Object Recognition Based on Nano Memristive Neural Network with Variable Structure

Abstract

Many Memristive neural network arrays that have been designed in recent years are simultaneously dealt with all of their synapses in working status. Therefore, when a relatively small-scale neural network is implemented with these memristor arrays, some of these synapses which are not used may cause errors in the result due to the impact of unexpected interference signals, and it can also cause some unnecessary energy consumption. In this paper, a memristive neural network with variable network structure is investigated. Based on this network, the number of synapses involved in the work can be flexibly adjusted to improve system performance. Two different scales of neural networks are simulated in Pspice to prove the feasibility and effectiveness of the proposed memristive neural network structure.