The features selected for this database come from the accelerometer and gyroscope 3-axial raw signals tAcc_XYZ and tGyro_XYZ. These time domain signals (prefix 't' to denote time) were captured at a constant rate of 50 Hz. Then they were filtered using a median filter and a 3rd order low pass Butterworth filter with a corner frequency of 20 Hz to remove noise. Similarly, the acceleration signal was then separated into body and gravity acceleration signals (tBodyAcc_XYZ and tGravityAcc_XYZ) using another low pass Butterworth filter with a corner frequency of 0.3 Hz.
Subsequently, the body linear acceleration and angular velocity were derived in time to obtain Jerk signals (tBodyAccJerk_XYZ and tBodyGyroJerk_XYZ). Also the magnitude of these three-dimensional signals were calculated using the Euclidean norm (tBodyAccMag, tGravityAccMag, tBodyAccJerkMag, tBodyGyroMag, tBodyGyroJerkMag).
Finally a Fast Fourier Transform (FFT) was applied to some of these signals producing fBodyAcc_XYZ, fBodyAccJerk_XYZ, fBodyGyro_XYZ, fBodyAccJerkMag, fBodyGyroMag, fBodyGyroJerkMag. (Note the 'f' to indicate frequency domain signals).
These signals were used to estimate variables of the feature vector for each pattern:
'_XYZ' is used to denote 3-axial signals in the X, Y and Z directions.
- tBodyAcc_XYZ
- tGravityAcc_XYZ
- tBodyAccJerk_XYZ
- tBodyGyro_XYZ
- tBodyGyroJerk_XYZ
- tBodyAccMag
- tGravityAccMag
- tBodyAccJerkMag
- tBodyGyroMag
- tBodyGyroJerkMag
- fBodyAcc_XYZ
- fBodyAccJerk_XYZ
- fBodyGyro_XYZ
- fBodyAccMag
- fBodyAccJerkMag
- fBodyGyroMag
- fBodyGyroJerkMag
The set of variables that were estimated from these signals are:
- mean: Mean value
- std: Standard deviation
- meanFreq: Weighted average of the frequency components to obtain a mean frequency
The complete list of variables of each feature vector is available in 'features.txt'