2

Syllabus
Lecture Note Introduction

Lecture1 Lecture2 Lecture3 Lecture4 Lecture5

Lecture6 Lecture7 Lecture8 Lecture9 Lecture10

Lecture11 Lecture12 Lecture13 Lecture14 Lecture15

Lecture16 Lecture17 Lecture18 Lecture19 Lecture20

Lecture21 Lecture22(complete) Lecture23(complete) Lecture24(complete) Lecture25(complete)

Problem Sets

PS 1 (Solution)

Use the following data files for Problem 3

Data: The true thetas are given so that you can compare your RLS results with the true values.

1.Test data: input output Theta=[5 -2 3]’

Test data is used for you to test your code.

2.Data for the questions

a) input(u) output(y) theta_1 theta_2 theta_3

b) input(u) output(y) theta_1 theta_2 theta_3

PS 2 (Solution)

Use the following data files for Problem 3

-If you have any questions on how to use the files, please contact Levi Wood (woodl@mit.edu)

Reference “Good” Signal

Data Set 1

Corrupted Data Noise Reference

Data Set 2

Corrupted Data Noise Reference

Notes:

1) Relationship between file names and labels in Figure 2 of Problem Set 3

File Names Figure 2 Labels

“GoodPPG” y_o

“CorruptedPPG” y

“acceleration” a

2) There is one “GoodPPG” reference signal that can be used to compare your

“recovered” signals, z, with for both data sets.

3) One of the data sets is “tricky.” Can you explain why you are getting strange results?

4) How do you think you should choose the filter order?

5) Levi Wood will be available to help you on Monday and Tuesday

Office Hours: room 3-351 Monday, Feb 27th 4-5pm

Tuesday, Feb 28th 4-5pm

By Appointment: woodl@mit.edu

Also feel free to ask questions by email.

PS 3 (Solution)

At = [0.8500 0.1500 0.0500; 0.0500 0.7000 0; 0.0500 -0.0500 0.7500]

Bt = [10.0000;0;14.2103]

Gt = [1;1;1]

Ht = [1 0 0]

Rt = 2.9862

Qt = 1.0129

input(u) output(y)

The actual X1(dTe) X2(dTw) X3(dL2) are given for comparison with the results estimated from the Kalman Filter.

PS 4 (Solution)

PS 5 (Solution)

All the data was derived from a yorkshire swine using a sampling rate of 100 Hz.

The input data (AorticFlowInput.txt) was measured in the root of the ascending aorta just outside the heart

and the output blood pressure data (BloodPressureOutput.txt) was measure in the radial artery (in the lower arm).

Standard units were used in collecting this physiological data.

The units for aortic flow are liters/minute and for blood pressure are millimeters of mercury (mmHg).

PS 6 PS 7 PS 8