RCS Optimization Analysis Method for Sea-Skimming Unmanned Aerial Vehicle Based on Back Propagation Neural Network Algorithm
The radar cross section (RCS) of sea-skimming unmanned aerial vehicle (UAV) can be influenced by the sea surface scattering under different sea conditions. In this paper, a composite model of the rough sea surface and sea-skimming UAV is established. A hybrid algorithm based on the application of ph...
Gespeichert in:
Veröffentlicht in: | Applied Computational Electromagnetics Society journal 2020-01, Vol.35 (5), p.519 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The radar cross section (RCS) of sea-skimming unmanned aerial vehicle (UAV) can be influenced by the sea surface scattering under different sea conditions. In this paper, a composite model of the rough sea surface and sea-skimming UAV is established. A hybrid algorithm based on the application of physical optics (PO) method and improved multilevel fast multipole algorithm (PO-IMLFMA) for solving the RCS of the composite model based on four-path model (FPM) is proposed. Compared with multilevel fast multipole algorithm (IMLFMA) and PO and method of moment (PO-MOM), PO-IMLFMA has the advantages of less memory consumption (about 295 MB) and faster solution speed (about 768 s) for solving the composite model. Furthermore, in view of the influence of sea surface on the RCS of sea-skimming UAV, a compensation scheme based on back propagation (BP) neural network for the RCS of UAV is proposed. The compensation scheme is analyzed for the monostatic RCS of sea-skimming UAV under different sea conditions. The compensation results show that the compensation errors under 1-scale, 3-scale and 5-scale sea conditions are less than 0.95 dBsm, 0.41 dBsm and 1.94 dBsm, respectively. In other words, the compensation scheme significantly reduces the influence of sea condition |
---|---|
ISSN: | 1054-4887 1943-5711 |