Pixel antennas expressed in the form of discrete variable’s structure (DVS) demonstrate immense potential in electromagnetic (EM) performance, but the design process relies on a huge full-wave simulation dataset. To efficiently design pixel antennas with high degrees of freedom (DoFs), in this study, a multi-head discrete action calibration proximal policy optimization (MDAC-PPO) reinforcement learning (RL) method is proposed. Specifically, a sophisticated multi-head discrete mechanism is developed to generate the placement status of all metal patches in a probabilistic form. Furthermore, a coupling calibration mechanism that considers local metal density is proposed to break down the intricate optimization target, expediting the iterative process. Additionally, to delve deeper into potential designs that might otherwise be neglected, an exclusive OR (XOR) operation state update manner is integrated into the PPO agent. Three design examples on a pixel antenna with a 2 ²⁵⁶ -sized solutions space, including optimization targets on impedance matching, gain, efficiency, and related bandwidth, are presented to show the optimization efficiency and performance of the MDAC-PPO method. The measured |S11| of the fabricated prototypes aligns well with simulations, further validating the reliability of the MDAC-PPO method in practical applications. Meanwhile, this method also provides a new perspective for designing other EM devices with DVS.

This article presents a defect detection method for diagnosing pseudo soldering problems in radio frequency (RF) circuit components. The algorithm integrates the source reconstruction method with the difference method (DM-SRM) to reconstruct the source distribution pinpointing the defect location effectively. Initially, the algorithm utilizes the difference between the radiated fields of the intact device and the defective one as input for the SRM. Subsequently, leveraging prior information regarding the sparsity of defects, the algorithm utilizes elastic regularization to address the ill-conditioned problems inherent in the SRM, thereby enhancing its noise robustness. The proposed method’s effectiveness and robustness are demonstrated through simulations on patch antennas and measurements using a near-field system for a microstrip power divider and a low-noise amplifier (LNA).

Near-field probes have been widely used in the measurement of radio frequency circuits and integrated circuits. It is an important process to correct the output distribution of the probe to the field distribution. However, near-field probes generally have certain receiving characteristics, which determine their performance. Therefore, it is very important to calibrate the receiving characteristics of near-field probes. In this article, we present an efficient probe calibration and probe compensation technique for electromagnetic near-field scanning. In the probe calibration process, the alternating direction method of multipliers is applied to flexibly solve the deconvolution problem, which facilitates reducing computational complexity in the spectral domain and elastic-net regularization in the spatial domain. For the probe compensation process, the second-order derivative constraint is introduced to improve the antinoise ability and the smoothness of the reconstructed results. Numerical examples and laboratory experiments show the effectiveness of the proposed method.

This paper presents a broadband near-field probe designed for measuring the normal magnetic field ($H_z$) in radio frequency (RF) circuits operating within a frequency range of 2–8 GHz. The proposed probe uses a cost-effective 4-layer printed circuit board (PCB) structure made with an FR-4 substrate. The probe primarily consists of an $H_z$ detection unit, a broadband microstrip balun, and a coaxial-like output. The broadband balun facilitates the conversion from differential to single-ended signals, thereby enhancing the probe’s common-mode rejection capability. This design ensures that the probe achieves both cost efficiency and high broadband measurement performance. Additionally, this work investigates the feasibility of employing microstrip lines as calibration standards for the $H_z$ probe. The probe’s structural parameters and magnetic field response were initially determined through simulations, and the calibration factor was subsequently verified by calibration experiments. In practical measurements, the field distributions above a microstrip line and a low-noise amplifier (LNA) were captured. The measured field distribution of the microstrip line was compared with simulation results to verify the probe’s performance. Meanwhile, the measured field distribution of the LNA was utilized to identify the radiating components within the amplifier.