Dataset Hub

RadioModRec-1 is a simulated dataset for Automatic Modulation Recognition (AMR) comprising fifteen digital modulation schemes, including 4QAM, 16QAM, 64QAM, 256QAM, 8PSK, 16PSK, 32PSK, 64PSK, 128PSK, 256PSK, CPFSK, DBPSK, DQPSK, GFSK and GMSK, which are widely used in modern wireless communication systems. The dataset supports Rayleigh and Rician channel models and additive white Gaussian noise (AWGN) conditions ranging from –20 dB to +20 dB in 5 dB steps. It was curated by Emmanuel Adetiba and Jamiu R. Olasina, with partial funding from Google’s TensorFlow Outreaches in Colleges program.

The dataset was generated using the open‑source TensorFlow‑based library Sionna for ray‑tracing simulations, covering reflection and scattering counts across different urban areas in China. The simulated area is a 620‑meter square, with both transmitters and receivers positioned at a height of 1.5 m. The dataset includes raw simulation output files, RSSI grayscale images, transmitter location images, and scene images.