Memistor

About: Memistor is a research topic. Over the lifetime, 608 publications have been published within this topic receiving 34905 citations.

A survey on memristor modeling and security applications

Abstract: With the recent advances in memristors as a potential building block for future hardware, it becomes an important and timely topic to study the role that memristors may play in hardware security. To address this issue, this paper presents a survey on research activities on memristor modelling and potential application of memristors in hardware security. First, we give an overview of the current literature on memristor experimentation, characterization, and modeling which includes Chua's original theoretical prediction model, more detailed models based on recent memristor implementations, and the SPICE simulation models. Then, we report the current research efforts on memristor-based security in three major areas: (1) memristor hardware primitives (e.g., physical unclonable function) that are based on the memristor effective resistance model, (2) encryption schemes that leverage the chaotic behavior of the memristor circuit, and (3) security concerns in memristor-based memory systems. We observe that most of these works have limited scope and are based on simplified memristor models which diminish their practical value in security applications. Security applications have strict demands on repeatability, reliability, robustness, unforgeability, cost, resilience, and so on. To address these deficiencies, we propose a list of research areas that need to be addressed for building memristor-based security applications. We also analyze how memristors, as a new hardware building block, will impact major challenges in hardware security.

Memristor based adders

Abstract: Currently memristors are being researched to offer logic and memory functions. Recently, ultra dense resistive memory arrays built from various two terminal semiconductor or insulator thin film devices have been demonstrated. This paper presents memristor-based designs of commonly used (ripple carry, conditional sum and parallel prefix) adders. The latency and area of these adders are compared.

Transmission characteristics study of memristors based Op-Amp circuits

Abstract: According to a physical memristor fabricated by nano technology, a monotone-increasing and piecewise-linear nonlinear memristor model is applied to study the transmission characteristics of memristors based inverting and noninverting op-amp circuits. Simulation results demonstrated that inverting amplification and noninverting amplification can be obtained in the memristor based op-amp circuits.

Ziksa: On-chip learning accelerator with memristor crossbars for multilevel neural networks

Abstract: Memristor crossbars support efficient realizations of spiking and non-spiking neural networks designs. In most of these designs off-chip/ex-situ training is used to set/update the state of the memrisitve devices. However, there is a growing need to design an efficient on-chip/in-situ learning for mobile autonomous systems. In this research, we propose an on-chip learning accelerator, known as Ziksa, that is integrated with the memristor crossbars. We demonstrate how regression and back-propagation in multi-level networks can be realized through Ziksa. The proposed accelerator is evaluated on a fabricated TiN-TaOx-TaTiN memristor crossbar. A 3-layer feedforward network was tested using Ziksa for classification. An accuracy of 95.3% was achieved on Wisconsin breast cancer dataset. The proposed learning accelerator can be envisioned as a core building block in a wide-range of cognitive algorithms that rely on on-chip online learning.

Memristors' Potential for Multi-bit Storage and Pattern Learning

Abstract: Memristor is a two-terminal device, termed as fourth element, and characterized by a varying resistance depending on the charge (current) flown through it. This leads to many interesting characteristics, including a memory of its past states, demonstrated in its resistance. Smaller area and power consumed by memristors compared to conventional memories makes them a more suitable choice for applications needing large memory. In this paper we explore one of the unique properties of memristors which extends their suitability by allowing storage of multi-bit data in a single memristor. Their ability of storing multi-bit patterns will be shown via a simplified proof and simulations. This characteristic can be advantageous for many applications. In this paper particularly, we briefly discuss its advantages in pattern learning applications.