PCM resistance prediction method based on artificial neural network

Abstract

The invention discloses a PCM resistance prediction method based on an artificial neural network. The PCM resistance prediction method comprises the following steps of S1, selecting sample data; S2, performing simulation processing on the data; S3, converting the simulation data into an array, and performing normalization processing; S4, dividing the processed data into a training set and a test set; S5, building a neural network model, and training the data of the training set through a neural network; and S6, testing the data of the test set through the trained neural network. According to the invention, proper phase change memory size parameters are selected as samples of the neural network; building a neural network model; the neural network is trained through the training set data sample, the neural network is tested through the test set data sample, the output accuracy of the neural network is ensured, the PCM resistance prediction is combined with the artificial neural network,the purpose of rapidly and accurately predicting the PCM resistance is achieved, and the PCM resistance prediction duration is shortened.

Description

technical field [0001] The invention relates to the technical field of memory, in particular to a PCM resistance prediction method based on an artificial neural network. Background technique [0002] With the rapid development of information technology in today's world, human society has entered the era of big data. People have gradually formed a huge data network through continuous creation and acquisition of information. How to save these data efficiently and completely has brought many challenges to the storage field. Therefore, In order to store massive amounts of information, people's demand for the development of high-speed memory is becoming more and more urgent. [0003] At present, mainstream storage devices on the market include Dynamic Random Access Memory (DRAM) and Flash Memory (FM), and DRAM and FM occupy most of the semiconductor storage market. Traditional DRAM has large capacity and fast read and write speed. , high reliability and other advantages, but the...

AI Technical Summary

Problems solved by technology

[0008] The present invention mainly solves the problem of long time-consuming prediction of PCM resistance in the existing technology; provides a PCM resistance prediction method based on artificial neural network, quickly predicts the resistance of phase change memory, and improves the phase change Memory Resistance Estimation Efficiency

Images