Sparse Reconstruction of SFCWSAR Based on an Approximate Observation Operator

Abstract

The traditional compressed sensing SFCWSAR (Stepped Frequency Continuous Wave Synthetic Aperture Radar) sparse reconstruction algorithm consumes a lot of computer memory and cannot compensate the range migration in the same pulse group. Based on this, this paper proposes a SFCWSAR sparse reconstruction algorithm based on an approximate observation operator. First, the algorithm replaces the accurate observation operator with the approximate observation operator, which greatly reduces the computer memory consumption while the algorithm is running and realizes the compensation of the range migration in the SFCWSAR pulse group. Furthermore, the SFCWSAR sub-band echo data under full sampling conditions are used to modify the important parameter of the Doppler center frequency of the approximate observation operator, which significantly improves the reconstruction accuracy of the scene. The SFCWSAR data show that, compared with the conventional sparse autofocus algorithm, the proposed algorithm takes less memory and can reconstruct scenes with high accuracy.

Dual‐Objective Optimization for Maximizing Photovoltaic Output Power in a Two‐Stage Single‐Phase Grid‐Tied System

“…The particle swarm optimization (PSO) algorithm encounters premature convergence problems and gets stuck in local power point. [30,31] Furthermore, defining the initial parameters (c 1 ,c 2 ,w) in PSO is complicated. Coyote optimization, [32] population-based DPVAR, [33] adaptive jellyfish algorithm, [34] marine predators algorithm, [35] modified Harris hawks optimizer (MHHO), [36] chaos map-based reconfiguration, [37] and simulated annealing [38] are different algorithms presented in the literature to enhance the power by reconfiguration.…”