压缩传感图像重建算法的FPGA实现

FPGA Implementation of Compressed Sensing Image Reconstruction Algorithm

摘要:

压缩传感理论允许以低于奈奎斯特定理规定的采样频率对信号进行采样并重构。近年来,压缩传感信号重建的研究大多用软件实现。文中设计一种基于FPGA的OMP重构算法的硬件实现,使用QR分解求解矩阵的伪逆。仿真结果表明应用FPGA得到的重构信号逼近原信号,且重建速度快。基于FPGA的压缩传感信号重建,减少了信号重构所需的时间,推进了压缩传感进入实际应用的进程。

Abstract:

Compressive sensing is a technology which allows us to sample at sub-Nyquist rate, then the signal can be recovered from the samples. In recent years, most research about compressive sensing focus on the domain of software. In this paper, presented is an architecture based on FPGA for the reconstruction of compressive sampled signals using orthogonal matching pursuit algorithm. QR-decomposition process is used to achieve the matrix pseudo inverse. Simulation results indicate that the architecture is available. Using the architecture based on the hardware, the signal reconstruction can no longer be confined to the software domain. This will reduce the signal reconstruction time and will expand the application field of the compressive sensing.