How to choose the appropriate low pass filter?

I am working on a project in which I have obtained the following readings from the accelerometer in the form of three columns (x,y,z):

0  0  0
-0.2  1.3  -0.3
-0.3  2.7  -0.5
-0.5  4  -0.8
-0.4  4.1  -0.8
-0.4  4.1  -0.8
-0.4  4.1  -0.8
-0.4  4  -0.8
-0.4  4  -0.8
-0.5  4.1  -0.7
-0.5  4.1  -0.7
-0.4  4.1  -1
-0.8  4.4  -1.1
On these readings I did normalization by using x1=x/sqrt(g*height of person).
Therefore a new matrix was now obtained as (x1,y1,z1). Now I have to pass it through the low pass filter to remove the noise. But I am not able to decide which filter to use and how to select the parameters of the filter (i.e order, fs,fc). I am using fdatool for the filter design. Can anyone help me out?

Answer 1

[Daniel kiracofe]

if you are new to filtering, and just want a simple low pass filter to remove noise, then start with a butterworth filter. It may not be the best, but it is easy to use. 4th order a is pretty good start.

Cutoff frequency depends on what you care about. i.e. what kind of noise you are trying to remove, how far away it is from the signal, what the signal looks like, etc. I'd suggest you just pick something, run it, and then plot the output versus your input data and see if it looks like what you want. If not, then try a different cutoff frequency.

Answer 2

[Star Strider]

I did my best about a week ago in my answer to What is the MATLAB code for applying a low pass filter on the readings of accelerometer and gyroscope in a sensor tag?, but you didn’t reply to my comment.

I need your help in deciding what part of your signal you want to work with and what part is noise, and what your sampling frequency is. I don’t know enough about your experimental conditions to make those decisions myself.

[NIHARIKA]

I am really very sorry, by mistake the button got clicked and the above ans was accepted. I still want your help. I have not got the solution to my problem. I just wanted to tell that the file 'accelerometer readings.csv' has the seconds written in it. It shows on the toolbaar in front of f(x) when we click on any of the time. I am sorry once again. I am not able to see any of your comments. How can i bring them back?