Font Size: a A A

Study Of Near Field Of Target Above The Half Space And High Efficient FDTD Parallel Algorithm

Posted on:2013-10-06Degree:MasterType:Thesis
Country:ChinaCandidate:P YinFull Text:PDF
GTID:2248330395456474Subject:Radio Physics
Abstract/Summary:PDF Full Text Request
The detection of Low-altitude target is one of the hot topics in the study ofelectromagnetic scattering. Considering the influence of the under ground, to solve thiskind of problem is very complex. This paper presents a semi-analytical FDTD methodfor handling the scattering problem of the target above the half space. The time domainreflected wave can be obtained by Fourier transform and the Fresnel formula when theplane EM wave impact on the ground. The computational domain of the traditionalmethod containing the target and under ground, but the computational domain of thisnew method only contains the target itself. So our new method can save a lot of memorywhen computing the scattering of an object located above the half space.It is often need to handle with electrically large complex target in practicalengineering problems. The process of modeling complex target always costs a lot ofmemory and computing time. Firstly, the general FDTD modeling methods areintroduced. Also, the advantages and disadvantages of these methods are analyzed. Andthen, the basic concept of MPI (Message Passing Interface) platform and parallel I/O arealso introduced briefly. Finally, FDTD modeling is accelerated by the method of parallelI/O. Numerical results show that this method greatly improves the efficiency ofmodeling.When analysis the characteristic of a target, it is need to compute the transientresponse of multi-direction sometimes. This paper implements multi-directionextrapolation in the computation of MPI based FDTD algorithm. The calculating resultsof Parallel program and serial program are good agreement with each other. Themulti-direction electromagnetic characteristics of an aircraft are analyzed by using ofour method.
Keywords/Search Tags:FDTD, Semi-analytical method, Parallel modeling, Parallel multi-extrapolation
PDF Full Text Request
Related items