I recently attended the Global Standards Leadership Conference organized by Justus Baron (Northwestern University), Kirti Gupta (Cornerstone Research, previously at Qualcomm), and Tim Pohlmann (LexisNexis® IPlytics) which took place on June 13, 2024, at the University of California, Berkeley, Haas School of Business. It was an interesting and well-attended conference with a number of different panels and a keynote speaker from UC-Berkeley. Here are some of the more noteworthy tidbits from the opening remarks and first panel of the day.
Opening Remarks
Tim Pohlmann was the introductory speaker. For those that do not know him, Tim essentially has a PhD in the relationship of patenting and standards (at least that is what his dissertation was about). He founded IPlytics (which was subsequently acquired by LexisNexis) to bring some transparency to the standards world by creating a platform to do data analytics at the intersection of patents and standards.
Because he is a data geek, he gave us a bunch of numbers to demonstrate examples of both over-declaration and under-declaration. Here’s some of the data he provided:
Over-Declaration:
According to several court ordered declaration reviews, on average only 20-30% of declared essential patents are, in fact, essential.
More recent expert studies have put that figure at only 10-15% of declared patents for 5G.
Under-Declaration:
Most of the companies that have submitted declarations for VVC [Versatile Video Coding, which is also known as H.266 and MPEG-I part 3 – I will try to use brackets for my comments about what was said] have submitted declarations that do not specify what the submitting company believes to comprise its essential patents.
Similarly, for Wi-Fi [IEEE 802.11], only about 8% of the submitted declarations for WiFi 4, 5 and 6 specify the patents to which the declaration relates. Almost all of the declarations are blanket declarations.
This means that, at least for some standards, there are more unknown patents than known patents and that makes every company’s share of the overall SEPs unknown.
First Panel of the Day – The Future of Wireless Technology
The first panel of the day was titled “The future of wireless technology.” The moderator was Kirti Gupta. The participants were individuals who have participated in the development of standards (primarily cellular standards): Raj Yavatkar, CTO at Juniper Networks; Irfan Ali, CTO Team, Mobility and IoT at Cisco Systems; Lorenzo Casaccia, VP Technical Standards and IP at Qualcomm Technologies, and Mang Zhu, Vice President and Head of IP at Interdigital (formerly at Motorola Mobility).
As one would expect from the list of participants, the panel focused on communications standards and technology – primarily 4G/5G/6G and WiFi. For purposes of comprehensibility, I have summarized and grouped the discussion according to the themes that emerged, rather than in the exact order in which they were made. Also, for shorthand, I’ll refer to the participants by their affiliation rather than by their name.[1]
Is 5G a Failure?
I have previously discussed the fact that the 5G rollout appears to have stalled, Stagnation and Innovation (sepessentials.com). This is of course a hot topic among companies that make and sell 5G products and the panel was no different. There was a lot of discussion about whether 5G is a failure and if so, why, and also a discussion about the ways 5G is different from 4G and whether those differences matter to anyone (e.g. to end users).
Qualcomm took the position that 4G was a “watershed” moment. On the other hand, 5G is about enabling new applications – many of which have not materialized. Nonetheless, Qualcomm was not ready to say that 5G was a failure. Instead, he pointed to China and how successful China has been in rolling out B2X (business to everything, including machine to machine, person to machine, machine to person) using 5G.
Juniper jumped in to say that China is not a good example because 5G is supported by the government in China, and China is not a market economy. 5G has not been successful elsewhere because it is very expensive to roll out and end users do not notice a difference between 4G and 5G service. The promises of 5G have not materialized for most end users. That means service providers cannot charge extra money for 5G services compared to 4G, making it uneconomical for them to roll 5G out.
He also said that it is really unclear whether 5G is enabling anything that cannot be enabled by 4G. He reminded the group that WiFi is getting better and better. Many of the new uses and new applications that 5G is intended to address can be done less expensively using WiFi, including machine to machine communications.
Cisco agreed that most consumers do not presently notice much of a difference between 4G and 5G, perhaps because, at least in the United States, there are so few pure 5G core network deployments. He also agreed that 5G use cases are taking more time than expected to materialize. But, he said that there have been some inroads in the deployment of private 5G enterprise networks and he expects the 5G use cases will eventually materialize. Consequently, it is too early to call 5G a failure. He also thought that the 5G rollout may prove useful to support service providers' upgrade to 6G when 6G is released.
Interdigital ducked the question of whether 5G is a failure, but did have a lot to say about how 5G is different from 4G.[2] According to her, 5G innovations result in higher data rates, lower latency and better reliability compared to 4G. According to her, this is mostly a result of eMBB [Enhanced Mobile Broadband] leading to wider and more flexible and adaptable bandwidth and improved MIMO [Multiple Input Multiple Output] with better beamforming. She said 5G is better for streaming video, for example [although I personally have not noticed this]. She did concede that mMTC [Massive Machine-Type Communications] has not really been used for much of anything yet.
What was fascinating about the discussion was how it revealed the complex interconnection between the end user experience and the willingness of service providers to adopt new technology. For cellular standards, service providers must see a clear benefit (e.g. they can make more money) in order to rollout a new cellular standard. Rollouts are expensive and, in order for service providers to use a new standard, the financial benefits must outweigh the very high price of rolling out a new network. That clear benefit is, for the most part, dependent on the experiences of the end users in using that new technology.
Although none of the participants said as much, what I took away from this part of the discussion was that the 5G developers spent too much time focusing on potential new uses and updates to 4G to enable those new uses and spent too little time focusing on improving the end user experience. That has resulted in a standard that has a lot of technology that most current users do not want or need, or do not want or need enough to justify the cost of transitioning away from the existing 4G and WiFi standards.
Will IoT adopt 5G?
Juniper thinks not. Juniper noted that, other than handset makers, most smart device makers do not care what communications technology is being used as long as it works well enough. Most IoT device makers are agnostic as to the technology – they just want seamless connectivity, and do not need a whole lot of bells and whistles [ok he did not say bells and whistles – that is my interpretation of what he said]. He did not think that most IoT device makers will voluntarily adopt 5G.
Qualcomm noted that WiFi sells to many different customers as compared to cellular technology which sells to just a few.[3] In other words, many more different device makers use WiFi in their products than use cellular.
How has the composition of standard developers changed?
Qualcomm had an interesting comment in response to the above question from the moderator. Qualcomm described the process of standards development as a mixed process of competition and collaboration. Companies participate when they will benefit from standard adoption even though they are competitors. Companies collaborate on the development but also compete on what technology to use in a standard. Qualcomm also noted that there are a lot more people and companies participating today than in the old days.
Cisco agreed and said that the service providers and a few major companies had been the primary drivers of the development of 4G. For 5G, there was a shift to a broader base of participation in the development process.
Interdigital noted that there had been a few significant changes in the regional make-up of the participants between 3G and 5G. She noted that the participants of the 3G development were primarily European and US companies. 4G also involved a lot of Europe and US companies but also a number of Korean and Japanese ones. 5G also involved a lot of Europe and US companies, some Korean ones, a large Chinese contingent but very few Japanese companies.
Interdigital also said that the 3G and 4G development efforts had been dominated by the service providers as well as Apple, Qualcomm and a few similar large companies. The development of 5G involved more wide-spread participation, including the car companies. She said that the car companies started participating in cellular standards development because the car companies had to pay for cellular licenses.
After listening to the participants’ remarks about 5G, the obvious question is, did the development of 5G suffer from having too many cooks in the kitchen? Was the very large number of participants, the attempt to include technology related to a multitude of possible new uses and the lack of clear service provider direction what lead to the The Bloatware That Is 5G (sepessentials.com).
What will 6G bring?
The panelists had a variety of different ideas about what 6G will bring. There was a lot of discussion about radio bands and release of spectrum and what new technologies will be needed depending on band. Some people thought that 28GHz was technically unfeasible; others that it is possible and would allow for much better performance. A number of the panelists discussed the need for seamless transition between networks (e.g. between WiFi and cellular) and improved interoperability.
Juniper suggested the need to transition cellular from a pure hardware play (which is very expensive to deploy) to a partially software-based technology. This would allow for less expensive and easier transitions between versions and generations.
Cisco said this was easier said than done – dedicated hardware can have issues, but virtualization also brings its own issues and does not always result in the same performance as hardware solutions, at least at first. Cisco thought that instead there should be more focus on improving the transport mechanisms, not just the radios, in order for 6G to be successful. Without better transport, at this stage of cellular development, radio improvements are just not that noticeable to most end users.
Qualcomm talked about the need for standardization in satellite communications. He believes that 6G might be used for that.
The panelists even disagreed about when 6G would be introduced. Qualcomm anticipates that 6G will be deployed in 2030. Juniper is less sanguine – he believes economic factors will delay 6G (similar to what has happened for 5G) for at least 5 additional years.
I concluded from this part of the discussion that there is not yet a consensus about what improvements to include, or even to focus on, for the development of 6G. One hopes that such consensus emerges soon if there is to be any chance of meeting the expected role out date.
[1] I note that several of the panelists said that their views did not necessarily reflect their employer’s views.
[2] Possibly shoring up arguments to be made later about the novelty of the 5G technology?
[3] Qualcomm is almost the only company that designs both cellular and WiFi chips and so (one assumes) knows the customer base for chips implementing each technology better than just about any other company.