| title | author | date |
|---|---|---|
CodeBook.md |
Kaie Kubjas |
25 Jul 2015 |
subject unique identifier assigned with the subject who performed the activity 1-30
activity activity performed by the subject WALKING WALKING_UPSTAIRS WALKING_DOWNSTAIRS SITTING STANDING LAYING
tBodyAcc-mean()-X tBodyAcc-mean()-Y tBodyAcc-mean()-Z average of mean values of the time domain body acceleration signal in the X, Y and Z directions respectively for each activity and each subject
tBodyAcc-std()-X tBodyAcc-std()-Y tBodyAcc-std()-Z average of standard deviations of the time domain body acceleration signal in the X, Y and Z directions respectively for each activity and each subject
tGravityAcc-mean()-X tGravityAcc-mean()-Y tGravityAcc-mean()-Z average of mean values of the time domain gravity acceleration signal in the X, Y and Z directions respectively for each activity and each subject
tGravityAcc-std()-X tGravityAcc-std()-Y tGravityAcc-std()-Z average of standard deviations of the time domain gravity acceleration signal in the X, Y and Z directions respectively for each activity and each subject
tBodyAccJerk-mean()-X tBodyAccJerk-mean()-Y tBodyAccJerk-mean()-Z average of mean values of the time domain body linear acceleration Jerk signal in the X, Y and Z directions respectively for each activity and each subject
tBodyAccJerk-std()-X tBodyAccJerk-std()-Y tBodyAccJerk-std()-Z average of standard deviations of the time domain body linear acceleration Jerk signal in the X, Y and Z directions respectively for each activity and each subject
tBodyGyro-mean()-X tBodyGyro-mean()-Y tBodyGyro-mean()-Z average of mean values of the time domain body angular velocity in the X, Y and Z directions respectively for each activity and each subject
tBodyGyro-std()-X tBodyGyro-std()-Y tBodyGyro-std()-Z average of standard deviations of the time domain body angular velocity in the X, Y and Z directions respectively for each activity and each subject
tBodyGyroJerk-mean()-X tBodyGyroJerk-mean()-Y tBodyGyroJerk-mean()-Z average of mean values of the time domain body angular velocity Jerk signal in the X, Y and Z directions respectively for each activity and each subject
tBodyGyroJerk-std()-X tBodyGyroJerk-std()-Y tBodyGyroJerk-std()-Z average of standard deviations of the time domain body angular velocity Jerk signal in the X, Y and Z directions respectively for each activity and each subject
tBodyAccMag-mean() average of mean values of the magnitude of the three-dimensional time domain body acceleration signal which is calculated using the Euclidean norm for each activity and each subject
tBodyAccMag-std() average of standard deviations of the magnitude of the three-dimensional time domain body acceleration signal which is calculated using the Euclidean norm for each activity and each subject
tGravityAccMag-mean() average of mean values of the magnitude of the three-dimensional time domain gravity acceleration signal which is calculated using the Euclidean norm for each activity and each subject
tGravityAccMag-std() average of standard deviations of the magnitude of the three-dimensional time domain gravity acceleration signal which is calculated using the Euclidean norm for each activity and each subject
tBodyAccJerkMag-mean() average of mean values of the magnitude of the the three-dimensional time domain body acceleration Jerk signal which is calculated using the Euclidean norm for each activity and each subject
tBodyAccJerkMag-std() average of standard deviations of the magnitude of the three-dimensional time domain body acceleration Jerk signal which is calculated using the Euclidean norm for each activity and each subject
tBodyGyroMag-mean() average of mean values of the magnitude of the time domain body angular velocity signal which is calculated using the Euclidean norm for each activity and each subject
tBodyGyroMag-std() average of standard deviations of the magnitude of the time domain body angular velocity signal which is calculated using the Euclidean norm for each activity and each subject
tBodyGyroJerkMag-mean() average of mean values of the magnitude of the time domain body angular velocity Jerk signal which is calculated using the Euclidean norm for each activity and each subject
tBodyGyroJerkMag-std() average of standard deviations of the magnitude of the time domain body angular velocity Jerk signal which is calculated using the Euclidean norm for each activity and each subject
fBodyAcc-mean()-X fBodyAcc-mean()-Y fBodyAcc-mean()-Z average of mean values of the frequency domain signal in X, Y and Z directions respectively obtained from the body acceleration signal by applying Fast Fourier Transform for each activity and each subject
fBodyAcc-std()-X fBodyAcc-std()-Y fBodyAcc-std()-Z average of standard deviations of the frequency domain signal in X, Y and Z directions respectively obtained from the body acceleration signal by applying Fast Fourier Transform for each activity and each subject
fBodyAccJerk-mean()-X fBodyAccJerk-mean()-Y fBodyAccJerk-mean()-Z average of mean values of the frequency domain signal in X, Y and Z directions respectively obtained from the body acceleration Jerk signal by applying Fast Fourier Transform for each activity and each subject
fBodyAccJerk-std()-X fBodyAccJerk-std()-Y fBodyAccJerk-std()-Z average of standard deviations of the frequency domain signal in X, Y and Z directions respectively obtained from the body acceleration Jerk signal by applying Fast Fourier Transform for each activity and each subject
fBodyGyro-mean()-X fBodyGyro-mean()-Y fBodyGyro-mean()-Z average of mean values of the frequency domain signal in X, Y and Z directions respectively obtained from angular velocity by applying Fast Fourier Transform for each activity and each subject
fBodyGyro-std()-X fBodyGyro-std()-Y fBodyGyro-std()-Z average of standard deviations of the frequency domain signal in X, Y and Z directions respectively obtained from angular velocity by applying Fast Fourier Transform for each activity and each subject
fBodyAccMag-mean() average of mean values of the frequency domain signal obtained from the magnitude of the three-dimensional body acceleration signal by applying Fast Fourier Transform for each activity and each subject
fBodyAccMag-std() average of standard deviations of the frequency domain signal obtained from the magnitude of the three-dimensional body acceleration signal by applying Fast Fourier Transform for each activity and each subject
fBodyBodyAccJerkMag-mean() average of mean values of the frequency domain signal obtained from the magnitude of the three-dimensional body acceleration Jerk signal by applying Fast Fourier Transform for each activity and each subject
fBodyBodyAccJerkMag-std() average of standard deviations of the frequency domain signal obtained from the magnitude of the three-dimensional body acceleration Jerk signal by applying Fast Fourier Transform for each activity and each subject
BodyBodyGyroMag-mean() average of mean values of the frequency domain signal obtained from the magnitude of the angular velocity signal by applying Fast Fourier Transform for each activity and each subject
fBodyBodyGyroMag-std() average of standard deviations of the frequency domain signal obtained from the magnitude of the angular velocity signal by applying Fast Fourier Transform for each activity and each subject
fBodyBodyGyroJerkMag-mean() average of mean values of the frequency domain signal obtained from the magnitude of the angular velocity Jerk signal by applying Fast Fourier Transform for each activity and each subject
fBodyBodyGyroJerkMag-std() average of standard deviations of the frequency domain signal obtained from the magnitude of the angular velocity Jerk signal by applying Fast Fourier Transform for each activity and each subject
Note: Data is collected from the accelerometers and gyroscope 3-axial raw signals tAcc-XYZ and tGyro-XYZ from the Samsung Galaxy S smartphone. These time domain signals 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).
In this data set, we consider averages of mean values and standard deviations of these variables for each subject and actvity.