Matching Items (4)

137504-Thumbnail Image.png

Edge Informed Fourier Reconstruction from Non-Uniform Spectral Data

Description

The reconstruction of piecewise smooth functions from non-uniform Fourier data arises in sensing applications such as magnetic resonance imaging (MRI). This thesis presents a new polynomial based resampling method (PRM)

The reconstruction of piecewise smooth functions from non-uniform Fourier data arises in sensing applications such as magnetic resonance imaging (MRI). This thesis presents a new polynomial based resampling method (PRM) for 1-dimensional problems which uses edge information to recover the Fourier transform at its integer coefficients, thereby enabling the use of the inverse fast Fourier transform algorithm. By minimizing the error of the PRM approximation at the sampled Fourier modes, the PRM can also be used to improve on initial edge location estimates. Numerical examples show that using the PRM to improve on initial edge location estimates and then taking of the PRM approximation of the integer frequency Fourier coefficients is a viable way to reconstruct the underlying function in one dimension. In particular, the PRM is shown to converge more quickly and to be more robust than current resampling techniques used in MRI, and is particularly amenable to highly irregular sampling patterns.

Contributors

Created

Date Created
  • 2013-05

152360-Thumbnail Image.png

Image processing using approximate data-path units

Description

In this work, we present approximate adders and multipliers to reduce data-path complexity of specialized hardware for various image processing systems. These approximate circuits have a lower area, latency and

In this work, we present approximate adders and multipliers to reduce data-path complexity of specialized hardware for various image processing systems. These approximate circuits have a lower area, latency and power consumption compared to their accurate counterparts and produce fairly accurate results. We build upon the work on approximate adders and multipliers presented in [23] and [24]. First, we show how choice of algorithm and parallel adder design can be used to implement 2D Discrete Cosine Transform (DCT) algorithm with good performance but low area. Our implementation of the 2D DCT has comparable PSNR performance with respect to the algorithm presented in [23] with ~35-50% reduction in area. Next, we use the approximate 2x2 multiplier presented in [24] to implement parallel approximate multipliers. We demonstrate that if some of the 2x2 multipliers in the design of the parallel multiplier are accurate, the accuracy of the multiplier improves significantly, especially when two large numbers are multiplied. We choose Gaussian FIR Filter and Fast Fourier Transform (FFT) algorithms to illustrate the efficacy of our proposed approximate multiplier. We show that application of the proposed approximate multiplier improves the PSNR performance of 32x32 FFT implementation by 4.7 dB compared to the implementation using the approximate multiplier described in [24]. We also implement a state-of-the-art image enlargement algorithm, namely Segment Adaptive Gradient Angle (SAGA) [29], in hardware. The algorithm is mapped to pipelined hardware blocks and we synthesized the design using 90 nm technology. We show that a 64x64 image can be processed in 496.48 µs when clocked at 100 MHz. The average PSNR performance of our implementation using accurate parallel adders and multipliers is 31.33 dB and that using approximate parallel adders and multipliers is 30.86 dB, when evaluated against the original image. The PSNR performance of both designs is comparable to the performance of the double precision floating point MATLAB implementation of the algorithm.

Contributors

Agent

Created

Date Created
  • 2013

157649-Thumbnail Image.png

Optimal sampling for linear function approximation and high-order finite difference methods over complex regions

Description

I focus on algorithms that generate good sampling points for function approximation. In 1D, it is well known that polynomial interpolation using equispaced points is unstable. On the other hand,

I focus on algorithms that generate good sampling points for function approximation. In 1D, it is well known that polynomial interpolation using equispaced points is unstable. On the other hand, using Chebyshev nodes provides both stable and highly accurate points for polynomial interpolation. In higher dimensional complex regions, optimal sampling points are not known explicitly. This work presents robust algorithms that find good sampling points in complex regions for polynomial interpolation, least-squares, and radial basis function (RBF) methods. The quality of these nodes is measured using the Lebesgue constant. I will also consider optimal sampling for constrained optimization, used to solve PDEs, where boundary conditions must be imposed. Furthermore, I extend the scope of the problem to include finding near-optimal sampling points for high-order finite difference methods. These high-order finite difference methods can be implemented using either piecewise polynomials or RBFs.

Contributors

Agent

Created

Date Created
  • 2019

152884-Thumbnail Image.png

Application of holomorphic embedding to the power-flow problem

Description

With the power system being increasingly operated near its limits, there is an increasing need for a power-flow (PF) solution devoid of convergence issues. Traditional iterative methods are extremely initial-estimate

With the power system being increasingly operated near its limits, there is an increasing need for a power-flow (PF) solution devoid of convergence issues. Traditional iterative methods are extremely initial-estimate dependent and not guaranteed to converge to the required solution. Holomorphic Embedding (HE) is a novel non-iterative procedure for solving the PF problem. While the theory behind a restricted version of the method is well rooted in complex analysis, holomorphic functions and algebraic curves, the practical implementation of the method requires going beyond the published details and involves numerical issues related to Taylor's series expansion, Padé approximants, convolution and solving linear matrix equations.

The HE power flow was developed by a non-electrical engineer with language that is foreign to most engineers. One purpose of this document to describe the approach using electric-power engineering parlance and provide an understanding rooted in electric power concepts. This understanding of the methodology is gained by applying the approach to a two-bus dc PF problem and then gradually from moving from this simple two-bus dc PF problem to the general ac PF case.

Software to implement the HE method was developed using MATLAB and numerical tests were carried out on small and medium sized systems to validate the approach. Implementation of different analytic continuation techniques is included and their relevance in applications such as evaluating the voltage solution and estimating the bifurcation point (BP) is discussed. The ability of the HE method to trace the PV curve of the system is identified.

Contributors

Agent

Created

Date Created
  • 2014