Real-Time, Over-the-Air Modulation Recognition using a 320 MS/s Spiking Neural Network on FPGA

This work presents a high-throughput, fully digital SNN-based AMR system implemented on an FPGA, achieving a 320 MS/s input rate and latency of just 3.8 μs, with minimal hardware footprint. The design is validated using both loopback and over-the-air measurements, demonstrating reliable real-time modulation recognition under practical RF conditions. The proposed architecture offers a promising direction for resource-constrained AMR in high data rate applications.