

Abstract:

The ZigBee specification provides a niche capability, extending the IEEE 802.15.4 standard to provide a wireless mesh network solution. ZigBee-based devices require minimal power and provide a relatively long-distance, inexpensive, and secure means of networking. The technology is heavily utilized, providing energy management, ICS automation, and remote monitoring of Critical Infrastructure CI operations it also supports application in military and civilian health care sectors. ZigBee networks lack security below the Network layer of the OSI model, leaving them vulnerable to open-source hacking tools that allow malicous attacks such as MAC spoofing or Denial of Service DOS. A method known as RF-DNA Fingerprinting provides an additional level of security at the Physical PHY level, where the transmitted waveform of a device is examined, rather than its bit-level credentials which can be easily manipulated. RF-DNA fingerprinting allows a unique human-like signature for a device to be obtained and a subsequent decision made whether to grant access or deny entry to a secure network. Two NI receivers were used here to simultaneously collect RF emissions from six Atmel AT86RF230 transceivers. The time-domain response of each device was used to extract features and generate unique RF-DNA fingerprints. These fingeprints were used to perform Device Classification using two discrimination processes known as MDAML and GRLVQI. Each process classifier was used to examine both the Full-Dimensional FD and reduced dimensional feature-sets for the high-value PXIe and low-value USRP receivers. The reduced feature-sets were determined using DRA for both quantitative and qualitative subsets. Additionally, each classifier performed Device Classification using a hybrid interleaved set of fingerprints from both receivers.

Subject Categories:

  • Biochemistry
  • Radio Communications

Distribution Statement:

APPROVED FOR PUBLIC RELEASE