Datasets

RadarCommDataset: Radar & Communication Modulation and Signal Classification Dataset

Abstract

RadarCommDataset is a wireless signal dataset released for public use under the Creative Commons Attribution - NonCommercial - ShareAlike 4.0 License (CC BY-NC-SA 4.0) [Additional info can be found at the end] by MR Lab at ANDRO. The lack of existing multitask labeled datasets for machine learning for wireless communication is the prime motivation urging this release. RadarCommDataset is the first of its kind multitask labeled dataset released to help the research community to advance machine learning for wireless communication. The dataset contains radar and communication waveforms. This repository supplements helper scripts to visualize as well as extract the dataset. Please contact for any other license.

Download link

GitHub Link

Format description

• Type - HDF5

• Key - {modulation,signal,snr,sample}

• Tensor - 256 x 1 - {inphase of length 128, quadrature of length 128}

• Sampling rate - 10 MS/s

• Total snapshots per waveform - 700

• SNR range - -20 dB to 18 dB in steps of 2 dB

• Modulation Classes - Pulsed Continuous wave, Frequency modulated Continuous wave, BPSK, AM-DSB, AM-SSB, ASK

• Signal Classes - Airborne detection radar, Airborne range radar, Air-Ground MTI radar, Ground mapping radar, Radar Altimeter, SATCOM, AM Radio, Short-range wireless

How to cite this dataset ?

• BibTeX

@inproceedings{JagannathMTL, title={{Multi-task Learning Approach for Automatic Modulation and Wireless Signal Classification}}, author={Jagannath, Anu and Jagannath, Jithin}, booktitle = {Proc. of IEEE International Conference on Communications (ICC)}, address = {Montreal, Canada}, month = {June}, year = {2021} }

• Plain Text

A. Jagannath, J. Jagannath. "Multi-task Learning Approach for Automatic Modulation and Wireless Signal Classification", in Proc. of IEEE International Conference on Communications (ICC), Montreal, Canada, June 2021.

RF-Fingerprinting-BT-IoT: Real-World Frequency Hopping Bluethooth Dataset from ioT Devices for RF Fingerpinting

Abstract

A real-world radio frequency (RF) fingerprinting dataset for commercial off-the-shelf (COTS) Bluetooth emitters under challenging testbed setups is presented in this dataset. The dataset includes emissions from 10 COTS IoT emitters (2 laptops and 8 commercial chips) that are captured with a National Instruments Ettus USRP X300 radio outfitted with a UBX160 daughterboard and a VERT2450 antenna. The receiver is tuned to record a 2 MHz bandwidth of the spectrum centered at the 2.414 GHz frequency. This is a first-of-its-kind dataset for fingerprinting Bluetooth emitters under challenging and diverse indoor laboratory setups. The dataset is split into two: Day1 and Day2 each of which is recorded in a different time frame, location, and testbed setup to enable critical generalization test of the trained deep learning (DL) model. The authors suggest training the DL model with the Day1 dataset which is a simpler setup with recordings under varied Bluetooth signal strengths followed by evaluating the generalization capability of the model with Day2 dataset which is a challenging and vastly different setup compared to Day1. This evaluation will validate the real-world deployment capability of the trained DL model. The dataset follows the SigMF specifications with certain field extensions to facilitate the fingerprinting application and include additional metadata fields. Each capture is of length 40 Mega Samples (MS) and is associated with a JSON metadata file.

Download link

IEEE Dataport

Instructions:

The dataset contains two separate Bluetooth datasets collected under different time frames, locations, and testbed setups.

1. Day1BT.tar.gz: Collected under line-of-sight conditions at varying distances per emitter as indicated in the associated JSON metadata file of each capture. The transmitter-receiver separation ranges from 1.6 ft to 9.8 ft in steps of 0.8 ft.

2. Day2BT.tar.gz: Collected under a challenging and rich multipath scattering scenario with maximum separation between the emitter and receiver being 24.2 ft. Each emitter is placed in the corners of the indoor laboratory as indicated by the lab layout picture with the receiver placed stationary at the center of the laboratory.

Each dataset contains a binary capture file (.dat) and an associated JSON metadata file (.meta.json). The capture file contains complex64 inphase-quadrature (IQ) samples and the JSON contains the metadata which characterizes the data capture environment.

How to cite this dataset? Please cite the original paper.

Bibtex:

@article{Jagannath2022, author = "Anu Jagannath and Jithin Jagannath", title = "{Embedding-Assisted Attentional Deep Learning for Real-World RF Fingerprinting of Bluetooth}", year = "2022", month = "9", url = "https://www.techrxiv.org/articles/preprint/Embedding-Assisted_Attentiona...", doi = "10.36227/techrxiv.20767315.v1" }

Additional License Info for ALL Datasets:

You must include the following in (a) any copy or distribution of this dataset or portion of this data set, (b) any Derivation (defined below) of this dataset or portion of this data set, or (c) any new data set or work that extracts from or modifies this data set or portion of this data set in any way:

Ownership and License. This data set is owned by ANDRO Computational Solutions, LLC (https://www.androcs.com/) and is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) located at https://creativecommons.org/licenses/by-nc-sa/4.0/ (the “Creative Commons License”)

Derivations. “Derivations” means that you create a new work or new data set that aggregates this data set or any portion of this dataset, uses this data set or any portion of this for any calculations, and/or is derived from or is based on this data set or any portion of this dataset.

Attribution. You must prominently display ANDRO Computational Solutions, LLC and the ANDRO logo: