Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
9
datasets available to search
ShareScore release 0.9.0
Dataset results
9 results for “FMCW radar”
Aircraft Marshaling Signals Dataset of FMCW Radar and Event-Based Camera for Sensor Fusion
<p><strong>Dataset Introduction</strong></p><p>The advent of neural networks capable of learning salient features from variance in the radar data has expanded the breadth of radar applications, often as an alternative sensor or a complementary modality to camera vision. Gesture recognition for command control is arguably the most commonly explored application. Nevertheless, more suitable benchmarking datasets than currently available are needed to assess and compare the merits of the different proposed solutions and explore a broader range of scenarios than simple hand-gesturing a few centimeters away from a radar transmitter/receiver. Most current publicly available radar datasets used in gesture recognition provide limited diversity, do not provide access to raw ADC data, and are not significantly challenging. To address these shortcomings, we created and make available a new dataset that combines FMCW radar and dynamic vision camera of 10 aircraft marshalling signals (whole body) at several distances and angles from the sensors, recorded from 13 people. The two modalities are hardware synchronized using the radar's PRI signal. Moreover, in the supporting publication we propose a sparse encoding of the time domain (ADC) signals that achieve a dramatic data rate reduction (>76%) while retaining the efficacy of the downstream FFT processing (<2% accuracy loss on recognition tasks), and can be used to create an sparse event-based representation of the radar data. In this way the dataset can be used as a two-modality neuromorphic dataset.</p><p><strong>Synchronization of the two modalities</strong></p><p>The PRI pulses from the radar have been hard-wired to the event stream of the DVS sensor, and timestamped using the DVS clock. Based on this signal the DVS event stream has been segmented such that groups of events (time-bins) of the DVS are mapped with individual radar pulses (chirps).</p><p><strong>Data storage</strong></p><p>DVS events (x,y coords and timestamps) are stored in structured arrays, and one such structured array object is associated with the data of a radar transmission (pulse/chirp). A radar transmission is a vector of 512 ADC levels that correspond to sampling points of chirping signal (FMCW radar) that lasts about ~1.3ms. Every 192 radar transmissions are stacked in a matrix called a radar frame (each transmission is a row in that matrix). A data capture (recording) consisting of some thousands of continuous radar transmissions is therefore segmented in a number of radar frames. Finally radar frames and the corresponding DVS structured arrays are stored in separate containers in a custom-made multi-container file format (extension .rad). We provide a (rad file) parser for extracting the data out of these files. There is one file per capture of continuous gesture recording of about 10s.</p><p>Note the number of 192 transmissions per radar frame is an ad-hoc segmentation that suits the purpose of obtaining sufficient signal resolution in a 2D FFT typical in radar signal processing, for the range resolution of the specific radar. It also served the purpose of fast streaming storing of the data during capture. For extracting individual data points for the dataset however, one can pool together (concat) all the radar frames from a single capture file and re-segment them according to liking. The data loader that we provide offers this, with a default of re-segmenting every 769 transmissions (about 1s of gesturing).</p><p><strong>Data captures directory organization (</strong><a href="https://zenodo.org/api/records/10359770/draft/files/radar8Ghz-DVS-marshaling_signals_20220901_publication_anonymized.7z/content">radar8Ghz-DVS-marshaling_signals_20220901_publication_anonymized.7z</a><strong>)</strong></p><p>The dataset captures (recordings) are organized in a common directory structure which encompasses additional metadata information about the captures.</p><p>dataset_dir/<stage>/<room>/<person>-<gesture>-<distance>/ofxRadar8Ghz_yyyy-mm-dd_HH-MM-SS.rad</p><p>Identifiers</p><ul><li>stage [train, test].</li><li>room: [conference_room, foyer, open_space].</li><li>subject: [0-9]. Note that 0 stands for no person, and 1 for an unlabeled, random person (only present in test).</li><li>gesture: ['none', 'emergency_stop', 'move_ahead', 'move_back_v1', 'move_back_v2', 'slow_down' 'start_engines', 'stop_engines', 'straight_ahead', 'turn_left', 'turn_right'].</li><li>distance: ['xxx', '100', '150', '200', '250', '300', '350', '400', '450'] (in cm). Note that xxx is used for none gestures when there is no person present in front of the radar (i.e. background samples), or when a person is walking in front of the radar with varying distances but performing no gesture.</li></ul><p>The test data captures contain both subjects that appear in the train data as well as previously <i>unseen</i> subjects. Similarly the test data contain captures from the spaces that train data were recorded at, as well as from a new <i>unseen</i> open space.</p><p><strong>Files List</strong></p><p><a href="https://zenodo.org/api/records/10359770/draft/files/radar8Ghz-DVS-marshaling_signals_20220901_publication_anonymized.7z/content">radar8Ghz-DVS-marshaling_signals_20220901_publication_anonymized.7z</a></p><p>This is the actual archive bundle with the data captures (recordings).</p><p><a href="https://zenodo.org/api/records/10359770/draft/files/rad_file_parser_2.py/content">rad_file_parser_2.py</a></p><p>Parser for individual .rad files, which contain capture data.</p><p><a href="https://zenodo.org/api/records/10359770/draft/files/loader.py/content">loader.py</a></p><p>A convenience PyTorch Dataset loader (partly Tonic compatible). You practically only need this to quick-start if you don't want to delve too much into code reading. When you init a DvsRadarAircraftMarshallingSignals class object it automatically downloads the dataset archive and the .rad file parser, unpacks the archive, and imports the .rad parser to load the data. One can then <i>request from it </i>a training set, a validation set and a test set as torch.Datasets to work with<i>.</i> </p><p><a href="https://zenodo.org/api/records/10359770/draft/files/aircraft_marshalling_signals_howto.ipynb/content">aircraft_marshalling_signals_howto.ipynb</a></p><p>Jupyter notebook for exemplary basic use of loader.py</p><p><strong>Contact</strong></p><p>For further information or questions try contacting first M. Sifalakis or F. Corradi.</p><p> </p>
Radar measurements on drones, birds and humans with a 77GHz FMCW sensor.
<p>This data set contains radar measurements on birds, humans and six different drones with a total of 75868 samples.</p> <p>The sensor was a frequency modulated continuous wave (FMCW) radar operating at 77 GHz with a mechanically scanning antenna.</p> <p>The 'ReadMe.txt' file contains a detailed description of the data.</p> <p>The data set is used in [1] where only FM-sweeps corresponding to azimuth index 54 to 203 are used (out of the provided 256), or 150 sweeps. </p> <p>When using this data set please refer to:</p> <p>[1] A. Karlsson, M. Jansson and M. Hämäläinen, "Model-Aided Drone Classification Using Convolutional Neural Networks," <em>2022 IEEE Radar Conference (RadarConf22)</em>, 2022, pp. 1-6, doi: 10.1109/RadarConf2248738.2022.9764194.</p> <p>The data in version 1.0 and 2.0 is identical apart from the format, ".mat" in 1.0 and ".npy" in 2.0</p>
Radar-based sensing of wind turbines blades based on 35 GHz FMCW sensors installed at operational wind turbine towers
<p>The dataset contains radar-based measurements of rotor blades from three operational wind turbines as part of a structural health monitoring system. For this purpose, a sensor box with a 35 GHz radar sensor (about 1 000 measurements per second) and a camera system (about 100 images per second), is mounted on each wind turbine tower at approximately 100 m height. In order to distinguish individual rotor blades, a machine-readable marker printed on a self-adhesive film was applied on the blade’s surface. When a rotor blade passes the sensor, the camera captures an image of the marker while the radar records a measurement. The marker is then identified and the recorded data is assigned to a particular rotor blade. The measurements demonstrate that the damage detection methodology can be transferred to an image processing problem. The challenge is to manage the strong influence from variable environmental and operational conditions, e.g. wind speed, azimuth orientation, that modify the rotor blade appearance in the radargram significantly. The dataset contains measurements from the intact turbine blade conditions, because it was not possible to introduce structural damage.</p>
Low-Range FMCW Insect Radar - Lab Experiments Data, Results, and Data Processing Software
<p>This research was funded by the Hessian State Ministry for Higher Education, Research and the Arts, Germany, as part of the LOEWE priority project Nature 4.0—Sensing Biodiversity.</p>
An IoT System for Smart Building Combining Multiple mmWave FMCW Radars Applied to People Counting
<p>This repository is related with the next paper. Please cite us if this code is useful to you.</p> <p>Barral, V., Dominguez-Bolano, T., Escudero, C. J., & Garcia-Naya, J. A. An IoT System for Smart Building Combining Multiple mmWave FMCW Radars Applied to People Counting.</p> <h2>Python Scripts</h2> <ul> <li><em>count_plot.py</em>: Generates a plot comparing image tracking estimation, radar with DBSCAN, and radar with OPTICS. Use example:</li> </ul> <pre><code>python count_plot.py count_video_full.log count_radar_moving_average_30_optics.log count_radar_moving_average_30_dbscan.log "Video" "Radar with OPTICS" "Radar with DBSCAN" average_count_full_optics_dbscan.pdf </code></pre> <h2>Radar measurements</h2> <ul> <li><em>test_0.bag</em>: FMCW mmWave radar measurements from three IWR6843 devices (ISK and AOP). Is a ROS (Robotic Operative System) log, can be played with <code>rosbag play test_0.bag</code></li> </ul> <h3>Radar people counting logs</h3> <ul> <li> <p><strong>count_radar_moving_average_30_dbscan.log</strong>: radar people counting estimation using a 30 seconds moving average and DBSCAN as clustering algorithm.</p> </li> <li> <p><strong>count_radar_moving_average_30_optics.log</strong>: radar people counting estimation using a 30 seconds moving average and OPTICS as clustering algorithm.</p> </li> </ul> <h2>Image tracking</h2> <h3>Videos</h3> <ul> <li><em>test_0_cam_0_anonymized.mp4</em>: Camera 0 capture.</li> <li><em>test_0_cam_1_anonymized.mp4</em>: Camera 1 capture.</li> </ul> <h3>Image tracking logs</h3> <ul> <li><em>test_0_cam_0_interp.txt</em>: People count in camera 0</li> <li><em>test_0_cam_1_interp.txt</em>: People count in camera 1</li> </ul> <p>The files have the following format:</p> <p>frame_number, track_id, bb_left, bb_top, bb_width, bb_height, conf, x, y, z The coordinates x, y, z are always -1 The bb fields define the bounding box of the detection If multiple people are detected in a frame, there are several lines with that frame_number, but with different track_ids.</p> <h3>Image people counting logs</h3> <ul> <li><strong>count_video_full.log</strong>: image tracking people counting estimation using a 30 seconds moving average.</li> </ul>
NOAA PSL Snow-level Radar FMCW Moments KettlePonds for SPLASH
<p>--------------------<br> Raw Output Files <br>--------------------</p> <p>Data Format for Radar FMCW Moments at BCK and KPS locations</p> <p>Note: Neither the FMCW radar at BCK (BrushCreek) nor KPS (KettlePonds) locations were calibrated. This means that the radar reflectivity derived from these systems can not be compared directly with the reflectivity factor (dBZ) from the NWS NEXRAD radar (e.g., near Grand Junction). In addition, the reflectivity-related variables (like signal power and radar signal-to-noise ratio) from each FMCW system will not be equivalent when sensing the same atmospheric backscattered reflectivity.<br> <br> <br>Header line 1 (space delimited fields from left to right):<br> Three-letter site identifier<br> Site latitude (degrees x 100)<br> Site longitude (degrees x 100)<br> Site altitude (m)<br> Operating frequency (Gigahertz x 10^5)</p> <p>Header line 2 (space delimited fields from left to right):<br> Beam index<br> Azimuth angle (degrees)<br> Elevation angle (degrees)<br> Last two-digits of year (start time of dwell)<br> Day of year (start time of dwell)<br> Hour of day (start time of dwell)<br> Minute (start time of dwell)<br> Second (start time of dwell)</p> <p>Header line 3 (space delimited fields from left to right):<br> Inter-pulse period (microseconds)<br> Pulse width (nanoseconds)<br> Range to first range gate (m)<br> Distance between range gates (m)<br> Number of ranges gates<br> Number of coherent integrations<br> Number of averaged spectra<br> Number of Fast-Fourier-Transform points<br> Number of code bits</p> <p>Data (space delimited columns from left to right for N rows, where<br> the first (Nth) row corresponds to the first (Nth) range gate):<br> Radial velocity (m/s)<br> Signal-to-noise ratio (dB) - not range corrected<br> Signal power (dB) - not range corrected<br> Spectral width (m/s)<br> Noise amplitude (dB)<br> QC value - not applicable (all zeros)<br> <br>Repeat Header lines 2-3 and Data for each succeeding radar dwell </p> <p> </p> <p>Year-Day-Hour</p> <p>Start: 21-272-17</p> <p>End: 23-249-15</p> <p> </p> <p>Missing data for KpsFmcwRaw</p> <p>Year-Day-Hour (# of files)</p> <p>21-272-18 to 21-273-18 (25)</p> <p>21-278-22 to 21-279-114 (17)</p> <p>21-288-11 to 21-293-21 (130)</p> <p>21-293-23 to 21-299-22 (144)</p> <p>21-358-14-17 (4)</p> <p>22-07-17 to 22-12-18 (122)</p> <p>22-14-17 to 22-15-22 (30)</p> <p>22-16-04 to 22-18-17 (62)</p> <p>22-38-01-17 (17)</p> <p>22-69-16 to 22-74-15 (120)</p> <p>22-89-05 to 22-90-23 (43)</p> <p>22-93-14 to 22-95-16 (51)</p> <p>22-96-07-16 (10)</p> <p>22-140-7 (1)</p> <p>22-157-20 to 22-201-20 (1056)</p> <p>22-255-13 to 22-256-17 (29)</p> <p>22-256-19 to 22-257-15 (21)</p> <p>22-257-20-23 (4)</p> <p>22-312-00-23 (24)</p> <p>22-233-16 to 22-234-19 (28)</p> <p>22-350-17 to 22-365-23 (367)</p> <p>23-54-14 to 23-55-21 (32)</p> <p>23-70-07 to 23-80-18 (252)</p> <p>23-148-00 to 23-149-23 (48)</p>
NOAA PSL Snow-level Radar FMCW Moments BrushCreek for SPLASH
<p>--------------------<br> Raw Output Files <br>--------------------</p> <p>Data Format for Radar FMCW Moments at BCK and KPS locations</p> <p>Note: Neither the FMCW radar at BCK (BrushCreek) nor KPS (KettlePonds) locations were calibrated. This means that the radar reflectivity derived from these systems can not be compared directly with the reflectivity factor (dBZ) from the NWS NEXRAD radar (e.g., near Grand Junction). In addition, the reflectivity-related variables (like signal power and radar signal-to-noise ratio) from each FMCW system will not be equivalent when sensing the same atmospheric backscattered reflectivity.<br> <br> <br>Header line 1 (space delimited fields from left to right):<br> Three-letter site identifier<br> Site latitude (degrees x 100)<br> Site longitude (degrees x 100)<br> Site altitude (m)<br> Operating frequency (Gigahertz x 10^5)</p> <p>Header line 2 (space delimited fields from left to right):<br> Beam index<br> Azimuth angle (degrees)<br> Elevation angle (degrees)<br> Last two-digits of year (start time of dwell)<br> Day of year (start time of dwell)<br> Hour of day (start time of dwell)<br> Minute (start time of dwell)<br> Second (start time of dwell)</p> <p>Header line 3 (space delimited fields from left to right):<br> Inter-pulse period (microseconds)<br> Pulse width (nanoseconds)<br> Range to first range gate (m)<br> Distance between range gates (m)<br> Number of ranges gates<br> Number of coherent integrations<br> Number of averaged spectra<br> Number of Fast-Fourier-Transform points<br> Number of code bits</p> <p>Data (space delimited columns from left to right for N rows, where<br> the first (Nth) row corresponds to the first (Nth) range gate):<br> Radial velocity (m/s)<br> Signal-to-noise ratio (dB) - not range corrected<br> Signal power (dB) - not range corrected<br> Spectral width (m/s)<br> Noise amplitude (dB)<br> QC value - not applicable (all zeros)<br> <br>Repeat Header lines 2-3 and Data for each succeeding radar dwell </p> <p> </p> <p>Year-Day-Hour</p> <p>Start: 21-354-23</p> <p>End: 23-248-20</p> <p> </p> <p>Missing data for BckFmcwRaw</p> <p>Year-Day-Hour (# of files)</p> <p>22-54-03 to 22-54-16 (14)</p> <p>22-58-01 (1)</p> <p>22-58-05 (1)</p> <p>22-58-17 to 22-58-18 (2)</p> <p>22-58-21 (1)</p> <p>22-58-23 to 22-59-02 (4)</p> <p>22-75-16 to 22-75-17 (2)</p> <p>22-79-17 to 22-79-18 (2)</p> <p>22-79-20,21,23 (3)</p> <p>22-80-0 to 22-97-17 (426)</p> <p>22-172-17 to 22-172-23 (7)</p> <p>23-102-7,11-15 (6)</p> <p>23-190-21,22 (2)</p> <p>23-204-19 (1)</p> <p>23-205-20,21 (2)</p> <p>23-233-21,22 (2)</p>
Breathing Rate and Heart Rate Dataset using Integrated mmWave FMCW Radar and Camera Steering System
<p>The presented dataset consists of raw and transformed CWT images of the breathing and heart waveforms obtained from a radar<br> and IP camera setup. The setup is a self-proposed setup with a mmWave radar mounted over an IP camera and can capture<br> the breathing and heart waveforms of the person in the room in front of the system in any orientation. The dataset can be used to estimate the vital signs using any machine learning model and important information about the respiration rate and pulse rate<br> can then be obtained. The dataset is a total of 1280 images- 720 raw and 720 processed. The processed dataset is labeled in six different classes. The first bifurcation is between breath and heart, breath signal is classified as low, normal, and high whereas the heart signal is classified as low, normal, and slightly low. The subject is oriented in differently so that the setup can steer towards the person and capture the breath and heart signals accordingly.</p>
CLPX-Ground: Ground-Based Frequency Modulated Continuous Wave (FMCW) Radar, Version 1
This data set is comprised of FMCW radar measurements, which are sensitive to electromagnetic discontinuities in the snowpack made during the 2002 and 2003 Cold Land Processes Field Experiment (CLPX).
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.