Datasets & Competitions
The MLC ETI is dedicated to foster the application of ML in communications by presenting datsets and competitions tailored for communication society. The target is to establish a set of common problems and corresponding datasets on which researchers can benchmark and compare their algorithms in a reproducible and credible way.
Datasets
- 1.25GHz localization dataset: From the IEEE CTW 2019 Challenge
- Body area network radio channel: Measurement set with transmit-receive link gain.
- CRAWDAD up/rf_recordings: RF recordings of several communication signals.
- CRAWDAD rutgers/noise: Received signal strength indicator from ORBIT testbed.
- DeepMIMO: A generic deep learning dataset for millimeter wave and massive MIMO applications
- Distributed Massive MIMO: Outdoor and outdoor-to-indoor measurements with 64 antennas and 18 users.
- Power Allocation in Multi-Cell Massive MIMO: Dataset for power allocation in a Massive MIMO network.
- RadioML: Recordings of digital and analog modulation types.
- Sussex-Huawei Locomotion Dataset: Annotated dataset for multimodal locomotion analytics of mobile users.
- RF WebLab: Access to an amplifier and measurement system for digital pre-distortion.
- ViWi: A deep learning dataset framework for vision-aided wireless communications.
- Device Identification: IoT device identification dataset.
- UWB Localization dataset: UWB localization data set contains measurements from four different indoor environments and can be used for range-based localization evaluation.
- 5G Performance: throughput performance, latency measurements, impact of mobility and obstructions, handoff analysis.
Open Dataset Initiative
The MLC ETI strongly encourage that new high-quality datasets are made openly available. If you have published a dataset that is not listed above, please contact the Datasets & Competitions Officer over email or slack with your interest. We can also help you with hosting the dataset at the IEEE MLC Datasets Server.
Data Competitions
- ITU AI/ML in 5G Challenge 2020: Applying Machine Learning in Communication Networks
- IEEE ICC 2020: Vision-Aided Beam Tracking for mmWave Systems
- IEEE CTW 2020: Self-Supervised Learning for User Localization
- IEEE CTW 2019: Positioning Algorithm Competition.