In linear IC's fabricated in a low-voltage CMOS technology, the reduction of the dynamic range due to the dc offset and low frequency noise of the amplifiers becomes increasingly significant. Also, the achievable amplifier gain is often quite low in such a technology, since cascoding may not be a practical circuit option due to the resulting reduction of the output signal swing. In this paper, some old and some new circuit techniques are described for the compensation of the amplifier's most important nonideal effects including the noise (mainly thermal and 1/f noise), the input-referred dc offset voltage as well as the finite gain resulting in a nonideal virtual ground at the input.

Circuit techniques for reducing the effects of op-amp imperfections: autozeroing, correlated double sampling, and chopper stabilization

A comprehensive survey of thinning methodologies is presented. A wide range of thinning algorithms, including iterative deletion of pixels and nonpixel-based methods, is covered. Skeletonization algorithms based on medial axis and other distance transforms are not considered. An overview of the iterative thinning process and the pixel-deletion criteria needed to preserve the connectivity of the image pattern is given first. Thinning algorithms are then considered in terms of these criteria and their modes of operation. Nonpixel-based methods that usually produce a center line of the pattern directly in one pass without examining all the individual pixels are discussed. The algorithms are considered in great detail and scope, and the relationships among them are explored. >

Thinning methodologies-a comprehensive survey

A new and accurate approach to charge-pumping measurements for the determination of the Si-SiO 2 interface state density directly on MOS transistors is presented. By a careful analysis of the different processes of emission of electrons towards the conduction band and of holes towards the valence band, depending on the charge state of the interface, all the previously ill-understood phenomena can be explained and the deviations from the simple charge-pumping theory can be accounted for. The presence of a geometric component in some transistor configurations is illustrated and the influence of trapping time constants is discussed. Furthermore, based on this insight, a new technique is developed for the determination of the energy distribution of interface states in small-area transistors, without requiring the knowledge of the surface potential dependence on gate voltage.

/pdf/a-reliable-approach-to-charge-pumping-measurements-in-mos-28u56t5opk.pdf

A reliable approach to charge-pumping measurements in MOS transistors

A new design methodology based on a unified treatment of all the regions of operation of the MOS transistor is proposed. It is intended for the design of CMOS analog circuits and especially suited for low power circuits where the moderate inversion region often is used because it provides a good compromise between speed and power consumption. The synthesis procedure is based on the relation between the ratio of the transconductance over DC drain current g/sub m//I/sub D/ and the normalized current I/sub D//(W/L). The g/sub m//I/sub D/ indeed is a universal characteristic of all the transistors belonging to a same process. It may be derived from experimental measurements and fitted with simple analytical models. The method was applied successfully to the design of a silicon-on-insulator (SOI) micropower operational transconductance amplifier (OTA).

A g/sub m//I/sub D/ based methodology for the design of CMOS analog circuits and its application to the synthesis of a silicon-on-insulator micropower OTA

Neuromorphic computing has come to refer to a variety of brain-inspired computers, devices, and models that contrast the pervasive von Neumann computer architecture This biologically inspired approach has created highly connected synthetic neurons and synapses that can be used to model neuroscience theories as well as solve challenging machine learning problems The promise of the technology is to create a brain-like ability to learn and adapt, but the technical challenges are significant, starting with an accurate neuroscience model of how the brain works, to finding materials and engineering breakthroughs to build devices to support these models, to creating a programming framework so the systems can learn, to creating applications with brain-like capabilities In this work, we provide a comprehensive survey of the research and motivations for neuromorphic computing over its history We begin with a 35-year review of the motivations and drivers of neuromorphic computing, then look at the major research areas of the field, which we define as neuro-inspired models, algorithms and learning approaches, hardware and devices, supporting systems, and finally applications We conclude with a broad discussion on the major research topics that need to be addressed in the coming years to see the promise of neuromorphic computing fulfilled The goals of this work are to provide an exhaustive review of the research conducted in neuromorphic computing since the inception of the term, and to motivate further work by illuminating gaps in the field where new research is needed

A Survey of Neuromorphic Computing and Neural Networks in Hardware.

A modified RSD algorithm has been implemented in a switched-current pipelined A/D converter. The offset insensitivity of the RSD Converter reduces the effect of several nonidealities proper to current copier cells. Moreover, the benefits resulting from the large tolerances inherent to the RSD algorithm and the pipelined architecture result in an improved conversion rate. Measurements on a first prototype give an integral nonlinearity error less than 0.8 LSB for 10-bit accuracy. Power dissipation is 20 mW and silicon area is 2.5 mm/sup 2/. The measured sampling rate is 550 kS/s. It is an improvement by a factor of twenty compared to known equivalent CMOS switched-current converters. It is nevertheless still well below the predicted conversion rate of 4.5 MHz, which should be obtained once this A/D converter is integrated into an analog front-end. Full compatibility with standard digital technologies makes this kind of converter attractive for low power, medium-fast converters with 10-bit accuracy. >

A 10-bit pipelined switched-current A/D converter

An implementation of a VLSI fully interconnected neural network with only two binary memory points per synapse is described. The small area of single synaptic cells allows implementation of neural networks with hundreds of neurons. Classical learning algorithms like the Hebb's rule show a poor storage capacity, especially in VLSI neural networks where the range of the synapse weights is limited by the number of memory points contained in each connection; an algorithm for programming a Hopfield neural network as a high-storage content-addressable memory is proposed. The storage capacity obtained with this algorithm is very promising for pattern-recognition applications. >

/pdf/neural-networks-for-high-storage-content-addressable-memory-3dib5i1hwe.pdf

Neural networks for high-storage content-addressable memory: VLSI circuit and learning algorithm

The basic requirements for electronic implementations of the fully connected Hopfield network are examined, highlighting the reasons why the authors regard analog implementations as more appropriate. Analog VLSI networks are then discussed, with particular reference to the selection of memory points and the design of the synapse, and experimental results are given. A test chip containing 14 neurons and 196 synapses is described. >

An analog VLSI implementation of Hopfield's neural network

The performances of micropower single-stage CMOS OTAs implemented in SOI or bulk technologies are compared. The improvements resulting from the superior device characteristics of fully-depleted SOI MOSFETs are discussed. Experimental verifications support the theoretical predictions.

Comparison of SOI versus bulk performances of CMOS micropower single-stage OTAs

Nowadays, real-time applications of Fuzzy Logic in different domains are being increasingly reported. ASIC-based analog hardware becomes an interesting solution for these kinds of applications because it benefits from: savings on silicon surface and power consumption, readily accomplishment with strict timing constraints and cost-effective production. This book focuses in-depth on the VLSI CMOS implementation and application of programmable analog Fuzzy Logic Controllers following a mixed-signal philosophy. This is to say, signals are processed in the analog domain whereas programmability is achieved by means of standard digital memories. This approach highlights the following crucial aspects: *The comprehensive study and analysis of the main analog fuzzy operators: Fuzzy Membership Functions, T-Norm, T-CoNorm and Defuzzifier circuits. *The study and development of mixed-signal Fuzzy Controllers architectures targeting the requirements for different applications. *The fabrication and test of full-ended demonstrators. *The partial fabrication and test of a prototype corresponding to a real-time Fuzzy Logic application in the field of Signal Processing.

Paul Jespers

Papers

A 10-bit pipelined switched-current A/D converter

Neural networks for high-storage content-addressable memory: VLSI circuit and learning algorithm

An analog VLSI implementation of Hopfield's neural network

Comparison of SOI versus bulk performances of CMOS micropower single-stage OTAs

Design of Analog Fuzzy Logic Controllers in CMOS Technologies: Implementation, Test and Application