CNN-Based Receiver Design for Chirp-Based Asynchronous Pulse Code Multiple Access for Massive IoT

Asynchronous Pulse Code Multiple Access (APCMA) is a wireless access scheme for massive IoT that offers strong robustness to large-scale packet transmissions by encoding information in pulse trains, as far as pulses are not lost. However, conventional threshold-based pulse detection suffers from pulse loss, false detection, and the need for careful parameter tuning, especially under low-SNR conditions. This letter proposes a machine learning (ML)-based pulse detection method for chirp spread spectrum-based APCMA. The proposed method is implemented on a GNU Radio/USRP receiver in the 920 MHz band, and its performance is evaluated in high-attenuation wired environments and large-scale wireless experiments. The results show that the proposed receiver achieves higher sensitivity without threshold tuning and yields consistently better performance than LoRa in collision-prone environments.