A diagnosis method of nonuniform and unmagnetized plasma parameters based on microwave reflection extremum algorithm (MREA) is proposed in this article. The continuous spectrum and extremal data of the plasma in microwaves are analyzed by the MREA to retrieve the parameters of the plasma. The MREA is composed of diagnostic process of plasma (DPP) and calibration of diagnostic results (CDRs). In the DPP, the plasma parameters are diagnosed by different extremes of the microwave signal, which is obtained by placing the plasma in the near field of the dual horn. In the CDR, the diagnostic accuracy of the DPP is verified and corrected by comparing with the preset value. The effectiveness and accuracy of the MREA are evaluated through simulation experiments. The simulated results show that the average diagnosis of electron density is 95.3%, the average diagnosis of collision frequency is 83.56%, and the average diagnostic accuracy of plasma overall parameters is 89.43%.

In this article, a physics-informed surrogate model (PISM) is proposed that integrates a transfer matrix circuit topology model (TM-CTM) for the efficient prediction of electromagnetic (EM) transmission characteristics in complex plasmas. The reliance on EM software for simulated plasma is eliminated by modeling the plasma using the TM-CTM. The strength Pareto evolutionary algorithm 2 (SPEA2) is employed to rapidly determine circuit component values, thereby enhancing simulation efficiency and generating a comprehensive dataset. Subsequently, a PISM is trained to map plasma parameters directly to S-parameters. The PISM effectively replaces time-consuming full-wave simulations. Validation results demonstrate that the PISM enables the rapid prediction of $S_{21}$ amplitude and phase, serving as a robust framework for analyzing complex plasma sheaths.

A high-gain antenna with a reconfigurable plasma metasurface (RPM) is proposed, which enables dual-functionality with high-gain transmission and reflection radiation. The metasurface is composed of a 29x29 array of sealed glass containers filled with inert gas. Plasma is generated through ionization by electrode plates positioned on both sides of the RPM. A reconfigurable plasma cell (RPC) is designed, exhibiting a dielectric loss below 3 dB and a 360 degrees phase response coverage in both transmission and reflection modes. Furthermore, by controlling the on/off state of the top electrode plate, flexible switching between the transmission and reflection modes of the RPM is achieved at the same frequency and polarization. Simulation results demonstrate that the proposed RPM antenna achieves low- sidelobe and high-gain radiation in both modes. In 16-20 GHz, the RPM can provide full-space radiation coverage, achieving a peak gain of 29.5 dBi in transmission mode and 30.3 dBi in reflection mode. The -3-dB gain bandwidths for the two modes are 22.2% and 19.4%, respectively. The maximum aperture efficiencies are 21.0% and 20.2%, respectively.

Radar cross section (RCS) computation of dynamic targets plays a crucial role in target recognition and battlefield awareness. However, traditional high-frequency methods are primarily designed for static targets and struggle to capture the real-time impact of attitude changes on scattering characteristics, with limited capabilities in result visualization. To address these challenges, this paper proposes a dynamic scattering modeling approach based on the Graphic Electromagnetic Computing (GRECO) framework. The proposed method integrates GPU rendering with off-screen framebuffer techniques to extract surface geometry and normal information in real time. Furthermore, ray tracing based on the Shooting and Bouncing Ray (SBR) method is employed to construct multiple scattering paths. A hybrid Geometrical Optics-Physical Optics (GO-PO) algorithm is then used to compute the RCS contributions from multiple scattering events in complex targets. Dynamic simulation results demonstrate that the method effectively handles dynamic scattering in complex multi-target scenarios. Within an elevation angle range of 60°-100°, the average absolute error reaches 4.59%, providing a viable technical solution for dynamic RCS analysis involving multiple targets.

This paper proposes a method for estimating the parameters of precessing cones based on the Genetic Algorithms (GA). The parameter dimension is reduced by the method of time-shift superposition, GA is introduced to achieve accurate estimation of smooth cone parameters. Firstly, a mathematical model of micro-Doppler signals is constructed for the cone target. The multi-component radar echo signals are separated using Independent Component Analysis (ICA), and subsequently, the separated signals are processed to extract the instantaneous micro-Doppler frequency features of the cone. Then, the rotation center position is removed by time-shifting the micro-Doppler frequency at the cone bottom, thereby reducing the parameter search dimension. Finally, after eliminating the influence of the rotation center, GA is utilized to optimize the remaining parameters, aiming to minimize the error between the theoretical model trajectory and the extracted feature trajectory, thus achieving accurate parameter estimation. Simulation results indicate that the proposed method achieves high accuracy and robustness in estimating micro-motion parameters of Precession cone targets. Compared to the traditional grid search method, it demonstrates improvements in estimation accuracy, stability, and efficiency.

This paper proposes a microbump patch antenna embedded with shorting wall. The antenna features a hemispherical microbump on the radiating patch, which significantly improves radiation efficiency. Two symmetrically placed shorting walls are introduced to suppress orthogonal radiation components, reduce cross-polarization (XP), and enhance the co-polarization to cross-polarization isolation (CP-XP isolation). The proposed design is optimized to improve gain, radiation efficiency, and polarization purity, thus ensuring better overall performance. Simulation results demonstrate that the proposed antenna exhibits high gain, excellent radiation efficiency, and outstanding polarization purity. Based on simulation results, a radiation efficiency of up to 97% and a polarization purity of 25.7 dB have been achieved by the proposed antenna. Additionally, the main lobe gain has been enhanced by 1.4 dB. These outcomes indicate that the antenna can be characterized as a high-performance antenna.

As the complexity of defense systems increases, similarity assessment methods have become crucial for optimizing decoy design and enhancing defense system performance. Traditional methods often rely on statistical features to analyze the fluctuation characteristics of RCS sequences, which limits a deeper understanding of the data and its underlying patterns. Because of these limitations, this paper proposes a novel Radar Cross Section (RCS) similarity evaluation method based on deep learning. First, five warhead and decoy models are constructed, and their RCS curves are simulated, which are then transformed into 2D images using the Gramian Angular Difference Field method. Following this, a similarity evaluation method is introduced, which utilizes an RCS-based target classification network with a hierarchical encoder-decoder architecture and DEConv layers to capture high-frequency information. The network is specifically designed to assess the RCS similarity by extracting feature vectors and calculating their similarities. Experimental results demonstrate the effectiveness of this method in both decoy evaluation.

Detecting multiple human targets in indoor scenarios using ultra-wideband (UWB) radar usually involves false detection results caused by the secondary reflections, which might reduce the target detection accuracy and cause a more severe deterioration when the number of targets increases. This article proposed a two-step accuracy improvement method for multitarget detection in environments with multiple human targets of more than three and strong secondary reflections by the surroundings, especially the walls. Based on the rough detection results acquired by the modified CA-CFAR (MCA-CFAR) processing, the first step achieves the primary false alarm suppression using a short-window accumulation in the time domain. Then, the second step applies the decision confidence on the detection results from the first step to assess the reliability of results for improved accuracy. The two-step accuracy improvement could thus have a higher accuracy through cascading false alarm suppression. The effectiveness and accuracy of the proposed algorithm are verified based on the experimental results.

The High Impedance Surface (HIS) or near-zero refractive index metamaterials are used extensively to enhance the antenna performance or radiate directly. However, the radiation characteristic of negative permeability metamaterial has not been focused. We verify the corresponding radiation mode by using the negative-permeability split-ring resonator (SRR), where the related radiation modes are the same as the resonant modes. And a modified rectangular waveguide is fed to the unit cells to indicate the radiation performance.

In this paper, a method for broadband and wideangle active cancellation stealth is proposed to address the issue of radar angle measurement errors. The method involves utilizing a phase shifter and an amplifier connected to the antenna in order to control the re-radiation field of the antenna and cancel out the scattering field of the target. Particle swarm optimization (PSO) is employed to calculate the values of phase shifters and amplifiers. The effectiveness of this method in achieving broadband and wide-angle active cancellation is confirmed through two monostatic simulation cases. Specifically, when using a flat plate as a case, it is demonstrated that the reduction in radar cross section (RCS) exceeds 10dB with a frequency width greater than 35MHz and an angle width greater than 15°. In the case of a missile, similar results are achieved with RCS reducing by 10dB when the frequency width exceeds 32MHz and the angle width exceeds 5°.

In the application of the wireless power transfer process for the space solar power station, the continuous exposure to the electromagnetic wave could lead to significant thermal effect on human health. Therefore, this paper investigates the accumulated electromagnetic radiation at 5.8 GHz by the temperature rise on a practical arm model which comes from a Chinese female (26-year-old, 162 cm high, and 50kg weight) and the physiological condition setting is included. By comparing the external electromagnetic distributions on the body surfaces facing and facing away from the illumination source, the corresponding differences of two benchmark lines are obtained for both the simulated and measured situations where the normalized values are close in the whole band. Through this indirect analysis consistency for the surface field distribution, the interior field distributions of three different layers (skin, fat, and muscle) are obtained depending on the simulated results. Then, the temperature rise effects are evaluated of which the fat has the highest temperature rise in a short long time than those of the other two layers, and it is the energy source to make all the layers eventually turn to be relatively steady for the 2-hour period.

A novel method for efficiently and accurately predicting the scattering response of coated targets is proposed. By using the scattering center model based on the geometrical theory of diffraction (GTD) as the transfer function, the scattering characteristics of coated targets could be predicted from those of PEC targets. First, a scattering center parameters extraction algorithm based on particle swarm optimization (PSO) is designed to obtain the scattering characteristics of PEC targets. Then, the reflection coefficient is calculated based on the electromagnetic parameters and thickness of the material, and the scattering response of coated targets is predicted by combining it with the scattering center model. Through analysis, the mean absolute error (MAE) of the proposed method for cube and the Slicy targets is 1.96% and 4.1% respectively. Furthermore, the proposed method saves 90% to 98% of the simulation time compared to the calculation time of commercial solver. The results demonstrate that the proposed method effectively enables accurate and rapid prediction of the scattering response of coated targets.

In this paper, the fractional area, which is the ratio of the area of metal surrounded to the substrate area for one cell, of the metamaterial cells is studied to find the relation between the gain enhancement and the fractional area, where the distance away from patch antenna is lambda(0)/2. Two cases of the split ring resonator are compared, one is that the structure and the relative size are kept while increases the cell number by increasing the fractional area, and the other is kept the fractional area but with different metal area and structure for two different cells of the same resonances. We find the gain doesn't always increase with the cell number, it will decrease when the fractional area is up to a certain degree; and the smaller cells, the higher gain enhancement.

In this paper, a microwave reflection method based on fusion algorithm is proposed to diagnose the peculiarity of plasma. This fusion algorithm is composed of genetic algorithm (GA) and neural network (ANN), that is GA-ANN. Only a single horn antenna is used to emit spherical waves to the preset plasma, and the scattering peculiarity of the horn antenna (impedance bandwidth, resonance point and gain) and the corresponding plasma characteristics (characteristic frequency, collision frequency) are obtained. The weight and bias of ANN are optimized by GA screening mechanism. Then GA-ANN is used to establish the mapping relationship between the scattering characteristics of antenna and plasma, which is finally used to diagnose the unknown plasma characteristics. The experimental results show that the relative error of plasma diagnosis using GA-ANN is 0.7 lower than that using ANN alone.

With a rather strong environment clutter and multi-target interference, while indoor monitoring using FWD radar in complex scenarios, the current CA-CFAR algorithm would inevitably introduce the variance peaks leading to the false detection results in multi-target detection. This paper proposed a Modified CA-CFAR based on the Sliding Window DC Removal and Adaptive Threshold method, using Sliding Window DC Removal to eliminate the static environment clutter to avoid the previous variance peaks involved by replacing the MTI and achieving the Adaptive Threshold through echo statistic estimation to reduce the false detection results. The effectiveness of the proposed algorithm has been verified through comparative analysis based on radar-measured data.

The knowledge of heart and respiratory rates (HRs and RRs) is essential in assessing human body static. This has been associated with many applications, such as survivor rescue in ruins, lie detection, and human emotion detection. Thus, the vital signal extraction from radar echoes after pre-treatments, which have been applied using various methods by many researchers, has exceedingly become a necessary part of its further usage. In this review, we describe the variety of techniques used for vital signal extraction and verify their accuracy and efficiency. Emerging approaches such as wavelet analysis and mode decomposition offer great opportunities to measure vital signals. These developments would promote advancements in industries such as medical and social security by replacing the current electrocardiograms (ECGs), emotion detection for survivor status assessment, polygraphs, etc.

Human vital signs such as respiratory and heart rates (RR and HR) are reflections of human body status, which are widely applied to rescuing or clinical estimation. In this study, an improved algorithm combining the Variational Mode Decomposition (VMD) and empirical wavelet transform (EWT) is proposed for human vital sign signal monitoring using Impulse Radio Ultra-Wide Band (IR-UWB) radar. The VMD processes the initial data and decomposes it into sub-signals to obtain the RR component. The HR-related ones with intermodulation components, determined by the power of the signal in the corresponding frequency band of HR, then become the input of the EWT. The processing results of the combined algorithm using both measured and simulated data reveal high accuracy.

An H-plane sliced rectangular patch antenna with two steps is proposed for wearable antenna. It comes from the conventional patch antenna fed by a coaxial line but the patch is sliced into 3 pieces along the H-plane direction and two steps with the same spacing are formed. The piece far away from the feed receives power by coupling from the middle piece of the top layer, which is connected to the feeding cable. The overlap area is large enough for the effective coupling, while the other piece receives power by the conduction current without overlap area where 11 posts are used to connect to the middle piece. The stepped antenna maintains almost the same characteristics with those of the conventional.

超表面因具有独特的操控电磁波的功能被广泛使用.但超表面设计需要优化单元结构以产生所需相位分布,这需要大量的时间和计算资源.提出了一种基于迁移学习的超表面设计方法,该方法可以预测复杂超表面单元的相位.使用ResNet34作为预训练模型的迁移学习网络,其中仅使用20000个样本进行训练网络,其准确率达到90%以上.模型实现了预测210×10不同结构的超表面单元的相位,与深度学习网络相比,减少训练样本且提升了准确率.为了验证该方法的有效性和可行性,在实际设计需求下成功设计了一个异常反射在45°角度的相位梯度超表面,并进行了仿真和实验验证.为超表面设计提供了一种新颖高效的方法,并为迁移学习在电磁领域内其他问题上的应用提供了参考.

The requirements of 5G/6G promote progress in the miniaturization of the antenna array, which promotes the development of a closely spaced decoupling technique. However, the present techniques face the common problem of beam tilt if the spacing is close. Thus, a pattern-corrected, closely-spaced technique is proposed in this paper for the two patch antennas with the λ0/20 edge-to-edge distance of the H-plane. The corresponding structure, which is inserted at the center of the spacing, consists of a vertical wall with a single substrate and two symmetrical T-type metals, and a slot at the center is reserved to adequately accommodate the vertical wall. The vertical strip at the other end of the T-type metal is connected to the ground of the patch antenna, while the parallel strip is placed exactly above the patch substrate. After an exact analysis, a prototype was fabricated and measured, and the results showed that the measurements agreed well with those of the simulations, the decoupling coefficients in the 5.8 GHz band were below −20 dB, and the measured radiation pattern at 5.81 GHz was corrected to the broadside from 28° and the maximum realized gain was 5.30 dB.