Plotting acceleration vs time on Matlab from accelerometer data that doesn't have a time column

Hello, I would like to plot x y and z axes of acceleration versus time from the csv file that was outputted from my sensor, the first two rows of the data of the csv file give the timestamp and the frequency of the reading therefore they should not be read by matlab and the columns represent x y and z axes respectively. The problem I am having is that the file does not have a time column and that I need to create an abstract one so that I could plot the acceleration vs time (or do I actually need one? is there a command that I can just use?) Thank you very much in advance!!

[David Hill]

You need to convert the time stamps into time. What is the format of the time stamp?

[Adviye Irem Yuceel]

There are no timestamps unfortunetly, there is only one timestamp that gives the time data collection started and then the data is sampled at 32 Hz

[Adviye Irem Yuceel]

The format of the given one timestamp is in unix or seconds from 1-1-1970 in UTC

Answer 1

[Mathieu NOE]

hello

as we know the sampling rate Fs and the amount of samples , there is no big difficulty to reconstruct a time vector :

samples = length(data);
dt = 1/Fs;
time_vector = (0:samples-1)*dt

Answer 2

[Duncan Po]

You can create a timetable, and just supply the start time and frequency. Timetable will compute the timestamps for your data. For example,

>> data = (1:10)';

>> starttime = datetime('now');

>> fs = 2; % 2Hz in this example

>> tt = timetable(data, 'StartTime', starttime, 'SampleRate', fs)

tt =

10×1 timetable

Time data

____________________ ____

08-Jun-2021 14:59:30 1

08-Jun-2021 14:59:30 2

08-Jun-2021 14:59:31 3

08-Jun-2021 14:59:31 4

08-Jun-2021 14:59:32 5

08-Jun-2021 14:59:32 6

08-Jun-2021 14:59:33 7

08-Jun-2021 14:59:33 8

08-Jun-2021 14:59:34 9

08-Jun-2021 14:59:34 10

You can then either plot directly using stackedplot, or extract the timestamps using tt.Time, and then call plot.

[Adviye Irem Yuceel]

I get this error message (some of my data values are negative because the sensor gives the opposite acceleration in negative):

Error using stackedplot (line 71)

Expected vars to be positive.

Error in stress (line 32)

stackedplot(tt,y)

Is there a way to get arround this? Also can I start the starttime from the time stamp given in the csv file with the data in unix format?

[Duncan Po]

To plot the timetable using stackedplot, you can simply pass in the timetable as the only input:

stackedplot(tt)

The error you encountered is due to the second input y. Apparently y is not a valid second input.

For unix time, there is a way to convert using the datetime function:

starttime = datetime(t, 'ConvertFrom','posixTime');