Coherent laser radar signal processing method based on GPU

Abstract

The present invention discloses a coherent laser radar signal processing method based on a GPU. Based on a CUDA (Compute Unified Device Architecture) algorithm of the NVIDIA Corporation, the method employs GPU (Graphic Processing Unit) parallel computation to replace FPGA or DSP operation in a traditional coherent wind lidar, and the CPU is employed to perform storage and further processing of a power spectrum output by the GPU. Compared to a traditional coherent wind lidar, the GPU computation is faster in speed and lower in cost, can achieve real-time wind speed inversion or real-time time-frequency conversion. The power spectrum data is stored to conveniently perform further processing of the data by employing other algorithms to extract more useful information such as atmospheric visibility, an atmospheric depolarization ratio and the like.

Description

technical field [0001] The invention relates to the technical field of laser radar, in particular to a GPU-based coherent laser radar signal processing method. Background technique [0002] Accurate atmospheric wind field measurement is of great significance for detecting atmospheric pollution, obtaining military environmental intelligence, improving aerospace safety, improving the accuracy of weather forecasts, and improving climate models. Wind lidar is an effective means of wind field measurement, which can be divided into direct detection wind lidar and coherent detection wind lidar. The direct detection wind measurement lidar uses an optical discriminator to convert the Doppler frequency shift information into the relative change of energy, and realizes the measurement of the atmospheric wind field; the coherent detection wind measurement lidar uses the coherence between the atmospheric echo signal and the local oscillator laser The beat frequency realizes the measurem...

AI Technical Summary

Problems solved by technology

With the computing power of 160G floating-point per second of TI's high-end DSP chip TMS320C6678, 8 parallel chips are needed to meet the computing requirements; currently the highest performance FPGA computing power is in the same order of magnitude as TMS320C6678, so the programming of coherent receivers is difficult, and the circuit complex structure

[0005] (2) The DSP and FPGA chips of the TMS320C6678 model are very expensive, and the single-core TMS320C6678 board is about 5,000 yuan

Therefore, coherent receiver circuits are expensive

[0006] (3) If the distance resolution of coherent lidar is improved, the output laser pulse width needs to be reduced. According to the relationship between pulse width and spectral broadening, it will lead to broadening of the echo signal power spectrum

The computing power of DSP or FPGA will become the bottleneck of the development of coherent lidar

Images