30 September 2022 in IEEE Journal on Multiscale and Multiphysics Computational Techniques
To facilitate the antenna design with the aid of computer, one of the practices in consumer electronic industry is to model and optimize antenna performances with a simplified antenna geometric scheme. The ease of handling multi-dimensional optimization problems and the less dependence on the engineers' knowledge and experience are the key to achieve the popularity of simulation-driven antenna design and optimization for the industry. In this paper, we introduce a flexible geometric scheme with the concept of mesh network that can form any arbitrary shape by connecting different nodes. For such problems with high dimensional parameters, we propose a machine learning based generative method to assist the searching of optimal solutions. It consists of discriminators and generators. The discriminators are used to predict the performance of geometric models, and the generators to create new candidates that will pass the discriminators. Moreover, an evolutionary criterion approach is proposed for further improving the efficiency of our method. Finally, not only optimal solutions can be found, but also the well trained generators can be used to automate future antenna design and optimization. For a dual resonance antenna design, our proposed method is better than the other mature algorithms.
28 December 2021 in IEEE Transactions on Microwave Theory and Techniques
In this article, we present a new diagnostic technique for printed circuit boards (PCBs) and integrated circuits (ICs). This technique consists of an array-probe scanning system and the phaseless inverse source method with a joint regularization. Different from the single probe scanning system, a ground plane for array probes is applied to protect the RF transmission path from interference by other radiation in the array probes scanning system. In order to consider the influence of both PCB ground and array probes ground on electromagnetic near-field distributions, we use the method of moments to solve the surface integral equations in the forward problem. In the inverse problem, we establish a nonlinear cost function between the magnitude-only fields and equivalent sources. Because of the ill-posedness of the inverse problem, an efficient iterative method with a joint regularization framework is applied to reconstruct the equivalent sources on the source plane above the PCB ground. Different levels of noise are added to the simulated data to validate the algorithm. Laboratory experiments and numerical examples show the effectiveness of the proposed method.
27 April 2020 in IEEE Transactions on Electromagnetic Compatibility
A new method to visualize the coupling path is proposed to help solve electromagnetic compatibility problems. By using the reciprocity theorem and the concept of reaction, the coupling density coefficient (CC) is defined. Integration of the CC on closed surfaces equals the coupled power at the receiver. By plotting the CC on a set of surfaces from the source to the receiver, the coupling path, which traces the coupled power flow, can be visualized. All visualization procedures can be performed in a full-wave simulation tool, thus making the proposed method convenient to use. To validate and demonstrate the potential benefits of the new method, visualization of the coupling path in a realistic automotive intentional electromagnetic interference problem is presented. In addition, a mitigation solution based on the visualization result was examined.
1 November 2024 in IEEE Transactions on Antennas and Propagation
Machine learning (ML) methods are widely employed in antenna design and optimization due to their strong performance in multi-dimensional optimization problems and reduced dependence on engineers’ experience. In this work, we propose an auxiliary classifier generative method inspired by the auxiliary classifier generative adversarial networks (AC-GANs) from the computer vision domain, integrated with an evolutionary strategy to facilitate class-based antenna design. By allocating suitable class labels, multiple design requirements can be efficiently met. This method categorizes antenna designs to enable simultaneous optimization of various antenna types within a unified geometric model. Utilizing a discriminator-generator framework typical of GANs, the discriminator predicts class labels from geometric models, while the generator creates new candidates to satisfy these classifications. An additional filter layer, a random forest classifier, refines the generator’s outputs, proving effective in our experiments. The incorporation of evolutionary criteria significantly enhances the design process, mirroring active learning principles to address the challenges of limited and expensive datasets. The final outcomes include both optimized geometric designs and a robust generator capable of producing these designs in large quantities. We substantiate the efficacy and benefits of our proposed method through an analytical case study and a practical dipole antenna example. The comparison between our method and other optimization methods highlights three distinct advantages of the proposed approach: versatility, scalability, and high efficiency.
17 November 2022 in Journal of Applied Geophysics
26 August 2020 in IEEE Transactions on Electromagnetic Compatibility
Source reconstruction using equivalent dipole moments to replace the original noise source is widely used in radio frequency interference. The conventional reconstruction methods only work well in the cases of free space or with a large ground plane. In this article a new method is proposed for source reconstruction over an arbitrarily shaped ground plane. It uses the field below the ground plane to eliminate the arbitrary ground effect. Comparing to the method of solving surface integral equation, it is simple and does not need the ground geometric information. The new method is validated in the full-wave simulations.