Research on Reconstruction Algorithm of Terahertz Signal Based on Compressed Sensing

Numerous features of terahertz make it increasingly prominent in the field of nondestructive testing, but terahertz time domain spectroscopy (THz-TDS) imaging system has the problems of long acquisition time and large amount of data. Compressed sensing theory points out that if the signal meets the sparsity condition in a domain, the original signal can be reconstructed with a number of sampling points far lower than that required by the Nyquist sampling theorem. In this research, the ideal terahertz signal and the measured terahertz signal are reconstructed by using the orthogonal matching pursuit algorithm and the sparsity adaptive matching pursuit algorithm. The results indicate that the ideal terahertz signal can be precisely reconstructed with a sampling rate of only about 15%, and the measured terahertz signal only needs about 25%. This research provides a reliable guarantee for the application of compressed sensing in THz-TDS imaging technology.