C 2 MI: A Covariance-Coupling Sparse Recovery Algorithm for Metasurface-Based Microwave Imaging

Antenna Measurement Techniques Association Symposium (AMTA), pp. 1–4

Abstract

In this work, we present$\mathbf{C}^{2}$MI (Covariance-Coupling Microwave Imaging), a second-order statistical formulation that enhances imaging under sparse measurements.$\mathrm{C}^{2}$MI models a second-order forward system,$\text{SL}=\mathrm{M}$, where S comprises upper-triangular elements of covariance matrices from permuted and transformed sensing patterns,$L$is the vectorized upper triangle of the scene covariance matrix, and M contains variances of measurement vectors. We solve this underdetermined system via$\ell_{1}$-regularized optimization with a structural constraint enforcing diagonal dominance: each offdiagonal$L_{i j}$must not exceed its corresponding diagonal terms$L_{i i}$and$L_{j j}$. This reflects the prior that strong inter-location coupling (captured by covariance off-diagonals) occurs only when their variances (diagonals) are high. The result is a sparse, interpretable, and structured estimate of scene covariance. Compared to$\ell_{2}+$total variation (TV) methods, C2 MI accurately reconstructs sparse targets using fewer masks-for instance, recovering 14 active pixels in an$8 \times 8$grid with only 20 frequencysweep masks, versus$28+$for$\ell_{2}+$TV. This improves efficiency, enabling real-time, portable, and low-cost microwave imaging. We validate$\mathbf{C}^{2}$MI in MATLAB simulations with metasurface antenna phase gradient measurements, showing promise for security, biomedical, and non-invasive sensing applications.

Authors 4

1. Firas Slewa Dawod

   University of Missouri–Kansas City

   Affiliation as printed

   School of Science and Engineering, University of Missouri-Kansas City,AERIAL Lab,MO,USA

2. M. Arif Aachen Chair of High Frequency Electronics

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Chair of High Frequency Electronics,Aachen,Germany

3. Renato Negra Aachen Chair of High Frequency Electronics

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Chair of High Frequency Electronics,Aachen,Germany

4. Sayan Kumar Roy

   University of Missouri–Kansas City

   Affiliation as printed

   School of Science and Engineering, University of Missouri-Kansas City,AERIAL Lab,MO,USA