Memristor-based Brain-like Reconfigurable Neuromorphic System

Abstract

It improves the data processing performance and efficiency notably that the memristor-based neuromorphic system is constructed by mimicking the features of brain. This paper proposes a memristor-based circuit implementation for brain-like reconfigurable neuromorphic system. This neuromorphic system contains memristor-based back-propagation neural network, memristor-based long short-term memory network, and memristor-based associative memory network. The reconfigurable circuit components can constitute the circuit hardware of the three memristor-based networks according to the task requirements or release the circuit hardware of the three memristor-based networks based on the forgetting mechanism along with time of the biology. Hence, this neuromorphic system has dynamic topology structure, which is similar as the feature of biological neural networks. A case study of the memristor-based brain-like reconfigurable neuromorphic system is presented in the paper. In the case study, the memristor-based back-propagation neural network and the memristor-based long short-term memory network are applied for the image recognition and speech recognition, respectively. The recognition results of the two memristor-based networks are input to the memristor-based associative memory network to recall the correlated information. This neuromorphic system has the potential to apply for the intelligent robot system.