The doi above is for this specific version of this dataset, which is currently the latest. Newer versions may be published in the future. For a link that will always point to the latest version, please use
doi: 10.4121/7cfe7f4f-e8ef-4a06-8359-ca5831e22810

Lopez Delgado, Ignacio Esteban; Wang, Dingyang; Francesco Fioranelli; Grajal, Jesus (2024): Dataset of human gait in a domestic setup collected with an LFMCW MIMO radar. Version 2. 4TU.ResearchData. software. https://doi.org/10.4121/7cfe7f4f-e8ef-4a06-8359-ca5831e22810.v2

Software

x choose version: version 1 - 2024-10-25

• Neurosciences
• Electrical and Electronic Engineering
• Medical Biotechnology
• Public Health and Health Services
• Biomedical Engineering
• Human Movement and Sports Science
• Engineering
• Technology
• Medical and Health Sciences

doppler signature, fmcw radar, gait, gait asymmetry, gait monitoring, mimo radar, radar

01/08/2024-30/09/2024

CC BY-NC 4.0

RefWorks, BibTeX, Reference Manager, Endnote, DataCite, NLM, DC, CFF

by Ignacio Esteban Lopez Delgado , Dingyang Wang , Francesco Fioranelli , Jesus Grajal

Radar technology is a strong candidate to analyze human gait. This dataset contains gait captures of 60 healthy individuals in a laboratory emulating a domestic scenario. The data is captured with the 24 GHz LFMCW MIMO radar Inras RadarBook 2 and with the Microsoft Azure Kinect camera. All the measurements are recorded with the approval of the ethics committee of TU Delft, under the project DARE-NERD. For further information regarding the radar and the Kinect configuration, and the experiment conditions, please refer to an article currently submitted for review.

The dataset contains a ".zip" file for each participant with:

• 3 ".mat" files, corresponding to each trial performed by the subject. Each file contains the radar data in the variable "tima_data". The dimensions are (fast time) x (slow time) x (channel index). Moreover, the radar configuration is contained in the variable "Ana".
• 3 ".csv" files, corresponding to each trial performed by the subject. Each file contains the 32 joint locations extracted with the Microsoft Azure Kinect camera for each time frame.
• A ".csv" file with the time shifts that need to be applied to the Kinect data to synchronize it with the radar data.

Moreover, the dataset provides

• An ".xlsx" file containg the identifier, gender, age, height, weight and shoe size of each participant.
• A matlab script to extract the radar data from the ".mat" file.
• A matlab script to extract the joint data from the ".csv" file.

The authors would like to thank the participants in the experiment for their availability.

• 2024-10-25 first online

• 2024-10-30 published, posted

4TU.ResearchData

Radar files: ".mat"; Kinect joint files: ".csv"

• Formación de Profesorado Universitario 2020 (grant code FPU20/06405) [more info...] Ministerio de Ciencia, Innovación y Universidades. Gobierno de España
• Ayudas complementarias de movilidad destinadas a beneficiarios del programa de formación del profesorado universitario (FPU) 2024 (grant code FPU20/06405 - EST24/00008) [more info...] Ministerio de Ciencia, Innovación y Universidades
• PD-RADAR: Cribado y monitorización del párkinson usando biometrías basadas en tecnología RADAR (PD-RADAR) (grant code TED2021-131688B-I00) Ministerio de Ciencia, Innovación y Universidades, Gobierno de España
• DARE (Distributed and Adaptive Radar for Enhanced sensing & classification) & NERD (Neuromorphic pErception fRamework for event-baseD radars) Dutch Research Council (NWO)

Information Processing and Telecommunications Center, Universidad Politécnica de Madrid
Microwave Sensing, Signals and Systems (MS3) Group, Delft University of Technology