This document summarizes a lab report for an electrical engineering course on linear control systems. The lab focused on using MATLAB to estimate multiple parameters of an input-output model and characterize the accuracy of those estimates. Specifically, the lab examined how the choice of input data can influence the correlation between two estimated parameters of a two-parameter model and how the number of model parameters affects statistical measures of the parameter estimates like mean and covariance. Experiments were conducted to estimate the parameters of a two-dimensional model using two different input data sets and the mean estimates for each parameter from the two experiments were reported.
This document summarizes a lab report for an electrical engineering course on linear control systems. The lab focused on using MATLAB to estimate multiple parameters of an input-output model and characterize the accuracy of those estimates. Specifically, the lab examined how the choice of input data can influence the correlation between two estimated parameters of a two-parameter model and how the number of model parameters affects statistical measures of the parameter estimates like mean and covariance. Experiments were conducted to estimate the parameters of a two-dimensional model using two different input data sets and the mean estimates for each parameter from the two experiments were reported.
This document summarizes a lab report for an electrical engineering course on linear control systems. The lab focused on using MATLAB to estimate multiple parameters of an input-output model and characterize the accuracy of those estimates. Specifically, the lab examined how the choice of input data can influence the correlation between two estimated parameters of a two-parameter model and how the number of model parameters affects statistical measures of the parameter estimates like mean and covariance. Experiments were conducted to estimate the parameters of a two-dimensional model using two different input data sets and the mean estimates for each parameter from the two experiments were reported.
EEL 4657 Linear Control Systems Laboratory Dawn Owens (N00152181) Lab 5 Least-Squares Estimation of Linear Multi-Parameter Models Abstract: The objective of this lab was to continue generalizing the insights from previous labs to input-output models that involve more than just one parameter to be estimated. Characterizing the accuracy of a multi-parameter predictor required utilizing fundamental statistical concept called the covariance, which is essentially a measure of the correlation between two random variables. Using the simplest setting of a twoparameter predictor, MATLAB will be used (i) to show that correlation between the two parameters can be influenced by the choice of input data and (ii) to examine the influence of the number of model parameters on second-order statistics.
Exercise L6-1 (Characterizing the Accuracy of a Two-Dimensional Parameter
Estimate) Statistic Mean Vector
Experiment 1 E[1] = -3.71 x 10-4 E[2] = 9.99 x 10-2