An Optimization-Based Deconvolution Approach for Recovering Time-Varying Phase Modulation Signal of Metasurface

Metasurfaces can alter radar echo characteristics by dynamically modulating the phase of incident electromagnetic waves, thereby degrading radar signal processing performance. Recovering the time-varying phase modulation signal of metasurfaces from radar echoes is crucial for understanding modulation patterns or suppressing modulation effects. To address this problem, we introduce an optimization-based deconvolution approach in this article. First, an optimization model for recovering phase modulation signals from metasurfaces is established by combining the least-squares (LS) deconvolution principle, total variation in the modulating signal, and imaginary part constraint. Furthermore, a composite optimization approach (COA) to solve the deconvolution optimization problem is developed, in which the alternating minimization and alternating direction method of multipliers (ADMM) frameworks are integrated to estimate the complex propagation loss and the modulation phase sequence. In addition, the convergence and computational complexity analysis of the proposed COA is conducted. Finally, the effectiveness of the proposed method is validated through simulation and measured experiments.