Showing papers in "IEEE Solid-State Circuits Magazine in 2015"
TL;DR: The StrongARM latch topology finds wide usage as a sense amplifier, a comparator, or simply a robust latch with high sensitivity as mentioned in this paper, and it consumes zero static power, produces rail-to-rail outputs, and its input-referred offset arises from primarily one differential pair.
Abstract: i»?The StrongARM latch topology finds wide usage as a sense amplifier, a comparator, or simply a robust latch with high sensitivity. The term âStrongARMâ commemorates the use of this circuit in Digital Equipment Corporationâs StrongARM microprocessor [1], but the basic structure was originally introduced by Toshibaâs Kobayashi et al. [2]. The StrongARM latch has become popular for three reasons: 1) it consumes zero static power, 2) it directly produces rail-to-rail outputs, and 3) its input-referred offset arises from primarily one differential pair. In this column, we study the circuit and its properties.
291 citations
TL;DR: The purpose of this article is to enumerate some of the pertaining progress slopes, explain their possible foundations, and speculate about their remaining lifetime.
Abstract: At the turn of this century, there was widespread concern that the performance of analog-to-digital converters (ADCs) might have reached a saturation point and would, in fact, deteriorate as we began to scale into deep submicron CMOS technology. The past 15 years of innovation have clearly refuted such fears. Driven by a combination of application pull, architectural modifications, and relentless optimization, we have seen steady improvements in several key performance metrics.
120 citations
TL;DR: An overview of the emerging field of analog-to-information conversion in light of various sub-Nyquist sampling techniques recently appearing in literature is given, to highlight some of the opportunities, challenges, and new applications such converters offer.
Abstract: Energy efficiency, long battery life, and low latency are key attributes of many emerging ultralow-power sensing and monitoring systems. Applications such as always-on reactive sensors for natural human?device interfaces, as well as multiple consumer and industrial applications for the Internet of Things (IoT), require ultralow-power designs beyond the promise of state-of-the art data converters.
65 citations
TL;DR: In this article, the authors study the role of field effect transistors (FETs) in nanometer switch topology and their role in the design of nanometer switches.
Abstract: Field-effect transistors (FETs) have been used as switches, particularly for analog signals, since the 1950s. In the early days of analog sampling, it was discovered that such devices exhibit an input-dependent on-resistance, thereby introducing distortion. This issue can be resolved by ?bootstrapping,? a circuit technique that minimizes the switch on-resistance variation in the presence of large input and output voltage swings. In this article, we study the bootstrapped switch topology and appreciate its role in nanometer designs.
55 citations
TL;DR: To minimize operating power and achieve maximum battery lifetime, the implementation of an ultra-lowpower wireless system requires an integrated design approach that considers many requirements including battery source, active and sleep mode energy requirements, system architectures, and circuit implementations.
Abstract: An ultra-low-power wireless system is one which has to operate for an extended period of time with only a limited power source available, and is typically constrained to a limited size (if size is not a limitation then a larger battery could be used). Clearly the term ?ultra-low-power wireless? covers a broad range of applications which may have different key drivers as illustrated in Figure 1. For a fitness device such as a heart rate monitor, the number one driver is often cost as these are basic consumer products. For a bio-implant such as a smart pacemaker, battery life is critical as battery replacement typically requires surgery. For a smart home system such as automatic climate control, cost and battery life are important but the system also needs to support a relatively large number of devices and the communication range should be large?throughout the whole building. Finally for a gaming application such as a wireless headset, high data rate and very low latency are key, so as not to ruin the high-speed gaming experience.
42 citations
TL;DR: This article provides a preview of topics presented at the International Solid-State Circuits Society conference in 2015, the conference theme is "silicon systems: small chips for big data".
Abstract: The International Solid-State Circuits Conference (ISSCC) is the flagship conference of the IEEE Solid-State Circuits Society. This year, for ISSCC, the theme is ?Silicon Systems: Small Chips for Big Data.? Big data is enveloping us: it is being generated by the Internet-of-Things (IoT), health care, and the Web, changing our society and our individual lives. Small silicon chips enable these changes through data sensing, gathering, processing, storing, and networking through wireless and wireline connectivity. Recent silicon-?system technologies, including ultra-low-power systems, high-performance circuits and systems, wireless power and data transmission, and three-dimensional (3-D) IC structures, will open the door to big data applications. Moreover, big data applications such as health care, machine learning, and sensor systems will challenge designers to consider new system architectures requiring advances in circuits and technology. ISSCC 2015 showcases novel circuit and system solutions that open new vistas for society, providing opportunities for new lifestyle transformations, all driven by big data technology.
38 citations
TL;DR: Energy-efficient near-threshold design has been proposed to increase energy efficiency across a wide range of applications as mentioned in this paper, and state-of-the-art near threshold techniques have been summarized that help overcome barriers to near threshold adoption, namely high variation at low voltage.
Abstract: i»?Energy-efficient near-threshold design has been proposed to increase energy efficiency across a wide range of applications. This article first provides a background motivating near-threshold and how it differs from super-threshold and subthreshold operation. Next, state-of-the-art near-threshold techniques are summarized that help overcome barriers to near-threshold adoption, namely high variation at low voltage. Last, example industrial and academic wide-voltage scaling systems are discussed.
38 citations
IBM1
TL;DR: This article describes the most common types of equalizers used in electrical links today, and the key to achieving dramatically higher data rates is to employ channel equalization.
Abstract: Another article in this issue, the introductory article by Tony Chan Carusone, discussed how the limited bandwidth of an electrical channel causes distortion (e.g., broadening) of the pulses traveling from transmitter to receiver. Without compensating for such distortion, the maximum data rate of a typical electrical link would be limited to only a few gigabits per second to avoid excessive intersymbol interference (ISI). The key to achieving dramatically higher data rates (up to 56 Gb/s in the latest proposed standards) is to employ channel equalization. This article describes the most common types of equalizers used in electrical links today.
37 citations
TL;DR: In this article, the cross-coupled pair (XCP) can serve as a negative resistance or a negative impedance converter in small-signal operation, or a regenerative circuit in large-Signal operation.
Abstract: In this article, we study applications of the cross-coupled pair (XCP) in analog and RF circuits The XCP can serve as a negative resistance or a negative impedance converter in small-signal operation, or a regenerative circuit in large-signal operation
36 citations
TL;DR: The bridged T-coil is a circuit topology that extends the bandwidth by a greater factor than does inductive peaking as discussed by the authors, and is used in many high-speed amplifiers, line drivers and input/output (I/O) interfaces in today?s wireline systems.
Abstract: The bridged T-coil, often simply called the T-coil, is a circuit topology that extends the bandwidth by a greater factor than does inductive peaking. Many high-speed amplifiers, line drivers, and input/output (I/O) interfaces in today?s wireline systems incorporate on-chip T-coils to deal with parasitic capacitances. In this article, we introduce and analyze the basic structure and study its applications.
30 citations
TL;DR: Among numerous Nyquist-rate analog-to-digital converter (ADC) architectures introduced throughout the years, three have found the widest usage: flash, pipelined, and successive-approximation register (SAR) topologies.
Abstract: Among numerous Nyquist-rate analog-to-digital converter (ADC) architectures introduced throughout the years, three have found the widest usage: flash, pipelined, and successive-approximation register (SAR) topologies. In this article, we focus on the last two and study their similarities and differences.
Abstract: Power consumption is a result of high speed and low noise requirements, and it can be minimized, provided proper operating points are selected for the input transistors. This is achieved by adopting a design procedure in which BSIM6/EKV model parameters are used to derive the gain and speed characteristics in asymptotic form. All three regions of operation i.e., strong and weak inversion and velocity saturation, are included. This design procedure is developed for channel lengths down to 5 nm. It is shown that inversion coefficients must be used around unity or (L/20 nm)2, depending on the actual channel length.
TL;DR: Digital I/O performance is now a headline specification in practically any high-performance electronic system, from consumer products to enterprise servers, as well as in chip-to-chip links.
Abstract: Digital I/O performance is now a headline specification in practically any high-performance electronic system, from consumer products to enterprise servers. Any future progress in integrated circuit computational capability must naturally be matched with progress on digital I/O. As a result, recent research has led to tremendous progress in the data rates communicated over individual chip-to-chip links.
IBM1
TL;DR: In this paper, the scaling principles were used to reduce the layout dimensions from 5?m to 1?m by using a single transistor and a capacitor and showed how well they worked by building some MOS transistors scaled to small dimensions.
Abstract: I think I was very fortunate to get a chance to work in microelectronics in its very early days. Those were very exciting times, particularly the day I found that I could build DRAM with just a single transistor and a capacitor. I wrote an internal paper and obtained a patent, but the first chance I had to work much on my idea came about four years later when the small team I worked with was challenged to make DRAM 25 times denser by reducing the layout dimensions from 5 ?m to 1 ?m. That is when we developed the scaling principles and showed how well they worked by building some experimental MOS transistors scaled to small dimensions. The progress in DRAM and all microelectronic devices and circuits has been amazingly successful since those first scaling principles were introduced. Many challenges have been met to achieve this, but today even more challenges have to be faced if progress is to continue.
TL;DR: There is a perception that classical semiconductor scaling is undergoing saturation and has recently not been delivering on the historical expectations for cost, power, and performance scaling, which has serious implications for the industry's ability to sustain the proliferation of increasingly complex systems at affordable prices and the consequent societal impacts.
Abstract: There is a perception that classical semiconductor scaling is undergoing saturation and has recently not been delivering on the historical expectations for cost, power, and performance scaling. Both the cost of technology node development and the cost to manufacture the newer nodes have been escalating, as has the nonrecurring engineering charge for new designs in these newer nodes. This has serious implications for our industry?s ability to sustain the proliferation of increasingly complex systems at affordable prices and the consequent societal impacts.
TL;DR: The need for better communication has been one of the driving forces behind technology advancements, including those in the field of data converters as discussed by the authors, and various methods were developed to allow communication over longer distances, such as drum beating and smoke signaling.
Abstract: The need for better communication has been one of the driving forces behind technology advancements, including those in the field of data converters. To extend communication beyond the limits of speech, various methods were developed to allow communication over longer distances, such as drum beating and smoke signaling. The fundamentals of coding and decoding are apparent in these early techniques.
TL;DR: Thanks to the bandwidth demand explosion in data centers and telecommunication infrastructures, the data-rate growth for wireline transceivers has been consistently increasing at about 2×/4 yr, as shown in Figure 1.
Abstract: Thanks to the bandwidth demand explosion in data centers and telecommunication infrastructures, the data-rate growth for wireline transceivers has been consistently increasing at about 2x/4 yr, as shown in Figure 1. The most-used standards are listed. Peripheral component interconnect express is used ubiquitously in almost all chip-to-chip interfaces. Optical Internetworking Forum (OIF)/Common Electrical Interface (CEI) and Ethernet usually come hand-in-hand for backplanes [long-reach (LR)], chip-to-module [short-reach (SR), or in some standards, medium-reach (MR)] applications. Common Public Radio Interface (CPRI) is the specialized standard for the wireless radio head, connecting between the top of the cell tower to the bottom of the tower, mostly in optical medium. Passive-Optical-Network (PON) distinguishes itself for the burst mode requirement.
TL;DR: This article is the second in a series based on a presentation for the Advanced RF CMOS Transmitter Techniques Forum at the 2015 International Solid-State Circuits Conference in San Francisco that provided a technical background upon which all the other forum presentations expanded greatly.
Abstract: The second part of this article series, which began in the previous issue [1], presents a quick survey of the many options available for power amplifier architectures.
TL;DR: In this paper, a simplified design of a current-based digital-to-analog converter (also known as IDAC) is presented, where the dc operation of a simple current mirror and how the process variation can affect the accuracy of the currents produced are discussed.
Abstract: The past three articles were related to the small-signal operation of MOS circuits. In this article, we step back and look at dc operation of a simple current mirror and how the process variation can affect the accuracy of the currents produced. We review this through a simplified design of a current-based digital to analog converter (also known as IDAC).
TL;DR: To be compliant with multistandard applications, the RF front end of a modern transceiver must satisfy several challenging tasks such as large operative bandwidth, low noise, and high linearity.
Abstract: i»?To be compliant with multistandard applications, the RF front end of a modern transceiver must satisfy several challenging tasks such as large operative bandwidth, low noise, and high linearity. Over the years, addressing such requirements has significantly changed the radio architecture toward an ultimate solution based on current mode signal processing and passive mixers. In this article, after a brief description of the typical structure of a wireless receiver, voltage and current mode signal processing will be compared showing why a fully current-mode approach is more suitable in deep-scaled CMOS technologies [1], [2].
TL;DR: Over the last several decades, digital communications technologies combined with integrated circuit scaling trends have enabled the microelectronic industry to dramatically scale the bandwidth of high-loss networks such as DSL and Ethernet, resulting in products with volumes in the billions of units.
Abstract: Over the last several decades, digital communications technologies combined with integrated circuit scaling trends have enabled the microelectronic industry to dramatically scale the bandwidth of high-loss networks such as DSL and Ethernet. These channel-limited applications depend on sophisticated equalization techniques to push well beyond the uncompensated bandwidth of the system. And in the last two decades, short-distance wireline links used for chip-to-chip communication applications have enjoyed equally impressive data rate scaling??from a few hundred megabits per second per lane to multigigabits per second in products with volumes in the billions of units.
TL;DR: The telephone is the first analog wireline communication system and also a long-lasting invention as mentioned in this paper, and it served our civilization for more than one and a half centuries, starting with the telegraph.
Abstract: Communication over wires began decades before over the air?in fact, in binary form. The device named ?the telegraph? by Claude Chappe was commercialized in the 1830s and served our civilization for more than one and a half centuries. Introduced by Alexander Graham Bell in the 1870s, the telephone is the first analog wireline communication system and also a long-lasting invention.
TL;DR: In this paper, the authors present an attempt to explore the future of wireline communications, focusing on chip-to-chip (C2C) communications in computer systems, and their focus will be limited to wireline technology.
Abstract: I believe that engineers and academics, whatever their field, should occasionally spend some time trying to forecast the future. They can then try and prepare for it by deciding what activities to prioritize now. This article represents an attempt to explore the future of wireline communications. To avoid making it too long, its focus will be limited to wireline technology for chip-to-chip communications in computer systems.
TL;DR: In this article, it is important to remember that much of the work we do becomes part of archive journals and to be very clear what the words we are using mean is vital to avoid ambiguity, to communicate clearly, and also to align with the history provided by the literature.
Abstract: All projects begin best at the definitions. Being very clear what the words we are using mean is vital to avoid ambiguity, to communicate clearly, and also to align with the history provided by the literature. It is very important to remember that much of the work we do becomes part of archive journals.
TL;DR: Circuit Intuitions provides insights and intuitions into circuit design and analysis aimed at undergraduate students but may serve the interests of other readers as well.
Abstract: i»?Welcome to âCircuit Intuitions!â This is the fifth article of a column series that appears regularly in this magazine. As the title suggests, each article provides insights and intuitions into circuit design and analysis. These articles are aimed at undergraduate students but may serve the interests of other readers as well. If you read this article, I would appreciate your comments and feedback, as well as your requests and suggestions for future articles in this series. Please send your e-mails to ali@ece.utoronto.ca.
TL;DR: In this paper, the authors reflect on data converter evolution through the lens of the converter market applications and provide some interesting perspectives as to how we got here and may even offer some insights about what might come next.
Abstract: There is a common debate in the circuits world about whether the market drives technology advances (embodied by Edison?s observation that ?necessity is the mother or invention?) or whether technical breakthroughs actually create new markets (embodied by the new invention looking for a ?killer app?). Converter innovation over the last 50 years has seen tremendous contributions from both technology push and market pull forces. Reflecting on data converter evolution through the lens of the converter market applications provides some interesting perspectives as to how we got here and may even offer some insights about what might come next.
TL;DR: Wang et al. as mentioned in this paper pointed out that China's most crucial import is not crude oil, but it is integrated circuits (ICs), which accounts for 55% of the worldwide semiconductor market.
Abstract: China is known for its exports, but what is China’s most crucial import? No, it is not crude oil It is integrated circuits (ICs)! So how important is the IC industry to China? Consider this: during the past ten years, ICs have been number one among all manufactured goods imported by China—more than cars and liquid crystal displays According to a recent report by PwCcom, in 2013 China consumed US$170 billion worth of ICs, which is equal to 55% of the worldwide semiconductor market If this economic factor is not sufficient, Edward Snowden’s 2013 revelations, which triggered a worldwide alert on national information security, has further convinced the Chinese government that China’s information technology infrastructure must not continue to rely so heavily on foreign imports
TL;DR: This article provides insights and intuitions into circuit design and analysis that are aimed at undergraduate students but may serve the interests of other readers as well.
Abstract: Welcome to the sixth article in the ?Circuit Intuitions? column series. As the title suggests, each article provides insights and intuitions into circuit design and analysis. These articles are aimed at undergraduate students but may serve the interests of other readers as well. I would appreciate your comments and feedback, as well as your requests and suggestions for future articles in this series. Please e-mail your comments to me at: ali@ece.utoronto.ca.