Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | English |
| Published: |
Zenodo
2026
|
| Subjects: | |
| Online Access: | https://doi.org/10.5281/zenodo.20144331 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <p>Abstract—In this paper, we apply a gradient descent-based<br>method to the synthesis of uniformly excited sparse phased<br>arrays. The effectiveness of gradient-based optimization in this<br>domain is highly dependent on the precision of the Peak Sidelobe<br>Level (PSLL) estimation. Conventional grid-based calculation<br>methods fail to provide the necessary accuracy, introducing<br>numerical noise that prevents efficient convergence. To address<br>this, we implement a grid-less analytical framework for radi-<br>ation pattern analysis. Our approach not only achieves near-<br>machine precision but also significantly enhances<br>computational performance compared to traditional discrete grid<br>search. Leveraging this high-precision engine, we develop an<br>upgraded gradient descent strategy capable of synthesizing large-<br>scale arrays. The proposed method effectively manages high-<br>dimensional search spaces and demonstrates superior perfor-<br>mance over heuristic algorithms in terms of both convergence<br>speed and the depth of sidelobe suppression.</p>