Tutorial 4 Instructions Spring 2019

PwnHero ELE829 Spring 2019 1 Tutorial # 4: STOCHASTIC MODELS FOR NOISE FILTERS Objective The objectives of this Tutorial are: 1) to introduce you to correlation function diagnostics to help determine noise profiles for several different 𝐻(𝑞) structures 2) to try out different MATLAB algorithms dealing with Time Series analysis (i.e. stochastic noise analysis), and 3) to continue to learn how to handle various MATLAB System Identifications commands, including data objects and time series (noise filters) identification. Logistics: Tutorial # 4 consists of identifying five different Noise Models. The Tutorial is a MATLAB simulation and can be completed outside of the Control Systems lab. In spring semesters, it has to be completed in one two-hour session. You will complete this Tutorial in pairs. Each pair will be working on a different set of data - you will be assigned a data set number in the first session. You should download the zipped data file and work with only the set of data corresponding to the assigned number. Each pair of partners will be briefly interviewed by the Course TA when handing in the Tutorial Report. Please refer to the posted Marking Sheet for the grading rubric. Theory: Recall from the previous Tutorials that MATLAB System Identification Toolbox, written by Lennart Ljung, introduces a general parametric model structure: ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( k e q D q C n k u q F q B k y q A k When the deterministic part of the system is disregarded, i.e. the input u(k) = 0, the resulting system output is just the coloured noise sequence, i.e. 𝑦(𝑘) = 𝑣(𝑘), or using Ljung's notation: ) ( ) ( ) ( ) ( ) ( k e q D q C k v k y PwnHero.com