Radar Emitter Identification Based on Novel Time-Frequency Spectrum and Convolutional Neural Network

Radar emitter identification (REI) is significant in both military and civilian application domains. A critical step for REI is signal feature extraction. Most radar emitter signals are non-stationary, and many studies apply time-frequency spectrum features for non-stationary signal analysis in rece...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE communications letters Jg. 25; H. 8; S. 2634 - 2638
Hauptverfasser: Xiao, Zhiling, Yan, Zhenya
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York IEEE 01.08.2021
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
<italic xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">k -means algorithm
Algorithms
Artificial neural networks
convolutional neural network
Emitters
Feature extraction
Fourier transforms
Frequency shift keying
Frequency spectrum
Gray-scale
Military applications
Neural networks
Performance evaluation
Radar
Radar emitter identification
Signal analysis
Signal to noise ratio
Spectrogram
Spectrograms
Spectrum analysis
Time-frequency analysis
ISSN:1089-7798, 1558-2558
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Radar emitter identification (REI) is significant in both military and civilian application domains. A critical step for REI is signal feature extraction. Most radar emitter signals are non-stationary, and many studies apply time-frequency spectrum features for non-stationary signal analysis in recent years. This letter proposes a novel spectrum calculation method for signal feature analysis using short-time Fourier transform (STFT) and <inline-formula> <tex-math notation="LaTeX">k </tex-math></inline-formula>-means algorithm. We first compute the time-frequency spectrograms of emitter signals by the proposed method. And we apply the convolutional neural network (CNN) for automatic identification based on the time-frequency images. In the experiment, we simulate different emitter signals for performance evaluation and compare our method with the spectrum analysis methods adopted in the literature. The results prove that our method can achieve excellent performance and has strong robustness in the condition of low signal-to-noise ratio (SNR), and our time-frequency analysis method works well in real-time.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1089-7798
1558-2558
DOI:10.1109/LCOMM.2021.3084043