EM Data Simulation System Based on GPU Acceleration
Zhang, Yisi (2019-12-19)
EM Data Simulation System Based on GPU Acceleration
Zhang, Yisi
(19.12.2019)
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
suljettu
Julkaisun pysyvä osoite on:
https://urn.fi/URN:NBN:fi-fe202001212893
https://urn.fi/URN:NBN:fi-fe202001212893
Tiivistelmä
In recent years, with the development of human beings to the ocean, the detection and identification of the sea surface and the target on it have become increasingly important. Therefore, studying the EM scattering characteristics of maritime targets is of great significance. In general, we can obtain the EM scattering characteristics of the target through actual measurement or simulation. However, the actual measurement needs to collect the information of the actual object and the environment. The amount of data required for research will lead to great overhead.Although the measured data is accurate enough, it is only used as a calibration parameter for the simulation method to improve accuracy. Therefore, the use of computer simulation to obtain the target radar cross section and other simulation parameters is the main means of EM scattering characteristics in the current research. For the target on the rough sea surface, it is difficult to obtain the simulation data by general method directly. It is necessary to model the whole system before extracting useful information from the echo data mixed with sea clutter. Based on the hybrid iterative algorithm of Physical Optics and Method of Moment, the EM scattering characteristics of the target under rough sea background are studied, and the Physical Optics method in the algorithm is accelerated by the graphical EM calculation method (GRECO). The accuracy of the overall algorithm is verified by numerical examples.
This paper first introduces the basic theoretical derivation of physical optics and moment method, and leads to the PO-MOM hybrid iterative method at multi-scale. The algorithm considers the multiple scattering between the PO region and the MoM region. The method can calculate the accurate results of the target under the complicated rough sea surface background of electric large size for radar. It also introduces the flow of the GPU rendering pipeline and various tests before rasterization. After the introduction of the depth test and the GRECO method, from the perspective of modeling the real sea surface, the principles and advantages and disadvantages of many commonly used sea spectrums are introduced. Elfouhaily sea spectrum is selected as the correlation height function of random sea surface generation. After generating the sea surface model, different wind speeds are analyzed for researching the impact on the sea surface. Finally, the EM scattering based on the GPU-accelerated PO method is calculated. We calculate the EM scattering characteristics of the rough sea surface and its upper target by the hybrid iterative method to verify the applicability and accuracy of the algorithm.
In this paper, we give the algorithm framework of GPU-accelerated PO method. By calculating the scattering echoes of rough sea surface under different sea conditions, the effectiveness of the method in the analysis of EM properties of rough sea surface is verified. At the same time, based on the hybrid iterative algorithm of physical optics and moment method, we study the EM scattering characteristics of the target on the sea surface under rough sea background, and verify the accuracy of the overall algorithm by numerical examples.
This paper first introduces the basic theoretical derivation of physical optics and moment method, and leads to the PO-MOM hybrid iterative method at multi-scale. The algorithm considers the multiple scattering between the PO region and the MoM region. The method can calculate the accurate results of the target under the complicated rough sea surface background of electric large size for radar. It also introduces the flow of the GPU rendering pipeline and various tests before rasterization. After the introduction of the depth test and the GRECO method, from the perspective of modeling the real sea surface, the principles and advantages and disadvantages of many commonly used sea spectrums are introduced. Elfouhaily sea spectrum is selected as the correlation height function of random sea surface generation. After generating the sea surface model, different wind speeds are analyzed for researching the impact on the sea surface. Finally, the EM scattering based on the GPU-accelerated PO method is calculated. We calculate the EM scattering characteristics of the rough sea surface and its upper target by the hybrid iterative method to verify the applicability and accuracy of the algorithm.
In this paper, we give the algorithm framework of GPU-accelerated PO method. By calculating the scattering echoes of rough sea surface under different sea conditions, the effectiveness of the method in the analysis of EM properties of rough sea surface is verified. At the same time, based on the hybrid iterative algorithm of physical optics and moment method, we study the EM scattering characteristics of the target on the sea surface under rough sea background, and verify the accuracy of the overall algorithm by numerical examples.