Nonlinear Dyn https://doi.org/10.1007/s11071-018-4291-1 ORIGINAL PAPER Experimental verification of a memristive neural network I. Carro-Pérez · C. Sánchez-López · H. G. González-Hernández Received: 12 July 2017 / Accepted: 16 April 2018 © Springer Science+Business Media B.V., part of Springer Nature 2018 Abstract This paper presents an electronic circuit able to emulate the behavior of a neural network based on memristive synapses. The latter is built with two flux-controlled floating memristor emulator circuits operating at high frequency and two passive resistors. Synapses are connected in a way that a bridge circuit is obtained, and its dynamical behavioral model is derived from characterizing memristive synapses. Analysis of the memristor characteristics for obtaining a suitable synaptic response is also described. A neural network of one neuron and two inputs is connected using the proposed topology, where synaptic positive and nega- tive weights can easily be reconfigured. The behavior of the proposed artificial neural network based on mem- ristors is verified through MATLAB, HSPICE simula- tions and experimental results. Synaptic multiplication is performed with positive and negative weights, and its behavior is also demonstrated through experimental results getting 6% of error. I. Carro-Pérez · H. G. González-Hernández Department of Mechatronics, Tecnológico de Monterrey, Campus Puebla, Via Atlixcáyotl 2301, Reserva Territorial Atlixcáyotl, 72453 Monterrey, Mexico e-mail: icarrop@itesm.mx H. G. González-Hernández e-mail: hgonz@itesm.mx C. Sánchez-López (B ) Department of Electronics, Autonomous University of Tlaxcala, Clzda. Apizaquito S/N, km. 1.5, 90300 Apizaco, Tlaxcala, Mexico e-mail: carlsanmx@yahoo.com.mx Keywords Neural network · Memristor · Synapse · Pinched hysteresis loop · Current conveyor 1 Introduction In recent years, special attention has been paid for developing brain-inspired computers capable of emu- lating energy consumption and memory density of bio- logical neural networks. Indeed, the human brain can perform about 10 quadrillion operations per second [1], consuming only about 20W of power and per- form complex tasks such as pattern association, voice and image recognition [2]. On the one hand, factors that influence on brain emulation are: level of emu- lation and the computational model of different types of neurons and neural process. For instance in 1952, Hodgkin and Huxley proposed a model and a compact circuit for the nerve axon membrane [3]. From this sim- ple model, new neuron models were evolved such that several effects observed in biological neurons could be taken into account [4–6]. On the other hand, to under- stand brain functional characteristics at low level, how- ever, it requires specialized technology and measure- ment instruments with high performances [7]. A plau- sible level of emulation is the neuron and synapse level. Chemical or electrical synapse are sites where the infor- mation is received of a neuron and sent to other neuron and it is also related to learning and memory. Consider- ing the aforementioned, an electronic device that can be used for biological synapse emulation is the memristor, 123