Contrastive Self-supervised Clustering for Specific Emitter Identification

Specific Emitter Identification (SEI) is crucial for attacking and defending Internet of Things (IoT) devices in untrusted scenarios or battlefield environments. However, existing SEI methods usually require annotation information, which is often unavailable in non-cooperative communications and unt...

Full description

Saved in:
Bibliographic Details
Published in:IEEE internet of things journal 2023-12, Vol.10 (23), p.1-1
Main Authors: Hao, Xiaoyang, Feng, Zhixi, Liu, Ruoyu, Yang, Shuyuan, Jiao, Licheng, Luo, Rong
Format: Article
Language:English
Subjects:
actual burst emitter
Algorithms
Annotations
Clustering
Data augmentation
Emitters
Feature extraction
Fingerprint recognition
Generative adversarial networks
Internet of Things
Long short term memory
Machine learning
Packets (communication)
Self-supervised learning
signal clustering
signal contrastive self-supervised clustering (SCSC)
signal data augmentation
Specific emitter identification (SEI)
Support vector machines
Task analysis
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Specific Emitter Identification (SEI) is crucial for attacking and defending Internet of Things (IoT) devices in untrusted scenarios or battlefield environments. However, existing SEI methods usually require annotation information, which is often unavailable in non-cooperative communications and untrusted scenarios. In this paper, we propose a signal contrastive self-supervised clustering (SCSC) method for unsupervised SEI applications. First, we propose SCSC with 1D fingerprint pyramid feature extractor (1D-FPFE) for obtaining hierarchical subtle features of emitter signals. Then, we propose a bit-pulse selection (BPS) strategy and several signal data augmentation methods. By constructing signal positive and negative instance pairs through data augmentation, our approach generates cluster preference representations in a contrastive self-supervised learning manner. Extensive experimental results based on communication burst emitter dataset show that SCSC achieves an accuracy improvement of about 26% over the current best communication signal clustering algorithm. Moreover, SCSC also exhibits good performance and generalization for 30 emitter clustering and few-shot unlabeled signal clustering.
ISSN:2327-4662
2327-4662
DOI:10.1109/JIOT.2023.3284428