<p>Techniques for a posteriori error estimation for finite element approximations of an elliptic partial differential equation are studied.This extends previous work on localized error control in finite element methods for linear elasticity.The methods are then applied to the problem of homogenization of periodic structures. In particular, error estimates for the effective elastic properties are obtained. The usefulness of these estimates is twofold.First, adaptive methods using mesh refinements based on the estimates can be constructed.Secondly, one of the estimates can give reasonable measure of the magnitude ofthe error. Numerical examples of this are given.</p>

Many object recognition systems use a small number of pairings of data and model features to compute the 3D transformation from a model coordinate frame into the sensor coordinate system. With perfect image data, these systems work well. With uncertain image data, however, their performance is less clear. We examine the effects of 2D sensor uncertainty on the computation of 3D model transformations. We use this analysis to bound the uncertainty in the transformation parameters, and the uncertainty associated with transforming other model features into the image. We also examine the impact of the such transformation uncertainty on recognition methods.

In many fields of statistical application the fundamental task is to quantify the association between some explanatory variables or covariates and a response or outcome variable through a suitable regression model. The accuracy of such quantification depends on how precisely we measure the relevant covariates. In many instances, we can not measure some of the covariates accurately, rather we can measure noisy versions of them. In statistical terminology this is known as measurement errors or errors in variables. Regression analyses based on noisy covariate measurements lead to biased and inaccurate inference about the true underlying response-covariate associations. In this thesis we investigate some aspects of measurement error modelling in the case of binary logistic regression models. We suggest a flexible parametric approach for adjusting the measurement error bias while estimating the response-covariate relationship through logistic regression model. We investigate the performance of the proposed flexible parametric approach in comparison with the other flexible parametric and nonparametric approaches through extensive simulation studies. We also compare the proposed method with the other competitive methods with respect to a real-life data set. Though emphasis is put on the logistic regression model the proposed method is applicable to the other members of the generalized linear models, and other types of non-linear regression models too. Finally, we develop a new computational technique to approximate the large sample bias that my arise due to exposure model misspecification in the estimation of the regression parameters in a measurement error scenario.

Techniques for a posteriori error estimation for finite element approximations of an elliptic partial differential equation are studied.This extends previous work on localized error control in finite element methods for linear elasticity.The methods are then applied to the problem of homogenization of periodic structures. In particular, error estimates for the effective elastic properties are obtained. The usefulness of these estimates is twofold.First, adaptive methods using mesh refinements based on the estimates can be constructed.Secondly, one of the estimates can give reasonable measure of the magnitude ofthe error. Numerical examples of this are given.

In SLCCA , the location and magnitude of significant coefficients are specified by the so-called significance map and magnitude respectively . As we know significance map is susceptible , error will propagate when data was deteriorated . This paper address this critical problem and provide an novel approach . In the significance map , the importance of data is interlaced . And our approach is to re-organize the significant map according to encoded symbol¡¦s characteristic . In SLCCA , four symbols are used to encode : POS , NEG , ZERO , LINK . POS or NEG represents the sign of a significant coefficient . ZERO represents an insignificant coefficient . LINK marks the presence of a significance-link . Symbol LINK is more important than POS NEG ZERO . Because when error happen in symbol LINK , it will lead to propagation error . Re-organized data is protected by differRS code . More important data are allocated more parity symbols .

Thesis--Copenhagen.

Ongoing technology improvements and feature size reduction have led to an increase in manufacturing-induced parameter variations. These variations affect various memory cell circuits, making them unreliable at low voltages. Memories are very dense structures that are especially susceptible to defects, and more so at lower voltages. Transient errors due to radiation, power supply noise, etc., can also cause bit-flips in a memory. To protect the data integrity of the memory, an error correcting code (ECC) is generally employed. Present ECC, however, is either single error correcting or corrects multiple errors at the cost of high redundancy or longer correction time. This research addresses the problem of memory reliability under adverse conditions. The goal is to achieve a desired reliability at reduced redundancy while also keeping in check the correction time. Several methods are proposed here including one that makes use of leftover spare columns/rows in memory arrays [Datta 09] and another one that uses memory characterization tests to customize ECC on a chip by chip basis [Datta 10]. The former demonstrates how reusing spare columns leftover from the memory repair process can help increase code reliability while keeping hardware overhead to a minimum. In the latter case customizing ECCs on a chip by chip basis shows considerable reduction in check bit overhead, at the same time providing a desired level of protection for low voltage operations. The customization is done with help from a defect map generated at manufacturing time, which helps identify potentially vulnerable cells at low voltage. An ECC based solution for tackling the wear out problem of phase change memories (PCM) has also been presented here. To handle the problem of gradual wear out and hence increasing defect rates in PCM systems an adaptive error correction scheme is proposed [Datta 11a]. The adaptive scheme, implemented alongside the operating system seeks to increase PCM lifetime by manifold times. Finally the work on memory ECC is extended by proposing a fast burst error correcting code with minimal overhead for handling scenarios where multi-bit failures are common [Datta 11b]. The twofold goal of this work – design a low-cost code capable of handling multi bit errors affecting adjacent cells, and fast multi bit error correction – is achieved by modifying conventional Orthogonal Latin Square codes into burst error codes. / text

No description available.

In many fields of statistical application the fundamental task is to quantify the association between some explanatory variables or covariates and a response or outcome variable through a suitable regression model. The accuracy of such quantification depends on how precisely we measure the relevant covariates. In many instances, we can not measure some of the covariates accurately, rather we can measure noisy versions of them. In statistical terminology this is known as measurement errors or errors in variables. Regression analyses based on noisy covariate measurements lead to biased and inaccurate inference about the true underlying response-covariate associations. In this thesis we investigate some aspects of measurement error modelling in the case of binary logistic regression models. We suggest a flexible parametric approach for adjusting the measurement error bias while estimating the response-covariate relationship through logistic regression model. We investigate the performance of the proposed flexible parametric approach in comparison with the other flexible parametric and nonparametric approaches through extensive simulation studies. We also compare the proposed method with the other competitive methods with respect to a real-life data set. Though emphasis is put on the logistic regression model the proposed method is applicable to the other members of the generalized linear models, and other types of non-linear regression models too. Finally, we develop a new computational technique to approximate the large sample bias that my arise due to exposure model misspecification in the estimation of the regression parameters in a measurement error scenario.

No description available.