Analyzing biological neural network activities is crucial for understanding neural communication and function. Conventional methods involve transmitting and processing large datasets offline, hindering real-time analysis. This paper presents a memristor-based reservoir computing (RC) system for real-time neural signal analysis. A perovskite halide-based memristor, driven by emulated neural spikes, reflects temporal features in spike trains. The RC system successfully recognizes firing patterns, monitors their transitions, and identifies neural synchronization states. This approach enables efficient neural signal analysis with high spatiotemporal precision and potential for closed-loop feedback control.

Xiaojian Zhu, Qiwen Wang, Wei D. Lu