Recently we used the onion peeling method to reconstruct the axisymmetric electric field distribution of point/plane electrodes from Kerr electro-optic measurements. The method accurately reconstructed the electric fi...
详细信息
Recently we used the onion peeling method to reconstruct the axisymmetric electric field distribution of point/plane electrodes from Kerr electro-optic measurements. The method accurately reconstructed the electric field from numerically generated data. However in the presence of experimental noise the performance was less satisfactory. The measurements were especially noisy and unstable near the needle tip which is also the most interesting region since most charge injection occurs here. In this paper we propose a new algorithm for Kerr electro-optic reconstruction of space charge in arbitrary geometries. The algorithm is built on the finite element method for Poisson's equation and will be called the finite element based Kerr electro-optic reconstruction (FEBKER) hereafter. FEBKER is tested on a point/plane like simple artificial axisymmetric geometry. Research is underway for application of the method to actual axisymmetric point/plane experiments. The method is readily extendible to arbitrary three dimensional geometries.
We study the problem of representing images within a multimedia Database Management System (DBMS), in order to support fast retrieval operations without compromising storage efficiency. To achieve this goal, we propos...
We study the problem of representing images within a multimedia Database Management System (DBMS), in order to support fast retrieval operations without compromising storage efficiency. To achieve this goal, we propose new image coding techniques which combine a wavelet representation, embedded coding of the wavelet coefficients, and segmentation of image-domain regions in the wavelet domain. A bitstream is generated in which each image region is encoded independently of other regions, without having to explicitly store information describing the regions. Simulation results show that our proposed algorithms achieve coding performance which compares favorably, both perceptually and objectively, to that achieved using state-of-the-art image/video coding techniques while additionally providing region-based support.
Given a group trellis and a partitioned signal set, we show how to find all geometrically uniform mappings from the trellis branches to the partition cells. The result is a procedure for constructing geometrically uni...
详细信息
We consider a particular approximate "water-pouring" scheme applied to an additive, colored Gaussian noise channel. The scheme allocates power uniformly over an optimized band of frequencies. Within the set ...
详细信息
We consider a particular approximate "water-pouring" scheme applied to an additive, colored Gaussian noise channel. The scheme allocates power uniformly over an optimized band of frequencies. Within the set of all noise power spectral densities yielding a fixed water-pouring capacity, we find the noise spectrum yielding the smallest mutual information between input and output when the transmitter is restricted to use the sub-optimal scheme. The loss at high SNR is a fraction of a bit per Hz; the relative loss at low SNR is less than 15%.
In this paper, we illustrate a methodology for determining the impact of interconnect pattern dependent variation on circuit performance. The methodology helps enable first pass prediction and can handle large layouts...
详细信息
In this paper, we illustrate a methodology for determining the impact of interconnect pattern dependent variation on circuit performance. The methodology helps enable first pass prediction and can handle large layouts using methods which are reasonably compatible with existing CAD tools. We illustrate the relative utility of the methodology using two case studies. Both studies are drawn from industrial relevant problems: unwanted skew in a balanced clock tree and capacitance variation of a critical net in an SRAM array.
Thin film bulk acoustic wave resonators(TFRs) show considerable promise as a silicon compatible integrated solution for RF bandpass filters with center frequencies in the 1-2 GHz range. This paper presents a VLSI comp...
详细信息
ISBN:
(纸本)0780333942
Thin film bulk acoustic wave resonators(TFRs) show considerable promise as a silicon compatible integrated solution for RF bandpass filters with center frequencies in the 1-2 GHz range. This paper presents a VLSI compatible sealed cavity TFR process using reactively sputtered AlN as the piezoelectric film. The devices are free-standing and have a fundamental longitudinal resonance at 1.36 GHz with an insertion loss of 3.5 dB, a K-eff(2) of .4%, and a Q(series) of 210.
Site specific VHF propagation models remain a subject of continuing interest. A number of approximate methods exist and have been applied successfully in specific cases, but the limitations of the approximations of th...
详细信息
Site specific VHF propagation models remain a subject of continuing interest. A number of approximate methods exist and have been applied successfully in specific cases, but the limitations of the approximations of these models remain unclear. A numerically exact model based on the banded matrix iterative approach (BMIA) to the method of moments is presented which enables the solution of practical propagation problems. Although this model remains computationally complex, use of the iterative method along with parallel computing techniques enables large scale propagation problems to be solved within reasonable amounts of computational time. A typical VHF propagation problem involves predicting the power measured by a receiver as a function of altitude at a given distance away from a like polarised transmitter and above an inhomogeneous terrain profile.
Use of the ac modulation technique which has an ac high voltage superposed onto a dc high voltage has increased the sensitivity of Kerr electro-optic measurements. The technique has been used when the electric field m...
详细信息
Use of the ac modulation technique which has an ac high voltage superposed onto a dc high voltage has increased the sensitivity of Kerr electro-optic measurements. The technique has been used when the electric field magnitude and direction have been constant along the light path. Recently the technique was extended with use of the Abel transformation to recover the non-uniform axisymmetric electric field magnitude along the ground plane and along the needle axis of a point/plane electrode geometry. In these measurements the direction of light propagation is constant along the light path. The experimental results were in good agreement with analysis for the space charge free case. In this work we extend the technique to also allow the electric field direction to vary along the light path. We show that it is possible to measure three "characteristic parameters", analogous to non-uniform photo-elastic stress analysis, that allow determination of the electric field magnitude and direction. These three parameters are related to the electric field through the matricant of the differential equations that govern the Kerr media and reduce to the electric field direction and light phase retardation proportional to the electric magnitude squared for the case where the electric field direction is constant.
Given a linear iteration of the form x := F(x), we consider modified versions of the form x := F(x + gamma d), where d is a fixed direction, and gamma is chosen to minimize the norm of the residual \\x + gamma d - F(x...
详细信息
Given a linear iteration of the form x := F(x), we consider modified versions of the form x := F(x + gamma d), where d is a fixed direction, and gamma is chosen to minimize the norm of the residual \\x + gamma d - F(x + gamma d)\\. We propose ways to choose d so that the convergence rate of the modified iteration is governed by the subdominant eigenvalue of the original. In the special case where F relates to a Markovian decision problem, we obtain a new extrapolation method for value iteration. In particular, our method accelerates the Gauss-Seidel version of the value iteration method for discounted problems in the same way that MacQueen's error bounds accelerate the standard version. Furthermore, our method applies equally well to Markov Renewal and undiscounted problems.
暂无评论