Radiometric identification based on maximum penalty graph learning

In this paper, we propose a novel fingerprint learning algorithm based on maximum penalty graph criteria for radiometric identification. The transient signal and preamble signal of emitters are used for fingerprint feature extraction. The proposed algorithm is based on characterizing the interclass...

Full description

Saved in:
Bibliographic Details
Main Authors: Jia, Yongqiang, Gan, Lu
Format: Conference Proceeding
Language:English
Subjects:
Detectors
Feature extraction
fingerprint learning
Fingerprint recognition
graph embedding
radio emitter identification
Radio frequency
Radiometry
Security
Transient analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we propose a novel fingerprint learning algorithm based on maximum penalty graph criteria for radiometric identification. The transient signal and preamble signal of emitters are used for fingerprint feature extraction. The proposed algorithm is based on characterizing the interclass separability and discriminative features. It extracts the synthesized output of all the factors that contribute to the minute differences between different emitters and can competently discriminate the emitters from different manufacturers. Our algorithm is superior to other competitive radio identification methods. Experiments on real data sets show the effectiveness of our algorithm for radio emitter identification tasks.
ISSN:2164-5221
DOI:10.1109/ICSP.2016.7878032