Efficient Near-Field Millimeter-Wave Sparse Imaging Technique Utilizing One-Bit Measurements

Current near-field millimeter-wave (MMW) imaging techniques are primarily designed for high-precision quantitative data. Nevertheless, high-precision sampling leads to challenges such as expensive hardware costs and huge data storage requirements. To address these issues, this article introduces a novel near-field MMW sparse imaging method utilizing one-bit measurements. One-bit sampling simplifies the collection and processing of received signals, resulting in a substantial reduction in hardware costs and data storage demands. We model one-bit measurements from a sparsity-driven perspective, based on compressed sensing (CS) theory, and introduce a convolutional reweighted
$l_1$
-norm constraint to promote the sparsity of clustered structures commonly found in near-field imaging. Furthermore, to circumvent the computational complexities associated with constructing, storing, and optimizing large-scale matrix-vector multiplications within CS theory, the proposed method utilizes the range migration algorithm (RMA) and its inverse operator as an alternative. The advantages of both CS and matched filtering (MF) approaches in imaging are successfully combined by this strategic integration, greatly reducing the computational and storage costs of using one-bit CS directly. Finally, thorough simulations and real-measured experiments are used to demonstrate the viability and efficacy of the proposed approach, which uses one-bit measurements for imaging.