.png){kind=icon}
.png){kind=icon}
Return
Jamming Pattern Recognition for Radar Based on Phase Space Reconstruction and Dynamic Characteristics
.png){kind=icon}
PRE
AI
T
F
W
DOI：10.1109/taes.2026.3716239.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
We propose a mechanism-guided radar jamming recognition framework based on phase space reconstruction. We formulate a discrete-time digital radio frequency memory state-machine model that represents memory write/read addressing, sampling/forwarding mode switching, numerically controlled oscillator (NCO) phase accumulation, digital gain control, and optional noise-branch activation. The model describes jamming patterns as control-word strategies of a common physical generator rather than as independent empirical equations. The emitted waveform is embedded in a reconstructed phase space, from which geometric descriptors and Gaussian mixture model phase-point distribution features are extracted and classified by a random forest. Simulation and measured-data experiments show that the proposed features improve robustness at low jamming-to-noise ratio compared with conventional time-, frequency-, and time-frequency-domain features, while providing competitive performance relative to deep learning methods and retaining physical interpretability.
Keywords:
Jamming
Modeling
Timing
Radar
Permission
Weighted sum model
Visualization
Delays
Technology
Accuracy
Journal
IF:
5.7
Papers:
651
Citations:
2.4W
Organization
Cited Papers