Jafait Junior Fodop Sokoudjou

This dataset is made of three subsets to train models for chipless RFID tags identification. There are two sets for training (one consisting of S21 recordings, and another one made of synthetically generated data) and one set made of S21 recordings for model testing. 

4800 + 2400 chipless RFID measurements of a population of 16 tags. Magnitude and phase are phase to allow DSP in the time domain. The measurement are made in a monostatic configuration with linearly-polarized antennas. The population of tags include circular ring resonators and square ring resonators. The dataset is used to trained convolutional neural networks. The inputs of the CNN is the continuous wavelet transform of the signals. The CWT is get from a previously selected portion of th signal in time domain.

The data are made of nine sets of measurements of four chipless RFID tags. The first two are made of 2600 measurements made at 160 centimeters (distance between the tag and the antenna) without (first) and with (second) initial background subtraction. The third set is made of 5600 measurements made in the range of 50 – 140 centimeters. The next six sets are measurements made to test models trained with the third set. All the measurements are made in the monostatic configuration.