An underwater ship radiated noise target recognition method
By constructing a multi-domain feature mapping and fusion network, the problems of low accuracy and poor robustness in underwater ship radiated noise identification in complex shallow sea environments are solved. High-precision identification is achieved in scenarios with multipath distortion and low signal-to-noise ratio, and it is suitable for underwater target identification.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WUHAN LINGJIU MICROELECTRONICS CO LTD
- Filing Date
- 2026-06-11
- Publication Date
- 2026-07-17
AI Technical Summary
Existing underwater ship radiated noise identification methods have low identification accuracy and poor robustness in complex shallow sea environments, cannot adapt to multipath distortion and low signal-to-noise ratio scenarios, and have weak engineering adaptability.
A target identification network for underwater ship radiated noise is constructed. It adopts a multi-domain feature mapping module, a three-branch multi-domain feature extraction module, a channel space joint attention module PCS-JAM with normal mode prior constraints, a physical feature track, an interactive enhancement track, and a modal confidence dynamic adaptive fusion module MCDE-AFM based on Bayesian inference to achieve comprehensive extraction and fusion of time domain, frequency domain, and time-frequency domain features.
It significantly improves recognition accuracy in low signal-to-noise ratio and multipath distortion scenarios, possesses excellent robustness and engineering practicality, and is suitable for various shallow water underwater target recognition scenarios.
Smart Images
Figure CN122413151A_ABST