The SHL dataset is released progressively as it is prepared.

  • SHL Preview: 3 users, 3 days per users, including 4 phone locations and timelapse camera.
  • SHL Complete User 1 – Hips phone: 1 user, all the recording days (about 7 months), comprising only the phone placed at the pocket (Hips) and timelapse camera. This is the part of the dataset used for the the 2018 SHL Challenge.


This dataset is made publicly available solely for non-profit and research purposes. For any other usage, please contact to discuss licensing terms.

Citation request

If you use the dataset, please cite at least:

[1] H. Gjoreski, M. Ciliberto, L. Wang, F. J. O. Morales, S. Mekki, S. Valentin, D. Roggen. “The University of Sussex-Huawei Locomotion and Transportation Dataset for Multimodal Analytics with Mobile Devices.” IEEE Access 6 (2018): 42592-42604.

[2] L. Wang, H. Gjoreski, M. Ciliberto, S. Mekki, S. Valentin, and D. Roggen, “Enabling reproducible research in sensor-based transportation mode recognition with the Sussex-Huawei dataset,” IEEE Access 7 (2019): 10870-10891.

If you use the App, please cite at least:

[1] M. Ciliberto, F. J. OrdoƱez Morales, H. Gjoreski, D. Roggen, S.Mekki, S.Valentin. “High reliability Android application for multidevice multimodal mobile data acquisition and annotation.” In Proc. ACM Conference on Embedded Networked Sensor Systems. ACM, 2017.

Dataset naming

We recommend to refer to the dataset as follows in your publications:

  • Use at least once the complete name: “The University of Sussex-Huawei Locomotion and Transportation Dataset” or “The Sussex-Huawei Locomotion and Transportation Dataset“. You may introduce the acronym of the dataset as well: “The University of Sussex-Huawei Locomotion-Transportation (SHL) Dataset“.
  • Subsequently, you may refer to the dataset with its acronym: “The SHL Dataset“.

Dataset documentation

SHL Dataset – Preview

This release comprises a part of the complete dataset. It includes:

  • Data from all three users and all four phone locations
  • Three recording-days per user
  • 59 hours of annotated recordings, corresponding to 227 hours of data for the 4 phone locations
  • All sensor modalities, except for audio
Trajectories covered by the users in the SHL Preview dataset
Distributions of the activities in the SHL Preview dataset

Release history

  • Version 1 (3 November 2017): initial release


Due to size, the download is split in 3 parts. Part 1 contains the data of user 1 and Matlab demonstration scripts. Part 2 and 3 contain the data of user 2 and 3 respectively.

Version 1 as zip: Part 1 (2.7GB), Part 2 (2.3GB), Part 3 (2.1GB)

Version 1 as tarball (tar.gz): Part 1 (3.7GB), Part2 (3.2GB), Part 3 (2.9GB)


  • Please use 7zip to extract the ZIP files.
  • A solution to the VideoReader error in MATLAB under Linux can be found here

SHL Dataset – Complete User 1 – Hips Phone

This release comprises a much larger part of the SHL dataset than the SHL Preview. It contains the original data which was used in the 2018 SHL Challenge. This release comprises:

  • Data from one user (User 1).
  • Data from one phone only, placed in trousers front pocket (Hips).
  • 391 hours of recordings over 7 months.
  • All sensor modalities (including timelapse video), except for audio.


Due to size, the download is split in multiple parts. Download all the parts in the same folder.

Use 7zip to extract the files from the archive: 7z x;;;;