Datasets
Standard Dataset
Raw_radar_fmcw_Dataset
- Citation Author(s):
-
WissalZarrami
Polytechnique Montréal - Submitted by:
- Wissal Zarrami
- Last updated:
- Tue, 04/15/2025 - 15:12
- DOI:
- 10.21227/fa4h-cj47
- Data Format:
- License:
27 Views- Categories:
- Keywords:
Abstract
This dataset contains raw FMCW radar signals collected for human localization and activity monitoring in indoor environments. The data was recorded using mmWave radar sensors across two different laboratory settings, designed to simulate real-life scenarios for human detection and localization tasks.
The dataset includes experiments with 1, 2, and 3 people, covering various positions, walking trajectories, and intermediate locations to evaluate model generalization. Ground truth information is embedded in the folder names, indicating the true distance and angle of the targets.
Each data folder contains 18 files: 9 files per antenna capturing complex-valued signals (.cf32 format) for multi-antenna processing. The signals are recorded without any preprocessing or feature extraction, enabling researchers to explore raw data learning, end-to-end modeling, or classical signal processing pipelines.
This dataset is especially suitable for research in radar-based human localization, privacy-preserving monitoring, signal enhancement, and machine learning on raw radar data.
This dataset contains raw FMCW radar signals collected for human localization and monitoring applications in indoor environments.
Each folder corresponds to a specific scenario based on:
-
Number of people (1, 2, or 3)
-
Target angle and distance (mentioned in folder name)
-
Environment (Lab1 or Lab2)
Inside each folder:
-
18 files: 9 files per antenna in
.cf32format (complex float32). -
Each
.cf32file contains raw complex IQ radar samples without any preprocessing.
To load the data in Python:
import numpy as np data = np.fromfile('rx1_0.cf32', dtype=np.complex64)
Ground truth labels are provided in folder names.
This dataset is suitable for radar signal processing, machine learning on raw signals, and human localization tasks.
More from this Author
