This repository offers smart-home wearable accelerometer and Radio Signal Strength Indicator (RSSI) data acquired : 1) with low-cost hardware; 2) with high-resolution location annotations; 3) from four UK homes. The data are intended to evaluate RSSI-based indoor localisation methods with activity measurements provided from a user-worn wearable device. Location labels are recorded automatically using a small camera which registers fiducial floor tags as the participant carries out their normal routines in a natural way. Approximately 14 hours of annotated wearable measurements are provided.

A user wears a wearable device on their wrist, which records accelerometer data at 25Hz. This data is transmitted, at 5Hz, towards a number of Bluetooth Low Energy access points (8-11) within the home.   The access points mark the packets with a Received Signal Strength Indicator (RSSI) measurement and note the accelerometer measurements. The user's actual location is derived from a camera which registers fiducial floor tags, placed at a metre apart, where the user pose and relative position is decoded using image processing code (also provided).

Getting started

1. Download and unzip the dataset.

2. Navigate to:
   ble-accelerometer-indoor-localisation-measurements/codes/load_dataset_py

3. Install prerequisites: pip -r install requirements.txt

4. Run the sample module to load and view the measurements : python3 src/load_data.py

Content

Read https://www.nature.com/articles/sdata2018168 for all the juicy details.
A guide to loading the data using python is provided at: /residentialwearabledatarepowithlocationlabelssub/codes/loaddataset_py/readme.txt

Acknowledgements

Data Authors: Dallan Byrne & Michal Kozlowski

Please cite:  Byrne, D., Kozlowski, M., Santos-Rodriguez, R., Piechocki, R. & Craddock, I. Residential wearable RSSI and accelerometer measurements with detailed location annotations. Sci. Data 5, 180168 (2018).
https://www.nature.com/articles/sdata2018168/

Thanks to:  Raul Santos-Rodriguez, Robert Piechocki, Ian Craddock, SPHERE IRC team, Beatriz Monsalve-Carcalen and Raimon Fransoy.

Funding: This work was performed under the Sensor Platform for HEalthcare in a Residential Environment (SPHERE) Interdisciplinary Research Collaboration (IRC) funded by the UK Engineering and Physical Sciences Research Council (EPSRC), Grant EP/K031910/1.