粒子群算法和遗传算法提取闪存器件模型参数的对比  

Comparison of particle swarm algorithm and genetic algorithm for extracting flash memory device model parameters

作　　者：闫国亮 丁洁[1] YAN Guoliang;DING Jie(Taiyuan University of Technology,Taiyuan 030024,China)

机构地区：[1]太原理工大学,山西太原030024

出　　处：《现代电子技术》2022年第6期29-34,共6页Modern Electronics Technique

基　　金：国家自然科学基金项目(61604105)。

摘　　要：为了有效指导闪存单元的模型参数提取,文中基于BSIM4模型,选取了合适的模型参数,利用粒子群算法和遗传算法作对比,找到了一种最佳的闪存器件模型参数提取方法。POM分子闪存器件是一种新型浮栅闪存器件,将作为研究对象被用于文中模型参数的提取工作。首先在BSIM4中选取要进行提取的参数;然后分别通过粒子群算法和遗传算法优化得到模型参数的最优值,使最终得到的模型仿真数据尽可能拟合对应的器件物理仿真数据,也就是最小化提取模型仿真数据和物理仿真数据之间的误差。结果表明:利用粒子群优化算法可以使模型仿真数据和物理数据较好地拟合,并且最终模型的参数提取误差均在2%以下,而遗传算法提取结果的最大误差为2.8963%;在参数提取误差和计算量方面,粒子群算法都优于遗传算法。In order to effectively guide the extraction of model parameters of flash memory cells,appropriate model parameters are selected based on the BSIM4 model.An optimal method for extracting model parameters of flash memory device is found by comparing the particle swarm algorithm and genetic algorithm.As a new type of floating gate flash memory device,POM molecular flash memory device is taken as the research object to extract the model parameters in this paper.The parameters to be extracted are selected in BSIM4,and then the optimal values of model parameters are optimized by means of the particle swarm optimization algorithm and genetic algorithm respectively,so that the model simulation data obtained finally can fit the corresponding device physical simulation data as much as possible,that is,the error between the extracted model simulation data and physical simulation data can be minimized.The results show that the particle swarm optimization algorithm can make the model simulation data and physical data fit well,and the parameter extraction error of the final model is less than 2%,while the maximum error of the extraction result of genetic algorithm is 2.8963%.In terms of parameter extraction error and calculation amount,the particle swarm optimization algorithm is better than the genetic algorithm.

关 键 词：模型参数提取 闪存器件模型 粒子群算法 遗传算法 参数优化 数据拟合 

分 类 号：TN919-34[电子电信—通信与信息系统] TP311[电子电信—信息与通信工程]