Search

The increasing amount of data processed on edge and the demand for reducing the energy consumption for large neural network architectures have initiated the transition from traditional von Neumann architectures towards in-memory computing paradigms. Quantization is one of the methods to reduce power and computation requirements for neural networks by limiting bit precision. Resistive Random Access Memory (RRAM) devices are great candidates for Quantized Neural Networks (QNN) implementations. As the number of possible conductive states in RRAMs is limited, a certain level of quantization is always considered when designing RRAM-based neural networks. In this work, we provide a comprehensive analysis of state-of-the-art RRAM-based QNN implementations, showing where RRAMs stand in terms of satisfying the criteria of efficient QNN hardware. We cover hardware and device challenges related to QNNs and show the main unsolved issues and possible future research directions.

Noise in image sensors led to the development of a whole range of denoising filters. A noisy image can become hard to recognize and often require several types of post-processing compensation circuits. This paper proposes an adaptive denoising system implemented using an analog in-memory neural computing network. The proposed method can learn new noises and can be integrated into or alone with CMOS image sensors. Three denoising network configurations are implemented namely, (1) single layer network, (2) convolution network, and (3) fusion network. The single layer network shows the processing time, energy consumption, and on-chip area of 3.2us, 21nJ per image, and 0.3mm^2 respectively, meanwhile, the convolution denoising network correspondingly shows 72ms, 236uJ, and 0.48mm^2. Among all the implemented networks, it is observed that performance metrics SSIM, MSE, and PSNR show a maximum improvement of 3.61, 21.7, and 7.7 times respectively.

In this work we theoretically propose scattering cancellation-based cloaks for heat waves that obey the Maxwell-Cattaneo equation. The proposed cloaks possess carefully tailored diffusivity to cancel the dipole scattering from the object that they surround, and thus can render a small object invisible in the near and far fields, as demonstrated by full-wave finite-element simulations. Mantle heat cloaking is further analyzed and proposed to simplify the design and bring this cloaking technology one step closer to its practical implementation, with promising applications in nanoelectronics and defense related applications., Comment: 19 pages, 4 figures

Surface plasmons polaritons were thought to exist only in metals near their plasma frequencies. The concept of spoof plasmons extended the realms of plasmonics to domains such as radio frequencies, magnetism, or even acoustic waves. Here, we introduce the concept of localized surface platonic modes (SPMs). We demonstrate that they can be generated on a two-dimensional clamped (or stress-free) cylindrical surface, in a thin elastic plate, with subwavelength corrugations under excitation by an incident flexural plane wave. Our results show that the corrugated rigid surface is elastically equivalent to a cylindrical scatterer with negatively uniform and dispersive flexural rigidity. This, indeed, suggests that plasmonic-like platonic materials can be engineered with potential applications in various areas including earthquake sensing, or elastic imaging and cloaking., Comment: 29 pages, 5 figures

In this letter, a reactance-less mono-stable oscillator is introduced for the first time using memristors. By replacing bulky inductors and capacitors with memristors, the novel mono-stable oscillator can be an area-efficient solution for on-chip fully integrated systems. The proposed circuit is described, mathematically analysed and verified by circuit simulations., Comment: This paper was submitted to Electronics Letters on the 28th of March 2012