Category Archives: Online Magazines

ISSCC 2013: Analog trends


February 14, 2013

Analog and digital integrated circuitsThe efficient control, storage, and distribution of energy are worldwide challenges, and are increasingly important areas of analog circuit research. While the manipulation and storage of information is efficiently performed digitally, the conversion and storage of energy must fundamentally be performed with analog systems. As a result, the key technologies for power management are predominantly analog. For example, there is much interest in wireless power transmission for battery charging applications, ranging from mobile handsets to medical implants, and increased efficiency in wireless power transmission is enabling faster charging over longer distances. There is also an explosion of technologies that permit energy to be collected from the environment via photovoltaic, piezoelectric, or thermoelectric transducers. A significant focus here is on analog circuits that are able to harvest sub-microwatt power levels from energy sources at 10’s of millivolts, to provide autonomy for remote sensors or to supplement conventional battery supplies in mobile devices. To achieve this, extremely low power must be consumed by the attendant analog circuits so that some energy is left over to charge a battery or super capacitor. Similarly, the power consumption of analog instrumentation amplifiers, oscillators, and audio power amplifiers is being scaled down to meet the demands of these low power systems. Fast power-up and -down is also desired from these circuits to permit high energy-efficiency during intermittent operation. Together, these technologies will permit devices to be powered indefinitely from sustainable sources, opening the door to ubiquitous sensing, environmental monitoring, and medical applications.

Analog circuits also serve as bridges between the digital world and the analog real world. Just like the bridges in our roads, analog circuits are often bottlenecks and their design is critical to overall performance, efficiency, and robustness. Nevertheless, digital circuits such as microprocessors drive the market; so semiconductor technology has been optimized relentlessly over the last 40 years to reduce the size, cost, and power consumption of digital circuits. Analog circuitry has proven increasingly difficult to implement using these modern IC technologies. For example, as the size of transistors has decreased, the range of analog voltages they can handle has decreased and the variation observed in their analog performance has increased.

These aspects of semiconductor technology explain two key divergent trends in analog circuits. One trend is to forgo the latest digital IC manufacturing technologies, instead fabricating analog circuits in older technologies, which may be augmented to accommodate the high voltages demanded by increasing markets in medical, automotive, industrial and high-efficiency lighting applications. Other applications dictate full integration of analog and digital circuits together in our most modern digital semiconductor technologies. For example, microprocessors with multiple cores can reduce their overall power consumption by dynamically scaling operating voltage and frequency in response to time-varying computational demands. For this purpose, DC-DC voltage converters can be embedded alongside the digital circuitry, driving research into the delivery of locally regulated power supplies with high efficiency and low die area, but without recourse to external components.

This and other related topics will be discussed at length at ISSCC 2013, the foremost global forum for new developments in the integrated-circuit industry. ISSCC, the International Solid State Circuits Conference, will be held on February 17-21, 2013, at the San Francisco Marriott Marquis Hotel.

Cabot Microelectronics to collaborate with SEMATECHSEMATECH announced today that Cabot Microelectronics Corporation has joined its Front End Processes, or FEP, program and will collaborate with SEMATECH to develop advanced solutions for emerging CMP applications.

“SEMATECH provides Cabot Microelectronics with excellent process capability to help identify and demonstrate emerging applications for CMP consumables,” stated Ananth Naman, Cabot Microelectronics’ Vice President of Research and Development. “We expect this collaboration to help enable Cabot Microelectronics support our customers’ emerging CMP requirements.”

As semiconductor device sizes shrink, new materials are introduced, and higher yields are targeted, achieving wafer scale planarity through CMP has become increasingly challenging. These issues are expected to continue to become more challenging in the context of low-power technologies.

“Cabot Microelectronics’ CMP processing solutions will complement our own device and process expertise,” said Paul Kirsch, SEMATECH’s director of Front End Processes. “We will work together to develop practical, manufacturable solutions to address the emerging needs of advanced transistor technologies.”

The goal of SEMATECH’s FEP program is to enable novel materials, processes, structural modules and electrical and physical characterization methods to support the continued scaling of logic and memory applications.

Headquartered in Aurora, Illinois, Cabot Microelectronics Corporation is a supplier of CMP polishing slurries and a growing CMP pad supplier to the semiconductor industry. Since becoming an independent public company in 2000, the company has grown to approximately 1,050 employees on a global basis.

SEMATECH is an international consortium of leading semiconductor device, equipment, and materials manufacturers for over 25 years.

GaN on GaN LEDs by SoraaSoraa announced yesterday the next generation of its high external quantum efficiency GaN on GaN LEDs. As described in Appl. Phys. Lett. 101, 223509, Soraa’s new LED outperforms the best-documented LED laboratory result by Nichia Chemical Co. at current densities of 100 A/cm2 and beyond.

“The record breaking performance from our next generation of GaN on GaN LEDs is a credit to the extremely talented research and development team at Soraa, and a testament to the vision of our founder and GaN on GaN pioneer, Dr. Shuji Nakamura,” said Mike Krames, CTO of Soraa. “But what’s amazing is that we have just scratched the surface in terms of performance gains from our GaN on GaN LED technology.”

Soraa’s GaN on GaN LEDs handle significantly more current and emit ten times more light per unit area of LED wafer material than conventional LEDs made by depositing GaN layers on cheaper foreign substrates like sapphire, silicon carbide or silicon. The company’s GaN on GaN technology leverages the advantages of the native substrate, including over a thousand times lower crystal defect densities that allow reliable operation at very high current densities (the same principle that enabled Blu-ray laser diodes). In addition to superior crystal quality, the native substrate’s optical transparency and high electrical and thermal conductivity enable a very robust, simple LED design that delivers maximum performance. Another advantage of the GaN on GaN approach is that it enables considerable flexibility in the choice of crystal growth plane.

 “We firmly believe that GaN on GaN is the future for LEDs and we’ve developed a comprehensive intellectual property program and patent portfolio covering the technology to maintain our leadership position.” added Krames.

Soraa leveraged the advantages of its first generation GaN on GaN LEDs by introducing in 2012 the true full-visible-spectrum LED MR16 lamps—a superior alternative to 50-Watt halogen MR16 lamps. Soraa’s LED MR16 lamps have a CRI of 95 and R9 of 95 (higher than most halogen lamps) and compared to halogen lamps, produce no UV or IR; last up to 10 times as long; use 75 percent less energy; run cooler; produce a much more consistent and efficient beam; and are compatible with existing lamp fixtures and lighting infrastructure.

CoorsTek, one of the largest technical ceramics manufacturers, today announced introduction of aluminum nitride substrates. Ideal for the rapidly growing LED market and other markets where high heat dissipation is useful, these ceramic substrates boast a thermal conductivity of 170 W/m K.

CoorsTek AlN (aluminum nitride) ceramic substrates feature a very high dielectric strength, are a non-toxic alternative to BeO (beryllium oxide), and exhibit a thermal expansion coefficient similar to Si, GaN, and GaAs semiconductors.

“While we already offer an extensive line of ceramic substrates, our new high performance aluminum oxide substrates cover high heat dissipation applications,” says Andrew Golike, electronics general manager for CoorsTek, Inc.

Established in 1910, CoorsTek is one of the largest technical ceramics manufacturer in the world with over 40 facilities in the Americas, Europe, and Asia.

Today, Research and Markets released the Gallium Nitride (GaN) Semiconductor Devices (Discretes & ICs) Market, Global Forecast & Analysis: 2012 – 2022. This research study on the GaN semiconductors market gives a detailed overview of the global GaN semiconductors market in the present scenario, and discusses the history, evolution, market by technology, market by products and devices, market by application segments, and by geography. Each classification done for the global GaN semiconductors market has an extensive segmentation with market estimates and forecasts till 2022 for each sub-market in terms of both – revenue and volume. The major GaN semiconductor products, namely the power semiconductors and opto-semiconductors, are analyzed in great detail throughout the research study in every type of classification.

GaN has turned out to be the choice for most of the power semiconductor applications and is quickly replacing the existing silicon technology, according to the report. The various properties of GaN,  such as wider bandgap, high break-down voltage, larger critical electric field, and higher thermal conductivity, let the GaN devices operate at higher voltages, high switching frequencies, handle higher power density, and offer enhanced power efficiency than the pure Si devices. These properties allow the GaN discretes like Schottky diodes, MOSFETs, and the other advanced transistors to operate at much higher voltage levels, which are difficult for the counterpart Si devices. GaN power semiconductors also help in reducing the conduction and switching losses, thereby offering higher efficiency in electronic systems. The major application segments of GaN power semiconductors currently are the inverters (& converters), RF devices, power supply modules, and motor drives being used across all the end user verticals.

In the aspect of opto-semiconductors, GaN has been increasingly used in LEDs, laser diodes, and optocouplers due to the primary characteristic of GaN showing much brighter emission characteristics than the other materials, such as Si, SiC, GaAs, and GaP. AlGaN, mixed with GaN, is used in opto-semiconductors designed for high-brightness and ultra-high brightness applications that operate at wavelengths <400 nm. This market is expected to have healthy demand due to the markets growing for green, blue, violet, purple, ultra-violet, and white LEDs. The growing market for solid state lighting applications in several areas of the consumer electronics, computers, automotive, industrial and military, aerospace and defense sectors is expected to boost the GaN opto-semiconductors market revenue over the years to come.

This report, based on the extensive research on the GaN semiconductors market and industry, is aimed at identifying the entire market for the GaN semiconductor devices, and all its sub-segments through extensively detailed classifications, in terms of both revenue and shipments. It focuses on giving a bird’s eye-view of the upcoming industry with regards to GaN semiconductor market with detailed market segmentations, combined with qualitative analysis at each and every aspect of the classifications done.

MEMS microphone market to doubleSilicon microphones are among a broad range of devices known as micro-electromechanical systems (MEMS), an emerging field in which various sensors and mechanical devices are constructed on a single wafer using processes developed for making integrated circuits (ICs). The chief advantage of micromachining silicon microphones is cost. Several sensors can be processed on a chip simultaneously and can be integrated with passive and active electronic devices.

According to a new market research study from Innovative Research and Products, or iRAP, titled MEMS Microphones – A Global Technology, Industry and Market Analysis (ET-118), silicon micro-machined microphones (also known as silicon microphones or MEMS microphones) have begun to emerge as a competitor technology to the electret condenser microphone (ECM). The global market for MEMS microphones has reached approximately $422 million in 2012. The market is predicted to increase to $865 million in 2017, with increasingly high uptake of MEMS microphones over alternatives for a variety of applications. Thanks to Apple Inc., which has spurred on this phenomenal growth by adopting MEMS microphones for their products, namely the iPhone, iPad and iTouch, hence paving the way for other smartphone and tablet manufactures to adopt the same.

MEMS microphones are more compact than traditional microphone systems, because they capture sound and convert it to a digital signal on the same chip. MEMS microphone solutions developed on the CMOS (complimentary metal oxide semiconductors) MEMS platform frees consumer electronic device designers and manufacturers from many of the problems associated with ECMs. CMOS MEMS microphones also integrate an analog-to-digital converter on the chip, creating a microphone with a robust digital output. Since the majority of portable applications will ultimately convert the analogue output of the microphone to a digital signal for processing, the system architecture can be made completely digital, removing noise-prone analogue signals from the circuit board and simplifying the overall design.

Report Highlights

The new iRAP study has focused on MEMS microphones that can be used in mobile phones, digicams, camcorders, laptops, automotive hands-free calling and hearing aids. It provides market data about the size and growth of the MEMS microphones application segments, new developments including a detailed patent analysis, company profiles and industry trends. The report also covered the underlying economic issues driving the MEMS microphones business, as well as assessments of new advanced MEMS microphones that are being developed.

Manufacturers of MEMS microphones expect competition to persist and intensify in the future from a number of different sources. Microphones are facing competition in a new, rapidly evolving and highly competitive sector of the audio communication market. Increased competition could result in reduced prices and gross margins for microphone products and could require increased spending by research and development, sales and marketing and customer support.

Micro-machined microphone chips can match and extend the performance of existing devices, for instance, by using sensor arrays. Silicon microphones also offer advantages to the OEM in the form of improved manufacturing methods (reliability, yield, assembly cost) combined with robustness. They also offer additional functionality, such as the ability to incorporate multiple microphones into portable electronic devices for noise suppression and beam forming.

The potential for smaller footprint components and resistance to electromagnetic interference also supports new cell phone designs. Moreover, MEMS microphones meet price points set by electret microphones by leveraging established high-volume silicon manufacturing processes. This combination of size, performance and functionality, and low cost are highly desirable for OEMs and consumers alike.

Many of these new “miniature” silicon microphones for consumer and computer communication devices are approximately one-half the size and operate on just one-third the power of conventional microphones.

The range of possible applications of these microphones derives from their important advantages as compared to conventional ECM technologies. Based on silicon MEMS technology, the new microphone achieves the same acoustic and electrical properties as conventional microphones, but is more rugged and exhibits higher heat resistance. These properties offer designers of a wide range of products greater flexibility and new opportunities to integrate microphones.

Report Conclusions

Major findings of this report are:

  • The MEMS microphones market is an attractive, and still growing, 100s of million-dollar market characterized by very high production volumes of MEMS microphones that are extremely reliable and low in cost.
  • Mobile phones would consistently have the largest share through 2017, followed by laptops and tablets, camcorders, hearing aids, headphones and automotive.
  • From 2012 to 2017, hearing aids will have the highest growth rate with AAGR at 27.46%, followed by headphones at 25% AAGR.
  • Regionally, North America had about 25.3% of the market in 2012, followed by Europe at 19.7 %, Japan at 15.7% and the rest of world at 39.5%.
  • In 2012, More than ten companies and institutions worldwide are active in the field of MEMS microphones, which can be divided in two different technological concepts – single-chip and two-chip. The number of active market participants is expected to double by 2017.
  • By 2017, MEMS microphones will achieve penetrations of 92% in the mobile phone market segment and 95% in PDAs, digicams and camcorders market.
  • In terms of technology, the largest share will be for two-chip integration.

MEMs in the medical fieldMicroelectromechanical (MEMS) devices are shaping the competitive landscape in the global medical device industry. Several factors are behind the increasing demand for and innovation in MEMS devices in the medical industry: growing number of MEMS applications in healthcare; innovations, revolution and growth in the personal healthcare market, including wireless implants; and rising awareness and affordability of healthcare.

Participants and would-be entrants must understand the medical MEMS device market in order to compete in it. Global Information (GII) highlights three major reports that present the key issues driving and constraining market growth, in addition to probable solutions that can address emerging concerns in the medical MEMS market. Report forecasts provide a quantitative assessment of the market for companies to benchmark their performance and plan for future high growth areas, while qualitative analyses provide both an overarching view and a detailed breakdown of the MEMS market.

MEMS Devices in Global Medical Markets

The use of MEMS devices by different stakeholders is driving market growth by adding to the demand of devices from different medical market segments as discussed above. This is also indirectly encouraging for medical sector market players (particularly big ones) that have diverse customer bases composed of different stakeholders and diverse product portfolios (such as diagnostics, research, and medical devices), as they can capitalize on the MEMS market by leveraging their existing resources to some extent. Moreover, a diverse set of devices catering to the needs of different stakeholders encourages new entrants into sectors of their choice to complement or suit their capabilities and potentials.

Integrated devices and advancements in inertial sensors, such as products for human motion analysis, are meeting the needs of the modernized healthcare delivery model, especially for the elderly patient sector, by adding the element of prevention. An example of product innovation is microneedles for drug delivery, which is gaining popularity by offering a pain-free and enhanced, accurate method of drug delivery. Similarly, the diagnostic devices have significantly reduced the sample testing time from hours to a few minutes, thus significantly adding value to the healthcare delivery model from different perspectives such as time efficiency, convenience, patient satisfaction, and ease of operations.

Microfluidic/lab on chip (LOC) is considered a revolutionary technology for the life sciences and healthcare industry. This technology enables the integration of assay operations, such as sample pretreatment and sample preparation, on a single chip. This is radically changing the pharmaceutical and life-sciences research sector by changing the way procedures, such as DNA analysis and proteomics, are conducted.

The microfluidic/lab on chip (LOC) segment is expected to rise to 72% of the market share of MEMS devices by 2017. Major growth drivers of this sector are research tools, which are expected to achieve significant growth of CAGR 28.8% from 2012 to 2017. A surge from 2012 to 2017 in research applications, such as proteomics, genomics, and cellular analysis, is also expected to boost this sector.

In terms of applications, the macro segments of the market include pharmaceutical and life-sciences research, in vitro diagnostics, home healthcare, and medical devices. Among all of these applications, research is expected to grow at the highest CAGR of 28.3% from 2012 to 2017.

BioMEMS

Expected to triple in size over the next five years, the bioMEMS market is expected to grow from $1.9 billion in 2012 to $6.6 billion in 2018. Microsystem devices have applications in four key healthcare markets: pharmaceutical, in-vitro diagnostics, medical devices and medical home care. Microsystem devices have become increasingly visible in the healthcare market by serving as solutions adapted to the requirements of various applications. The usefulness of these devices is two-fold: they improve medical device performance for the patient; and secondly, they offer competitive advantages to system manufacturers. For example, the introduction of accelerometers in pacemakers has revolutionized the treatment of cardiac diseases.

BioMEMS devices examined in the report include: pressure sensors, silicon microphones, accelerometers, gyroscopes, optical MEMS and image sensors, microfluidic chips, microdispensers for drug delivery, flow meters, infrared temperature sensors, and emerging MEMS including RFID, strain sensors, and energy harvesting.

The Global MEMs Device, Equipment, and Materials Markets: Forecasts and Strategies for Vendors and Foundries

A significant portion of MEMS manufacturing technology has come from the IC industry. MEMS devices can be made using silicon wafers and the manufacturing process can incorporates semiconductor manufacturing processes such as sputtering, deposition, etching and lithography. This report analyzes the market for MEMS devices and the equipment and materials to make them.

This report provides forecasts for the following key MEMS device applications: ink jet head, pressure sensor, silicon microphone, accelerometer, gyroscope, MOEMS, Micro Display, Microfluidics, RF MEMS, Micro Fuel Cells, and more.

InfiniLED’s latest MicroLEDs, or µLEDs, have produced record optical beam intensity. This new device is capable of producing up to 1mW of light from a single 20µm pixel at 405nm. This is equivalent to a light output density of more than 300 W/cm2 – the highest recorded for a commercially available LED type device.

“These results highlight the capabilities of the MicroLED,” said Dr. Bill Henry, chief commercial officer of InfiniLED, “This device can be seen as a cross-over between the power and collimation of a laser and the simplicity of an LED. The unique devices enable a range of applications. This was achieved without the need for external optics indicating the potential for further improvement of the performance.”

The MicroLED combines the benefits of a laser and a LED to produce ultra-high light output. The MicroLED provides the wavelength flexibility, drive characteristics and simplicity of a LED as well as the power and collimated beam of a laser. The ability to produce such light intensity and control directly from the chip enables the light to be efficiently used in a range of applications.

InfiniLED has achieved this record performance using the patented MicroLED structure. A parabolic reflector is etched into the semiconductor material during the fabrication process. This places an optical component directly at the site of light generation and at the most effective position for control of the light produced. Not only has the light been shown to be extracted in ultra-high intensity but also at high efficiencies. By directing all the generated light through a single surface of the semiconductor, it can be efficiently collected and used in the wider system.

The MicroLED is currently being used in a range of applications included life sciences, consumer electronics and OEM equipment. The MicroLED (µLED) can be fabricated as a single pixel, large clusters of pixels or as addressable arrays where each pixel is individually switchable. The single pixels can be used to produce high intensity, collimated light over a small area or to produce useable light at ultra-low currents. The single pixels produce light with a few nanoamps of current. To produce larger amounts of light, clusters of tightly packed MicroLEDs are available. This results in high light density and collimated emission over a wider area. MicroLEDs (µLEDs) are also available as addressable arrays of pixels. The collimation from each pixel results in high packing densities and minimal crosstalk between the devices.

Additionally, the high current densities achievable and low capacitance allows the MicroLEDs to be switched at very high speeds. Experimental work is on-going with the Tyndall National Institute and the results will be announced shortly.

 “The applications for the MicroLED are many and varied,” Henry added. “InfiniLED is developing light sources for use in areas such as diagnostics, printing and battery powered consumer electronics. We are particularly focused on applications where the efficient use and control of light is of greatest importance. The first products with MicroLEDs incorporated will be on the market shortly and we look forward to new releases in the near future.” 

InfiniLED will demonstrate this technology at BiOS and Photonic West in San Francisco this February.

Introduction

This year has shown increased innovation, integration, and technical maturity across RF frequency bands. This document outlines emerging RF trends that will be covered at ISSCC 2013.

ISSCC 2013 authors will present an ongoing drive toward increasing levels of integration. This trend can be seen in all areas of RF design from mm-Wave, to cellular, to imaging, to wireless sensors. In mm-Wave designs, higher system complexity (front-end, synthesizer, and baseband) is increasingly being integrated onto a single die. In cellular, the push for integration has led to a strong trend of architectures allowing better linearity and co-existence of these multiple bands and standards. In a related trend, there has been much research the last few years into various ways to remove costly and bulky SAW filters and duplexers. Some of these research areas include highly linear blocker-tolerant receivers, mixer-first receivers, feedback blocker cancellation, feed-forward blocker cancellation, N-path filters, and electrical balance of hybrid transformers. Strong work continues in the effort to integrate CMOS PAs while delivering viable power/efficiency performance. Finally, a clear trend this year was a significant number of chips demonstrated in 65nm CMOS compared to other technology nodes. This observation was noted across all frequency ranges and circuit topologies. The chips presented at ISSCC 2013 confirm that RF devices will continue to see larger levels of integration at the chip- and package level for years to come.

Over the past decade, the papers submitted to ISSCC have indicated clear trends in the continuing push to higher frequencies of operation in CMOS and BiCMOS. This trend has continued this year for oscillators, mm-Wave amplifiers, and PAs. An emerging trend is the increasing complexity of systems operating in the 60-to-200GHz range. The push to ever-higher frequencies is being pursued by both industry and academia for various applications. An important application is high-data-rate communication. With the low-GHz frequency spectrum already overcrowded, researchers are continuing to target frequencies above 60GHz. Two other applications for products operating in these frequency bands are imaging and radar. These frequencies are desirable for such products due to their high spatial resolution and enablement of small antenna dimensions, allowing efficient beam-forming arrays. Another continuing trend is the integration of mm-Wave antennas into silicon substrates.

The combination of these two trends (that is, increased integration and the push to higher frequencies) has enabled a new class of fully-integrated application-driven systems. With the availability of many RF and mm-Wave building blocks in CMOS and BiCMOS, fully integrated solutions for specific emerging domains are appearing, both in the RF and the mm- Wave frequency range. These systems are built on a foundation of circuit-block innovations that have been developed over the past few years. Single-chip radars in RF and mm-Wave frequencies with improved resolution, improved efficiency, showing increasing levels of integration are being demonstrated. Similarly, new systems are being developed for ultra-wideband radar and mm-Wave wireless sensing. Demonstrations in the biomedical field are clearly moving from simple electrical measurements towards real medical measurements in realistic environments using systems-in-package (SiP).

We now discuss these two trends supported with data from chips to be presented at ISSCC 2013.

Complexity and maturity in the mm-Wave and sub-mm-Wave ranges

The high cutoff frequency of bipolar transistors and highly downscaled MOS transistors enables the realization of circuits and systems operating in the mm-Wave range. In the last few years, high-data-rate communication in the 60GHz band and car radar around 77GHz have garnered much attention. While the integration level in these domains is already quite high, we see an improvement of the performance of the building blocks (e.g. output power of PAs, spectral purity and tuning range of VCOs).

The 100GHz barrier for the operating frequency of silicon circuits has been broken a few years ago. Whereas initially elementary building blocks like a VCO and an amplifier operating above 100GHz have been realized, we now witness the trend of increasing complexity in circuits operating above 100GHz. Meanwhile, the electrical performances at the building block level improve: the output power of mm-Wave and sub-mm-Wave sources and PAs increases and VCOs can operate at ever-increasing frequencies with a higher tuning range.

Co-existence and efficiency for cellular applications

RX and TX linearization: In the past few years there has been increasing interest in techniques to improve the linearity of transmitters and receivers. Improved linearity in the receivers will ease the requirements on the RF filtering of out-of-band blockers that can be accomplished, for instance, by placing a programmable notch filter in the RF path. TX linearity improvements will benefit performance parameters such as error-vector magnitude (EVM), ACLR and spectral purity.

Efficiency: PA efficiency improvements demonstrated this year will directly impact the battery life in portable applications. These efficiency improvement techniques include analog and digital pre-distortion, dynamic biasing and envelope tracking.

Digitally-assisted RF: The trend towards digitally-assisted RF continues and is increasingly applied to mm-Wave chips. More digitally assisted calibration techniques are being demonstrated in order to improve the overall performance of the transceiver by reducing the impact of analog impairments at the system level. These techniques include: spur cancelation/reduction, IIP2 improvements, and digital pre-distortion.

VCOs: There is a continuing trend toward improvements in phase-noise figure-of-merit (FOM) and power consumption due to circuit techniques like Class-C and Class-D VCOs. This year’s ISSCC shows clear contributions to this field.

This and other related topics will be discussed at length at ISSCC 2013, the foremost global forum for new developments in the integrated-circuit industry. ISSCC, the International Solid-State Circuits Conference, will be held on February 17-21, 2013, at the San Francisco Marriott Marquis Hotel.

Demand for ubiquitous mobile functionality to achieve enhanced productivity, a better social-networking experience, and improved multimedia quality, continues to drive innovation in technologies that will deliver to these objectives in an energy and cost-efficient manner. While the performance of embedded processors has increased to meet the rising demands of general-purpose computations, dedicated multimedia accelerators provide dramatic improvements in performance and energy efficiency of specific applications. Energy harvesting is another area of growing importance, leading to technologies that leverage non-volatile logic-based SoC’s for applications that do not have a constant power source or handheld devices with very limited battery capacity.

Technology scaling continues to be exploited to deliver designs capable of operating at lower voltages, resulting in reduced energy per operation, as well as reducing the area required to implement specific functions. Processors unveiled at ISSCC 2013 are built on a variety of technology nodes, with best-in-class results accomplished along the axes of integration, performance/watt and functional integration, as well as a few industry-first implementations. These are demonstrated in various process nodes ranging from 0.13μm down to 28nm bulk, and SOI CMOS technologies.

Emerging medical applications require a significant reduction in the standby power over state-of-the-art commercial processors. This drives the exploration of new leakage-reduction techniques in both logic and on-chip memories, targeting orders of magnitude reduction in leakage currents. Fast wake-up time requirements drive the need for saving and restoring the processor state.

In the late 1990s, a GSM phone contained a simple RISC processor running at 26MHz, supporting a primitive user interface. After a steady increase in clock frequency to roughly 300 MHz in the early 2000s, there has been sudden spurt towards 1 GHz and beyond. Moreover, following trends in laptops and desktops, processor architectures have become much more advanced, and recent smart phones incorporate dual and even quad-core processors, running up to 2GHz frequencies. Battery capacity, mostly driven by the required form factor, as well as thermal limits imply a power budget of roughly 3W for a smartphone. From this budget, also the power amplifier (for cellular communication) and the displays have to be powered. The available power budget for everything digital is in the range of 2W (peak) to 1W (sustained). As a result, energy efficiency has become the main challenge in designing application processors, graphics processors, media processors (video, image, audio), and modems (cellular, WLAN, GPS, Bluetooth). For video and image processing, the trend has been towards dedicated, optimized hardware solutions. Some new areas where dedicated processors are particularly needed include gesture-based user interfaces, and computational imaging, to name a few. For all digital circuits, the limited power budget leads to more fine-grained clock gating, various forms of (adaptive) voltage-frequency scaling, a variety of body-bias schemes, and elaborate power management strategies.

Interestingly, cellular links, wireless LAN, as well as short links consistently show a 10× increase every five years, with no sign of abating. With essentially constant power and thermal budgets, energy efficiency has become a central theme in designing the digital circuits for the involved signal processing. Historically, CMOS feature sizes halve every five years. For a brief period in the 1990s, CMOS scaling (a.k.a. Dennard scaling) provided a 23 (α-3) increase in energy efficiency per five years, almost matching the required 10×. During the past decade; however, CMOS scaling offers a roughly 3× improvement in energy efficiency every five years. The resulting ever-widening gap has led to alternative approaches to improve energy efficiency, namely, new standards, smarter algorithms, more efficient digital signal processors, highly-optimized accelerators, smarter hardware-software partitioning, as well as the power management techniques mentioned above.

This and other related topics will be discussed at length at ISSCC 2013, the foremost global forum for new developments in the integrated-circuit industry. ISSCC, the International Solid-State Circuits Conference, will be held on February 17-21, 2013, at the San Francisco Marriott Marquis Hotel.