mmWave: The Battle of the Bands

Overview

Every year the number of wireless enabled devices and the amount of data consumed continues to grow at an exponential rate (53% CAGR[1].) As these devices create and consume growing amounts of data, the wireless communications infrastructure that connects these devices must evolve to support the demand. Increasing the spectral efficiency of 4G-based network is not enough to deliver the step function in data rate necessary for the three high level 5G use cases as defined by 3GPP[2], shown in Figure 1, with the goal of providing ubiquitous, instantaneous mobile broadband data. Recognizing this, researchers have been looking to higher frequencies as a potential solution. Positive results from early channel sounding work is refocusing the world’s wireless standardization bodies on how next generation 5G wireless systems might incorporate and benefit from these new frequencies and wider bandwidths.

Contents

Defining the KPIs for 5G

Each use case is designed to allow a future wireless standard to address new applications underserved by existing wireless standards, each requiring a new and different set of key performance indicators (KPIs).  The Enhanced Mobile Broadband (eMBB) as defined by the IMT 2020 use case envisions a peak data rates of 10 Gb/s, 100x over 4G[3].  Data rates are empirically linked to available spectrum according to the Shannon Hartley theorem which states that capacity is a function of bandwidth (ie spectrum) and channel noise[4].  With spectrum below 6 GHz fully allocated, spectrum above 6 GHz, specifically in the mmWave range presents an attractive alternative to address eMBB use case.

 

Figure 1: Three high level 5G use cases as defined by 3GPP and IMT 2020

 

mmWave: The story of 3 frequencies

Service operators around the world have paid billions of dollars for spectrum to service their customers.  Auction pricing for spectrum highlights its value in the market and the scarcity of this precious resource.  Opening up new spectrum could enable service operators to accommodate more users while also delivering a higher performance mobile broadband data experience.  Compared to spectrum below 6GHz, mmWave is plentiful and lightly licensed, meaning it could be made accessible to service operators around the world.  Advances in silicon manufacturing have reduced the prices of mmWave equipment dramatically to a point where it is feasible for consumer electronics.  The challenges impacting mmWave adoption now lie primarily in the unanswered technical questions regarding this largely uninvestigated spectrum.

Service operators have begun investigating in mmWave technology to evaluate the best candidate frequencies for use in mobile applications.  The International Telecommunication Union (ITU) and 3GPP have aligned on a plan for 2 phases of research for 5G standards.  The first phase defines a period of research for frequencies under 40 GHz to address the more urgent subset of the commercial needs completing September 2018.  The second phase is slated to begin in 2018 and completing December 2019 to address the KPIs outlined by IMT 2020.  This second phase focuses on frequencies up to 100 GHz.
In an effort to globally align the standardization of mmWave frequencies, ITU released a list of proposed globally viable frequencies between 24 GHz and 86 GHz after the most recent World Radiocommunications Conference (WRC)[5]:

 

24.25–27.5GHz                                        31.8–33.4GHz

37–40.5GHz                                             40.5–42.5GHz

45.5–50.2GHz                                           50.4–52.6GHz

66–76GHz                                                      81–86GHz


Shortly after ITU proposal, the Federal Communications Commission (FCC) in the United States issued a Notice of Proposed Rule Making (NPRM)on October 21st 2015 that proposed new flexible service rules among the 28 GHz, 37 GHz, 39 GHz, and 64-71 GHz bands[6].

 



Figure 2: FCC Bands Proposed for Mobile Use[6]



While the ITU, 3GPP, and other standards bodies have decided on 2020 as the deadline for the 5G standard to be defined, cellular providers are working on an accelerated schedule for delivering 5G service.  In the United States, Verizon and AT&T are aiming to roll out an early version of 5G in 2017.  Korea is aiming to roll out 5G trials at the Olympics in 2018, and Japan wants to demonstrate 5G technologies at the Tokyo Olympics in 2020.  Through these varying groups and motivations, a set of frequencies are beginning to emerge as the candidates for 5G: 28 GHz, 39 GHz, and 72 GHz.

These 3 frequency bands have emerged for a number of reasons.  First, unlike 60 GHz which has approximately 20 dB/km loss due to oxygen absorption[7], they have much lower oxygen absorption rates, shown in the figure below, making them more viable for long distance communications.  These frequencies also function well in multipath environments and can be used for non-line-of-sight communications.  Using highly directional antennas in combination with beam forming and beam tracking, mmWave can provide a reliable and very secure link.  Research into the channel properties and potential performance has already been started for 28 GHz, 38 GHz, and 73 GHz by Dr. Ted Rappaport and his students at NYU Polytechnic School of Engineering.  They have published several papers with propagation measurements and studies on potential service outages at these frequencies.  This existing data and research at these frequencies combined with the availability of spectrum world-wide makes these 3 frequencies the starting point for mmWave prototyping.

 

Figure 3: Atmospheric absorption across mmWave frequencies in dB/km[7]

 

28 GHz

As mentioned above, service providers are eager to access the extensive unallocated mmWave spectrum, and they will be key influencers for which frequencies will be used in the mmWave spectrum.  In February 2015, Samsung performed their own channel measurements and have been able to show that 28 GHz is a viable frequency for cellular communications.  Their measurements validated the expected path loss for urban environments (path loss exponents is 3.53 in Non Line-of-Sight (NLoS) links) and Samsung claims that this data suggests a mmWave communications link can be supported for over 200 meters of distance[8].  Their research also includes work with phased array antennas.  They have begun characterizing designs that could fit intricate phased arrays inside cell phones.  In Japan, NTT Docomo partnered with Nokia, Samsung, Ericsson, Huawei and Fujitsu to do their own successful field trials at 28 GHz (along with other frequencies).

In September 2015, Verizon announced that they, along with key partners including Samsung, will be conducting field trials in the United States in 2016.  This is 4 years earlier than the proposed 2020 date for the 5G standards to be set, making Verizon an early player in the 5G market.  In November 2015 Qualcomm ran experiments at 28 GHz with 128 antennas to demonstrate mmWave technology in a dense urban environment and how directional beam forming can be used for non-line-of-sight communications.  With the announcement from the FCC that the 28 GHz spectrum can be used for mobile communications, further experiments and field trials in the US are expected to continue.  Verizon has also announced a deal to lease the 28 GHz spectrum from XO communications with the option to purchase the spectrum at the end of 2018.

It should be noted, however, that the 28 GHz band is not included in the ITU’s list of globally viable frequencies.  Whether or not it will be the long term frequency for 5G mmWave applications is yet to be determined. The spectrum’s availability in the US, Korea, and Japan, along with US service providers’ commitment to early field trials could push 28 GHz into US mobile technology regardless of the global standards.  Korea’s desire to show 5G technology at the 2018 Olympics could also push 28 GHz into consumer products before the standards bodies finalize the 5G standards.  The fact that this frequency was not on the International Mobile Telecommunications (IMT) spectrum list did not go unnoticed and has drawn some attention from FCC commissioners.  Commissioner Jessica Rosenworcel said in a speech she gave in Washington in February 2016:


"There are some places where when we look high, I believe the United States will need to go it alone.  This includes the 28GHz band…Unfortunately, at the World Radio Conference in Geneva last year this band was left off the table. It was not included in the study list for 5G spectrum. But because this band has a global mobile allocation I think the United States should continue to explore this spectrum frontier. Tests in this band are already underway in South Korea and Japan. So I don't think this is the time to hold back. I think we need to move ahead -- on our own -- and have a framework in place for the 28GHz band by the end of the year.”


Commissioner Michael O’Rielly went so far as to write an entire blog article for the FCC expressing his displeasure with the 2015 World Radio Conference (WRC) results:


“This leads me to contemplate the practical effect of what happened at WRC-15 and its impact on the ITU role going forward.  There is a real possibility that these practices undermined the value of future WRCs and increased the risk that the ITU will become a tool for governments and incumbent spectrum users to halt spectral efficiency and technological progress[9]”


Whether or not 28 GHz becomes a widely adopted frequency for 5G is yet to be seen, but it is clear that it is important right now.

 

73 GHz

In parallel to the work being done around 28 GHz, E-band frequencies have been of interest for mobile communications in the last several years.  The channel measurements taken by NYU at 73 GHz were leveraged by Nokia to begin research at this frequency.  At National Instrument’s yearly conference, NI Week, in 2014, Nokia was able to show their first over-the-air demo operating at 73 GHz using National Instruments prototyping hardware.  This system has continued to evolve as research has continued, with consistent public demonstrations to display new technological achievements.  By Mobile World Congress 2015, the prototyping system was capable of over 2 Gigabits per second (Gbps) data throughput using a lens antenna and beam tracking.  A MIMO version of this system was showing at the Brooklyn 5G Summit in 2015 operating at over 10 Gbps and less than a year later at Mobile World Congress (MWC) the prototype demonstrated a 2 way over the air link operating above 14 Gbps.

Nokia was not alone showing a 73 GHz demo at MWC 2016.  Huewei also showed a prototype with Deutsche Telekom operating at 73GHz.  This demo, using Multi-user (MU) MIMO, displayed high spectrum efficiency and potential for more than 20 Gbps throughput rates for individual users.

Some research has begun at 73 GHz and more is anticipated in the next 3 years.  One of the defining characteristics of 73 GHz that sets it apart from 28 GHz and 39 GHz is the available contiguous bandwidth.  There is 2 GHz of contiguous bandwidth available for mobile communications at 73 GHz, the widest of the proposed frequency spectrums.  By comparison, 28 GHz offers 850 MHz of bandwidth and there are 2 bands around 39 GHz offering 1.6 GHz and 1.4 GHz bandwidth in the US.  And as mentioned earlier, per Shannon, more bandwidth equates to more data throughput, and this gives 73 GHz a big advantage over the other frequencies mentioned.

 

38 GHz

With the least amount of currently ongoing publicly available research, 38 GHz still has potential to be part of the 5G standards.  It was listed by the ITU as a globally viable frequency.  There is existing channel data proving that it is a viable frequency done by NYU.  One of the challenges with this frequency versus 28 GHz or 73 GHz is that there are more existing uses for this frequency.  The FCC has proposed spectrum for potential mobile use, helping facilitate future research in the band in the US.

Verizon, while focusing on 28 GHz for its initial field trials in 2016, already has a plan in place to do testing at 39 GHz.  XO Communications owns substantial licenses at 39 GHz in addition to the licenses it owns at 28 GHz.  With such a large investment from a service provider into this spectrum and its place on the IMT list, it is a contender for the 2020 5G standards.

 

mmWave Prototyping

To capitalize on the promise of mmWave for 5G, researchers must develop new technologies, algorithms and communications protocols as the fundamental properties of the mmWave channel are different from current cellular models, and are relatively unknown.  The importance of building mmWave prototypes cannot be overstated, especially in this early time frame.  Building mmWave system prototypes demonstrates the viability and feasibility of a technology or concept in a way that simulations alone cannot.  mmWave prototypes communicating in real-time and over the air in a variety of scenarios will unlock the secrets of the mmWave channel, and enable technology adoption and proliferation.

There are several challenges to building a complete mmWave communications prototype.  Consider a baseband subsystem capable of processing a multi-GHz signal.  Most of today’s LTE implementations typically use 10 MHz channels (20 MHz maximally), and the computation load increases linearly with bandwidth.  In other words, the computational capacity must increase by a factor of 100 or more to address the 5G data rate requirements.  For mmWave system physical layer computation, FPGAs are essential for prototyping.

Building custom hardware that is capable of prototyping mmWave applications is a daunting task.  One of the reasons mmWave frequencies are so appealing for communications is the large amounts on contiguous bandwidth.  Finding an off the shelf hardware transmitter or receiver with the 1 to 2 GHz of required bandwidth for 5G applications is very expensive, or impossible depending on the frequency needed.  Even if one can find this hardware, having the ability to configure and process the raw data is limited if it is available at all.  Because of this, designing custom FPGA processing boards becomes an appealing option.  The amount of engineering time needed to design the hardware for an FPGA board may not seem significant, but when coupled with the amount of time needed to develop a software interface to communicate with it even the most skilled engineer could easily spend one year or more on this design process.  And this is only one piece of the prototyping system.

In addition to FPGA boards, a mmWave prototyping system needs to utilize state of the art DACs and ADCs to capture between 1-2 GHz of bandwidth.  There are some RFICs on the market today that include chips that convert between baseband and mmWave frequencies, but these options are limited and mostly cover the 60 GHz band.  An IF and RF stage can be used as an alternative to RFICs.  Once an engineer has a baseband and IF solution, there are more options available from vendors for mmWave radio heads than the baseband RFICs, but still not many.  Developing a mmWave radio head requires RF and microwave design expertise, an entirely different skill set than developing FPGA boards, meaning that a team with a diverse set of skills is needed to develop all of the necessary hardware.  FPGAs must be considered a core component to a mmWave baseband prototyping system, and programming a multi-FPGA system capable of processing multi-GHz channels presents increased system complexity.  To address the system complexity and software challenges faced by service providers and communications researchers, National Instruments provides a configurable set of mmWave prototyping hardware along with a mmWave physical layer in source code that accounts for the fundamental aspects of a mmWave system baseband and also provides abstractions for data movement and processing across multiple FPGAs to simplify the task.  These tools are designed to assist in the transition of new prototypes into systems and products that will be crucial to the development 5G technology.

 

Summary

While the future of exactly how 5G will be implemented is not yet clear, it is clear that mmWave will be one of these technologies.  The large amount of contiguous bandwidth available above 24 GHz is needed to meet data throughput requirements, and researchers have already been able to show through prototyping that mmWave technology can be used to deliver rates above 14 Gbps.  The biggest unanswered question is which mmWave frequency band that is used for mobile communications.  The ITU may be able to play a role in setting one frequency for mobile use with 5G.  Handset makers and consumers would benefit from the reduced cost of developing and using only one set of silicon instead of the multiple chips needed in today’s phones for global coverage.  However, frequency incumbents are costly to relocate.  Finding a single band that can be agreed upon globally is an admirable goal to strive for but ultimately might not be achieved.  To meet aggressive timelines, service providers in different regions are choosing to ignore the recommendations of the ITU and use the spectrum most readily available even if it doesn’t scale globally. They are also taking advantage of the ability to accurately prototype 2-way communications links in field testing and trials, a vital part of 5G development, allowing researchers to demonstrate this new technology and move it towards standardization faster than ever before.

Despite many unknowns, one thing is very clear: mmWave technology is going to be deployed, and it will be deployed quickly.  The next generation of wireless communications is on the horizon and the world is watching and waiting to see how exactly how it will be implemented.

 

Next Steps

Learn more about NI mmWave Transceiver System

Explore how Nokia rapidly prototyped a mmWave system

 

References

[1] CISCO VNI 2016: http://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/mobile-white-paper-c11-520862.html

[2] RAN 5G Workshop, Sep 19, 2015   http://www.3gpp.org/news-events/3gpp-news/1734-ran_5g

[3] IMT 2020 https://www.itu.int/dms_pubrec/itu-r/rec/m/R-REC-M.2083-0-201509-I!!PDF-E.pdf

[4] Taub, H., & Schilling, D. L. (1986). Principles of Communication Systems. McGraw-Hill.

[5] Resolution Com6/20, Provisional Final Acts WRC-15. WRC-15 (pp. 424-426). Geneva: ITU. http://www.itu.int/dms_pub/itu-r/opb/act/R-ACT-WRC.11-2015-PDF-E.pdf

[6] Use of Spectrum Bands Above 24 GHz for Mobile Radio Services, GN Docket No. 14-177, Notice of Proposed Rulemaking, 15 FCC Record 138A1 (rel. Oct. 23, 2015)

[7] T. S. Rappaport, J. N. Murdock, and F. Gutierrez, ‘‘State of the art in 60 GHz integrated circuits & systems for wireless communications,’’ Proc. IEEE, vol. 99, no. 8, pp. 1390–1436, Aug. 2011.

[8] Samsung “5G Vision”, p. 7, http://www.samsung.com/global/business-images/insights/2015/Samsung-5G-Vision-0.pdf page 7

[9] O’Rielly, M. (2016, January 15).  2015 World Radiocommunication Conference: A Troubling Direction https://www.fcc.gov/news-events/blog/2016/01/15/2015-world-radiocommunication-conference-troubling-direction