Xianliang Wu

Application of Compressed Sensing to solving 3D electromagnetic scattering problems over a wide angle

This paper introduces Compressed Sensing (CS) into Method of Moments (MoM) to solving 3D electromagnetic scattering problems over a wide angle. In traditional MoM, the calculation of the problem over a wide angle is implemented with iterative operations at a finer angle step, which must results to low efficiency. After applying CS technique to MoM results over any angles can be calculated with only several measurements. This method can be implemented by the following steps: obtaining measurements of induced currents by MOM with a sparse transform, and finally reconstructing the real induced currents by a recovery algorithm. Numerical examples showed that the results computed by CS could keep high accurate with low CPU time consumed.

A new solution of Schrödinger equation based on symplectic algorithm

An explicit three dimensional symplectic finite-difference time-domain method is introduced to solve the Schrodinger equation. The method is obtained by using a symplectic scheme with a propagator of exponential differential operators for the time direction approximation and fourth-order collocated finite differences for the space discretization. Firstly, a high-order symplectic framework for discretizing the Schrodinger equation is presented. Secondly, the comparisons on numerical stability, dispersion are also provided between the new symplectic scheme and other commonly used methods. Finally, numerical examples are given to evaluate the performance of the proposed method and the advantages on the accuracy and stability are also further demonstrated.

Study of periodic dispersive structures at oblique incidence by split-field FDTD method

In this paper, the periodic structures are simulated by the split-field finite difference time domain (FDTD) method. By using this method, a set of auxiliary elements are introduced to represent the discretization of the Floquet transformed Maxwells equations with oblique incidence. We here extend the split-field method to the study of 2D dispersive structures of Drude model combining with the Z transformation method. Some numerical experiments are presented to validate the code, and the obtained results are in good with the analytic results.

Optimized Operator-Splitting Methods in Numerical Integration of Maxwell's Equations

Optimized operator splitting methods for numerical integration of the time domain Maxwells equations in computational electromagnetics (CEM) are proposed for the first time. The methods are based on splitting the time domain evolution operator of Maxwells equations into suboperators, and corresponding time coefficients are obtained by reducing the norm of truncation terms to a minimum. The general high-order staggered finite difference is introduced for discretizing the three-dimensional curl operator in the spatial domain. The detail of the schemes and explicit iterated formulas are also included. Furthermore, new high-order Pade approximations are adopted to improve the efficiency of the proposed methods. Theoretical proof of the stability is also included. Numerical results are presented to demonstrate the effectiveness and efficiency of the schemes. It is found that the optimized schemes with coarse discretized grid and large Courant-Friedrichs-Lewy (CFL) number can obtain satisfactory numerical results, which in turn proves to be a promising method, with advantages of high accuracy, low computational resources and facility of large domain and long-time simulation. In addition, due to the generality, our optimized schemes can be extended to other science and engineering areas directly.

A 0.6–2.4 GHz Broadband GaN HEMT Power Amplifier with 79.8% Maximum Drain Efficiency

A highly efficient and broadband 10 W GaN HEMT power amplifier (PA) is presented, which employs the hybrid PA mode, transferring between continuous Class-F, continuous Class-B/J and continuous inverse Class-F. A GaN PA is designed and realized based on this mode-transferring operation using low-pass filter output matching network. The maximum theoretical efficiency of this hybrid continuous modes PA is more than 78.5%. Specifically, the operating bandwidth is determined by the low pass filter output matching network and the theoretical bandwidth can achieved multi-octave. The proposed design strategy is experimentally verified by a 0.6–2.4 GHz PA design with 79.8% maximum drain efficiency and 10 W output power. The footprint of the fabricated PA is 75 mm (times ) 40 mm.

Comparative study on recovery algorithms for solving electromagnetic scattering problems by compressive sensing

A new incident source is formed based on compressive sensing (CS) theory, in which much information from different incident angle is included. By solving the integral equations under this new incident source, one can obtain the measurements of the induced currents without calculating the problems with a finer angle increment. Especially, different recovery algorithms of CS are studied as used to solve electromagnetic problems over a wide incident angle. Numerical simulation results have shown that the efficiency of the new method can be greatly improved, since the total number of measurements required can be reduced drastically by choosing suitable recovery algorithm.

Numerical simulation of GPR based on 3-D high-order FDTD

The high-order finite-difference time-domain (HO-FDTD) technique is used in the simulation of ground-penetrating radar modeling in three dimensions (3-D), which can improve accuracy and reduce the error caused by numerical dispersion effectively. To absorb waves reflected from edges we implement convolutional perfectly matched layer (CPML) absorbing boundaries. It can efficiently absorb the reflections and greatly increase the computation efficiency. The surface-based reflection and cross-hole GPR modeling are simulated, and numerical results show the efficiency of the method.

Design of image-reject hairpin filter applied for Ku-band LNB

An image-reject hairpin bandpass filter used in Ku-band low noise block (LNB) is designed, and optimized in the paper. The designed detail of the filter is explained and the image-reject is illustrated theoretically. According to the research, under the condition of reducing the size as small as possible, the hairpin filter used in Ku-band LNB we designed not only can satisfy the demands from bandwidth and return loss, but also has good image-reject. In the end, the results of the simulation and measurement are presented, which prove the feasibility of our design.