I am going to use Spectral Analysis to analyze the types of movement. There are two types of movement, which are related to the positive and negative values of the Y-axis. When the Y-axis value is positive, it is one movement, and when the Y-axis value is negative, it is another movement. But Spectral Analysis can only analyze the effective value of a value from the perspective of RMS, but cannot judge the positive or negative of the value. Is there any way to solve this problem?
You could use a Flatten block and check for mix/max values of the Y-axis. Do you have other types of movements in your dataset or is it just checking Y-axis positive and negative values? If the latest ML may not be the most suitable option