.png){kind=icon}
Return
Efficient Multiscale Point Cloud Transformer for Time-Modulated Array Synthesis
.png){kind=icon}
PRE
AI
DOI：10.1109/lawp.2026.3734638.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The synthesis of time-modulated arrays (TMAs) is a challenging problem due to its strong nonlinearity and high dimensionality. This letter employs a dual-coupled equivalent excitation to reformulate the TMA synthesis problem as a generalized phased array optimization task. Based on this reformulation, an efficient inverse design framework using a multiscale point cloud transformer (MPCT) is proposed. By modeling radiation patterns as point cloud-like angular samples and incorporating direction embedding with subtractive attention, the MPCT effectively captures the intrinsic geometric relationships among directions, enabling accurate and stable excitation prediction. Numerical results verify the superior performance and high computational efficiency of the proposed method for various TMA synthesis cases.
Keywords:
time-modulated array
array synthesis
multiscale point cloud transformer (MPCT)
radiation pattern
Journal
IF:
4.8
Papers:
1.0W
Citations:
2.8W
Organization
Cited Papers
No cited papers available