[CI4R-TOBB4]

4 GHz CW Human Activity Dataset

Details about radar utilized:

• ​Radar measurements were made in an indoor laboratory environment spanning a range of 1–5 m using an NI-USRP2922 model software-defined radio platform programmed to transmit a continuous wave signal at 4 GHz.
  
• Two SAS-571 antennas having a 48 deg. azimuthal beam width were mounted along with the USRP 1 m above the ground.
  
• Measurements were taken such that the direction of motion were directly aligned with the center of the antenna beam pattern.

• # Classes: 12
  
• # Samples/class: A total of1007 samples were acquired distributed for each class as
  • Walking [71]
  • Jogging [72]  
  • Crawling [74]
  • Limping [104]
  • Cane [123]   
  • Falling [53]   
  • Wheelchair [149]
  • Crutches [74]   
  • Sitting [50]   
  • Walker [121]   
  • Falling from Chair [60]   
  • Creeping [56]
• GitHub Link: https://github.com/ci4r/4-GHz-Human-Activity-Dataset-​
  
• ​Publications to be cited:​
  • ​M.S. Seyfioglu, A.M. Ozbayoglu, S.Z. Gurbuz, “Deep convolutional autoencoder for radar-based classification of similar aided and unaided human activities,” IEEE Transactions on Aerospace and Electronic Systems, vol. 54, iss. 4, August 2018.​
  • M.S. Seyfioglu and S.Z. Gurbuz, “Deep neural network initialization methods for micro-Doppler classification with low training sample support,” IEEE Geoscience and Remote Sensing Letters, vol. 14, iss. 12, December 2017.