ACES Publication Search





There are 9 search results for:



Title: A DUAL NORMAL MODE REPRESENTATION FOR ELECTROMAGNETIC SCATTERING: SOME INITIAL CONSIDERATIONS
Abstract: This paper presents a new technique for characterizing the electromagnetic interaction with and scattering from objects of finite extent. By exploiting the structure of the operators (in this case matrices) associated with the interaction, it is shown that both the system (impedance) matrix and the transfer (admittance) matrix can be partitioned into a sum of two matrices such that each of the resulting matrices can be decomposed into a set of normal modes. As a result, the number of parameters needed to describe the interaction is now significantly reduced and the parameters identified by this technique are model independent, i.e. they are measurable parameters. Potential applications of the technique include EM computations, compact descriptions of scatterers and antennas, interpretation of measured data, and algorithm development applicable to scattering and inverse scattering problems. [Vol. 3, No. 2, pp. 28-56 (1988)]
Author(s): T. H. Lehman
File Type: Journal Paper
Issue:Volume: 3      Number: 2      Year: 1988
Download Link:Click here to download PDF     File Size: 2097 KB

Title: COMPARISON OF METHODS FOR FAR ZONE SCATTERING FROM A FLAT PLATE AND CUBE
Abstract: Different high frequency methods are used to analyze the backscatter and bistatic scattering from a flat plate and a cube. The results are compared and their validity is checked against method of moments and measurements. A newly developed far zone corner diffraction coefficient based on the latest equivalent current and PTD solutions cast in UTD form is discussed. [Vol. 3, No. 2, pp. 57-78 (1988)]
Author(s): R. J. Marhefka, T. J. Brinkley
File Type: Journal Paper
Issue:Volume: 3      Number: 2      Year: 1988
Download Link:Click here to download PDF     File Size: 1446 KB

Title: ON THE COMPARISON OF NUMERICAL METHODS
Abstract: Serious comparisons of numerical methods are important for scientists who develop new codes as well as for those who use programs. Historical considerations show some errors which were made in the past and should be avoided in the future. Every numerical code is based not only on numerical but also on analytical considerations. Both of them have to be taken into account. As a result, benchmarks for complicated topics (numerical calculations of electromagnetic fields) should give more information than just numbers like 'speed', 'memory requirement', etc. [Vol. 3, No. 2, pp. 79-82 (1988)]
Author(s): Ch. Hafner
File Type: Journal Paper
Issue:Volume: 3      Number: 2      Year: 1988
Download Link:Click here to download PDF     File Size: 508 KB

Title: IMPLEMENTATION OF GEMACS 3.3 ON PERSONAL COMPUTERS
Abstract: GEMACS 3.3 is a powerful MM/GTD hybrid package which can model a wide range of antenna and scattering problems. It is intended for use on large mainframe computers but an implementation on a personal computer has advantages in the possibilities of interactive use and graphical output. An approach to this implementation is presented, together with benchmark test results from typical PCs, a mainframe and a supercomputer. [Vol. 3, No. 2, pp. 83-87 (1988)]
Author(s): Adel F. Armanlous, Peter S. Excell
File Type: Journal Paper
Issue:Volume: 3      Number: 2      Year: 1988
Download Link:Click here to download PDF     File Size: 479 KB

Title: OBTAINING SCATTERING SOLUTIONS FOR PERTURBED GEOMETRIES AND MATERIALS FROM MOMENT METHOD SOLUTIONS
Abstract: In this paper, we present an efficient method for computing the solution to scattering problems using a perturbation scheme based on the solution of related original problems. Assuming the radar cross section has been computed for a particular scatterer associated with a moment method matrix B, we call the computation of the radar cross section of a slightly perturbed scatterer a "perturbed problem of B". If the original problem has n unknowns, and the perturbed problem is formed by changing p cells of the original problem, then our method requires an operation count of O(n2p + p3) while a direct moment method solutions requires an operation count of O(n3). Our method involves application of the Sherman-Morrison-Woodbury formula for inverses of perturbed matrices. We show that the method can be easily implemented in any moment method code, and the user does not have to learn a new input procedure. Further, the modified code can provide a basis for a non-linear optimization procedure which minimizes the radar cross section of an obstacle by varying the surface impedance's. An appropriate objective function in this problem depends on the radar cross section at the angles and frequencies of interest. Let n be the number of cells in the obstacle and let p be the number of cells with variable impedance, with n>>p. Then application of the Sherman-Morrison-Woodbury formula results in objective function evaluations requiring an O(np+p3) operation count. In contrast, application of the classical moment method results in objective function evaluations requiring an O(n3) operation count. Numerical results from large practical problems demonstrate the efficiency and stability of the new method. [Vol. 3, No. 2, pp. 95-118 (1988)]
Author(s): Ellzabeth Yip, Brian Tomas
File Type: Journal Paper
Issue:Volume: 3      Number: 2      Year: 1988
Download Link:Click here to download PDF     File Size: 1321 KB

Title: THE EFFECTS OF HEAVY CHARGED PARTICLE IRRADIATION OF MOSFET DEVICES
Abstract: lonizing cosmic particle radiation poses a serious threat to electronic devices (such as metal-oxide semiconductor, field effect transistors -- MOSFETs) that are used in outer space. The physical process in which a bombarding ion creates electron-hole pairs in the SiO2 layer of a MOSFET, the subsequent collection of charge at the SiO2-substrate interface and its effect on the operating characteristics of the transistor is modeled with two second order, coupled differential equations. The coupled equations are solved using the finite difference technique known as the Alternating Direction Implicit Method, ADI. Preliminary verification of the computer code was performed using a low energy proton accelerator. The measured change in MOSFET operating characteristics compared favorably with the predicted results. The results show that the damage due to ionizing particles is greatly dependent on the energy of the bombarding particle, its angle of incidence, and the magnitude of the bias applied to the MOSFET. [Vol. 3, No. 2, pp. 119-130 (1988)]
Author(s): Wlilliam Eichinger, Patrlck O'Reilly, Christopher Lehner
File Type: Journal Paper
Issue:Volume: 3      Number: 2      Year: 1988
Download Link:Click here to download PDF     File Size: 966 KB

Title: HF GROUND CONSTANT MEASUREMENTS AT THE LAWRENCE LIVERMORE NATIONAL LABORATORY (LLNL) FIELD SITE
Abstract: The SRI International open-wire-line (OWL) kit was used 3AE5 July 1987 to measure the HF ground constants at the Lawrence Livermore National Laboratory (LLNL) field site in Livermore, CA. Data were acquired at 11 locations about 250 ft west of the LLNL facility fence in the vicinity where a longwire and broadband dipole were erected in August 1987 for making impedance measurements for the purpose of validating the Numerical Electromagnetics Code (NEC). An additional location was measured to the north of the antenna site where field strength data were to be taken. Several samples were taken at most locations. Best estimates of the conductivity, relative permittivity (relative dielectric constant), dissipation factor and skin depth were computed as the median values versus frequency for 2 through 30 MHz. Data were acquired at l-MHz intervals from 2 MHz through 8 MHz, and the interval was increased to 2 MHz from 8 MHz to 30 MHz. The maximum and minimum values were also determined as bounds on the conductivity and relative permittivity values for use in parameter sensitivity analyses. The conductivity values for the relatively dry, densely packed light brown clay fell between those typical of pastoral land and rich agricultural land at about 4xlO-2S/m. The relative permittivity values exhibited more variation with frequency. At the low end of the HF band, the relative permittivity values exceeded those of a non-flooded rice paddy (e.g., about 150 at 2 MHz); whereas, at the high end of the band, the relative permittivity approximated values typical of rich agricultural land (about 17 at 30 MHz). The skin depth varied from about 2 m at 2 MHz to 0.7 m at 30 MHz. The dissipation factor was about 1.5, so the soil acted almost as a semiconductor rather than as a lossy conductor or a lossy dielectric. Both the relative dielectric constant and conductivity are important in modeling antennas and propagation over the ground at the LLNL site. Data from nearby wells indicated that the water table was at least 20 m below the surface. Therefore, a one layer slab model adequately described the ground at this site for HF down to the skin depth. [Vol. 3, No. 2, pp. 131-165 (1988)]
Author(s): George H. Hagn
File Type: Journal Paper
Issue:Volume: 3      Number: 2      Year: 1988
Download Link:Click here to download PDF     File Size: 2258 KB

Title: NUMERICAL ELECTROMAGNETICS COMPUTATION USING THE INMOS T800 TRANSPUTER ON AN OLIVETTI M24 PERSONAL COMPUTER
Abstract: As is known to most engineers and scientists interested in computational electromagnetics analysis, G.J. Burke and A.J. Poggio developed a code called NEC2 at the Lawrence Livermore Laboratory in California [1]. This code was developed using a mainframe computer and was written in Fortran. The code has subsequently been used on a VAX785 mainframe at this University.
Author(s): J.J. LE ROUX
File Type: Journal Paper
Issue:Volume: 3      Number: 2      Year: 1988
Download Link:Click here to download PDF     File Size: 662 KB

Title: NUMERICAL INTEGRATION OF MARCUSE'S POWER LOSS FORMULA
Abstract: D. Marcuse [1] has derive a power loss formula to calculate the power losses from an electromagnetic wave traveling down a tapered dielectric rod. He first considers the losses at a step and then approximates the tapered rod by a series of steps. He assumes that, when the radius of the rod is a, the constants describing the bound mode will be the same as for an infinite rod. When the radius changes, so do the coefficients. Further, there is an additional change due to power taken from the bound mode and converted into the radiation modes. For a full discussion of the model, the interested reader is referred to [1]. Here, we are concerned only with the numerical integration of his formula
Author(s): JAMES P. COUGHLIN, ALBERT D. KRALL, ROBERT H. BARAN
File Type: Journal Paper
Issue:Volume: 3      Number: 2      Year: 1988
Download Link:Click here to download PDF     File Size: 1230 KB