[paper] SPICE PCM Model

A SPICE Model of Phase Change Memory for Neuromorphic Circuits

Xuhui Chen¹, Huifang Hu¹, Xiaoqing Huang¹, Weiran Cai², Ming Liu³ (Fellow, Ieee), Chung Lam⁴,  Xinnan Lin¹ (Member, IEEE), Lining Zhang⁵ (Senior Member, IEEE)

and Mansun Chan⁶ (Fellow, IEEE)

¹The Shenzhen Key Lab of Advanced Electron Device and Integration, ECE, Peking University Shenzhen Graduate School, Shenzhen 518055 CN
²Institute of Microscale Optoelectronics, Shenzhen University, Shenzhen 518061 CN
³Key Laboratory of Microelectronics Devices and Integration Technology, Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, and the University of Chinese Academy of Sciences, Beijing 100049 CN
⁴Jiangsu Advanced Memory Technology Co., Ltd, Huaian 223302 CN
⁵School of Electronic and Computer Engineering, Peking University, Shenzhen 518055, CN
⁶HKUST Shenzhen Research Institute, Shenzhen 518057, China, and Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology, HK

doi: 10.1109/ACCESS.2020.2995907

Abstract: A phase change memory (PCM) model suitable for neuromorphic circuit simulations is developed. A crystallization ratio module is used to track the memory state in the SET process, and an active region radius module is developed to track the continuously varying amorphous region in the RESET process. To converge the simulations with bi-stable memory states, a predictive filament module is proposed using a previous state in iterations of nonlinear circuit matrix under a voltage-driven mode. Both DC and transient analysis are successfully converged in circuits with voltage sources. The spiking-timedependent- plasticity (STDP) characteristics essential for synaptic PCM are successfully reproduced with SPICE simulations verifying the model’s promising applications in neuromorphic circuit designs. Further on, the developed PCM model is applied to propose a neuron circuit topology with lateral inhibitions which is more bionic and capable of distinguishing fuzzy memories. Finally, unsupervised learning of handwritten digits on neuromorphic circuits is simulated to verify the integrity of models in a large-scale-integration circuits. For the first time in literature an emerging memory model is developed and applied successfully in neuromorphic circuit designs, and the model is applicable to flexible designs of neuron circuits for further performance improvements. 

FIG: Schematic diagram of commonly used PCM mushroom structure

URL: https://IEEExplore.IEEE.org/stamp/stamp.jsp?tp=&arnumber=9097232&isnumber=6514899

Multiphysics Simulation of Biosensors

M. Madec, L. Hébrard, J. Kammerer, A. Bonament, E. Rosati and C. Lallement

Multiphysics Simulation of Biosensors Involving 3D Biological Reaction-Diffusion Phenomena in a Standard Circuit EDA Environmen

IEEE Transactions on Circuits and Systems I: Regular Papers.

doi: 10.1109/TCSI.2018.2885223

Abstract: The topic of this paper is the development of biological models for 3D reaction-diffusion phenomena that can be used in any circuit electronic design automation environment for the simulation of biosensors. Biological systems that involve such 3D phenomena are described by partial differential equations. Our approach consists in discretizing these equations according to the finite-difference method and converting the resulting ordinary differential equations into an assembly of elementary electronic equivalent circuits that are directly simulated with SPICE. The main interest of this approach is the ability to couple such models with third-party SPICE models of electronic circuits, sensors, and transducers, i.e., models from any physical domain ruled by Kirchhoff laws, allowing modeling and simulation of any multi-physics systems in a conventional circuit design environment, here CADENCE. The tool is validated on simple problems for which analytical solutions are known. Then, the interest of the approach is illustrated on the study of a biosensor.

URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8602461&isnumber=4358591