.png){kind=icon}
.png){kind=icon}
Return
A Real-Time Beam Pattern Synthesis Method for Variable Curvature Cylindrical Conformal Array Based on Neural Network
.png){kind=icon}
PRE
AI
Y
S
W
W
DOI：10.1109/LAWP.2026.3656707.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
To address the degradation in radiation performance caused by external deformations in variable-curvature cylindrical conformal antenna arrays, this letter proposes a real-time beam pattern synthesis method based on a multibranch neural network. Using a cylindrical flexible array with a curvature radius ranging from 100 mm to 300 mm as an example, a five-branch neural network framework is constructed to enable both low-sidelobe and adaptive null beam synthesis. The average inference times of the two neural networks are 2.42 ms and 2.51 ms, respectively, which ensures real-time performance. The effectiveness of the proposed method is validated through full-wave simulations and experimental measurements.
Keywords:
Adaptive null
array synthesis
conformal array
low-sidelobe
neural network
Journal
IF:
4.8
Papers:
1.0W
Citations:
2.8W
Organization
Cited Papers