How to detect trigger point of signal rising

Hi,

I have below time series data(one data per seconds), each column is one variable. I want to know which variable triggers to rise or fall the signal in last column(Meaning: after which variable's signal rise or fall will triggers the last column signal start changing from zeor to 0.1).

0  0  0  0  0  12  0
0  0  0  3  3  12  0
0  0  0  3  3  12  0
0  0  0  3  3  12  0
0  0  0  3  3  12  0
0  0  0  3  3  12  0
0  0  3  3  3  12  0
0.3  0  3  3  3  12  0.1
0.3  0  3  3  3  12  0.1
0.3  1  3  3  3  12  0.1
0.3  1  3  3  3  12  0
0.3  1  3  3  3  12  0.2
0.3  1  3  3  3  12  0.2
0.3  1  3  3  3  12  0.2

Answer 1

[Bernhard Suhm]

You could apply the CUSUM algorithm. You can find an example with code at https://dsp.stackexchange.com/questions/30039/detect-to-rising-stable-and-falling-point-in-non-smooth-rectangular-wave

[Kanakaiah Jakkula]

Sir,

I know the point (in last column) where it is changed its direction for first time, and so on. But I want to detect after which column(column1~6) data fall or raise or settled, then column7 (last column) data chage its direction(rise or fall)for first time.

Answer 2

[Fatih Erden]

This may be thought of as a change point detection problem. Depending on the type of your signal, you can use many techniques including Bayesian change point detection, fractal dimension or variance analysis. There is also a new function called findchangepts(x). Using this function you can find the change points (specified by the user) where the mean/rms/std values change the most (at https://www.mathworks.com/help/signal/ref/findchangepts.html).