[CI4R-HAR77]

77 GHz FMCW Human Activity Dataset

• Details about radar utilized: 
  • ​TI AWR1642 BOOST (77 GHz FMCW with 4 GHz bandwidth)
    
• # Classes:   14

1. VCMF (Vacuuming)                         
2. WCANE (Walking with cane)                       
3. CRTH (Walking with crutches)                   
4. LD1H (Carry a load in one hand)                                
5. LD2H (Carrying a load in both hands)       
6. BOOK (Putting books on shelf)
7. WTOES (Walking on toes)
8. SCIS (Scissors gait)
9. SKIP (Skipping)                                   
10. LIMP (Limping)
11. DRAGF (Dragging furniture)             
12. FWDM (Forward marching)
13. WALK (Walking)                                 
14. SSTEP (Short steps)

• # Samples/class: ~100
  
• GitHub Link: https://github.com/ci4r/15_Class_Activity_Dataset
  
• ​Publications to be cited:
  • ​M. M. Rahman, S. Z. Gurbuz and M. G. Amin, “Physics-Aware Generative Adversarial Networks for Radar-Based Human Activity Recognition,” in IEEE Transactions on Aerospace and Electronic Systems, 2022, doi: 10.1109/TAES.2022.3221023.