The module will explain Autocorrelation and its function and properties. Also, examples will be provided to help you step through some of the more complicated statistical analysis.
Signals with stochastic, or random components may have an informative power spectral density (PSD). Estimation of the PSD of a stochastic signal requires averaging. The periodogram averages the magnitude spectra of smaller blocks of the signal to estimate the PSD. An alternative approach implicitly averages via windowing the auto-correlation of the signal to produce a low-variance smoothed spectral estimate.
An overview of DSP optimization techniques, from the design stage, through high-level implementations, to assembly implementations.
An overview of DSP optimization techniques, from the design stage, through high-level implementations, to assembly implementations.
Describes signals that cannot be precisely characterized.
You will investigate the effects of windowing and zero-padding on the Discrete Fourier Transform of a signal, as well as the effects of data-set quantities and weighting windows used in Power Spectral Density estimation.
Students are to implement and optimize a power spectral density estimator, a pseudo-noise (PN) sequence generator, and an IIR filter.
Students are to implement and optimize a power spectral density estimator, a pseudo-noise (PN) sequence generator, and an IIR filter.
This is a processor exercise in which students implement a spectrum analyzer using mixed C and assembly code. Students are to acquire a block of 1024 samples, apply a Hamming window, compute a length-1024 Discrete Fourier Transform using provided Fast Fo
This is a processor exercise in which students implement a spectrum analyzer using mixed C and assembly code. Students are to acquire a block of 1024 samples, apply a Hamming window, compute a length-1024 Discrete Fourier Transform using provided Fast Fo
This is a processor exercise in which students implement a spectrum analyzer using mixed C and assembly code. Students are to acquire a block of 1024 samples, apply a Hamming window, compute a length-1024 Discrete Fourier Transform using provided Fast Fo
This module introduces the interactive optimizing compiler combination VPO/VISTA. VPO (Very Portable Optimizer) is a retargetable compiler developed at the University of Virginia. VISTA (VPO Interactive System for Tuning Applications), developed at Florida State University, is a GUI that allows a user to interactively select optimization steps with VPO feedback. Instructions are given on how to apply VPO/VISTA optimization to three routines provided in the power spectral density (PSD) estimator of the previous lab. Students are then asked to optimize these routines twice: once using VISTA and once using any method desired.
You will write MATLAB code to compute the autocorrelation sequence of a simple signal. Then you will implement the Levinson-Durbin algorithm in MATLAB and analyze a recording of your own voice.
You will write code to compute the autocorrelation or autocovariance of an input signal. Then you connect a microphone to the DSP and write code to detect the beginning of a speech segment. Finally, you will combine the two programs and compare results with MATLAB.
Speech analysis and synthesis with Linear Predictive Coding (LPC) exploit the predictable nature of speech signals. Cross-correlation, autocorrelation, and autocovariance provide the mathematical tools to determine this predictability.
The course introduces statistical theory to prepare students for the remainder of the econometrics sequence. The emphasis of the course is to understand the basic principles of statistical theory. A brief review of probability will be given; however, this material is assumed knowledge. The course also covers basic regression analysis. Topics covered include probability, random samples, asymptotic methods, point estimation, evaluation of estimators, Cramer-Rao theorem, hypothesis tests, Neyman Pearson lemma, Likelihood Ratio test, interval estimation, best linear predictor, best linear approximation, conditional expectation function, building functional forms, regression algebra, Gauss-Markov optimality, finite-sample inference, consistency, asymptotic normality, heteroscedasticity, and autocorrelation.
