Design of a Robust Memristive Spiking Neuromorphic System with Unsupervised Learning in Hardware

2021 ◽  
Vol 17 (4) ◽  
pp. 1-26
Author(s):  
Md Musabbir Adnan ◽  
Sagarvarma Sayyaparaju ◽  
Samuel D. Brown ◽  
Mst Shamim Ara Shawkat ◽  
Catherine D. Schuman ◽  
...  
Keyword(s):  
Unsupervised Learning ◽  
Recognition Task ◽  
Single Layer ◽  
Process Variations ◽  
Spike Timing ◽  
System Level ◽  
Design Parameters ◽  
Digital Pulse ◽  
Neuromorphic System ◽  
The Impact

Spiking neural networks (SNN) offer a power efficient, biologically plausible learning paradigm by encoding information into spikes. The discovery of the memristor has accelerated the progress of spiking neuromorphic systems, as the intrinsic plasticity of the device makes it an ideal candidate to mimic a biological synapse. Despite providing a nanoscale form factor, non-volatility, and low-power operation, memristors suffer from device-level non-idealities, which impact system-level performance. To address these issues, this article presents a memristive crossbar-based neuromorphic system using unsupervised learning with twin-memristor synapses, fully digital pulse width modulated spike-timing-dependent plasticity, and homeostasis neurons. The implemented single-layer SNN was applied to a pattern-recognition task of classifying handwritten-digits. The performance of the system was analyzed by varying design parameters such as number of training epochs, neurons, and capacitors. Furthermore, the impact of memristor device non-idealities, such as device-switching mismatch, aging, failure, and process variations, were investigated and the resilience of the proposed system was demonstrated.

scholarly journals Dynamic and Static Calibration of Ultra-Low-Voltage, Digital-Based Operational Transconductance Amplifiers

Electronics ◽  
2020 ◽  
Vol 9 (6) ◽  
pp. 983 ◽  
Author(s):  
Pedro Toledo ◽  
Paolo Crovetti ◽  
Hamilton Klimach ◽  
Sergio Bampi
Keyword(s):  
Pulse Width Modulation ◽  
Low Voltage ◽  
Process Variations ◽  
Pulse Modulation ◽  
Silicon Area ◽  
Operational Transconductance Amplifiers ◽  
Static Calibration ◽  
Digital Pulse ◽  
Transconductance Amplifiers ◽  
The Impact

The calibration of the effects of process variations and device mismatch in Ultra Low Voltage (ULV) Digital-Based Operational Transconductance Amplifiers (DB-OTAs) is addressed in this paper. For this purpose, two dynamic calibration techniques, intended to dynamically vary the effective strength of critical gates by different modulation strategies, i.e., Digital Pulse Width Modulation (DPWM) and Dyadic Digital Pulse Modulation (DDPM), are explored and compared to classic static calibration. The effectiveness of the calibration approaches as a mean to recover acceptable performance in non-functional samples is verified by Monte-Carlo (MC) post-layout simulations performed on a 300 mV power supply, nW-power DB-OTA in 180 nm CMOS. Based on the same MC post-layout simulations, the impact of each calibration strategy on silicon area, power consumption, and OTA performance is discussed.

scholarly journals Determining the Lyapunov Spectrum of Continuous-Time 1D and 2D Multiscroll Chaotic Oscillators via the Solution ofm-PWL Variational Equations

2013 ◽  
Vol 2013 ◽  
pp. 1-11 ◽  
Author(s):  
Jesus Manuel Munoz-Pacheco ◽  
Luz del Carmen Gómez-Pavón ◽  
Olga Guadalupe Félix-Beltrán ◽  
Arnulfo Luis-Ramos
Keyword(s):  
Lyapunov Exponents ◽  
Dynamic Range ◽  
Piecewise Linear ◽  
Lyapunov Spectrum ◽  
System Level ◽  
Design Parameters ◽  
Chaotic Oscillator ◽  
Chaotic Oscillators ◽  
Variational Equations ◽  
The Impact

An algorithm to compute the Lyapunov exponents of piecewise linear function-based multidirectional multiscroll chaotic oscillators is reported. Based on themregions in the piecewise linear functions, the suggested algorithm determines the individual expansion rate of Lyapunov exponents fromm-piecewise linear variational equations and their associatedm-Jacobian matrices whose entries remain constant during all computation cycles. Additionally, by considering OpAmp-based chaotic oscillators, we study the impact of two analog design procedures on the magnitude of Lyapunov exponents. We focus on analyzing variations of both frequency bandwidth and voltage/current dynamic range of the chaotic signals at electronic system level. As a function of the design parameters, a renormalization factor is proposed to estimate correctly the Lyapunov spectrum. Numerical simulation results in a double-scroll type chaotic oscillator and complex chaotic oscillators generating multidirectional multiscroll chaotic attractors on phase space confirm the usefulness of the reported algorithm.

scholarly journals Non-linear Memristive Synaptic Dynamics for Efficient Unsupervised Learning in Spiking Neural Networks

2021 ◽  
Vol 15 ◽  
Author(s):  
Stefano Brivio ◽  
Denys R. B. Ly ◽  
Elisa Vianello ◽  
Sabina Spiga
Keyword(s):  
Neural Networks ◽  
Classification Performance ◽  
Spike Timing ◽  
Spiking Neural Networks ◽  
System Level ◽  
Training Time ◽  
Background Theory ◽  
Non Linear ◽  
Synaptic Dynamics ◽  
The Impact

Spiking neural networks (SNNs) are a computational tool in which the information is coded into spikes, as in some parts of the brain, differently from conventional neural networks (NNs) that compute over real-numbers. Therefore, SNNs can implement intelligent information extraction in real-time at the edge of data acquisition and correspond to a complementary solution to conventional NNs working for cloud-computing. Both NN classes face hardware constraints due to limited computing parallelism and separation of logic and memory. Emerging memory devices, like resistive switching memories, phase change memories, or memristive devices in general are strong candidates to remove these hurdles for NN applications. The well-established training procedures of conventional NNs helped in defining the desiderata for memristive device dynamics implementing synaptic units. The generally agreed requirements are a linear evolution of memristive conductance upon stimulation with train of identical pulses and a symmetric conductance change for conductance increase and decrease. Conversely, little work has been done to understand the main properties of memristive devices supporting efficient SNN operation. The reason lies in the lack of a background theory for their training. As a consequence, requirements for NNs have been taken as a reference to develop memristive devices for SNNs. In the present work, we show that, for efficient CMOS/memristive SNNs, the requirements for synaptic memristive dynamics are very different from the needs of a conventional NN. System-level simulations of a SNN trained to classify hand-written digit images through a spike timing dependent plasticity protocol are performed considering various linear and non-linear plausible synaptic memristive dynamics. We consider memristive dynamics bounded by artificial hard conductance values and limited by the natural dynamics evolution toward asymptotic values (soft-boundaries). We quantitatively analyze the impact of resolution and non-linearity properties of the synapses on the network training and classification performance. Finally, we demonstrate that the non-linear synapses with hard boundary values enable higher classification performance and realize the best trade-off between classification accuracy and required training time. With reference to the obtained results, we discuss how memristive devices with non-linear dynamics constitute a technologically convenient solution for the development of on-line SNN training.

scholarly journals Simulation of the Spiking Neural Network based on Practical Memristor

2018 ◽  
Vol 173 ◽  
pp. 01025
Author(s):  
Xi Zhu ◽  
Yi Sun ◽  
Haijun Liu ◽  
Qingjiang Li ◽  
Hui Xu
Keyword(s):  
Neural Network ◽  
Recognition Task ◽  
Single Layer ◽  
Spiking Neural Network ◽  
Biological Structure ◽  
Biologically Inspired ◽  
The Impact ◽  
The Brain ◽  
Pattern Recognition Task ◽  
Significant Attention

In order to gain a better understanding of the brain and explore biologically-inspired computation, significant attention is being paid to research into the spike-based neural computation. Spiking neural network (SNN), which is inspired by the understanding of observed biological structure, has been increasingly applied to pattern recognition task. In this work, a single layer SNN architecture based on the characteristics of spiking timing dependent plasticity (STDP) in accordance with the actual test of the device data has been proposed. The device data is derived from the Ag/GeSe/TiN fabricated memristor. The network has been tested on the MNIST dataset, and the classification accuracy attains 90.2%. Furthermore, the impact of device instability on the SNN performance has been discussed, which can propose guidelines for fabricating memristors used for SNN architecture based on STDP characteristics.

Information Driven Product Development: A Framework for Product Design and Analysis Integration

2007 ◽  
Author(s):  
Angran Xiao
Keyword(s):  
Product Development ◽  
Product Information ◽  
Information Model ◽  
Product Lifecycle Management ◽  
System Level ◽  
Design Parameters ◽  
Analysis Model ◽  
Element Analysis ◽  
Product Information Model ◽  
The Impact

New paradigms and accompanying software systems are necessary to support the integration of system level design and discipline level analysis activities for the implementation of product lifecycle management. In this paper, we present an information driven product development method for the integration in the context of multidisciplinary product realization. The method contains three constituents: product information model which represents the associativities among design requirements, product components, and design parameters; compromise Decision Support Problem which maps the information model directly into design problems; and knowledge based Finite Element Analysis which generates analysis model automatically from the information model. Information driven product development uses product information model as a communication media between design and analysis activities, hence provides an effective way to trace the impact of design changes, facilitates the reuse of analyses models, and supports collaborative decision-making. An electronic chip package design and analysis scenario is presented to illustrate and demonstrate this method.

Comparative assessment of morphological mechanisms of maintaining structural liver homeostasis in the dynamics of exposure to coal-rock dust and sodium fluoride on the body

2020 ◽  
pp. 189-194
Author(s):  
M. S. Bugaeva ◽  
O. I. Bondarev ◽  
N. N. Mikhailova ◽  
L. G. Gorokhova
Keyword(s):  
Sodium Fluoride ◽  
Morphological Changes ◽  
The Body ◽  
System Level ◽  
Production Factor ◽  
Coal Rock ◽  
The Impact ◽  
Structural Homeostasis ◽  
Rock Dust

Introduction. The impact on the body of such factors of the production environment as coal-rock dust and fluorine compounds leads to certain shift s in strict indicators of homeostasis at the system level. Maintaining the relative constancy of the internal environment of the body is provided by the functional consistency of all organs and systems, the leading of which is the liver. Organ repair plays a crucial role in restoring the structure of genetic material and maintaining normal cell viability. When this mechanism is damaged, the compensatory capabilities of the organ are disrupted, homeostasis is disrupted at the cellular and organizational levels, and the development of the main pathological processes is noted.The aim of the study is to compare the morphological mechanisms of maintaining structural homeostasis of the liver in the dynamics of the impact on the body of coal-rock dust and sodium fluoride.Materials and methods. Experimental studies were conducted on adult white male laboratory rats. Features of morphological mechanisms for maintaining structural homeostasis of the liver in the dynamics of exposure to coal-rock dust and sodium fluoride were studied on experimental models of pneumoconiosis and fluoride intoxication. For histological examination in experimental animals, liver sampling was performed after 1, 3, 6, 9, 12 weeks of the experiment.Results. The specificity of morphological changes in the liver depending on the harmful production factor was revealed. It is shown that chronic exposure to coal-rock dust and sodium fluoride is characterized by the development of similar morphological changes in the liver and its vessels from the predominance of the initial compensatory-adaptive to pronounced violations of the stromal and parenchymal components. Long-term inhalation of coal-rock dust at 1–3 weeks of seeding triggers adaptive mechanisms in the liver in the form of increased functional activity of cells, formation of double-core hepatocytes, activation of immunocompetent cells and endotheliocytes, ensuring the preservation of the parenchyma and the general morphostructure of the organ until the 12th week of the experiment. Exposure to sodium fluoride leads to early disruption of liver compensatory mechanisms and the development of dystrophic changes in the parenchyma with the formation of necrosis foci as early as the 6th week of the experiment.Conclusions. The study of mechanisms for compensating the liver structure in conditions of long-term exposure to coal-rock dust and sodium fluoride, as well as processes that indicate their failure, and the timing of their occurrence, is of theoretical and practical importance for developing recommendations for the timely prevention and correction of pathological conditions developing in employees of the aluminum and coal industry.The authors declare no conflict of interests.

Making Up 3D Bodies

2021 ◽  
Vol 4 (2) ◽  
pp. 1-9
Author(s):  
Kiona Hagen Niehaus ◽  
Rebecca Fiebrink
Keyword(s):  
Software Tool ◽  
The Body ◽  
Design Parameters ◽  
Artistic Practice ◽  
Human Form ◽  
Modeling Systems ◽  
Exploratory Approach ◽  
Disability Status ◽  
High Level ◽  
The Impact

This paper describes the process of developing a software tool for digital artistic exploration of 3D human figures. Previously available software for modeling mesh-based 3D human figures restricts user output based on normative assumptions about the form that a body might take, particularly in terms of gender, race, and disability status, which are reinforced by ubiquitous use of range-limited sliders mapped to singular high-level design parameters. CreatorCustom, the software prototype created during this research, is designed to foreground an exploratory approach to modeling 3D human bodies, treating the digital body as a sculptural landscape rather than a presupposed form for rote technical representation. Building on prior research into serendipity in Human-Computer Interaction and 3D modeling systems for users at various levels of proficiency, among other areas, this research comprises two qualitative studies and investigation of the impact on the first author's artistic practice. Study 1 uses interviews and practice sessions to explore the practices of six queer artists working with the body and the language, materials, and actions they use in their practice; these then informed the design of the software tool. Study 2 investigates the usability, creativity support, and bodily implications of the software when used by thirteen artists in a workshop. These studies reveal the importance of exploration and unexpectedness in artistic practice, and a desire for experimental digital approaches to the human form.

scholarly journals Simulation Tests of a Cow Milking Machine—Analysis of Design Parameters

Processes ◽  
2021 ◽  
Vol 9 (8) ◽  
pp. 1358
Author(s):  
Ewa Golisz ◽  
Adam Kupczyk ◽  
Maria Majkowska ◽  
Jędrzej Trajer
Keyword(s):  
Mathematical Model ◽  
Simplified Model ◽  
Design Parameters ◽  
Degree Polynomial ◽  
Fourth Degree ◽  
Local Loss ◽  
Linear Drag ◽  
Milking Machine ◽  
Machine Analysis ◽  
The Impact

The objective of this paper was to create a mathematical model of vacuum drops in a form that enables the testing of the impact of design parameters of a milking cluster on the values of vacuum drops in the claw. Simulation tests of the milking cluster were conducted, with the use of a simplified model of vacuum drops in the form of a fourth-degree polynomial. Sensitivity analysis and a simulation of a model with a simplified structure of vacuum drops in the claw were carried out. As a result, the impact of the milking machine’s design parameters on the milking process could be analysed. The results showed that a change in the local loss and linear drag coefficient in the long milk duct will have a lower impact on vacuum drops if a smaller flux of inlet air, a higher head of the air/liquid mix, and a higher diameter of the long milk tube are used.

scholarly journals Impact of COVID-19 pandemic on the oncologic care continuum: urgent need to restore patients care to pre-COVID-19 era

2021 ◽  
pp. 1-11
Author(s):  
Ernest Osei ◽  
Ruth Francis ◽  
Ayan Mohamed ◽  
Lyba Sheraz ◽  
Fariba Soltani-Mayvan
Keyword(s):  
Cancer Care ◽  
Healthcare Professionals ◽  
Negative Impact ◽  
System Level ◽  
Care Continuum ◽  
Cancer Services ◽  
Number Of Patients ◽  
Prevention Programmes ◽  
Screening And Prevention ◽  
The Impact

Abstract Background: Globally, cancer is the second leading cause of death, and it is estimated that over 18·1 million new cases are diagnosed annually. The COVID-19 pandemic has significantly impacted almost every aspect of the provision and management of cancer care worldwide. The time-critical nature of COVID-19 diagnosis and the large number of patients requiring hospitalisation necessitated the rerouting of already limited resources available for cancer services and programmes to the care of COVID-19 patients. Furthermore, the stringent social distancing, restricted in-hospital visits and lockdown measures instituted by various governments resulted in the disruption of the oncologic continuum including screening, diagnostic and prevention programmes, treatments and follow-up services as well as research and clinical trial programmes. Materials and Methods: We searched several databases from October 2020 to January 2021 for relevant studies published in English between 2020 and 2021 and reporting on the impact of COVID-19 on the cancer care continuum. This narrative review paper describes the impact of the COVID-19 pandemic on the cancer patient care continuum from screening and prevention to treatments and ongoing management of patients. Conclusions: The COVID-19 pandemic has profoundly impacted cancer care and the management of cancer services and patients. Nevertheless, the oncology healthcare communities worldwide have done phenomenal work with joint and collaborative efforts, utilising best available evidence-based guidelines to continue to give safe and effective treatments for cancer patients while maintaining the safety of patients, healthcare professionals and the general population. Nevertheless, several healthcare centres are now faced with significant challenges with the management of the backlog of screening, diagnosis and treatment cases. It is imperative that governments, leaders of healthcare centres and healthcare professionals take all necessary actions and policies focused on minimising further system-level delays to cancer screening, diagnosis, treatment initiation and clearing of all backlogs cases from the COVID-19 pandemic in order to mitigate the negative impact on cancer outcomes.

scholarly journals System Level Simulation of Microgrid Power Electronic Systems

Electronics ◽  
2021 ◽  
Vol 10 (6) ◽  
pp. 644
Author(s):  
Michal Frivaldsky ◽  
Jan Morgos ◽  
Michal Prazenica ◽  
Kristian Takacs
Keyword(s):  
Smart Grid ◽  
Simulation Model ◽  
Real Time ◽  
Power Converter ◽  
System Level ◽  
Time Behavior ◽  
Operational Parameters ◽  
System Level Simulation ◽  
Time Operation ◽  
The Impact

In this paper, we describe a procedure for designing an accurate simulation model using a price-wised linear approach referred to as the power semiconductor converters of a DC microgrid concept. Initially, the selection of topologies of individual power stage blocs are identified. Due to the requirements for verifying the accuracy of the simulation model, physical samples of power converters are realized with a power ratio of 1:10. The focus was on optimization of operational parameters such as real-time behavior (variable waveforms within a time domain), efficiency, and the voltage/current ripples. The approach was compared to real-time operation and efficiency performance was evaluated showing the accuracy and suitability of the presented approach. The results show the potential for developing complex smart grid simulation models, with a high level of accuracy, and thus the possibility to investigate various operational scenarios and the impact of power converter characteristics on the performance of a smart gird. Two possible operational scenarios of the proposed smart grid concept are evaluated and demonstrate that an accurate hardware-in-the-loop (HIL) system can be designed.

Sign in / Sign up

Export Citation Format

Share Document