The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
In this work, crossbars with memristors down to 40x40 nm2 in size were fabricated by hybrid process combining optical and e-beam lithography. The developed process provides ~90% <<yield of forming-free>> nanodevices exhibiting memristive properties with endurance up to 105 cycles, making this approach suitable for neuromorphic applications.
The high density of on-chip nonvolatile memory provided by memristive elements is highly desirable for many applications. However, it raises concerns about finding the best programming strategies to limit the energy consumption of such systems. Here, we highlight the case of magnetic memory, where several unconventional programming strategies can reduce energy consumption, especially for applications...
In this present work, initial results of the growing of neuronal like cells on the memristive substrate are going to be presented. SH-SY5Y line cells where chosen for this test, thanks to their features similar to neurons and they were grown on Polyaniline (PANI) multilayer. PANI is a well known conductive polymer and it's also the active layer of a special class of device, the organic memristors...
In the past few years many significant innovations in the fabrication of solid-state nanopores have triggered a large number of new biophysics applications, due to the small dimension (5-20 nm) reached using a large scale of membrane materials. These significant improvements catalyzed a deeper understanding in the mechanism of charge diffusion inside nanopores (concerning the ionic flux). To correlate...
The paper presents a set of black-box modules for emulating higher–order elements (HOEs) from Chua´s periodical table, based on special mutators. For the emulation of arbitrary HOE, four kinds of modules with various directions of movement across the table are necessary: one for horizontal direction, one for vertical direction, and two for diagonal directions. Another four mutators are needed in order...
Perceptron is an artificial neural network that can solve simple tasks such as invariant pattern recognition, linear approximation, prediction and others. We report on the hardware realization of the perceptron with the use of polyaniline-based memristive devices as the analog link weights. An error correction algorithm was used to get the perceptron to learn to implement the NAND and NOR logic functions...
Layer-by-layer polyelectrolyte self assembling is a very promising techniques for the realization of organic films with nm resolution in thickness. We report here the application of the technique for the realization of active layers of organic memristors and for the realization of stochastic systems with learning properties. In particular, active channels of the memristive devices were fabricated...
The conduction characteristics of La1/3Ca2/3MnO3 (LCMO) films grown by pulsed laser deposition were investigated. The devices exhibit bipolar resistive switching effect with intermediate conduction states achievable by the application of positive and negative voltage ramps. The I-V curves are modeled using a nonlinear memristive approach based on the double-diode equation and the solution of the generalized...
The paper is dedicated to the hardware realization neuromorphic networks mimicking some properties of the nervous system. In particular, we will consider memristive devices-based logic gates with memory, elementary perceptron, and deterministic and stochastic networks, where memristive devices are used as electronic analogs of biological synapses.
It’s been quite a while since scientists are seeking for the ancestor of von Neumann computing architecture. Among the most promising candidates, memristor demonstrates advantageous characteristics, which open new pathways for the exploration of advanced computing paradigms. In this work we propose the design of a novel crossbar geometry, which is heterogeneous in terms of its cross-point devices,...
In this work the electrical performance of a Rad- Hard designed 1T-1R device based on the combination of an Enclosed Layout Transistor (ELT) and a TiN/HfO2/Ti/TiN based resistor is presented for the first time. Moreover, an architectural solution for 1Mbit radiation hard RRAM array implementation is proposed.
The paper presents a SPICE implementation of a behavioral model of the TiO2 memristor which is equivalent to the Pickett model, but it is easier to simulate with respect to convergence and matrix size. The proposed port equation approximates the Simmons model of tunneling barrier. The state equation has been modified in order to prevent exponential overflows during simulation. A full PSpice netlist...
The scaling roadmap of devices under a “more than Moore” scenario is resulting in the emergence of new types of devices. Among them, memristors seem to be promising candidates to be suitable for various areas of application such as in memories and neuromorphic computing chips. However, memristive devices still face some challenges to be resolved before becoming a mainstream. This work analyzes the...
The emergence of memristive devices is currently driving an increasing interest in neuromorphic computing, which could complement and provide enhanced functionalities to existing CMOS/Von Neumann processors. Various plasticity mechanisms, analogous to synaptic plasticity in the brain, have indeed been implemented in emerging memristive systems. Additionally, we have recently demonstrated experimentally...
In this work a simple network composed by a first 25 sensory neurons layer and a second 10 output neuron layer connected by 250 memristor synapses is proposed. The system was simulated in PSPICE in order to recognize 5X5 pixel binary images.
Resistive RAM (RRAM) are of great interest to the silicon microelectronics industry, offering the possibility of low programming energy per bit, rapid switching, and very high levels of integration. Moreover a great effort has been devoted to exploring the potential of RRAM in neuromorphic applications. Here we present the study of silicon-rich silica films to establish the switching properties, chemical...
This paper discusses implementations of gradientdescent based learning algorithms on memristive crossbar arrays. The Unregulated Step Descent (USD) is described as a practical algorithm for feed-forward on-line training of large crossbar arrays. It allows fast feed-forward fully parallel on-line hardware based learning, without requiring accurate models of the memristor behaviour and precise control...
We analyzed ReRAM-based memristors by using the charge-flux relations instead of the traditional current-voltage approach. We used simulated and experimental data to develop a circuit model. Simulations of devices with different conductive filament sizes were employed to fit a 3-parameter model, later on the relations between the model parameters were characterized in-depth. Finally, we used the model...
In this brief, the memristive behavior of a very simple device, that of a tungsten filament bulb, is reported. In specific, this kind of element operates as a non-ideal memristor, thus demonstrating a Type-II non-crossing, pinched hysteretic loop. A brief description of this experimental behavior from the perspective of memristive properties, is apposed.
The switching dynamics of filamentary Pt/HfO2/TiN memristive devices is managed through sub-threshold pulses in order to display gradual resistance decrease useful for analog logic computation based on spiking networks. Such memristive devices are known to display abrupt set transitions (resistance decrease) that require current limitation because of the triggering of a threshold switching event....
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.