​We attended the operator and vendor consortium of 5G Americas.  The themes of the show were: 5G, spectrum, cell siting, Asia-Pacific operator progress.  For the second time in the past couple weeks, we saw FCC Commissioner Michael O'Reilly present, and his key messages were similar both times, focusing on CBRS, C-Band and 6 GHz.  In attendance from the North American service provider side were AT&T, T-Mobile US, Shaw, and Sprint (we focused on NA operators mainly in this write-up).  Notable vendors included Cisco, Commscope, Ericsson, Intel, Kathrein, Mavenir, Nokia, Qualcomm, and Samsung.  We would say the most important theme from the show is the surge in interest in unlicensed spectrum, both for the use of mobile operators, as well as competing carriers, as well as by enterprises both for indoor and outdoor applications.  For this write-up, we are focusing primarily on comments made by some of the leading operators who attended the conference.

AT&T discussed mmWave, future 3GPP releases, 5G phones, Mobile Edge Computing and indoor cellular, mid-band spectrum strategies, 5 GHz spectrum usage, Mobile Edge Computing (MEC), StandAlone (SA), among other topics.  AT&T views mmWave as just a tool in the toolkit, so to speak, and not the only spectrum that is useful in 5G.  It considers mmWave to be most helpful in urban and potentially indoor settings.  Representatives said that future 5G-oriented Releases 16 & 17 are expected to be software upgrades to existing hardware and won't require new equipment to incorporate these new capabilities which will include network slicing.  AT&T is making a big deal about its Mobile Edge Computing (MEC) initiative.  At the conference, it emphasized MEC as having two main parts: a) expansion to about 100 edge sites (mostly Central Offices) from about 20 central locations in the LTE era and initially supporting packet core, and b) Microsoft Azure services managed end-to-end by AT&t.  The company also emphasized that it plans to pursue some indoor cellular opportunities, some that currently leverage 5 GHz using LAA technology, some that will leverage CBRS and some that will leverage mmWave.  We get the impression from AT&T that it is open in how it pursues future mid-band spectrum strategies.  Its strategy could change based on: a) the timing of the CBRS PAL licenses (currently slated for June 25, 2020), b) the potential for C-Band  private auctions (potentially in the mid 2020 timeframe), c) the potential for some or all of the 6 GHz spectrum availability (where Wi-Fi 6 would co-occupy), as well as other factors.  We learned that, at least in certain regions, the company is making very ample use of 5 GHz spectrum using LAA techniques.  AT&T seeing its picocells (small cells) get around 100 Mbps from LAA out of a total 130 Mbps inclusive of around three other licensed spectrums.  We were surprised the company makes such ample use of unlicensed spectrum where Wi-Fi currently exists.  The 5 GHz experience of AT&T leads us to think that 6 GHz, which promises to offer far more spectrum that the 5 GHz swath presently available, could be very beneficial to mobile operators and their consumers, as well as the Wi-Fi industry, and its consumers.  AT&T expects that by this time next year, it will be "pushing" 5G to all its customers, part as a result of handsets adopting 5G capabilities, part the result of its network seeing nationwide coverage.  Of the services that AT&T operates, it is installing mainly Packet Core in its MEC systems.  AT&T is also planning to run Microsoft Azure services in its MEC locations.   It expects that both Packet Core and Azure will see a 10-20 ms latency reduction by being located in MEC locations.  AT&T says that StandAlone (SA) is "just new software," and downplayed the significance of the upgrade from EPC/NonStandAlone (NSA) to SA.

Sprint "is all-in on 2.5 GHz mid-band deployments for 5G services."  Given the company's potential merger with T-Mobile USA, we view its network-build-out choices as being somewhat limited.  It has limited options because it increases its near-term value to its acquirer, T-Mobile, if it deploys 5G in 2.5 GHz.  Likewise, it is doesn't implement in mmWave, this reduces overlap with T-Mobile, who is deploying there.  The company reiterated that it had launched 5G in 9 markets.  It is seeing its peak speeds on 5G (aided by the fact that it has simultaneously upgraded hardware to Massive MIMO) be about 3-5 times that of its 8T8R LTE systems.  It currently covers 11M POPs and 2,100 square miles with 5G.  Sprint also shared that it sees RFPs from customers to replace Wi-Fi with 5G, though it didn't share more about this topic.  The company's experience is that in upgrading its macro base stations to Massive MIMO 64T64R capabilities, it is getting 3-4x faster throughput than its 8T8R systems, though in the field these measurements vary widely.  Additionally, Sprint said that its Massive MIMO systems relative to earlier systems show "generally the same coverage," with 1-2 dB better sometimes.  Sprint is exploring ORAN and vRAN but "not adopting near term."

Shaw (Canada) presented its mobile LTE and 5G efforts and plans.  Shaws plans are interesting because the company has significant cable services deployed in Canada.  The company said nearly all the mobile technology it has installed in the past three years are "5G-ready."  It will use 5G first in 600 MHz, then in mid-band (probably in 3.5 GHz) and the last in mmWave. Shaw expects that low-band 5G handsets will be available in 2020, and, similar to what AT&T said, it expects that is when 5G mobile will start in earnest in Canada.  Shaw admitted that it is behind where the US operators are in deploying 5G, but offered no apologies, as it felt it is where it needs to be from a competitive standpoint in Canada.  Almost laughing, Shaw explained that it would never consider deploying mmWave along highways, and that only high-density locations would get mmWave coverage.  Shaw's view that mmWave is for high-density locations was shared universally by other operators in attendance, including AT&T, Sprint and T-Mobile US.

T-Mobile US spokespersons explained that mmWave has seen some challenges, relative to initial expectations and that while it does get mmWave to operate beyond near-line-of-sight, the view of T-Mobile is that mmWave is "just part of 5G."  T-Mobile expects 3GPP Release 16 to be completed in 2020, but that it will be 2021 before it deploys Release 16, which won't require "a massive hardware refresh" and which will incorporate industrial and connected vehicles features.  T-Mobile views 5G as being appropriate for indoor installations because while mmWave has challenges penetrating glass and concrete, but when 5G operates in low and mid-band spectrums, the "issue goes away."  By 2020, T-Mobile expects StandAlone packet core to be ready, but since its current EPC/NonStandAlone (NSA) systems are already virtualized, the upgrade to SA is "not a forklift" upgrade.  T-Mobile says virtual RAN (vRAN) "will take time," and that it will "need accelerators," which we take to mean FPGA-based Network Interface Cards (NICs) or the like to allow servers to operate faster than just x86 processors will allow.

​We attended the CBRS Alliance event in Washington DC today, and by our rough estimate, about 350-400 people were in attendance representing groups such as regulators, legislators, lawyers, technology vendors, property owners, service providers, investors, media and analysts.  We were impressed with the widespread interest in the new shared spectrum technology and services running in the 3.5 GHz band that is now called “OnGo.”  We have researched CBRS for many years and found several acronyms and CBRS-specific terminology to be blossoming.  We found several themes at the CBRS Alliance event and a follow-on event at Federated Wireless, a SAS service provider, of special note: a) the OnGo experience will serve as a mold for regulators, operators and other interested parties not just in the US, but also the rest of the world, b) Tier 1 operators and WISPs appear focused on Fixed Wireless Access (FWA) deployments in CBRS spectrum, at least initially, c) many presenters focused on the “OnGo backhaul to gateways” use-case, at least as an initial opportunity, d) interested parties have a concern that PAL licenses may become very expensive when the auctions occur, and e) there were a very large number of devices supporting OnGo at this event.

Acronym soup.  The CBRS Alliance did its best to explain the various acronyms and how the various players work together.  It would take at least six pages to cover just the top-level details.  The idea here is that the 150 MHz of spectrum in the 3.5 GHz range was previously used exclusively by the US Department of Defense and is now going to be shared using a three-tier process, where the military (the incumbent) will have use of it when it needs, then private license holders will get next dibs (PAL), followed by general users (GAA).  Starting today, GAA users will begin use of the spectrum in the Initial Commercial Deployment (ICD) that was announced today, starting at 9 AM Eastern.  A group of service providers called Spectrum Access System (SAS) providers have been authorized to install radios on the US coastline that sense when the military is using the spectrum and send channel-use information to equipment that is operating in the CBRS spectrum.  These SAS providers will, therefore, coordinate the frequencies between incumbent, PAL, and GAA users.
​
Our view on why OnGo and “Shared spectrum” matters.  We expect that by sharing spectrum amongst various parties, more traffic can move across a smaller range of frequency than by using the more common method of auctioning off frequency bands to be used exclusively by one entity.  We estimate that shared frequency will carry ten-times more traffic than frequency bands licensed for the exclusive use of single entities.  Thus, it is for the greater good that this OnGo / CBRS experience go the distance and allow a public demonstration of whether multi-tiered shared spectrum can succeed or not.  Already, we have the experience of shared spectrum in the 2.4 GHz and 5 GHz bands used by WiFi – there is no doubt this has been successful; in fact, most public estimates show about 80% of smartphone traffic is carried by WiFi rather than cellular systems, all of which as of yesterday was carried on licensed spectrum.  At the CBRS Alliance event, guest speaker, US FCC Commissioner Michael P. O’Reilly said that based on the success of OnGo, he expects similar models could be applied to additional spectrum (and he implied this might the sequential order of launch): C-band (3.7-4.2 GHz), 3.45-3.55 GHz, 3.1-3.45 GHz and 7 GHz (which we understand is meant to be the same thing as what is being discussed at 6 GHz by the WiFi community).

FWA opportunity is front and center.  Charter and AT&T focused their comments on their plans to deploy fixed broadband systems.  AT&T shared some impressive statistics about the performance of recent trials using Massive MIMO cell sites using distributed RAN over CBRS spectrum, which is connected to indoor baseband over fiber optics to the radio sites and then connects wirelessly to customer premises equipment mounted at the roofline: it said it achieved 140x12 Mbps at slightly over one mile over line of sight using 20 Mhz channels. Charter discussed it had deployed its first commercial FWA in Davidson City, NC to rural locations.  It also discussed how it uses dual SIM systems to allow customer coverage to Verizon’s cellular network. Charter also discussed private LTE, neutral host, and Industrial IoT use cases.  The Wireless Internet Service Provider’s Association (WISPA) President spoke about its members’ enthusiasm for OnGo and explained that 100’s of WISPs used the 3.65 GHz spectrum and expects more will use the 3.5 GHz / CBRS spectrum.  Currently, WISPA says WISPS in the US have 6 million customers.

OnGo as a backhaul.  We detected a theme that seems durable: CBRS spectrum can be used by enterprises with far-flung operations to save costs by reducing the installation of wired / optical cables and associated infrastructure.  There was an impressive list of vendors who had equipment at the show, a number of which were gateway devices that made connections between CBRS and other well-known protocols such as Ethernet and WiFi, to name a couple.  While OnGo/CBRS support is not as widespread on devices today, IoT devices supporting other wired and wireless systems certainly are, the list of which includes WiFi, Zigbee, Bluetooth, Ethernet and more.  We were taken by how compelling some presenters made a case for using CBRS simply assuming a reduction in new cabling to enable new systems such as kiosks, surveillance, digital signage, farming, and so on.  Many of these examples would increase the deployment of existing protocols like WiFi, Zigbee, Bluetooth, and Ethernet, instead of reducing their demand.  The idea that OnGo/CBRS competes with existing systems may be incorrect.

PAL auctions.  Commissioner O’Reilly said PAL auctions are scheduled for June 25, 2020.  In our formal and informal interviews, we understand there is a growing concern that CBRS spectrum auctions could be aggressively pursued not only by existing Tier 1 mobile operators but also by other players, not least of which could include MSOs and maybe even “Big Tech” companies.  Since the 3.5 GHz spectrum is where many countries besides the US have begun deploying 5G services, making equipment in these frequency bands commonplace, there is ample reason to want to use this spectrum in the US.  Bidders may raise the price high enough that enterprises will choose not to compete, and won’t view the CBRS spectrum as attractive as they had hoped.  In this case, PAL would look quite a bit more like a typical licensed spectrum, similar to other auctions.

OnGo devices abound.  At the show, the following vendors had devices on show (see pictures):  Sercomm, MultiTech, Sierra WIreless, Zyxel, Encore, Cradlepoint, AMIT Wireless, Commscope / Ruckus, Accelleran, Bai Cells, Cambium, Samsung, Google, LG Electronics, Sequans, Telit, JMA Wireless, Motorola Solutions, Cisco, BEC Technologies, Ericsson, ip access, BLINQ, Comba Telecom, and Westell.

We attended the Wi-Fi NOW conference in Redwood City, CA this week and attended some interesting presentations.  We are writing about our observations and notes from the Google, Quantenna, Mist Systems and Mojo Networks presentations.  

Google Station presentation.  "GOOGLE STATION: PUBLIC WI-FI TO CONNECT THE NEXT BILLION INTERNET USERS."  Monica Garde and Erika Wool made an interesting presentation.  The jist of the presentation, from our viewpoint, is that Google is partnering with service providers and enabling these service providers to monetize the Wi-Fi network through a revenue sharing system that is based primarily upon advertising.  The company shared some statistics, which we have in the accompanying slide.

Quantenna presentation.  James Chen, VP Product Line Management presented "GREAT INNOVATIONS PART ONE: MASSIVE MIMO & DUAL-BAND 802.11AX".  Chen made the the case that 8x8 WiFi (that Quantenna calls Massive MIMO) outperforms 4x4 systems.  For instance, in its tests, at 85 RSSI and through a wall, performance was 1.6x greater using 8x8 compared to 4x4.  The company also made the case that Massive MIMO has greater throughput compared to non Massive MIMO, as well; the company has demonstrated >1 Gbps throughput in a typical home.  The company showed that Massive MIMO alleviates the "Sticky Client" using a 1x1 Samsung Galaxy Tab Active2 device.  The company did not talk about 802.11ax, unfortunately, other than to say that 8x8 is relevant for 802.11ax, as well.

Mojo Networks presentation.  Mojo CEO Rick Wilmer made the point that simply enabling Cloud-managed Wi-Fi has been done already, implying this is cloud 1.0, and that this message is boring.  The company explained that its cloud architecture is cloud 2.0 because it takes advantage of the capabilities in the cloud and enables - Cognitive Wi-Fi.  Cognitive Wi-Fi, as far as Mojo is concerned, has to do with big data (store key client parameters and run ML algorithms) and smart edge APs.  The company didn't go into deep science of ML/AI, but explained the ML workflow: 1-data collection, 2-training the classifier model, 3-trained model in action, 4-result.  

Mojo explained that it has lots of data to perform Machine Learning on.  It has 1/2M APs deployed.  The company shared that using 1 week of data of a subset across only 4 verticals (enterprise, education, mfg, retail & hospitality): 237K clients, 31M associations, 400+ applications.  Separately, the company said it has obtains 50M associations per week (in a press release).  A significant amount of the data that is delivered to the cloud has been pre-processed in the Mojo APs; the APs cache 2 days of data.  The point of these statistics, according to Mojo, is that it has more data than other Wi-Fi vendors to train its Machine Learning system on.  

According to Mojo, using inference engine, automatically fixes everything possible.  Wilmer says that this makes interacting with the User Interface less necessary because it takes care of problems automatically.  Was Mojo serious or joking when it said, "the UI may disappear as we know it."  Time will tell.

The company shared some other information that was interesting:

• 130K APs deployed at a single customers - a very large customer.  Very scalable.
• Many Mojo customers are decomissioning their EThernet infrastructure (including cables) in lieu of using Wi-Fi from Mojo. 
• Mojo deploys 'a lot' of 3 radio APs.  With third radio, a listening radio, it watches RF environment and does wireless intrusion scanning.  Every 5 seconds, it feeds critical information to the ML environment.  So, if users physically move, the AI system can make adjustments to the system so there are fewer performance issues.  The third radio can also be used as a client, to simulate performance.  This client capability can run the specific application that will used in an environment and simulate whether the Wi-Fi will perform adequately before an event (like a test at a university, for instance).
• Open Standards.  Believe all APs from all vendors should interoperate like light bulbs.  That would accelerate the Wi-Fi market.  Wilmer highlighted that the OCP event, earlier this year, Mojo 'lit up' the entire trade show floor with non Mojo APs using the Mojo system.
• Mojo will announce AP providers soon that are not Mojo APs that you can plug.  No margin stack from an OEM, purchased directly from an Asian ODM.  

Mist Systems.  Bob Friday, of Mist made a presentation on May 17, 2018.  In addition to the content from his presentation, I interviewed other at Mist personnel at the show.  The company claims it is focusing and having success in selling to large enterprises.  We learned that Mist uses Broadcom WiFi chips and has a custom-designed Bluetooth antennae array (shown at the show).  The company highlights its location services as a unique capability, and it draws upon its Bluetooth capabilities to deliver location.  However, the company's main message is its AI capabilities; in some ways, it has become the poster-child for AI amongst startups in the networking industry.  Mist's presentation at the show reiterated the same point - that it is an AI company.  

Stepping back, Mist has been shipping commercially for a year now.  In our observation and research, the company's efforts to take share from competitors has landed it on the map - over the past two quarters, its larger competitors have taken notice of Mist and see it competing at large enterprise accounts.  

During the Q&A part of the presentation by Bob Friday, Mist CTO and founder was asked something that we found very interesting; the question was what kinds of algorithms does Mist use in its system, and do they all need to learn?  The answer was to the effect that many different types of algorithms are used, linear optimization, decision tree analytics, neural networks, etc.  Friday made the case that there are just certain things you just know about how a WiFi network will and should work, so why go an have a machine learn about it when you already know it.  This begs the question - how necessary is AI in the first place, especially if the vendor and its IT workers or VARs have gobs of experience and can design and implement a Wi-Fi network right in the first place.  Looking at the problem differently, what this means is that some vendors may have had different backgrounds than competitors and can design Wi-Fi systems that know how to work under a variety of working conditions.  Friday was also asked another question - given that Mist is focusing so much on AI, does this mean that far fewer IT workers will become employed?  Bob's answer was diplomatic, but probably true - he said that no, we'll need the same number of workers in the near-term, and that AI Wi-Fi will simply allow the same number of IT workers to make better decisions.  Still, the question makes it clear - the audience is concerned about job loss as AI works its way into the IT industry.