اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدLevel-Shifted Neural Encoded Analog-to-Digital Converter
82
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
This paper presents the new approach in implementation of analog-to-digital converter (ADC) that is based on Hopfield neural-network architecture. Hopfield neural ADC (NADC) is a type of recurrent neural network that is effective in solving simple optimization problems, such as analog-to-digital conversion. The main idea behind the proposed design is to use multiple 2-bit Hopfield NADCs operating as quantizers in parallel, where analog input signal to each successive 2-bit Hopfield ADC block is passed through a voltage level shifter. This is followed by a neural network encoder to remove the quantization errors. In traditional Hopfield NADC based designs, increasing the number of bits could require proper scaling of the network parameters, in particular digital output operating region. Furthermore, the resolution improvement of traditional Hopfield NADC creates digital error that increases with the increasing number of bits. The proposed design is scalable in number of bits and number of quantization levels, and can maintain the magnitude of digital output code within a manageable operating voltage range.
قيم البحث
اقرأ أيضاً
The inherent stochasticity in many nano-scale devices makes them prospective candidates for low-power computations. Such devices have been demonstrated to exhibit probabilistic switching between two stable states to achieve stochastic behavior. Recen
tly, superparamagnetic nanomagnets (having low energy barrier EB $sim$ 1kT) have shown promise of achieving stochastic switching at GHz rates, with very low currents. On the other hand, voltage-controlled switching of nanomagnets through the Magneto-electric (ME) effect has shown further improvements in energy efficiency. In this simulation paper, we first analyze the stochastic switching characteristics of such super-paramagnetic nanomagnets in a voltage-controlled spintronic device. We study the influence of external bias on the switching behavior. Subsequently, we show that our proposed device leverages the voltage controlled stochasticity in performing low-voltage 8-bit analog to digital
High-speed high-resolution Analog-to-Digital Conversion is the key part for waveform digitization in physics experiments and many other domains. This paper presents a new fully digital correction of mismatch errors among the channels in Time Interlea
ved Analog-to-Digital Converter (TIADC) systems. We focus on correction with wide-band input signal, which means that we can correct the mismatch errors for any frequency point in a broad band with only one set of filter coefficients. Studies were also made to show how to apply the correction algorithm beyond the base band, i.e. other Nyquist zones in the under-sampling situation. Structure of the correction algorithm is presented in this paper, as well as simulation results. To evaluate the correction performance, we actually conducted a series of tests with two TIADC systems. The results indicate that the performance of both two TIADC systems can be greatly improved by correction, and the Effective Number Of Bits (ENOB) is successfully improved to be better than 9.5 bits and 5.5 bits for an input signal up to the bandwidth (-3dB) range in the 1.6-Gsps 14-bit and the 10-Gsps 8-bit TIADC systems, respectively. Tests were also conducted for input signal frequencies in the second Nyquist zone, which shows that the correction algorithms also work well as expected.
Ferroelectric field effect transistors (FeFETs) are being actively investigated with the potential for in-memory computing (IMC) over other non-volatile memories (NVMs). Content Addressable Memories (CAMs) are a form of IMC that performs parallel sea
rches for matched entries over a memory array for a given input query. CAMs are widely used for data-centric applications that involve pattern matching and search functionality. To accommodate the ever expanding data, it is attractive to resort to analog CAM for memory density improvement. However, the digital CAM design nowadays based on standard CMOS or emerging nonvolatile memories (e.g., resistive storage devices) is already challenging due to area, power, and cost penalties. Thus, it can be extremely expensive to achieve analog CAM with those technologies due to added cell components. As such, we propose, for the first time, a universal compact FeFET based CAM design, FeCAM, with search and storage functionality enabled in digital and analog domain simultaneously. By exploiting the multi-level-cell (MLC) states of FeFET, FeCAM can store and search inputs in either digital or analog domain. We perform a device-circuit co-design of the proposed FeCAM and validate its functionality and performance using an experimentally calibrated FeFET model. Circuit level simulation results demonstrate that FeCAM can either store continuous matching ranges or encode 3-bit data in a single CAM cell. When compared with the existing digital CMOS based CAM approaches, FeCAM is found to improve both memory density by 22.4X and energy saving by 8.6/3.2X for analog/digital modes, respectively. In the CAM-related application, our evaluations show that FeCAM can achieve 60.5X/23.1X saving in area/search energy compared with conventional CMOS based CAMs.
We study and analyze the fundamental aspects of noise propagation in recurrent as well as deep, multi-layer networks. The main focus of our study are neural networks in analogue hardware, yet the methodology provides insight for networks in general.
The system under study consists of noisy linear nodes, and we investigate the signal-to-noise ratio at the networks outputs which is the upper limit to such a systems computing accuracy. We consider additive and multiplicative noise which can be purely local as well as correlated across populations of neurons. This covers the chief internal-perturbations of hardware networks and noise amplitudes were obtained from a physically implemented recurrent neural network and therefore correspond to a real-world system. Analytic solutions agree exceptionally well with numerical data, enabling clear identification of the most critical components and aspects for noise management. Focusing on linear nodes isolates the impact of network connections and allows us to derive strategies for mitigating noise. Our work is the starting point in addressing this aspect of analogue neural networks, and our results identify notoriously sensitive points while simultaneously highlighting the robustness of such computational systems.
Time to Digital Converters (TDCs) are very common devices in particles physics experiments. A lot of off-the-shelf TDCs can be employed but the necessity of a custom DAta acQuisition (DAQ) system makes the TDCs implemented on the Field-Programmable G
ate Arrays (FPGAs) desirable. Most of the architectures developed so far are based on the tapped delay lines with precision down to 10 ps, obtained with high FPGA resources usage and non-linearity issues to be managed. Often such precision is not necessary; in this case TDC architectures with low resources occupancy are preferable allowing the implementation of data processing systems and of other utilities on the same device. In order to reconstruct gamma-gamma physics events tagged with High Energy Tagger (HET) in the KLOE-2 (K LOng Experiment 2), we need to measure the Time Of Flight (TOF) of the electrons and positrons from the KLOE-2 Interaction Point (IP) to our tagging stations (11 m apart). The required resolution must be better than the bunch spacing (2.7 ns). We have developed and implemented on a Xilinx Virtex-5 FPGA a 32 channel TDC with a precision of 255 ps and low non-linearity effects along with an embedded data acquisition systems and the interface to the online FARM of KLOE-2.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
