Signal Processing Project For this project, you will be applying several signal processing algorithms to several instances of a given type of one-dimensional data set of your choice. Possible choices include: A) ECG B) EEG C) speech D) ultrasound-based 1D dataset You will analyze at least 3 waveforms (1D) from the data set of your choice using 3 (or more) algorithms of your choice for a particular purpose. This makes a minimum of 3x3=9 analyses. That purpose could be to detect a feature or attribute of interest (like heart rate or the relative t wave amplitude for ECG), to analyze frequency content, or to reduce noise and enhance the signal, for instance. Possible algorithms include: a) peak finding b) frequency filtering c) smoothing d) noise reduction f) spectrogram g) short-time Fourier Transform (STFT) h) frequency content analysis You will share your preliminary results with a classmate during class and include a summary in your project report (for part of the grade). The final report (3-5 pages, not including code) will cover your own analysis. Specifically, include in a few paragraphs on each algorithm you used, describe: • what you did to implement the algorithm? . what were the results? (include figures) why/when is this useful? In your final project submission include: • analysis.m files data files (if not built in or provided on eLearning) project report document (.docx, .pdf) If you wish to propose a challenging project of your own talk to me individually. Ground rules: You need to do independent original work. While you may use built-in functions from MATLAB, do not copy the examples or use code directly copied from the web. It is OK to use such code for inspiration. Do not use a project from another course. You can get started with waveforms built into MATLAB (some are not available in latest MATLAB versions, so this also available as a .zip in eLearning): load wecg; % ECG signal sampled at 180 Hz load mit200% The ECG data and annotations are taken from the MIT-BIH Arrhythmia Database. The data are sampled at 360 Hz. load mit203 % The ECG data and annotations are taken from the MIT-BIH Arrhythmia Database (with noise). The data are sampled at 360 Hz. load Espiga3; % 23 channel EEG data sampled at 200 Hz Or, you can use waveforms from a UTD lab you are associated with, or from the web (Kaggle.com is a one source).