Microwave Journal
www.microwavejournal.com/articles/5191-the-next-wireless-wave-is-a-millimeter-wave

The Next Wireless Wave is a Millimeter Wave

The past few years has witnessed the emergence of CMOS-based circuits operating at millimeter-wave frequencies. Integrated on a low cost organic packaging, this is the promise for high volume fabrication, lowering the cost and opening huge commercial i...

August 3, 2007

In the past few years, the interest in the millimeter-wave spectrum at 30 to 300 GHz has drastically increased. The emergence of low cost high performance CMOS technology and low loss, low cost organic packaging material has opened a new perspective for system designers and service providers because it enables the development of millimeter-wave radio at the same cost structure of radios operating in the gigahertz range or less.

In combination with available ultra-wide bandwidths, this makes the millimeter-wave spectrum more attractive than ever before for supporting a new class of systems and applications ranging from ultra-high speed data transmission, video distribution, portable radar, sensing, detection and imaging of all kinds.


While at a lower frequency the signal can propagate easily for dozens of kilometers, penetrate through construction materials or benefit from advantageous reflection and refraction properties, one must consider carefully the characteristics (in particular strong attenuation and weak diffraction) of the millimeter-wave propagation, and exploit them advantageously. The free-space loss (FSL) (after converting to units of frequency and putting them in decibel form) between two isotropic antennas can be expressed as1

FSL = 92.4 + 20 log F + 20 log D

where

F = frequency in gigahertz and
D = line-of-sight distance in kilometers

As an example, at 60 GHz the free-space loss is much more severe than at the frequencies usually used for cell phone and wireless applications. The link budget at 60 GHz is 21 dB less than the one at 5 GHz under equal conditions.2 In addition, other loss and fading factors increasingly affect the millimeter-wave transmission, such as gaseous (see Figure 1), rain, foliage, scattering and diffraction losses.

Beside the huge and unexploited bandwidth availability and the perspective of multi-gigabit to terabit networks, the potential of the millimeter-wave spectrum has many others attributes: enabling densely packed communication link networks, from very short range to medium range; leveraging frequency reuse to its paroxysm while increasing the security level of each link; integrating high efficiency radiating elements at the millimeter scale, leading to compact, adaptive and portable integrated systems; exploiting quasi-unlimited and unique electromagnetic signatures for detection, diagnostic or imaging.

Recently, the availability of standard CMOS technology enabling the design of MMIC circuits operating efficiently up to 100 GHz has revived the interest and investment in the 7 GHz of bandwidth unlicensed band in the 60 GHz spectrum. The specificity of the 60 GHz spectrum is the attenuation characteristics due to atmospheric oxygen absorption in the order of 10 to 15 dB/km over a bandwidth of about 8 GHz.

This attenuation precludes long-range communications, but provides an extra spatial isolation that is beneficial for frequency re-use in an indoor dense local network, reduces co-channel interference and provides extra safety for secure short-range point-to-point links. In addition to supporting multi-gigabit networks, this makes the 60 GHz spectrum a great opportunity for indoor ultra-high speed short-range wireless communications, targeting multimedia applications and others.

Similarly, extremely fast growing opportunities for low cost commercial millimeter-wave systems are exploited at even higher frequencies, such as 77 GHz for automotive radar, 71 to 76 and 81 to 86 GHz for outdoor 10 Gbps networks, and 94 GHz for medical and security imaging. This just preludes terabits systems operating beyond 120 GHz and above.

The Multimedia Trend

The emergence of a multitude of “bandwidth hungry” multimedia applications has definitely had a leading role in the renewal of interest in the millimeter-wave spectrum. The conventional WLAN systems (802.11a, b and g) are limited to a data rate of, at best, 54 Mb/s. Alternative solutions such as UWB and MIMO systems will start becoming available to extend the speed up to 600 Mb/s, targeting 1 Gb/s and above in the near future. It is noteworthy that wireless networks tend to lag at least one generation behind wired LAN interconnect technology.3-4

Two primary types of applications are driving the requirement for even higher data rates: ultra-fast file sharing and uncompressed high definition video streaming. Figure 2 illustrates the projected average storage capacity of PCs (desktop and laptop), reaching nearly 300 Gbytes in 2010, as well as the average storage capacity of embedded hard-drives and flash products. In the case of portable devices, especially in the case of smart cell phones, one can note a clear migration from micro-hard-drive toward high speed flash memory technology, exhibiting capacity up to 100 Gbytes and access speed exceeding the Gb/s in the horizon of 2010. It is obvious that today high speed wireless systems will lead to prohibitive synchronization time.

Figure 3 illustrates the data throughput requirement for uncompressed video streaming. It appears again that the data throughput requirement is well in excess of 1 or 2 Gbps, following a progression from 5 to 10 Gb/s and above.

This demand has since pushed the development of technologies and systems operating at millimeter-wave frequencies, while maintaining a cost structure similar to the one of conventional WLAN systems. These throughput requirements of multimedia systems are dictated by interconnect and interface technologies such as PCI-express, High Definition Multimedia Interface (HDMI), Display Port (DP) or Unified Display Interface (UDI), as shown in Figure 4.

Two major standardization bodies, IEEE 802.15.3c and Ecma International TC32-TG20,5-6 are specifically considering these requirements, in the particular case of the 60 GHz spectrum, for applications ranging from very low cost peer-to-peer interface up to high performance Wireless Personal Area Networks (WPAN), including high definition uncompressed video streaming. Back-compatibility should also be considered to provide seamless connectivity across the technologies that will support the coming 4G communications infrastructure (see Figure 5).

CMOS-FR4: A Low Cost Millimeter-wave Radio Platform

Since the mid-90s, many examples of MMIC chipsets have been reported for millimeter-wave radio applications using GaAs FET and InP PHEMT technologies.7 More recently, SiGe BiCMOS technology has also been demonstrated to be a viable alternative.8 Despite their commercial availability and their performance, however, these technologies struggle to enter the market because of their prohibitive cost and their limited capability to integrate advanced baseband processing.

The steadily increasing frequency range of CMOS process technologies has now made the design of low cost, highly integrated 24 and 60 GHz millimeter-wave radio possible in silicon.9-10 Proof of concept has been validated using CMOS 130 nm technology; however, CMOS 90 nm is the first technology node that enables high performance and power efficient implementation of 60 GHz transceivers suitable for high volume products.

In addition, the optimum combination and co-design of CMOS technology with low cost FR4-based packaging technology is a requisite to ensure the minimal cost structure possible, the key for the successful deployment of ultra-high speed, high capacity, 60 GHz WPAN and video streaming applications.

Finally, innovative PHY, MAC, ADC and signal processing approaches are required to provide simultaneously ultra-high bandwidth, very high PHY-MAC efficiency at an affordable price and an acceptable power budget. As depicted in Figure 6, the convergence of module, CMOS MMIC, signal processing and high efficiency PHY-MAC technologies are the necessary key enablers of the coming generation of low cost, high performance millimeter-wave systems.

Millimeter-wave CMOS Technology

The CMOS technology has advanced to a point that a complete chipset for millimeter-wave applications can be implemented using silicon. In a standard 90 nm CMOS technology it is now possible to achieve an Ft and Fmax beyond 150 GHz. Proper transistor geometry and layout, as well as complete and accurate modeling and optimized parasitic extraction methods up to the millimeter-wave frequency of interest are the entry point for such designs (see Figure 7).

The use of millimeter-wave low loss micro-strip line and micro-inductors for matching purposes are very characteristic of this new generation of millimeter-wave designs leading to more compact area and higher performance than its co-planar waveguide (CPW) counterpart. Power gain is in excess of 8 dB at 60 GHz and at a current density of 0.2 mA/mm enables reliable and low power circuit design. In addition, noise figures of 5.5 dB are achievable for similar biasing conditions, which make the optimization of low noise amplifiers easier. P1dB compression points of 4 to 7 dBm are reachable with fairly straightforward power amplifier designs. Fundamental frequency cross-coupled VCOs exhibiting phase noise better than –95 dBc/Hz at 1 MHz offset guaranties proper transmission and demodulation of multi-gigabit/s modulated signals. Figure 8 shows an example of a V-band CMOS 90 nm chipset developed for multi-gigabit short-range multimedia applications.

Comparable figures of merit are also achievable at higher frequencies with the introduction of high volume production 65 and 45 nm CMOS technology, enabling now the design of low power E-band transceiver and targeting a high level of integration for systems such as 77 GHz automotive radar, 71 to 76 and 81 to 86 GHz 10 Gbps outdoor links, and 94 GHz imaging.

The research efforts at the Georgia Electronic Design Center have been focused on the development of a millimeter-wave CMOS fully integrated single chip radio suitable for multi-Gb/s applications. A super-heterodyne architecture using high IF frequency has been chosen and optimized to support wideband modulated signals. In addition, low power mixed-signal circuit techniques and innovative high speed analog-to-digital conversion are used to enable the integration of very low power PHY operating at multi-gigabit and multi-giga samples/s.

FR4-LCP-Based Module and Antenna Technology

Liquid Crystal Polymer has emerged as a promising low cost alternative for millimeter-wave module implementation. It combines uniquely outstanding microwave performances at low cost and large area FR4 PWB processing capability. It appears as a platform of choice for the packaging of the future 60 GHz gigabit radio. 24 x 18 inch FR4-LCP multi-layer substrates are fabricated using high volume standard PWB production lines. An example of a large panel area FR4-LCP multi-layer substrate is shown in Figure 9.

Compact filter designs using planar and integrated waveguide (IWG) techniques have been validated and measured, exhibiting less than 2 dB minimum insertion for a relative bandwidth of 8 percent at 61.5 GHz, and a rejection greater than 20 dB at 6 GHz offset.6-11 A wideband millimeter-wave feed-through transition exhibiting less than 0.2 dB insertion loss has also been implemented.

One of the obvious attractiveness of the millimeter-wave is the small wavelength, allowing the integration of multiple radiating elements in an array configuration while occupying a minimum space (see Figure 10). Numerous antenna array solutions have been developed to address various application scenarios ranging from VSR (very short reach) omni-directional to point-to-point link.12-13

Such generic packaging platforms provide a path of choice toward the low cost integration of scalable SISO-MIMO radio systems (SM radio) using compact multi-sector phased-array architecture that overcomes simultaneously the fundamental limitations of millimeter-wave signal propagation and CMOS technology. The multi-sector architecture can either be integrated on a single large panel or in a compact 3D integrated millimeter-wave module, including an embedded filter and antenna phased array, as shown in Figure 11. Extended azimuth and elevation coverage, provided by conformal multi-sector configuration, and extended range (including non-LOS scenario) provided by high gain adaptive phased-array technology, are the breakthrough attributes of future commercial millimeter-wave systems.

15 Gbps and HD-Video Millimeter-wave Test-bed

The GEDC has established an experimental millimeter-wave wireless test-bed, using 60 GHz as a demonstrator vehicle to study the channel characteristic of a real indoor environment. Researchers recently established a new world record for the highest data rate transmitted wirelessly at 60 GHz, achieving a peak data transfer rate of 15 gigabit/s at a distance of 1 meter, 10 Gigabit/s at a distance of 2 meters and 5 gigabit/s at a distance of 5 meters. In addition, high definition video streaming running at 1.485 Gb/s has been demonstrated through a one-inch thick wood table. Special efforts have been dedicated to the complete transceiver module implementation operating at a power budget well below the one hundred pico-joules range. Figure 12 shows the demodulated transmission of the multi-gigabit signal and the experimental set-up of the video transmission through a one-inch thick wood table.

Conclusion

The development of millimeter-wave radios at the same cost structure of radios operating in the microwave region opens a new field of innovation for system designers. The convergence of a FR4-based module, CMOS MMIC, signal processing and high efficiency PHY-MAC technologies becomes today’s reality, enabling the coming generation of low cost high performance millimeter-wave systems. The feasibility of ultra high speed wireless transmission beyond 10 Gbps has been demonstrated on a low power, low cost platform. A power budget well below the one hundred pico-joules/bit range has been achieved, already looking at the next level of innovation targeting 100 Gbps transmission and the femto-joule/bit power budget.

The spreading of millimeter-wave technology in the consumer electronic market place is on its way, leveraging bandwidth availability at various frequencies, ranges and levels of system complexity. Peer-to-peer ultra fast synchronization and adaptive WPAN, for data and video distribution, will drive the cost down and further eases the adoption of low cost CMOS-based millimeter-wave platforms for automotive radar, outdoor point-to-point/point-to-multi-point links, portable radar, security, sensing and imaging systems, including numerous medical applications.

References

1. M. Marcus and B. Pattan, “Millimeter-wave Propagation: Spectrum Management Implications,” IEEE Microwave Magazine, June 2005, pp. 54–62.

2. P. Smulders, “Exploiting the 60 GHz Band for Local Wireless Multimedia Access: Prospects and Future Directions,” Communications Magazine, Vol. 40, Issue 1, January 2002, pp. 140–147.

3. J.P. Ebert, E. Grass, R. Irmer, R. Kraemer, G. Fettweis, K. Strom, G. Tränkle, W. Wirnitzer, R. Witmann, H.J. Reumerman, E. Schulz, M. Weckerle, P. Egner and U. Barth, “Paving the Way for Gigabit Networking,” Global Communications Newsletter, April 2005.

4. G. Fettweis, “WIGWAM – Wireless Gigabit With Advanced Multimedia Support,” Wireless World Research Forum (WWRF), New York, NY, October 27–28, 2003.

5. IEEE 802.15 Working Group for WPAN; http://www.ieee802.org/15/.

6. http://www.ecma-international.org/ memento/TC32-TG20-M.htm.

7. K. Ohata, et al., “Wireless 1.25 Gb/s Transceiver Modules Utilizing Multi-layer Co-fired Ceramic Technology,” ISSCC Digest, February 7–9, 2000.

8. S.K. Reynolds, B.A. Floyd and T. Zwick, “60 GHz Transceiver Circuits in SiGe Bipolar Technology,” IEEE ISSCC Digest Technical Papers, February 2004, pp. 442–443.

9. L.M. Franca-Neto, R.E. Bishop and B.A. Bloechel, “64 GHz and 100 GHz VCOs in 90 nm CMOS Using Optimum Pumping Method,” IEEE International Solid-State Circuits Conference Digest Technical Papers, February 2004, pp. 444–45.

10. C. Doan, et al., “Design Considerations for 60 GHz CMOS Radios,” IEEE Communications Magazine, December 2004, pp. 132–140.

11. K.S. Yang, S. Pinel, I.K. Kim and J. Laskar, “Low Loss Integrated Waveguide Passive Circuits Using Liquid Crystal Polymer System-on-Package (SOP) Technology for Millimeter-wave Applications,” accepted for publication in IEEE Transactions on Advanced Packaging, 2007.

12. I.K. Kim, S. Pinel, J. Laskar and J.G. Yook, “Circularly and Linearly Polarized Fan Beam Patch Antenna Arrays on Liquid Crystal Polymer Substrate for V-band Applications,” APMC 2005, Vol. 4, December 4–7, 2005.

13. T. Ihara, et al., “Switched Four-sector Beam Antenna for Indoor Wireless LAN Systems in the 60 GHz Band,” Topical Symposium on Millimeter Waves, July 7–8, 1997, pp. 115–118.

Joy Laskar received his BS degree in computer engineering from Clemson University in 1985. He received his MS and PhD degrees in electrical engineering from the University of Illinois at Urbana-Champaign in 1989 and 1991, respectively. Prior to joining Georgia Tech in 1995, he held faculty positions at the University of Illinois and the University of Hawaii. At Georgia Tech, he holds the Schlumberger Chair in Microelectronics, is the director of Georgia’s Electronic Design Center, and heads a research group of 25 members with a focus on integration of high frequency mixed-signal electronics for next generation wireless and wired systems. He has authored or co-authored more than 400 papers, several book chapters (including three textbooks in development), numerous invited talks and has more than 25 patents pending. Most recently his work has resulted in the formation of two companies. In 1998 he co-founded an advanced WLAN IC company, RF Solutions, which is now part of Anadigics. In 2001 he co-founded a next generation analog CMOS IC company, Quellan, which is developing collaborative signal processing solutions for the enterprise, video, storage and wireless markets. He is a 1995 recipient of the Army Research Office’s Young Investigator Award, a 1996 recipient of the National Science Foundation’s CAREER Award, the 1997 NSF Packaging Research Center Faculty of the Year, the 1999 co-recipient of the IEEE Rappaport Award, the faculty advisor for the 2000 IEEE MTT International Microwave Symposium (IMS) Best Student Paper award, the 2001 Georgia Tech Faculty Graduate Student Mentor of the year, the recipient of a 2002 IBM Faculty Award, the 2003 Clemson University College of Engineering Outstanding Young Alumni Award, the 2003 recipient of the Outstanding Young Engineer of the Microwave Theory and Techniques Society, and he has been named IEEE Fellow from 2005. For the 2004-2007 term, he has been appointed an IEEE Distinguished Microwave Lecturer for his seminar entitled “Recent Advances in High Performance Communication Modules and Circuits.” He was also named an IEEE Electron Devices Society Distinguished Lecturer in 2006.

Stéphane Pinel received his BS degree from Paul Sabatier University, Toulouse, France, in 1997, and his PhD degree in microelectronics and microsystems from the Laboratoire d’Analyse et d’Architecture des Systemes, Centre National de la Recherche Scientifique (CNRS), Toulouse, France, in 2000. He has worked on an UltraThin Chip Stacking (UTCS) European Project for three years involving Alcatel Space and IMEC (Belgium). He has been on the research faculty at the Georgia Electronic Design Center (GEDC) of the Georgia Institute of Technology, Atlanta, GA, since 2000. He has authored or co-authored over 125 journals and proceeding papers, two book chapters, numerous invited talks, participated and organized numerous workshops at international conferences such IMS, and holds four patents. He was the recipient of the first price of the SEE 1998 award, the second prize of the IMAPS 1999 award, and the 2002 International Conference on Microwave and Millimeter-wave Technology Best Paper Award (Beijing, China). His research interests include CMOS, SiGe and SOI RF and millimeter-wave circuit design, advanced 3D integration and packaging technologies, RF and millimeter-wave embedded passives (filters, antenna arrays) design using organic (Liquid Crystal Polymer) and ceramic materials (LTCC), RF-MEMS and micromachining techniques, system-on-package for RF and millimeter-waves front-end module. He is currently leading research efforts for the development of a 10 Gbps wireless test-bed at GEDC.

Debasis Dawn received his B.Eng. degree from Jadavpur University, India, in 1986, his M.Tech. degree from IIT Kanpur, India, in 1989, and his PhD degree from Tohoku University, Japan, in 1993, all in electrical engineering. He was a research associate at Tohoku University for five years and was engaged in various government funded research projects in the development of millimeter-wave circuits and millimeter-wave photonic circuits. He was a research engineer at Fujitsu Laboratories Ltd. in Japan for five years and was engaged in MMIC development and device modeling using GaAs pHEMT process technology. He also held a senior MMIC design engineer position at SONY Corp, Japan. He has been on the research faculty at the Georgia Electronic Design Center (GEDC) of the Georgia Institute of Technology, Atlanta, GA, since 2006, and is currently engaged in the development of RF front-end circuits for wireless personal area network (WPAN) applications using CMOS process technology. He has published more than 20 papers in international journals and conferences and holds two US patents. He is a member of the IEEE.

Saikat Sarkar received his B.Tech. degree in electronics and electrical communication engineering from the Indian Institute of Technology, Kharagpur, India, in 2003, and his MS degree from the Georgia Institute of Technology, Atlanta, GA, in 2005. He worked at Intel Corp., Hillsboro, OR, as a summer intern in 2004. He is currently pursuing his PhD degree at the Georgia Institute of Technology, where he is a member of the Microwave Applications Group at the Georgia Electronic Design Center (GEDC). He has authored or co-authored over 15 journal and conference papers. His research interests include silicon-based millimeter-wave front-end circuit design, system integration and the analysis of millimeter-wave high data rate low power wireless systems.

Bevin George Perumana received his B.Tech. degree in electrical engineering from the Indian Institute of Technology, Kharagpur, India, in 2002, and his MS degree in electrical engineering from the Georgia Institute of Technology, Atlanta, GA, in 2005, where he is currently pursuing his PhD degree in electrical engineering. In 2002 he became a research consultant with the Advanced VLSI Design Laboratory at the Indian Institute of Technology. He has held internship positions with the Communications Technology Laboratory of Intel Corp., Hillsboro, OR, and with PCS, Mumbai, India. He has been a graduate research assistant with the Microwave Application Group at the Georgia Institute of Technology, where his main research focus is CMOS wireless transceiver circuits.

Padmanava Sen received his B.Tech. degree in electronics and electrical communication engineering from the Indian Institute of Technology, Kharagpur, India, in 2003, and his MS degree from the Georgia Institute of Technology, Atlanta, GA, in 2005. He worked for IBM, Burlington, VT, as a co-op for seven months in 2004 and three months in 2006. He is currently pursuing his PhD degree from the Georgia Institute of Technology, where he is a member of the Microwave Applications Group at the Georgia Electronic Design Center (GEDC). He has authored or co-authored over 15 IEEE journal and conference papers. His research interests include the analysis and development of millimeter-wave silicon-based transmitters.