Home > Projecten > Technische Universiteit Delft > Circuits en systemen >
Jaarcongres 2011
Nieuws
Agenda
Over STW
Folder STW
Kennisexploitatie
Praktijkvoorbeelden
Logos
Organisatie
Adres en routebeschrijving
Jaarverslagen
Utilisatierapporten
Address and route description
English brochure
STW publicaties
Infobalie
Algemeen
Aanvragers
Referenten en Juryleden
Projectleiders
Gebruikers
Projecten
Programma's
Vacatures
Links
English
Login
Contact

Millimetre-wave integrated circuits for silicon technology (DCS.6422)

Project nummer: dcs6422

Omschrijving van het onderzoek

Fast and effective information transfer is a cornerstone of a knowledge based economy. Future broadband wireless networks will utilize millimeter-wave integrated circuits (MMICs) that process radio frequency (RF) information signals with wavelengths of just a few millimetres. These circuits will operate at tens of gigahertz, or over ten times the clock speed of the latest Pentium microprocessor. The need for RF circuitry operating at ever increasing frequencies is twofold. On the one hand, lower frequency bands are congested by the continually increasing number of subscribers for mobile and wireless communications networks. Also, data transfer speeds must increase in order to support faster computing technology (e.g., the next generations of PCs) and broadband internet services. MMICs reduce the need for external components to an absolute minimum, which reduces costs and makes them easier to manufacture in large quantities with improved reliability. At present, most MMICs are fabricated using gallium arsenide (GaAs) or indium phosphide (InP) materials, which are 3 to 5 times more expensive than silicon. From the economic and strategic points of view, integration of millimeter-wave circuits on silicon can further reduce costs because the radio or RF circuitry can then (potentially) be combined with computing circuits on the same silicon chip.
This paves the way for new applications incorporating wireless and computing technologies in products affordable to the average consumer. Integration of MMICs in silicon is becoming feasible as transistor dimensions continue to shrink with each new technology generation. This provides faster circuits with increased data processing capabilities. Within the next 5 years, silicon transistors will be capable of operation beyond 200GHz, bringing high-speed and highperformance RF circuits within reach.
Rapid implementation of MMICs in silicon technology is hindered by two problems. Firstly, the short wavelengths (just a few millimetres) are close to the length of a wire on a silicon chip, which changes the behaviour of the signals being processed considerably. Therefore, to predict the impact of these changes with computer-aided design tools (CAD), new computer models are required to model on-chip interconnect and components, such as inductors and transformers constructed from on-chip wire connections. Secondly, signal attenuation is considerable on semiconducting substrates like silicon. However, new technologies for on-chip interconnect, such as the slow-wave magnetic components that have been developed by members of this research group, do not suffer this impairment. This technology also allows RF electronics to operate below 1 Volt with reduced current and power consumption and improved sensitivity and fidelity. These benefits are more compelling at millimetre-wave frequencies, because the size of each passive element on-chip can be reduced in proportion to the shrinking wavelength so that it consumes less chip area.
Scalable CAD models of passive slow-wave components will be developed as part of this project. However, the overall goal of the proposed research is to develop and demonstrate silicon ICs that both operate effectively at millimetre-wave frequencies and exploit the emerging silicon technologies with transistor bandwidths in excess of 100 GHz.
The proposed applications contexts are: a broadband distributed amplifier (50 GHz bandwidth) to drive lasers in the next generation of multi-gigabit/second glass fibre communication systems, and key elements of a 77 GHz radar transceiver. The transceiver combines transmitter and receiver on-chip, and is intended for automotive collision avoidance radar applications. The production volume for these radars are in the millions per year, as multiple units are built into every new automobile. The cost/volume benefits of silicon IC technology can enable this to happen if adequate performance is realized through our efforts. Dutch industry will benefit from application of integrated high-performance RF and digital circuitry on the same silicon substrate as it opens up valuable new consumer markets in the communication and transportation sectors.
The demand for broadband information from the internet via wired and wireless networks is fuelled by the exponential growth in processing speed and data storage capabilities of computer hardware. It requires on-going improvements in network bandwidth and radio frequency allocations. As a result, new frequency allocations in the 2-11 GHz and 10- 66 GHz range are now being made to support future wireless network growth, and 40Gb/s optical fibre networks are in the development phase. The results of this project supports these two important areas by providing models and key components needed to implement broadband and narrowband integrated circuits on a silicon chip operating between 10 and 77 GHz. This effort removes the main obstacle to utilisation of silicon as a millimetre-wave technology, by making new slow-wave passive components and models for computer-aided design widely available. It will also provide an integration path for these models into current CAD systems for chip designers. When linked into these tools, design of a wide variety of silicon microwave ICs should be possible, and we will prove this with our two demonstration circuits.
Integration of our designs and models with standard tools provides a path for other users to access and design microwave circuits that are less expensive, more reliable and smaller, and it also makes them potential consumer items. For the end-user, this makes existing products cheaper and opens the door to exciting new applications such as collision avoidance radar for automobiles and Gbit/s data rate wireless LANs. It also supports the position of the Dutch semiconductor industry in a huge global market that is increasingly dependent on broadband communications technology. This research project has a very high potential for industrial takeup in new and existing applications, as it builds on silicon technologies and commercial CAD tools that are widely available.

Gebruikers

Er zijn nog geen gebruikers geïdentificeerd, interesse?

Projectleider

Prof.dr. J.R. Long Technische Universiteit Delft
Elektrotechniek, Wiskunde en Informatica
Postbus 5031
2600 GA Delft

Status van het project

Gestart: 01-09-2004
Einddatum: 04-10-2007
.

  Print | Over deze site |  Sitemap | Voorbehoud | Gewijzigd 7-3-2006
Nieuws uitgelicht
Nieuwsbrief Technologiestichting STW, januari 2012
31 januari 2012
Elke maand stuurt Technologiestichting STW haar relaties een link naar de web-based nieuwsbrief. Hierin staat een maandelijks overzicht van het jongste nieuws van de bestuurstafel, onderzoeksnieuws, o... [meer]