For the last five years, I have spent considerable time turning my mind to the issues of good and bad coexistence between Wi-Fi and competing technologies, such as LTE‑U, LAA, and NR‑U, in unlicensed spectrum. I have undertaken my thinking in the context of discussions and debates about coexistence in multiple forums, including IEEE 802, 3GPP, ETSI BRAN, and the Wi-Fi Alliance. I even ended up as the Chair of the IEEE 802.11 Coexistence Standing Committee.
Recently (June 2022), I was asked to summarize the state of play in an IEEE 802 Tech Talk and my conclusions about what all this discussion means in practice. The slides of this talk are available, as is a recording. This blog is a written version of the IEEE 802 Tech Talk, better suited to those who don’t want to listen to my dulcet tones for 45 minutes. Even in reading this blog, you have a choice; you can stop at the Executive Summary, or you can wade through eleven pages of detailed discussion… enjoy!
The short version: The Executive Summary
In the not-so-distant past, communications services were expensive, complicated & monopolized by a small number of service providers. Fortunately, technology has now enabled a new world of communications that is highly competitive and very functional.
While this change has been driven by many factors, the emergence of unlicensed spectrum was a key enabler. Wi-Fi, as the most widely used communications access solution, has driven unlicensed spectrum to fulfill its potential and to do so in a way that makes very efficient use of spectrum, a scarce and valuable community resource. Wi-Fi’s spectrum efficiency is easily demonstrated by observing that current use patterns show that Wi-Fi networks carry about twenty times more data than cellular networks, despite having access to roughly the same amount of spectrum.
Wi-Fi has enabled anyone, anytime, anywhere to set up and operate a simple, cheap, and flexible wireless network in their home or office that just works! This outcome was partially achieved by Wi-Fi’s use of a sophisticated diversity of sharing mechanisms that helped avoid coexistence chaos in many challenging sharing environments.
Wi-Fi’s success in unlicensed spectrum (and the new unlicensed spectrum justified by that success) has triggered the cellular community to propose new cellular-based technologies for use in unlicensed spectrum, including LAA and NR‑U. The problem is that these systems only use a small subset of Wi-Fi’s diverse and sophisticated sharing mechanisms, with early deployment measurements showing poor coexistence with Wi-Fi systems (and probably between LAA/NR-U systems too). The likely poor coexistence of these cellular technologies with Wi-Fi would be a major economic (and social) problem, given Wi-Fi’s importance to consumers and businesses and its multi-trillion-dollar global economic value … every year!
There may be nothing to worry about if LAA and NR‑U are not widely deployed, which is a distinct possibility. If they are widely deployed, significant effort will be required to mitigate the potential adverse impact on Wi-Fi’s efficient and shared use of unlicensed spectrum today. Interestingly, one mechanism to avoid coexistence issues between Wi-Fi and LAA/NR‑U is to allocate 1200 MHz for unlicensed spectrum so that the different technologies can more easily avoid operating in the same channel. A better overall solution to maximize consumer benefit is to focus on using Wi-Fi in unlicensed spectrum & 3GPP’s new NR-based technologies in licensed spectrum. This will optimize the use of both licensed and unlicensed spectrum, avoiding coexistence issues for free!
As regulators search for the most efficient use of spectrum resources, Wi-Fi’s superior sharing characteristics, widespread adoption, and ability to flexibly meet users’ needs all stand out as part of a compelling case to allocate more unlicensed spectrum, particularly in the 6 GHz band.
There are a key few messages that I would like you to take away from this blog. They are that:
- User needs: Whatever wireless networking solutions are deployed in licensed or unlicensed spectrum, they must meet users’ needs while making efficient use of spectrum
- Licensed spectrum: Unlicensed spectrum (usually using Wi-Fi technology) is the most efficient way of meeting user’s needs in most environments (particularly home and enterprise), but licensed spectrum (using cellular technology) is also necessary to meet user’s needs in some environments (particularly in the wider area and at higher physical speeds)
- 6 GHz band is needed: With the introduction of broadband connections that provide more than 10 Gb/s, the full 1200 MHz of the 6 GHz band is urgently required so that Wi-Fi can continue to distribute all of the data delivered to end-users and enable new applications including Virtual Reality and automation.
- Focus on Wi-Fi: While any technology is allowed to use unlicensed spectrum, a focus on Wi-Fi is desirable to avoid coexistence issues that will arise from the use of unsophisticated sharing mechanisms by technologies like LAA and NR-U.
If buy into my conclusions then I suggest you stop reading now to save 10 minutes of your life. If you disagree with my conclusions then I would ask you to dive into the detail articulated in the rest of this blog. If you still disagree please feel free to contact me to start a discussion.
The long version: Focus on Wi-Fi in expanded unlicensed spectrum in 6 GHz band!
Communications used to be expensive, complicated & monopolized
One does not need to go very far back in history to a time when most means of communication across any distance were expensive, complicated, and relied on services typically provided by monopolistic service providers (often government-owned).
I recall phoning my grandparents in South Africa from Australia each Christmas when I was a child (OK, it was half a century ago), where we had to book days in advance for a 3-minute call, and then spent most of the call talking about how expensive it was. Even a few years ago, many telecommunications customers were charged vast sums by the minute (or byte), in a manner that had very little to do with actual fixed and variable costs.
Communications is now highly competitive and functional
The telecommunications world has changed, and for the better in so many dimensions. Competition and technology have helped drive the marginal cost of voice and data communications (including POTS-like, e-mail, and Webex-like services) across the globe to almost zero. For many people, the costs are zero, offset against revenues to providers from advertising or other similar business models.
The complexity of global communications has also significantly decreased, although I suspect my parents and mother-in-law might disagree. The reality is that the services themselves are less complex, but the choices they provide have massively expanded, sometimes making everything seem more complicated. They are certainly more functional.
While traditional service providers still dominate the provision of facilities-based cellular services, they face far more competition in the provision of broadband services to consumers and enterprises. Of course, the desire for monopolistic power never goes away, with a new generation of monopolistic wan-a-bees operating higher up the value chain.
Unlicensed spectrum is a key driver of the modern communications
There are many reasons for this new world of communications, depending on what part of the ecosystem one focuses on. Certainly, the silicon-based technology revolution that really took off in the 1980s has had a major impact on the ecosystem.
However, with my Wi-Fi bias, I am going to claim that something that started in 1987 was just as important. 1987 was the year that the FCC in the US started an experiment that opened the 2.4 GHz junk band for unlicensed use. Many older readers may recall the days when setting up home or business networks involved running many meters of cable between routers, desktop computers, and other connected devices. This new spectrum, along with Wi-Fi technology, enabled anyone, anytime, anywhere to set up and operate a simple, cheap, and flexible wireless network in their home or office that just worked, using an amazing and still growing variety of devices & applications. The FCC experiment was so successful that it has now spread globally, also expanding into parts of the 5 GHz band and (hopefully) all the 6 GHz band.
Wi-Fi is the most successful technology in unlicensed spectrum
The success of Wi-Fi can be measured in various ways. A year or so ago, the Wi-Fi Alliance commissioned serious economists to estimate the economic value of Wi-Fi in specific countries and globally. They determined the economic value of Wi-Fi was 3.3 trillion dollars per annum in 2021, likely to grow to 4.9 trillion dollars per annum in 2025.
Personally, as a measure of Wi-Fi’s success, I have always liked the surveys undertaken by the Wi-Fi Alliance when I was the Chairman of the Board of Directors more than ten years ago. At that time, it was discovered people would rather give up coffee or beer than their Wi-Fi. Coffee and beer are very high bars. However, this expressed love for Wi-Fi makes perfect sense when one considers that about 20 times more data traffic today goes across Wi-Fi networks compared to traditional cellular-based networks, which themselves are a vital component of the modern communications revolution. This bias towards Wi-Fi occurs despite Wi-Fi having access to about the same amount of low and mid-band spectrum as the various cellular technologies.
Avoiding coexistence chaos
Wi-Fi has sidestepped coexistence chaos in unlicensed spectrum by avoiding other technologies, diverse Wi-Fi on Wi-Fi sharing mechanisms and through self-interest.
Wi-Fi’s tremendous success highlights a fascinating question. Why does the use of multiple technologies (including Wi-Fi) controlled by a multitude of unmanaged users operating in the same spectrum not result in coexistence chaos? There are three main answers from a historical perspective:
- Different channels: Wi-Fi generally doesn’t compete with other technologies in the same channels at the same time; technologies like Bluetooth in 2.4 GHz usually avoid Wi-Fi by operating between & beyond the usual Wi-Fi channels
- Diverse sharing mechanisms: Wi-Fi includes a diversity of sharing mechanisms (e.g., Listen Before Talk, exponential back-off, Preamble Detection, NAV, RTS/CTS, etc) that are specifically designed to enable good coexistence between independently managed Wi-Fi networks
- Self-interest: The IEEE 802.11 Working Group, the Wi-Fi Alliance and the Wi-Fi industry more generally have a vested interest in managing coexistence with the deployed Wi-Fi base as new Wi-Fi generations are specified.
Wi-Fi uses a diversity of sharing mechanisms to promote coexistence
Wi-Fi technology was built from the very beginning to handle environments where multiple devices with different owners had to operate in the same channel. There is no perfect solution to handle this very difficult problem. Wi-Fi’s technical solution is to use a diversity of sharing mechanisms (as noted above), built on top of the basic Listen Before Talk (LBT) mechanism.
LBT avoids transmissions in a channel if someone else is already using the same channel nearby. This is a concept that we are all used to when talking around the dinner table; don’t talk if someone is already talking (I wish my family would follow this protocol!). It is not a perfect mechanism because a potential transmitter can only hear the state of the channel at the transmitter and not where the transmission may cause interference. Using the dinner table analogy again, the people at the two ends of a long table might start talking because they can’t hear anyone else talking but the person in the middle ends up hearing both of them, and so neither of them properly (this is often caused the hidden station problem).
The power of LBT, as practiced by Wi-Fi, is that does not rely on just a single listening mechanism. A Wi-Fi device defers to other Wi-Fi devices at a very low energy level when it is sure the energy comes from another Wi-Fi device, by detecting a special Wi-Fi packet preamble. This is known as Preamble Detection and Wi-Fi typically uses a preamble energy detection threshold of -82 dBm. A Wi-Fi device also defers when it hears energy without hearing a preamble. This is a backup listening mechanism that helps avoid interfering with other devices (using any technology, including Wi-Fi), but also recognizes that any transmission made by the Wi-Fi device in such an environment is unlikely to be received properly anyway.
These two listening mechanisms are used to drive a sharing mechanism based on a very old and well-proven technique called CSMA/CA (Carrier Sense Multiple Access with Collision Avoidance). The Wi-Fi version is called EDCA. The idea is that if another device is transmitting (or another person is talking at the dinner table) when your device wants to transmit then your device defers for a random period (that increases exponentially with each unsuccessful transmission). Again, this is something we are all used to doing when attempting to insert ourselves into a conversation around the dinner table (my family uses a different technique … they just shout louder).
But there is more. Wi-Fi also allows reserving the channel for limited periods using a special field that is included in Wi-Fi packets (the special field is called NAV – or Network Allocation Vector – and is sent as part of the MAC packet, for those technically inclined). This avoids the need to rely solely on listening, which, as noted earlier, is imperfect. For example, the NAV reservation technique enables a special RTS/CTS (Request to Send/Clear to Send) mechanism, which helps mitigate the classic hidden station problem. In our dining room analogy, NAV with RTS/CTS allows the two people at each end of our long table to coordinate when they transmit, even though they can’t hear each other so that the person in the middle can successfully receive from both ends of the table.
Cellular carriers are now using Wi-Fi in unlicensed spectrum
The cellular community has decided that additional unlicensed spectrum for Wi-Fi unlicensed spectrum couldn’t be ignored and left for Wi-Fi to dominate.
Unlicensed spectrum is technically available for use by any technology, as long as it follows basic rules set by regulators in various countries. Some regulators apply quite strict and detailed rules, whereas others are more laisse faire. However, unlicensed spectrum (especially the 5 GHz & 6 GHz bands) is often called “Wi-Fi spectrum” (even by many in the cellular community) because up until now Wi-Fi has been the main, and often the only, successful user of this unlicensed spectrum.
However, many cellular stakeholders have now decided that access to unlicensed 5 GHz & 6 GHz bands is an opportunity too valuable (or too cheap) to leave for Wi-Fi. After some false starts, many cellular carriers are now using Wi-Fi in unlicensed spectrum to offload traffic. They are often using Wi-Fi offload to provide services in indoor locations typically inaccessible to cellular services or just to avoid deploying very expensive cellular technologies.
Some cellular stakeholders are now discussing using variations of licensed technology in unlicensed spectrum as an alternative, despite Wi-Fi mostly meeting their needs already. One area where Wi-Fi might not meet the needs of some cellular stakeholders is that they do not dominate the ownership of the required IPR in the way they do for cellular technologies. In recent times, four new cellular-related technologies have been proposed to use unlicensed spectrum:
- LTE‑U (LTE – Unlicensed) from the LTE‑U Forum
- MulteFire from the MulteFire Alliance
- LAA (License Assisted Access) & NR‑U (New Radio – Unlicensed) from 3GPP
The proponents of all four technologies have asserted wonderous benefits compared to Wi-Fi, related to throughput, latency, and reliability, as well as integration with licensed technologies. The truths of these claims are for discussion at another time, although I will note that many of the claims have proven to be more hype rather than reality.
The first two technologies are no longer being promoted by the cellular industry because they took too long to be specified and introduced to the market. In the case of LTE‑U, part of the delay was driven by a controversy that arose because LTE‑U was based on a very strange way of sharing with other technologies. Rather than using a distributed sharing mechanism like Wi-Fi where everyone is roughly equal, it used a selfish sharing mechanism whereby the LTE‑U system decided what access it wanted and only then gave the remainder for use by other technologies.
3GPP-defined LAA & NR‑U to share unlicensed spectrum using a subset of Wi-Fi’s sharing mechanisms
In more recent years, 3GPP has approved two new specifications for use in unlicensed spectrum; LAA and NR‑U. LAA is a complement to systems operating in licensed spectrum, focusing on using unlicensed spectrum to provide supplementary downlink capacity. NR‑U can operate independently of systems operating in licensed spectrum, just like Wi-Fi. In addition, NR‑U seems to address very similar use cases to Wi-Fi. The main difference between NR‑U and Wi-Fi is that the former is based on technology derived from traditional 3GPP-defined technology, which tends to be more complex and more expensive.
LAA and NR‑U resemble Wi-Fi in another respect, which is very important when discussing the coexistence between the different systems. After much debate in 3GPP, strongly influenced by access rules defined by ETSI BRAN in Europe, LAA and NR‑U were specified to use similar sharing mechanisms as Wi-Fi. All three systems use an EDCA-based version of LBT with exponential back-off and the same timing parameters. However, LAA and NR‑U only listen for other systems using energy detection (not preamble detection), albeit at a lower threshold of -72 dBm, and neither LAA nor NR‑U use any of the NAV-based sharing mechanisms that have proved so valuable to Wi-Fi over the last 20+ years.
In fairness, not using Wi-Fi’s preamble detection or NAV mechanisms makes good sense for LAA and NR‑U, because the adoption of these mechanisms would require fundamental changes to the traditional cellular technologies leveraged in the specification of LAA and NR‑U. Indeed, their adoption would essentially require LAA and NR‑U to become Wi-Fi, which would defeat one purpose of the cellular community defining an alternative to Wi-Fi.
Using only a subset of sharing mechanisms results in poor coexistence outcomes
Early measurements of deployed LAA systems show that using only a subset of sharing mechanisms results in poor coexistence outcomes in real environments.
Extensive simulations over many years in 3GPP and other forums suggested that LAA and NR‑U systems using just a subset of Wi-Fi’s diverse sharing mechanisms would coexist fairly with Wi-Fi systems, at least in the average case. The big question is does this claim hold up in real deployments in unlicensed spectrum, which often are not very average? Unfortunately, there have not been many measurements of real commercial deployments because LAA, while deployed by a limited number of operators globally, is not used extensively. In the case of NR‑U, there are no known commercial deployments to measure as of mid-2022.
The measurements that have occurred so far suggest that LAA use of just a subset of Wi-Fi’s sharing mechanisms results in poor coexistence outcomes. This conclusion is based on work in 2020 and 2021 conducted by researchers at the University of Chicago. Their results were summarized in two presentations in 2021 to the IEEE 802.11 Coexistence Standing Committee (which I chair):
- LAA/Wi-Fi Coexistence Issues: Wi-Fi Client Association and Data Transmission by Vanlin Sathya, Muhammed Iqbal Rochman & Monisha Ghosh, 13 Jan 2021 (see 11-20-1973)
- LAA/Wi-Fi Coexistence Experiments: preliminary results by Monisha Ghosh & Muhammed Iqbal Rochman, 15 Nov 2021 (see 11-21-1858)
These measurement-based studies suggest a variety of potential underlying problems that are not obvious in average simulated environments but are rudely exposed in real-world environments. It appears these problems cause LAA (and probably NR‑U too) systems to adversely (and somewhat unreasonably) impact Wi-Fi operations:
- Whereas Wi-Fi systems often defer at energy levels of -82 dBm (using preamble detection), LAA systems only defer at energy levels above -72 dBm
- Whereas Wi-Fi systems have hidden station mitigation mechanisms based on the use of NAV with RTS/CTS mechanisms, LAA systems do not
The University of Chicago researchers must confirm their results at their new home at the University of Notre Dame, and independent research is also needed. However, in my opinion, the results are likely to be broadly correct because they can be easily explained.
Preparing for poor coexistence issues
Poor coexistence between LAA/NR‑U & Wi-Fi is only an issue if LAA/NR‑U are deployed at scale but industry needs to be ready for this possibility.
The poor coexistence between LAA & Wi-Fi highlighted by the University of Chicago’s studies is interesting from an academic perspective. However, they are only important from a practical perspective if LAA and NR‑U are deployed widely.
The good news so far is that the deployment trajectory of LAA seems to have stalled globally. Data from GSA shows that the number of operators globally deploying LAA has remained constant and low (only nine) for the last 2.5 years. It seems likely that even these nine operators have not deployed LAA very widely.
There are a variety of reasons that might drive the limited long-term deployment of LAA by operators, including:
- Operators determine LAA is not compelling enough (or cheap enough) compared to other technologies, particularly Wi-Fi
- Operators are waiting for NR‑U, a more flexible technology that does not require complementary access to licensed spectrum
- Operators understand the coexistence issues between LAA and Wi-Fi and do not want to adversely impact Wi-Fi, which is widely used by operators (and their customers)
- Operators are hoping to persuade regulators to specify 700 MHz in the upper 6 GHz band as licensed spectrum, thus avoiding coexistence issues with Wi-Fi, or any need to use LAA or NR‑U technology.
We don’t know which of these possibilities will represent reality. The bottom line is that there are currently many uncertainties and unknowns. There is still a real possibility that the wide deployment of LAA or NR‑U will cause serious damage to the multi-trillion-dollar economic and social impact of Wi-Fi on society. In this context, it is vital to continue work to understand the impact of LAA and NR‑U on Wi-Fi operations and begin to develop mitigations. I strongly encourage anyone interested in these topics to participate in the IEEE 802.11 Coexistence Standing Committee (contact me for details on how to do this).
Using Wi-Fi in unlicensed spectrum & NR in licensed spectrum provides the best overall solution while avoiding coexistence problems
There is a better solution to avoid coexistence issues between LAA/NR‑U and Wi-Fi in unlicensed spectrum. This better solution is based on an understanding and recognition that:
- Wi-Fi is a technology that already meets the needs of billions of people and will continue to do so as it is refined by the IEEE 802.11 Working Group and the Wi-Fi Alliance in the future
- The IEEE 802.11 Working Group is developing the next generation of Wi-Fi as IEEE 802.11be, which will probably be certified by the Wi-Fi Alliance as Wi-Fi 7 as early as 2024
- The IEEE 802.11 Working Group is also starting to discuss the following generation, which is likely to be certified by the Wi-Fi Alliance as Wi-Fi 8
- Wi-Fi’s multi-level and diverse sharing mechanisms have proved, over 20+ years, as more than sufficient to provide good (albeit not perfect) coexistence between independent systems in unlicensed spectrum.
- In contrast, it is no surprise that the one-dimensional energy detection focused sharing mechanism used by LAA and by NR‑U is inadequate.
The better solution is to focus on Wi-Fi as the main WLAN technology in unlicensed spectrum because it has been proven to provide an excellent balance between performance, costs, and coexistence, allowing anyone, anywhere, any place to set up a network that just works! Of course, other technologies (such as Bluetooth, UWB, or WiSun) can be used to support niche applications (noting niche does not mean small) where it can be shown there is good coexistence. The use of Bluetooth in the 2.4 GHz band today is a perfect example of a successful and sensible niche use.
This then leaves licensed spectrum for use mainly by the cellular technologies being developed by 3GPP as 5G/6G and enabled by ITU-T as IMT. 5G/6G will mainly be used in wide areas, high power use, and outdoor use cases for which they are well suited, Wi-Fi will mainly be used for local areas, low power, and indoor use cases for which it is so well suited. There will, of course, be some overlap between these use cases. Rather than worrying about coexistence at the PHY level, this concept just requires everyone to focus on integrating 5G/6G solutions with Wi-Fi further up the network stack, with the ultimate goal of providing the same high performance to consumers no matter where they are or what they are doing.
There may also be a case for locally licensed spectrum that can be used by either cellular technologies (locally licensed spectrum will be just like licensed spectrum for these technologies) or by Wi-Fi (locally licensed spectrum will be just like a well-managed unlicensed spectrum from the perspective of Wi-Fi).
There is a significant need for additional unlicensed spectrum to support future demand for data traffic in a cost-effective and efficient manner
A coexistence solution based on an acceptance that Wi-Fi is the primary technology in unlicensed spectrum leads us straight into another contemporary debate relating to the allocation of the 6 GHz band for unlicensed use (mainly by Wi-Fi) or licensed use (mainly by 5G/6G licensed technologies under the IMT banner). In most parts of the world, it has already been agreed that the lower 500 MHz of the 6 GHz band will be allocated for unlicensed use. This response to the reality that Wi-Fi traffic is doubling every 3 years, and the existing 2.4 GHz and 5 GHz bands are quickly becoming congested. As new data-intensive applications like AR/VR become more widespread, the demands on the existing unlicensed spectrum could become unsustainable. Even with Wi-Fi’s flexibility and robust coexistence capabilities, new spectrum bands are necessary to address this increasing demand.
The main current debate is about the upper 700 MHz of the 6 GHz band (which extends into the 7 GHz band). In some countries, the upper 6 GHz band has already been allocated for unlicensed use (e.g., the US, Brazil, and Saudi Arabia). In other countries (e.g., across Europe), there is an ongoing debate that will probably not be resolved until the ITU-R WRC-2023 conference at the end of 2023.
The debate about the allocation of the upper 6 GHz band is multi-dimensional, with lots of handwaving and self-serving arguments, often based on unsupported and very enthusiastic assumptions. Both sides of the debate assert that more spectrum is needed to support growing user needs for more data. At least on that point, there is agreement:
- Those arguing for the upper 6 GHz band to be licensed for use by 5G/6G (or IMT) assert that only operators using licensed spectrum can provide the services needed by users (presumably for an oligopoly-defined fee). Interestingly, in making this argument they seem to be abandoning the idea that their own LAA/NR‑U technology can service users in unlicensed spectrum.
- Those arguing for the upper 6 GHz band to be unlicensed, mainly for use by Wi-Fi, assert that doing so will allow for anyone, anywhere, any place to set up a network that just works, inherently meeting users’ needs (at a cost-driven towards zero by competition and choice).
Rather than relying on the handwaving by both sides to resolve this debate, it might be better to rely on some facts. In 2022, using Europe as an example, on the basis that a resolution of this debate in the context of Europe will have a significant impact globally:
- Technologies, like 4G/5G/6G, that use licensed spectrum have access to just more than 1000 MHz of low band and mid-band spectrum …
- … and technologies, like Wi-Fi, that use unlicensed spectrum have access to about 1000 MHz of low band and mid-band spectrum (including the lower 6 GHz band but excluding the upper 6 GHz band)
- … and yet (using 2021 data from the German Ministry for Economic Affairs & Climate) the vast bulk of data today is transmitted over Wi-Fi
- Mobile networks in Germany in 2021: 4 Exa Bytes (5%)
- Fixed networks (mostly Wi-Fi) in Germany in 2021: 76 Exa Bytes (95%)
- … this is likely to grow in the future, supported by fixed broadband backhaul links to homes and businesses increasingly supporting rates of more than 10 Gb/s.
These facts clearly demonstrate that unlicensed spectrum, using technologies like Wi-Fi, satisfies the needs of most users today, in a manner that is significantly more efficient (up to 20 times more efficient) and flexible than licensed spectrum. It typically does so at a much lower cost to the user. It is not difficult to also conclude that the benefit of providing another 700 MHz for unlicensed use has a significantly greater “bang for the buck” than allocating it to 5G/6G operators for licensed use.
It is possible none of this rational and evidence-based argument matters very much to those in the cellular community who are advocating for the upper 6 GHz band (or even the whole 6 GHz band) to be licensed. Maybe they recognize unlicensed spectrum really is the better and more flexible way to satisfy users’ needs and Wi-Fi will be even more able to support users’ needs by making all of the 6 GHz bands unlicensed. However, maybe their real goal has nothing to do with satisfying users’ needs but instead is all about blocking Wi-Fi in whatever way they can in an effort to protect their cellular technology/licensed spectrum-based oligopolies.
Allocating the entire 6 GHz band for unlicensed use has one final benefit, taking us all the way back to the question of coexistence. More unlicensed spectrum in the 6 GHz band makes it easier for LAA/NR‑U to operate in different channels from Wi-Fi, thus avoiding poor coexistence outcomes between the different technologies. This is like how we avoid poor coexistence outcomes between Wi-Fi and Bluetooth in the 2.4 GHz band. Alternatively, we could use just Wi-Fi in 6 GHz unlicensed bands, focusing on NR technology in the licensed bands …
A Conclusion: Allocate the entire 6 GHz for unlicensed use, mainly by Wi-Fi to avoid coexistence issues!
Congratulations on getting this far. I put my conclusions up front in the Executive Summary, but I have been told I need to put my conclusions at the end too! So here they are again …
- Whatever wireless networking solutions are deployed in licensed or unlicensed spectrum, they must meet users’ needs while making efficient use of spectrum
- Unlicensed spectrum (usually using Wi-Fi technology) is the most efficient way of meeting user’s needs in most environments (particularly home and enterprise), but licensed spectrum (using cellular technology) is also necessary to meet user’s needs in some environments (particularly in the wider area and at higher speeds)
- With the introduction of broadband connections that provide more than 10 Gb/s, the full 1200 MHz of the 6 GHz band is urgently required so that Wi-Fi can continue to distribute all of the data delivered to end-users and enable new applications, including Virtual Reality and automation.
- While any technology is allowed to use unlicensed spectrum, a focus on Wi-Fi is desirable to avoid coexistence issues that will arise from the use of unsophisticated sharing mechanisms by technologies like LAA and NR-U.