Hu Jun

University of Electronic Science and Technology of China, Chengdu, Sichuan Sheng, China

Are you Hu Jun?

Claim your profile

Publications (27)2.47 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, two novel strategies are proposed to determine the multipole number of multilevel fast multipole algorithm (MLFMA). With the modified multipole number, the computation of RCS of 3D electrically large structures is realized. Compared with traditional MLFMA, The proposed methods in this paper can reduce the multipole number, CPU time and the memory requirement. Numerical results show that the modified methods have the complexity of O(NlogN) both for the computation of matrix-vector multiplication and the memory requirement, thus yields more efficiency for scattering problems of 3D electrically large structures.
    Radar Conference, 2009 IET International; 05/2009
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: As the fastest integral equation solver to date, the multilevel fast multipole algorithm (MLFMA) has been applied successfully to solve electromagnetic scattering and radiation from 3D electrically large objects. But for very large-scale problems, the storage and CPU time required in MLFMA are still expensive. Fast 3D electromagnetic scattering and radiation solvers are introduced based on MLFMA. A brief review of MLFMA is first given. Then, four fast methods including higher-order MLFMA (HO-MLFMA), fast far field approximation combined with adaptive ray propagation MLFMA (FAFFA-ARP-MLFMA), local MLFMA and parallel MLFMA are introduced. Some typical numerical results demonstrate the efficiency of these fast methods.
    Journal of Systems Engineering and Electronics 01/2008; 19(2):252-258. · 0.38 Impact Factor
  • Rui Xi, Hu Jun, Nie Zaiping
    [Show abstract] [Hide abstract]
    ABSTRACT: To speed up the solution of scattering from multiple conducting objects with complex structures, a hybrid multi-level fast multi-pole algorithm with a generalized forward-and-backward method is developed. The region involving multiple conducting objects is divided into finite uniform sub-regions. The induced current in each sub-region is solved iteratively by a generalized forward-and-backward method. The interactions between sub-regions and the iterative current in each sub-region are computed by the multi-level fast multi-pole algorithm. The present method has only the memory requirement of O(NmaxlogNmax), where Nmax is the maximum number of unknowns located in these sub-regions. Numerical results demonstrate the efficiency and accuracy of the present method for multi-object scattering problems.
    Electromagnetics 01/2008; 28(8):572-581. · 0.73 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A full-wave analysis of the electromagnetic problem of a three-dimensional (3-D) antenna radiating through a 3-D dielectric radome is preserued. The problem is formulated using the Poggio-Miller-Chang-Harrington-Wu(PMCHW) approach for homogeneous dielectric objects and the electric field integral equation for conducting objects. The integral equations are discretized by the method of moment (MoM), in which the conducting and dielectric surface/interfaces are represented by curvilinear triangular patches and the unknown equivalent electric and magnetic currents are expanded using curvilinear RWG basis functions. The resultant matrix equation is then solved by the multilevel fast multipole algorithm (MLFMA) and fast far-field approximation (FAFFA) is used to further accelerate the computation. The radiation patterns of dipole arrays in the presence of radomes are presented. The numerical results demonstrate the accuracy and versatility of this method.
    Journal of Systems Engineering and Electronics 01/2008; 19(1):81-87. · 0.38 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper introduces the research work on the extension of multilevel fast multipole algorithm (MLFMA) to 3D complex structures including coating object, thin dielectric sheet, composite dielectric and conductor, cavity. The impedance boundary condition is used for scattering from the object coated by thin lossy material. Instead of volume integral equation, surface integral equation is applied in case of thin dielectric sheet through resistive sheet boundary condition. To realize the fast computation of scattering from composite homogeneous dielectric and conductor, the surface integral equation based on equivalence principle is used. Compared with the traditional volume integral equation, the surface integral equation reduces greatly the number of unknowns. To compute conducting cavity with electrically large aperture, an electric field integral equation is applied. Some numerical results are given to demonstrate the validity and accuracy of the present methods.
    Journal of Systems Engineering and Electronics 01/2008; · 0.38 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Based on the distance symmetry of the translation operators in multi-level fast multipole algorithm (MLFMA), this paper presents a one-dimension local interpolation algorithm for fast computation of translators. Theoretical analysis and numerical results show that the presented method is about three times faster than the two-dimension interpolation method.
    01/2007;
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a parallel modified electric field integral equation (MEFIE) with GMRES solver. GMRES is applied in solving the MEFIE, which improves the computing efficiency greatly. Then, a parallel scheme for MEFIE is developed and conjuncted with the parallel GMRES and MLFMA. Numerical results show the strong ability of the proposed parallel scheme in solving scattering from electrically large open structures.
    01/2007;
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, a novel strategy of the multipole numbers of the MLFMA -modified multipole number method - is presented for solving the electromagnetic scattering problem of the electrically large PEC targets. The truncation error of the Green's functions is one of the dominating errors in FMM and MLFMA. The modified method greatly reduces the multipole numbers, and then the CPU time need for the iterative is greatly reduced. Also, this method keeps the high accuracy of the traditional MLFMA.
    Microwave Conference Proceedings, 2005. APMC 2005. Asia-Pacific Conference Proceedings; 01/2006
  • [Show abstract] [Hide abstract]
    ABSTRACT: A modified electric field integral equation (MEFIE) is proposed to realize fast and accurate solution of 3D electromagnetic scattering. By adding the principle value term of magnetic field integral equation (MFIE) operator into original electric field integral equation (EFIE) operator, a well-conditioned modified EFIE operator is constructed. To avoid the loss of accuracy, a appropriate iterative technique with stable convergence is developed. The present method reduces greatly the condition number of EFIE, attains much faster convergence than EFIE. And, a reasonable accuracy is also achieved by few iterations. Numerical results demonstrate the validity and efficiency of the MEFIE.
    Microwave Conference Proceedings, 2005. APMC 2005. Asia-Pacific Conference Proceedings; 01/2006
  • [Show abstract] [Hide abstract]
    ABSTRACT: Fast multipole algorithm (FMA) based on method of moment (MOM) can speed up the computation and reduce the storage requirements and computational complexity. But it is still a difficult problem to simulate RCS of extremely electrically large objects, while physical optics (PO) is useful. In this paper, PO and FAM was combined to get both accuracy and efficiency. And CRWG basis is developed on the model made by commercial software named ANSYS.
    Microwave Conference Proceedings, 2005. APMC 2005. Asia-Pacific Conference Proceedings; 01/2006
  • [Show abstract] [Hide abstract]
    ABSTRACT: The solution of the electromagnetic scattering from target with electrically large aperture cavity is very important and challenging. The electric field integral equation (EFIE) with partial coupling principle presented in this paper to compute equivalent magnetic current on aperture surface is easy to understand and can expand the aperture electrical size greatly. Furthermore, a new usage of multilevel fast multipole algorithm (MLFMA) speeds up the computation.
    Microwave Conference Proceedings, 2005. APMC 2005. Asia-Pacific Conference Proceedings; 01/2006
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: As the fastest integral equation solver up to now, multilevel fast multipole algorithm (MLFMA) has been applied successfully to solve electromagnetic scattering and radiation from 3D electrically large object. But for very large scale problems, the storage and CPU time required in MLFMA are still expensive. In this paper, a local multilevel fast multipole algorithm (LMLFMA) is proposed to further speed up the efficiency of MLFMA in conjugate gradient (CG) iteration. In the LMLFMA, only the local interactions between the subscatters are taken into account. And, the interaction regions in iteration are varying adaptively with iterative current density. With decrease of iterative error, iterative current density tends to real one, the interaction regions required are diminishing. When the iterative error is less than a critical iteration error, only the interaction between nearby regions at the finest level is considered. Numerical results show that the LMLFMA has good accuracy, and much better efficiency than traditional MLFMA.
    Microwave, Antenna, Propagation and EMC Technologies for Wireless Communications, 2005. MAPE 2005. IEEE International Symposium on; 09/2005
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, a new method IPMLFMA is presented. This new method has been used to solve the scattering from the PEC plate; during very few iteration steps we can obtain the real solution of the current. Two typical numerical results prove the validity of the presented method.
    Antennas and Propagation Society International Symposium, 2005 IEEE; 08/2005
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A local multilevel fast multipole algorithm (LMLFMA) is proposed to further speed up the efficiency of the multilevel fast multipole algorithm (MLFMA), which is used to expedite the computation of matrix-vector multiplication in conjugate gradient iteration. In the present method, the coarsest level is determined by the iterative error. And at the coarsest level, the interaction between far regions is omitted. When the iterative error is less than the critical iteration error, only the interaction between nearby regions is considered. Numerical results show that the LMLFMA has a reasonable accuracy, and higher efficiency, compared with the MLFMA combined with the partly approximation iteration (PAI) technique. So it is very efficient for solution of scattering from 3D targets with electrically large sizes.
    Antennas and Propagation Society International Symposium, 2005 IEEE; 08/2005
  • [Show abstract] [Hide abstract]
    ABSTRACT: The characteristic basis function method (CBFM) is applied to analyze electromagnetic scattering problems. According to the concepts of grouping and high level effects, characteristic basis functions (CBFs) in different levels are constructed on large blocks and used to generate a size-reducing matrix equation. The paper gives some observations on CBFM and also proposes a new approach for generation of the CBF matrix equation, which can simplify the calculation of the CBF impedance matrix, important for reducing computational time and storage. Numerical results show the validation of the proposed approach.
    Computational Electromagnetics and Its Applications, 2004. Proceedings. ICCEA 2004. 2004 3rd International Conference on; 12/2004
  • [Show abstract] [Hide abstract]
    ABSTRACT: As the fastest numerical solver for 3D electromagnetic scattering up to now, the multilevel fast multipole algorithm (MLFMA) has excellent properties. The computational complexity and the storage requirement is O(NlogN) and O(N), respectively, for an N unknowns problem. For a given object, MLFMA has different properties and accuracy when the discretization density and the grouping technique of the MLFMA change. This work investigates in detail the computational complexity, the storage requirement and the accuracy of MLFMA, in the case of conducting sphere scattering. It shows good properties, including the complexity and accuracy which can be achieved when a suitable discretization density and grouping size are chosen.
    Computational Electromagnetics and Its Applications, 2004. Proceedings. ICCEA 2004. 2004 3rd International Conference on; 12/2004
  • [Show abstract] [Hide abstract]
    ABSTRACT: The MLFMA (multilevel fast multipole method) has been the most popular method in electronic field computation, but because of the limitations of computers, both in CPU speed and memory size, we still can not use this method to solve real targets. We introduce an MLFMA based on the MPI (message passing interface) in cluster computing.
    Computational Electromagnetics and Its Applications, 2004. Proceedings. ICCEA 2004. 2004 3rd International Conference on; 12/2004
  • [Show abstract] [Hide abstract]
    ABSTRACT: A multilevel fast inhomogeneous plane wave algorithm (MLFIPWA) is applied to realize fast and accurate solutions of electromagnetic scattering from a conducting object above half space. It reduces dramatically the computational complexity of the IEM for layered medium problems, and is suitable for solving not only objects with large varying longitudinal sizes, but also stratified media with arbitrary layers. Numerical results show the validity and efficiency of the MLFIPWA.
    Radio Science Conference, 2004. Proceedings. 2004 Asia-Pacific; 09/2004
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, efficient methods are discussed to calculate radiation or scattering from three-dimensional perfectly electrically conducting objects located in the half-space environment. Firstly, optimization for impedance-matrix-filling is accomplished to avoid repeated evaluation of Green's functions or Sommerfeld integrals. Then the appropriate evaluation methods for Sommerfeld integrals, such as the method of integration along folded Sommerfeld integral path (FSIP), the method of integration along steepest descent path (SDP), the discrete complex image method (DCIM) and so on, are applied to the different regions with different observation-to-source distance. By these methods, the Sommerfeld integrals can be calculated accurately and fast. Finally, tabulation and interpolation for Green's functions are used to further accelerate the process of impedance-matrix-filling. Some examples are presented for validation. The numerical results have shown that these methods can be used to model the radiation and scattering from complex objects located in a half-space environment accurately and efficiently.
    Radio Science Conference, 2004. Proceedings. 2004 Asia-Pacific; 09/2004
  • [Show abstract] [Hide abstract]
    ABSTRACT: The multilevel fast multipole algorithm (MLFMA), combined with the diagonal element approximation method, is used to realize a fast solution of scattering from a 3D coated object. The impedance boundary condition (IBC) is used to approximate the real boundary condition. MLFMA based on conjugate gradient iteration is applied to expedite greatly the field computations from electric current sources and magnetic current sources. The diagonal element approximation (DEA) method is used to achieve an efficient solution of the sparse matrix equation. Numerical results show the present method is efficient for solving scattering from a 3D coated object with arbitrary shape.
    Antennas and Propagation Society International Symposium, 2004. IEEE; 07/2004

Publication Stats

21 Citations
2.47 Total Impact Points

Institutions

  • 1998–2009
    • University of Electronic Science and Technology of China
      • School of Electronic Engineering
      Chengdu, Sichuan Sheng, China
  • 2004
    • Nanjing University of Science and Technology
      Nan-ching, Jiangsu Sheng, China
  • 2000
    • University of Science and Technology of China
      Luchow, Anhui Sheng, China