Terahertz Near-Field Sparse Imaging Method Based on Multiple Transmitter and Multiple Receiver Arrays

Abstract

The invention relates to a terahertz near field sparse imaging method based on a multi-input multi-output array. Terahertz near field sparse imaging of the multi-input multi-output array is realized by setting an imaging scene, acquiring terahertz echo, discretizing the echo, estimating scattering point number through Akaike information amount criterion and finally estimating target point coordinates through a sparse reconstruction algorithm. By the method, nonlinearity of target information in path delay is retained, one-one mapping relation between observation matrix atom serial numbers and target scenes is utilized, and target reconstruction is realized on the basis of greed idea of sparse reconstruction technology. In order to improve dependence of a classic greed algorithm on target sparseness (namely scattering point number), the Akaike information amount criterion is utilized to perform self-adaptive estimation on sparseness of a target on the basis of remaining amount when the algorithm is utilized to iteratively select atom sequence each time, so that universality and practicability of the target reconstruction algorithm are further improved.

Description

technical field [0001] The invention relates to an imaging technology, in particular to a terahertz near-field sparse imaging method based on a multi-input and multi-receive array. Background technique [0002] Terahertz electromagnetic waves have the characteristics of low energy and coherent measurement, which make terahertz imaging valuable in many fields; and terahertz waves have strong penetrating ability and are directional, so terahertz imaging in some environments Has many advantages. [0003] As we all know, in the theory of microwave imaging based on synthetic aperture, the two quantities of distance resolution and Nyquist sampling rate jointly determine the complexity of the imaging system. The upper limit of range resolution is determined by the bandwidth of the transmitted signal, and according to the Nyquist sampling law, this requires the sampling frequency of the system to be at least twice the signal bandwidth. This means that to improve the distance resol...

AI Technical Summary

Problems solved by technology

[0005] The present invention aims at the problem that the traditional target reconstruction algorithm based on Fourier transform increases the approximation error and narrows the target reconstruction range in the terahertz near-field multi-station imaging, and proposes a terahertz near-field based multi-transmission and multi-reception array. Field sparse imaging method, using a sparse reconstruction algorithm for adaptive estimation of sparsity to solve the problem of increased approximation error and narrowed target reconstruction range in the traditional Fourier transform-based target reconstruction algorithm in terahertz near-field multi-station imaging problem, as well as the problem of low resolution of terahertz imaging at low sample rate, to achieve near-field sparse imaging of linear compact terahertz multi-input multi-receiver array

Images