DeepVerse 6G
DeepVerse 6G is a digital twin dataset designed to generate synthetic, high-fidelity, multi-modal sensing and communication data for 6G research.
- Enables flexible generation by defining datasets via scenarios and parameter sets
- Ensures dataset reproducibility for benchmarking and comparisons
- Combines accurate wireless ray-tracing with photo-realistic visuals
- Generates synchronized wireless comm., vision, LiDAR, radar, and position/IMU data
- Provides synthetic environments that replicate real-world DeepSense 6G scenarios
- Several indoor/outdoor scenarios with high diversity
DeepSense 6G
DeepSense 6G is a real-world multi-modal dataset that comprises coexisting multi-modal sensing and communication data, such as mmWave wireless communication, Camera, GPS data, LiDAR, and Radar, collected in realistic wireless environments
- World’s first large-scale real-world sensing-communication dataset
- 1Million+ multi-modal sensing-communication data samples
- Modalities include mmWave comm., GPS, Image, LiDAR and Radar
- 30+ scenarios targeting various applications
- Several indoor/outdoor locations with high diversity
- Data collected at different times of day and weather conditions
- Tens of thousands of manually and auto labeled data
DeepMIMO
DeepMIMO is a framework for generating large-scale wireless MIMO datasets based on accurate 3D Ray-tracing
- Completely defined by the ray-tracing scenario and the set of parameters
- Enables a wide range of machine learning communication and sensing applications
- Allows dataset reproducibility for benchmarking and comparisons
- Generating datasets compatible with 5G NR channel models and numerologies
- Several indoor/outdoor scenarios with high diversity
- New scenarios are added on a regular basis