SMotionDataSet: Published by Harvard Dataverse (Download -> Dataset:)

Abstract: The dataset is devised to benchmark techniques dealing with human behavior analysis based on multimodal inertial measurement wearable shimmer sensors.

Data Set Characteristics:  Multivariate, Time-Series Number of Instances: 114 Area: Computer
Attribute Characteristics: Real Number of Attributes: 14 Date updated 2019-04-20
Associated Tasks: Classification

Pattern analysis


Dar ul Sukoon, Elderly Care Home, Karachi

EDHI Foundation Old age home, Karachi

Department of Special Education, University of Karachi

Department of Computer Science ,Federal Urdu University of science Art and Technology

Email to whom correspondence should be addressed:

Data Set Information: 

The dataset comprises body motion recordings for several volunteers of diverse profile while performing certain physical activities. Sensors placed on the subject’s waist is used to measure the motion experienced by diverse body parts, namely, acceleration and rate of turn. Data is divided into five age and weight groups categories.

S. No. Age Groups Male Female Total
1 41 – 50 yrs. 3 3 6
2 51 – 60 yrs. 33 30 63
3 61 – 70 yrs. 13 6 19
4 71 – 80 yrs. 13 3 16
5 >80 yrs. 5 5 10
Total 67 47 114

See the db file with groups


Activities: 3
Sensor devices: 1
Subjects: varies


The collected dataset comprises body motion and vital signs recordings for several volunteers of diverse profile while performing 3physical activities (Table 1). Shimmer3 wearable sensors were used for the recordings. The sensors were respectively placed on the subject’s waist attached by using elastic straps (as shown in the figure in attachment). The use of multiple sensors permits us to measure the motion experienced by diverse body parts, namely, the acceleration, the rate of turn, thus better capturing the body dynamics. All sensing modalities are recorded at a sampling rate of 50 Hz(normal range), which is considered sufficient for capturing human activity. few session was recorded using a video camera. The activities were collected in an out-of-lab environment with no constraints on the way these must be executed, with the exception that the subject should try their best when executing them.


The activity set is listed in the following:
L1: Standing still (5sec)
L4: Walking (1 min)
L11: Stand to Sitting (3 steps) (time varies)
NOTE: In brackets are the number of repetitions (Nx) or the duration of the exercises (min).

A complete and illustrated description (including table of activities, sensor setup, etc.) of the dataset is provided in the papers presented in the presentation section.

Attribute Information:

The data collected for each subject is stored in a different log file: ‘shimmer XXX.xls'(XXX will be number between 001 to 999). Each file contains the samples (by rows) recorded for all sensors (by columns). The labels used to identify the activities are similar to the abovementioned (e.g., the label for walking is ‘4’).

The meaning of each column is detailed next:
Column 1: Time stamp raw
Column 2: Time stamp in millisecond
Column 3: acceleration raw (X axis)
Column 4: Acceleration cal (X axis)
Column 5: acceleration raw (Y axis)
Column 6: Acceleration cal (Y axis)
Column 7: : acceleration raw (Z axis)
Column 8: Acceleration cal (Z axis)
Column 9: gyro raw (X axis)
Column 10: gyro cal (X axis)
Column 11: gyro raw (Y axis)

Column 12: gyro cal (Y axis)
Column 13: gyro raw (Z axis)
Column 14: gyro cal (Z axis)

*Units: Acceleration (m/s^2), gyroscope (deg/s).

Relevant Papers:

Citation Request:

Use of this dataset in publications must be acknowledged by referencing the following :

A.Nadeem, K. Rizwan and N.Mehmood, “SMotion dataset’, developed under the project of fall detection system for elderly, available at
We would appreciate if you send us an email (adnan.nadeem at fuuast dot edu dot pk) to inform us of any publication using this dataset.