In this paper, we introduce an innovative deep learning (DL) methodology designed for real-time quantitative microwave imaging (MWI). Our approach is centered around the utilization of a deep convolutional asymmetric encoder-decoder structure (DCAEDS), which requires only a single-frequency far-field measurement of the electromagnetic (EM) scattered field as input and subsequently predicts the contrasts (permittivities) of the target materials. During the offline training process, we incorporate an EM forward solver specifically crafted to compute the EM scattered field generated by the predicted target contrasts (permittivities) produced by the DCAEDS. The DCAEDS is seamlessly integrated with this EM forward solver to optimize the loss function. This loss function comprises two fundamental components: (1) Data-induced loss, directly quantifying the dissimilarity between the predictions of our proposed DCAEDS and the actual labels for the target contrasts (permittivities); (2) Physics-induced loss, evaluating the distinctions between the measured EM scattered field and the computed EM scattered field derived from the predicted target contrasts (permittivities) generated by the DCAEDS. Our DL approach excels in delivering precise results, even for high-contrast targets, overcoming limitations associated with conventional methods, such as computational cost and ill-conditioning. Numerical benchmarks using dielectric targets underscore the practicality and effectiveness of our DL-based approach.

Decoupling capacitors are essential components for power integrity optimization of the power distribution networks (PDNs). In practical designs, capacitors, and their parasitic inductance have statistical deviations from nominal values, due to fabrication tolerances, temperature, aging and applied DC and AC voltage levels. This work presents a framework to evaluate the impact of such uncertainties on Zinput, the input impedance of the PDN at the evaluation port (normally at the die side of the system). Closed-form expressions are derived for the mean and variance of the input impedance, Zinput, by modeling capacitance (C) and equivalent series inductance (ESL) as random variables. The framework explicitly captures parameter sensitivities, showing that equivalent series resistance (ESR) plays a critical role for the sensitivity control. The results demonstrate that increasing ESR suppresses resonance peaks and reduces the variance, suggesting that optimal ESR values can stabilize the PDN without excessively raising baseline impedance. Furthermore, analysis of multi-port PDNs shows that variance decreases rapidly with the first few decaps and then saturates, consistent with the law of large numbers, thereby identifying a practical trade-off point for decap count. A mathematical proof is presented to show that the variance decreases vs 1/Neff, where Neff is the effective number of decaps. Finally, the method is extended to non-Gaussian parameter distributions (lognormal, gamma, and beta), showing that the conclusions remain valid beyond Gaussian assumptions. Overall, the proposed approach provides a computationally efficient and robust tool for statistical PDN analysis and offers design guidelines for the decap selection, ESR tuning under variations in decap properties.

In this paper, a novel deep learning (DL) approach has been proposed to realize high-resolution electromagnetic (EM) inversion imaging. The newly proposed approach is based on the deep convolutional double-module structure (DCDMS), consisting of the pixel-interpolating module and the corresponding quality-improving module. While the pixel-interpolating module roughly increases the ‘resolution’ of the initial input, the following quality-improving module realizes quantitative EM imaging in high resolution. The input of the proposed DCDMS adopts the mixed input scheme, consisting of the received EM scattered field and the initial reconstruction in much low resolution computed from Gauss-Newton method. The output of the proposed model is the high-resolution contrast (permittivity) ‘image’ of the target domain. In such manner, the proposed DL approach can make use of much less measurement to realize high-resolution EM inversion imaging accurately and efficiently even for high-contrast scatterers, which can hardly be realized by conventional methods. The training of DCDMS is based on the simple synthetic dataset. Numerical benchmarks are offered to illustrate the excellent performance of DCDMS, which provides a novel thinking for conducting the real-time quantitative EM inversion imaging.

In this research work, a dual-band metasurface has been proposed in order to increase the isolation between two closely packed dual-band dielectric resonator antennas (DRAs). The metasurface has been placed only 0.08λ₀ away from the antennas at the center frequency of the lower band. The proposed dual-band metasurface unit cell consists of two copper traces of different lengths printed on both sides of a printed circuit board (PCB). High isolation of more than 20 dB has been achieved for 3.4 and 4.9 GHz 5G bands with impedance bandwidths of more than 200 MHz and 1.4 GHz respectively. The antenna exhibits excellent compactness by reducing the spacing between radiating elements to 0.12λo at 3.5 GHz. The proposed isolation technique can be useful for the applications where PCB of the antenna system is not allowed to be modified.

In this work, we introduce an electromagnetic source imaging (EMSI) approach utilizing deep learning (DL) techniques, which employs deep convolutional conditional denoising diffusion probabilistic model (DCCDDPM). Conventional EMSI methods often struggle with various challenges, such as low accuracy and ill-posedness. The newly proposed EMSI method, named as DCCDDPM-EMSI, includes the forward diffusion process and the reverse diffusion process. Its diffusion process starts with EM equivalent sources on targets and adds Gaussian noise in a series of steps. On the counterpart, its reverse diffusion process step-by-step predicts the noise by using DL-based noise prediction network combined with the EM scattering measurements as conditional inputs. In contrast to traditional DDPM frame, the proposed DCCDDPM-EMSI introduces model-based loss term, which measures the discrepancy between the true EM sources and those predicted ones, in the diffusion process during its training process. Consequently, the proposed DCCDDPM-EMSI can reconstruct EM equivalent source of targets from the measured EM scattered field data. Unlike conventional EMSI methods, DCCDDPM-EMSI allows for higher fidelity EM source reconstruction without incurring excessive computational cost. Numerical benchmarks demonstrate that DCCDDPM-EMSI cannot only ensure the high accuracy but also the excellent generality, which offers significant advancements for DL-inspired quantitative EM imaging.

This article introduces a recursive parallel dynamic mode decomposition (RPDMD) scheme tailored for multidimensional harmonic retrieval (MHR), specifically applied to MIMO wireless channel sounding. The RPDMD algorithm is devised to address the complexities inherent in multidimensional scenarios, leveraging the dynamic mode decomposition (DMD) framework within a recursive parallel structure. Initially, the observed tensorial multidimensional harmonic data is transformed into a 2-D matrix format along the rth dimension. Subsequently, DMD dissects this matrix data into eigenvalues and their associated modes. The real and imaginary components of the DMD eigenvalues yield damping factors and frequencies in the rth dimension, respectively. Furthermore, recursive DMD is employed to scrutinize each mode independently for parameter retrieval across the remaining dimensions, enabling parallel analysis. Ultimately, this high-dimensional correlated decomposition scheme delivers paired damping factors and frequencies for all tones. Notably, the proposed approach can ascertain the number of tones in undamped sinusoidal signals, making it particularly suitable for MHR even without prior knowledge of the source count. Numerical experiments demonstrate the accuracy and robustness of the RPDMD scheme, with comparative analysis indicating that RPDMD outperforms similar methods, achieving optimal results with minimal mean square error in high signal-to-noise ratio scenarios. This work presents an effective data-driven solution for the MHR problem in MIMO wireless channel sounding.

Epstein-Barr virus (EBV) is a well-recognized oncogenic virus that promotes several lymphoid and epithelial cancers. The Epstein-Barr nuclear antigen 1 (EBNA1), which is known to be expressed in all EBV-positive cancers, plays a vital role in viral genome replication and maintenance and is therefore emerged as an attractive target for clinical intervention. Several EBNA1 inhibitors have shown potency in the growth inhibition of EBV-positive cancers, yet low bioavailability and in vivo unmonitored nature hamper their further implementation. Here a novel EBNA1 nano-inhibitor based on EBNA1-specific peptide inhibitor (P4) functionalized ZnGa2O4:Cr3+ ultra-small near-infrared persistent luminescent (NIR-PL) nanoparticles (ZGOC-P4) is developed. Owing to the specific binding to EBNA1, ZGOC-P4 nano-inhibitor can quickly achieve nuclear internalization in EBV-positive nasopharyngeal carcinoma (NPC) cells (C666-1) and selectively inhibit their growth. In sharp contrast, ZGOC-P4 nano-inhibitor shows no inhibition effect on EBV-negative NPC cells (HK-1). Moreover, the results indicate that the well-designed nano-inhibitor enables efficient tumor-targeting accumulation in NPC xenograft model under the monitoring of autofluorescence interference-free NIR-PL imaging in vivo and suppresses EBV-associated tumor growth with an inhibition rate of 61.6%. This work highlights the potency of ZGOC-P4 on NPC treatment and may provide new sight into future research on EBV-associated diseases.

Methods of optimizing decoupling capacitor placement on power distribution networks (PDNs) are often limited due to the computational complexity required to calculate the impact of connecting loads to an impedance matrix with hundreds of rows and columns. This work proposes that by removing all but one member of the impedance matrix before calculating, checking the impact of adding capacitors to the matrix can be done efficiently, and optimization methods can be viable even when requiring millions of impedance calculations.

This article presents a hybrid data-driven method, termed moving average-Hankel-dynamic mode decomposition (MAHankDMD), for joint direction of arrival (DOA) and frequency estimation in environments affected by both radio frequency interference (RFI) and Gaussian white noise. The proposed approach integrates two key components: 1) a moving average-dynamic mode decomposition (DMD) filter that effectively mitigates Gaussian white noise and separates RFI from the source signal, and 2) a Hankel-DMD method that accurately estimates the DOA of the filtered signal and associates it with the corresponding frequency. The moving average-DMD stage first enhances the signal-to-noise ratio and improves the robustness of the estimation process through noise and inference mitigation, while the subsequent Hankel-DMD stage enables reliable parameter extraction even for overlapping sginals or strong interference conditions. Numerical simulations demonstrate the robustness of moving average-Hankel-DMD (MAHankDMD), showing its ability to precisely estimate both DOA and frequency under challenging conditions involving RFI and Gaussian white noise interference. The proposed algorithm thus provides an effective solution for channel parameter estimation in complex noisy environments.

The two-port shunt configuration is often heralded as the gold standard for low-impedance measurements. However, this measurement method is not without its own issues. The shunt configuration inherently creates a ground loop between the measurement device's reference plane and the reference of the device under test (DUT). Additionally, probes often must be oriented in such a way that allows inductive coupling to occur. This work uses microprobes to measure the shunt impedance of a non-ideal short to explore the limitations of this measurement method in terms of both frequency and impedance. It highlights the importance of ground loop isolation and common-mode rejection ratio (CMRR) as well as limitations resulting from the mutual inductance between probes. The conclusions of these experimental measurements aim to find a path to further optimize shunt impedance measurements for power distribution networks (PDNs) as the industry points towards further lowering PDN impedance to meet increased current demands.

Estimating channel parameters such as azimuth, elevation, Doppler shift, and delay is a key challenge in wireless communication, often formulated as a multidimensional harmonic retrieval (MHR) problem. To address this, we propose a high-order dynamic mode decomposition (HODMD) framework for robust frequency estimation from high-dimensional signals in noisy environments. The HODMD approach combines high-order singular value decomposition (HOSVD) to decompose tensor data into a core tensor and mode matrices, with dynamic mode decomposition (DMD) to extract frequencies from the imaginary parts of the DMD eigenvalues. Simulation examples validate the effectiveness of the proposed method, demonstrating its efficiency in solving MHR problems.

Epstein-Barr nuclear antigen 1 (EBNA1), a sequence-specific DNA binding protein of Epstein-Barr virus (EBV), is essential for viral genome replication and maintenance and is therefore an attractive target for the therapeutic intervention of EBV-associated cancers. Several EBNA1-specific inhibitors have demonstrated the ability to block EBNA1 function in vitro, but practical delivery strategies for these inhibitors in vivo are still lacking. Here, we report an intelligent hierarchical targeting theranostic nanosystem (denoted as mZGOCS@MnO2-P5) that integrates an azide (N3) terminal dual-targeting peptide (N3-P5), a tumor microenvironment-responsive degradable MnO2 nanosheet, and a mesoporous ZnGa2O4:Cr3+, Sn4+ near-infrared persistent luminescence (NIR-PL) nanosphere (mZGOCS). In our design, mZGOCS@MnO2-P5 enables primarily targeting of the EBV-specific oncoprotein LMP1 (an EBV-encoded transmembrane protein) via the LMP1 targeting motif within P5. Once internalized into cells, the MnO2 nanosheet would be degraded in the acidic and reducing tumor microenvironment, simultaneously releasing P5 and recovering the NIR-PL of ZnGa2O4:Cr3+, Sn4+ initially quenched by the MnO2 nanosheet, thereby providing an autofluorescence interference-free NIR-PL imaging signal for monitoring the delivery efficacy of P5. The released P5 can secondarily target EBNA1 via the EBNA1 binding motif, blocking its function and thus inhibiting the growth of EBV-positive tumors. The feasibility of our developed hierarchical targeting theranostic nanosystem is well demonstrated both in vitro and in vivo, highlighting the huge translational potential of mZGOCS@MnO2-P5 in EBV-associated cancer therapy.

A electromagnetic-circuital-thermal-mechanical multiphysics numerical method is proposed for the simulation of microwave circuits. The discontinuous Galerkin time-domain (DGTD) method is adopted for electromagnetic simulation. The time-domain finite element method (FEM) is utilized for thermal simulation. The circuit equation is applied for circuit simulation. The mechanical simulation is also carried out by FEM method. A flexible and unified multiphysics field coupling mechanism is constructed to cover various electromagnetic, circuital, thermal and mechanical multiphysics coupling scenarios. Finally, three numerical examples emulating outer space environment, intense electromagnetic pulse (EMP) injection and high power microwave (HPM) illumination are utilized to demonstrate the accuracy, efficiency, and capability of the proposed method. The proposed method provides a versatile and powerful tool for the design and analysis of microwave circuits characterized by intertwined electromagnetic, circuital, thermal and stress behaviors.

Time-resolved electromagnetic near-field scanning plays a pivotal role in antenna measurement and unraveling complex electromagnetic interference and compatibility issues. However, the rapid acquisition of high-resolution spatio-temporal data remains challenging due to physical constraints, such as moving the probe position and allowing sufficient time for sampling. This article presents a novel hybrid approach combining Kriging for sparse spatial measurement, compressed sensing (CS) for sparse temporal sampling, and dynamic mode decomposition (DMD) for comprehensive analysis of the dual-sparse sampling electromagnetic near-field data. We leverage CS to optimize sparse sampling in the time domain and Latin hypercube sampling to guide the probe position and realize sparse measurement in the space domain. By leveraging the inherent sparsity within electromagnetic radiated signals, CS reliably represents time-domain signals while reducing the required time samples. Then, DMD is used to extract meaningful insights from the resulting sparse spatio-temporal data, resulting in the sparse dynamic modes and temporal evolution information. Next, the Kriging is employed to infer missing spatial measurements for each sparse dynamic mode. Finally, the entire spatio-temporal signals are reconstructed based on interpolated dynamic modes and temporal evolution information. An example using crossed dipole antennas as the device under test is provided to validate the proposed method. It is found that the proposed Kriging-CS-DMD framework effectively reconstructs electromagnetic fields with precision while simultaneously reducing the measurement workload in both the time and space domains. This methodology could be further employed for various applications, such as space-time-modulated electronic devices.

The suppression of the large resonance peak that may appear due to the equivalent parallel circuit between the package capacitance and PCB inductance is discussed. Such resonance may be amplified if the decoupling capacitors are not appropriately selected. The relevant parameters involved in the PDN design and a feasible solution strategy are presented based on the identification of a simplified equivalent circuit that is able to replicate the resonant behavior. The optimization of the relevant parameters of such circuit are able to suggest the best strategy for identifying the decoupling capacitors with appropriate values of parasitic inductance and resistance.

This paper proposes a novel approach to uncertainty quantification (UQ) in the partial equivalent element circuit (PEEC) method through a physics-informed neural networks (PINNs)-based polynomial chaos expansion (PCE) scheme. Initially, the PEEC method is formulated via the electrical field integral equations and continuity equations. Subsequently, random parameters are introduced to construct stochastic equations, generating input and output observations for training data. The PCE method is then employed to establish a mapping function. To calculate the coefficients of polynomial bases, a PINNs-based method is applied, utilizing the constructed matrix derived from the training data. Finally, this methodology enables the determination of stochastic parameters for quantities of interest within the PEEC method. The numerical example involving a transmission line is provided to verify the efficiency of the proposed method. It is found that the uncertainty is well quantified in all cases.

This paper presents a data-driven methodology that utilizes Dynamic Mode Decomposition (DMD) for the time-domain (TD) electromagnetic (EM) modeling of microwave devices. As an unsupervised machine learning technique, DMD leverages a limited set of unlabeled spatio-temporal electromagnetic (EM) data to determine DMD eigenvalues and eigenmodes. Then, the obtained DMD model reconstructs the dynamics as a series of exponential terms based on linear assumptions. The effectiveness of this approach is demonstrated through the TD EM modeling of photonic crystal waveguides. Comparative analysis with the finite-difference time-domain (FDTD) method shows that the DMD model not only achieves precise modeling but also facilitates robust short-term forecasting.

A novel broadband dual-polarised transmitarray antenna (TA) utilising tightly coupled cross dipole cells is proposed in this work. The transmitarray cell using the tightly coupling wideband principle comprises two radiation patches designed as two orthogonal planar dipoles with four interdigital capacitors, two meandered parallel plate transmission lines, and a ground. Each cell has a square shape and a dimension of approximately 0.28 lambda c ${\lambda }_{\mathrm{c}}$ where lambda c ${\lambda }_{\mathrm{c}}$ is the wavelength at central frequency 5.5 GHz. The transmitarray cell can achieve 475 degrees phase shift at central frequency and transmission magnitude better than -2.5 dB within the working band. To verify the feasibility of this design, a tightly coupled dual-polarised transmitarray antenna (TCDPTA) is modelled and manufactured. The transmitarray aperture size is approximately 4.1 lambda c ${\lambda }_{\mathrm{c}}$ x ${\times} $ 4.1 lambda c ${\lambda }_{\mathrm{c}}$. The simulation and measurement illustrate that the TCDPTA has stable and distortion-free main beams whose side lobe levels are generally below -10 dB in the band of 3.0-8.0 GHz. The measured gain at central frequency is 16.2 dBi and peak gain is 19 dBi at 7.5 GHz. The working bandwidth is 90.9% and 3 dB gain bandwidth is 66.7%. The measured cross-polarisation levels are below -15 dB at axial direction. This TA has potential applications for high-date-rate communication and high-revolution radar imaging systems at C-band.

This article proposes a novel deep-learning (DL)-based source reconstruction method (SRM). The proposed DL-based SRM employs the deep convolutional conditional generative adversarial network (DCCGAN), which only demands one-transmitter single-frequency far-field measurement on electromagnetic (EM) scattered field as its input and further predicts the equivalent source on target scatterers. The proposed DCCGAN includes the generator ( $G)$ with an EM forward simulator and the corresponding discriminator ( $D)$ , both consisting of the complex-valued deep convolutional neural networks (DConvNets). During the offline training, the generator learns the distribution between the measured scattered field data and the corresponding equivalent source on target scatterers, while the discriminator determines whether the presented equivalent sources are real or fake. Therefore, the proposed DCCGAN can generate the unknown equivalent source from measured scattered field data, by learning the distribution between the known equivalent sources and their corresponding field. Furthermore, the proposed DL-based SRM can overcome the limitation of conventional methods, involving high computational cost and strong ill-conditions. Consequently, the proposed DL-based SRM can realize the reconstruction of the equivalent source with higher accuracy and lower computation complexity. Numerical examples have demonstrated the feasibility of the proposed DL-based SRM, which opens the new path for DL-based EM computation approaches.

In this paper, we propose to use a deep learning approach to detect unit failure in array antennas. Due to natural machine life cycle and/or unexpected accidents, antenna units unavoidably suffer from the risk of failure, leading to the deterioration of array performance. To realize the detection of unit failure, the far-field radiation patterns are used as the input of the deep convolutional neural network (DConvNet) for antenna array diagnosis learning. The proposed DConvNet consists of continuous functional groups of convolution, batch normalization, and activation layers, followed by a fully connected layer to realize recognition, i.e., the fault diagnosis of antenna array. Different from conventional diagnosis techniques, the main advantage of the proposed deep learning approach does not require intensive computations based on Green’s function. The training data are collected by the electromagnetic simulation tool. Additionally, the Gaussian noise is added to the training data to imitate the interference in real application scenarios. The proposed DConvNet for array diagnosis is verified by three numerical benchmarks and demonstrates that it can diagnose antenna array in a complex environment with generality.