Survey Process
Graphics processing units (GPUs) have been advancing rapidly over the past decade. The hardware used first for the image processing displayed on the screen has been converted into a device for calculating parallel (general- purpose GPU). The GPU can be used to perform radar data processing during the signal processing phase or during the data processing stage. Considering that radar data are processed on a large scale, it is washed away and allows the computational process to be parallel. Previously, radar data processing was done using a special digital signal processor (DSP) device and/or field programmable gate array (FGGA). But the confiscation required for both types of devices is more expensive than the GPU. In addition, both have low scaling.
Today, high-performance embedded computer applications pose significant challenges. Traditionally, the application has been realized in the use of application-specific integrated circuit (ASIC) and/or DSPs. The advantages of ASIC in terms of performance and power requirements can often not keep up with the increase in fiction costs. GPUs, on the other hand, offer superior computer performance while maintaining their versatility.
Radar is a system that detects the presence, direction, distance, and speed of aircraft, ships, and other objects (moving and stationary objects) by sending radio waves that reflect from object to source. Radar transmits a short pulse modified by a directional antenna, which rotates at an uncontrolled speed. If the antenna leads to the target, a small portion of the transmitted energy will be reflected from the target as it travels to the antenna. The waiting distance between sent and received legumes is proportional to the distance. The direction of the antenna (azimuth antenna) determines the azimuth of the target in the process of placing the gingerbread. Image produced by radar helps in locating an object that reflects the signal. Traditionally, source images are provided with a special CRT view tabbed project-level indicator. Currently, source images are provided using computers and raster displays.
In high-frequency (HF) radar systems, the classical platform for signal processing uses DSP and field programmable gate array (FPGA). However, the seizure of these two sites is very high, and the scalability is bad for both sites. It is therefore important to find new ways to process HF radar signals. Inspired by the software-limited radar concept and the excellent hardware speed performed in parallel processing to the GPU, we explored how to implement the HF radar algorithm on a GPU on a regular PC platform. Numerical tests were carried out to demonstrate the effectiveness of the proposed method.
