Regularization Imaging Algorithm With Accurate G Matrix for Near-Field MMW Synthetic Aperture Imaging Radiometer

Abstract

Abstract-In order to improve the reconstruction accuracy of near-field SAIR, a novel regularization imaging algorithm based on an accurate G matrix is proposed in this paper. Due to the fact that the regularization reconstruction is usually an underdetermined problem, inaccurate operation matrix G will lead to great reconstruction error in the imaging results, or even the normal imaging cannot be obtained. In this paper, we establish an accurate G matrix based on the accurate imaging model of near-field SAIR. Compared with the traditional G matrix with some unnecessary approximations, the proposed G matrix without approximation can improve the reconstruction accuracy effectively. For improving the accuracy of matrix G further, the corresponding parameters are corrected according to the RMSE between the imaging results of the regularization method and modified FFT method which is not sensitive to the parameters' change. The effectiveness of this calibration method has been tested by 1D simulation experiments. Moreover, the 2D simulation experiments demonstrate that the proposed accurate G matrix can improve the imaging accuracy of regularization method effectively. Finally, the 1D imaging experiment is performed to test the effectiveness of the proposed method for the actual synthetic aperture imaging further.

A Real-Time Monitoring Method for Civil Aircraft Take-Off and Landing Based on Synthetic Aperture Microwave Radiation Technology

“…The actual detection system will have channel, antenna, position, pattern, and temperature errors, etc., [ 15 ]. The common method is to correct the G -matrix through standard point sources in the laboratory [ 21 , 22 , 23 ]. However, in airport applications, moving the system to the laboratory regularly to correct the G -matrix is impractical.…”

Analysis of the Noise Components for Affecting the Imaging Performance of the Synthetic Aperture Interferometric Radiometer (Sair)

“…Based on the principle of the interferometry, the SAIR system measures the data (namely visibility samples) by using the cross-correlation between multiple receiving channels and then recovers the image of the observation scene with the measured visibility samples [1,2,11,12].…”

“…where P s includes the power of the Gaussian signal of interest and the Gaussian noise generated by the receivers, and P f and P drif t are the power of the flicker noise and random-walk noise, respectively. With the visibility samples measured by the SAIR system, the image of the interest scene can be reconstructed by using the inverse Fourier transform [1,2,11,12]:…”

A Robust Reweighted L1-Minimization Imaging Algorithm for Passive Millimeter Wave SAIR in Near Field

“…To reduce multiplicative errors, RRIA adopts an accurate processing of the distance $R_n^c$ and $R_n^c$ and modifies the traditional G matrix which is established by the Taylor expansion to establish a new accurate G matrix suitable for both far field and near field [18]. Thus, without being processed approximately by Taylor expansion, Equation (2) can be expressed as: ${boldnormalV}_{c,l}={\displaystyle \sum _{n=1}^{N}T(n)F_c{F}_l^{*}{e}^{-jK[\sqrt{{(x_n-X_c)}^2+(y_n-Y_c)+h^2}-\sqrt{{(x_n-X_l)}^2+(y_n-Y_l)+h^2}]}}$…”

“…Different from the application and the choice of the weight factor in the [8,18], according to the signal model of RRIA for SAIR imaging, the SAIR imaging system measures the correlation value directly and reconstructs images from the visibility function, which means G matrix is essential for recovery no matter whether SAIR imaging is sparse enough or not. Unweighted L 1 -minimization attempts to find the solution with the smallest sum of the magnitudes of nonzero terms.…”