ECG Signal Filtering Using MATLAB (FIR/IIR)

Objective is to design appropriate digital filters on MATLAB and apply filtering technique(s) on a recorded raw ECG signal which is contaminated by many unknown noises. The desired ECG signal's bandwidth is 0.5-80 Hz. and the raw signal has been digitized with Sampling Frequency of 2000 Hz. I am providing you both the clean ECG signal and noisy ECG signal so that after you remove noise from raw ECG signal, you should calculate the accuracy of your filtering process by measuring the difference (error in terms of root mean square error, rmse) between filtered signal and original clean signal. You are free to use any type of filter. i.e. either FIR or IIR of any order, but you must explain why you use a particular filter. The Results in your report must include the following graphs and tables Frequency Analysis plot (PSD or Spectrum Estimation where x-axis will be Frequency in Hz) of all three signals: Raw, Clean and Filtered Time-domain plot (Time in Sec) of all three signals: Raw, Clean and Filtered Frequency Response plot of your designed filter (Both Phase and Magnitude Response). Table showing Filtering Performance vs. different order of the filter Table showing Filtering Performance vs. different window functions used in the filter (applicable for FIR filter). Table showing Pass-band and Stop-band Specifications (ripple, frequency, etc.) of the filter Finally selected after optimization. Performance Evaluation of your Designed Filter: You should use the following two performance metrics: corrcoef (built-in function in MATLAB): To check the Similarity between Clean and Filtered Signal in time domain. mscohere (built-in function in MATLAB): To check the magnitude squared coherence (Similarity between Clean and Filtered Signal in Frequency domain). rmse (attached .m file): To check the Root Mean Square Error with respect to Clean Signal before (i.e. raw) and after (i.e. filtered) filtering is performed.