​Mediatek (2454.TT) the 2nd largest company in Taiwan (market cap) and the largest producer of 4G and 5G chipsets (units) is preparing to list its Bluetooth chip subsidiary Dafa/Airoha (pvt) and is making the applications for listing on the TSE Emerging Market board, which is a prerequisite for listing on the TSE.  The company is expected to apply for full listing on the TSE after the necessary 6 months on the Emerging Market board, with its performance setting the tone for pricing on the TSE.  The listing is targeted toward attracting engineers and workers in the semiconductor space to Dafa, as competition from other semiconductor companies in Taiwan for talent has become quite fierce.  As bonuses are roughly evenly divided across all divisions at Mediatek, regardless of performance, the idea of an independent public company takes ‘sharing’ out of the bonus mix and opens the potential for higher rewards for performance.
Dafa provides Bluetooth devices to Sony (SNE), Apple (AAPL), Xiaomi (1810.HK) and JBL (pvt[1]) among others and a 10% stake in the company was sold to a group of VCs for NT$650/share, which was the largest semiconductor deal in the Taiwan VC world over the last few years.  Given 145m shares, the valuation for that transaction would be ~$3.3b US.  The Emerging Market listing is expected in May or June.


[1] JBL is owned by Harman International, who is owned by Samsung.

5G is all the rage.  Faster service and more bandwidth are the calling cards of 5G advertisements, along with the coverage maps that tout coverage across much of the US, but as we have noted in the past, much of the 5G coverage that is offered to consumers is piggybacked on 4G infrastructure and is of the sub6 variety, the ugly stepsister of 5G.  The problem with 5G is that as the frequencies get higher, which improves speed and bandwidth, the distance a 5G signal can travel decreases, and less distance means more base stations to keep customers from losing the 5G signal.  As each base station is a cost to carriers, the need to tradeoff between speed and cost has kept much 5G capacity in the low to mid sub6 spectrum.
While 5G even at these frequencies is a step up from 4G (in most cases), it does not exploit the true nature of 5G as an improvement in technology and sets the bar on the low side of what 5G can really do.  Then there is mmWave, the princess of 5G, with all the speed and bandwidth that the technology can promise, but there is a catch, and it’s the same one that plagues 5G technology only more so.  MMWave 5G signals, which operate between 24.25 GHz and 43.5 GHz, are even more susceptible to the distance characteristics mentioned above and need to be retransmitted every few hundred meters, making the service far too expensive for a carrier mobile network.  That said, there are many applications that would benefit from mmWave and are situated where the necessity for multiple base stations would not be burdensome.
Stadiums are such a situation where mmWave 5G can be easy adapted to provide ultra-high speed capabilities that stadium goers can use to view camera angles that they might be far from or detail that might not be available on a large LED display.  Businesses however would be the true beneficiaries of mmWave 5G, as the ability of a factory where sensors are on a mmWave 5G network to process real-time information across a large production line, would be vastly improved both in speed of data collection and bandwidth available using mmWAve 5G and the concept of relatively limited transmission spread works toward keeping that information secure.  All in mmWave 5G, while not yet viable for a large mobile network, is perfect for specific instances where speed, low latency, and bandwidth are most important and coverage is not the key factor.
One problem however is that mmWave frequencies need to be assigned to carriers and many countries are still in the throes of figuring out how to deal with sub6 5G frequencies, so mmWave is still in its infancy across the globe, but some countries have recognized the value in allocating mmWave spectrum, allowing carriers to work through the mechanics of making viable use cases for the technology, some more than others.  We have put together a list of those countries that have assigned mmWave frequencies or are in test/trial mode.  While the list contains all countries that meet those criteria, we have arranged them in order of the level of assignment, meaning how much of the mmWave spectrum has been assigned.  Surprisingly South Korea is absent from the list as is the Netherlands, where politics have been more of a determining factor than the prospects for expanding 5G service.  The UAE is the only country that has assigned all mmWave capacity to date.

While the datasets for 5G are still relatively small, which leaves us without reliable historic data, as 5G matures, we would expect growth rates to fall into more seasonal patterns similar to the mobile phone market.  January 5G data seems to indicate that we are beginning to see such a pattern, albeit at much higher y/y growth levels, as typically slower smartphone growth months January and February begin to affect 5G device expansion.  January saw only 1.5% 5G m/m device growth, the slowest monthly growth since 5G deployments began, while 5G smartphone m/m growth was 3.1%, slower than January ’21 but still up 115.3% y/y, with most other 5G device types seeing little or no growth m/m.  We would expect February results to be similar, while we would expect March to see a return of m/m growth across many of the device categories.

​Few of us worry about cell service, with the US covered by 2G, 3G, 4G and now 5G coverage in all but the most remote locations, and smartphones are enough of a status symbol that unless you keep your vintage smartphone covered, it would be hard to keep from being publicly embarrassed when using an old smartphone.  But there are those that do, and that will be getting more difficult soon.  As of February 2022 AT&T (T) will shut down the remainder of its 3G network, T-Mobile (TMUS) will shut down Sprint’s 3GCDMA network on March 31, and its own 3G network on July 1, while Verizon (VZ) will end its 3G network on December 31 of this year.
The bulk of 3G network users would be older smartphones but there are also older tablets, watches, medical devices, SOS (“I’ve fallen and I can’t get up”) services, and security systems that still operate on 3G, so it is incumbent for folks to check the operation of such services to make sure they are not disconnected.  Here’s a list (certainly not complete) of phones that will not work by the end of this year (or earlier).  AT&T provides a way you can look up your phone’s model to see if it will still work on their network, while T-Mobile says it will contact all affected subscribers.  Verizon gives a number to call to check.

• HTC Desire 500.
• Samsung Galaxy Trend Lite.
• Archos 53 platinum.
• Samsung Galaxy Trend II.
• Samsung Galaxy S3 mini.
• Caterpillar Cat B15.
• Sony Xperia m.
• Wiko Sync Five.
• Wiko Darknight.
• Samsung Galaxy Xcover 2 phone.
• Huawei Ascend G740.
• ZTE Grand S Flex.
• Lenovo A820.
• Huawei Ascend Mate.
• ZTE V956.
• UMi X2.
• Huawei Ascend D2.
• Samsung Galaxy Core.
• Faea F1.
• THL W8.
• ZTE Grand X Quad v987.0
• ZTE Grand Memo.
• Samsung Galaxy Ace 2.
• LG Lucid 2.
• LG Optimus F7 phone.
• LG Optimus L3 II Dual.
• LG Optimus F5 phone.
• LG Optimus L5 II.
• LG Optimus L5 II Dual.
• LG Optimus L3 II.
• LG Optimus L7 II Dual.
• LG Optimus L7 II.
• LG Optimus F6 phone.
• LG Enact.
• LG Optimus L4 II Dual.
• LG Optimus F3 phone.
• LG Optimus L4 II.
• LG Optimus L2 II.

·        LG Optimus F3Q phone
·        Iphone 5 or older
 

Last week the Ninth Circuit Court of Appeals issued a ruling in reference to a case from the US District Court of the Eastern District of California.  That decision, which denied the plaintiff’s (American Cable Association, CTIA, NCTA, The Broadcast Association & the US Telecom Association) request for a preliminary injunction and upheld the lower court’s decision, was a step forward in California’s battle to reinstate the California Internet Consumer protection and Net Neutrality Act of 2018 (SB-822), a state regulation that would be restored after the FCC, under the Trump administration, decided to move broadband internet services from operating under common-carrier regulations (Title II), to ‘information services’ (Title I), which terminated  federal net neutrality rules.
The plaintiffs had sought an injunction to keep California’s attorney General from enforcing the state’s Net Neutrality rules, which codified the original FCC Net Neutrality Rules, under the conclusion that the FCC lacked the authority to preempt state rules because they had changed the classification of broadband services from Title II (regulated) to Title I (unregulated) which also caused them to lose the authority to regulate the states.  This was sort of a catch 22, as the deregulation of broadband services made in 2018 by the FCC actually took away their ability to supersede state broadband regulations, which leaves the state’s to decide on a state-by-state basis, whether they want to reestablish net neutrality rules.  Under the ruling, if it stands, the only way the FCC could force the states to abide by its 2018 ruling ending net neutrality would be to reinstate broadband to Title II, which would also reinstate federally mandated net neutrality, hence the Catch-22.
What the ruling does however, is to open the idea of reinstating net neutrality to state governments, who will likely face challenges from carrier associations that promote the idea that services such as speed or price should be applied based on the customer’s content, equipment, location or other factors.  This puts an end to carriers being able to ‘throttle’ certain customers who are using high bandwidth levels but paying lower fees and giving that bandwidth to its highest paying customers.  The battle is certainly not over, but the pathway toward net neutrality is once again unblocked for now.  The next step would be the Supreme Court, who will also look at the concept of net neutrality from a legal standpoint rather than a practical one, as opposed to the way the average user looks at internet services.  The general public has indicated in polls that it favors net neutrality but that will have little influence on the Supreme Court’s decision, if it makes it that far, as their decision will be based on precedent, or at least their interpretation of precedent, which means that a Catch 22 like the one above could set the tone for the internet for years to come, rather than viewing it as a public service.  Justice can be blind…

​Huawei (pvt) has begun arbitration proceedings against the Kingdom of Sweden after the country targeted Huawei specifically by excluding them as a supplier of 5G network products and services and further requiring the company to remove all of the company’s installed equipment by January 1, 2025.  The arbitration will take place in the World Bank Group’s International Center for Settlement of Investment Decisions after an appeal in the Swedish court confirmed the decision made in June of 2021 by Sweden’s Post & telecom Authority.  With Sweden the 2nd country in Europe (the UK was 1st) to officially ban the company, relations between China and Sweden have been strained with Ericsson (ERIC) expecting some retaliation from the Chinese government, especially as the suit begins.  While no damages were specified in the suit, public media says the initial compensation sought was over $500m but could go considerably higher if Huawei can make the case that Sweden has violated its international responsibilities toward Huawei’s international business.

According to a report by the European Court of Auditors, an independent external auditor that checks that funding from EU taxpayers are collected, used correctly, and achieve ‘value for money’, who answers to the European Parliament and Council, the EU is and has been facing delays in the deployment of 5G networks, both from a lack of government sponsorship and security issues.  Based on a 2017 study the report shows that the introduction of 5G across four key industrial sectors (automotive, health, transport, and energy) could generate revenue as much as €113b/yr. ($127.5b US), while creating 2.3m jobs for EU members and a 2021 addendum report estimated that 5G could add up to €1t ($1.13t US) to the European GDP between 2021 and 2025.
That said, the report does indicate that such deployments carry risks, such as privacy issues, threats to national security, supply chain dependencies, and cyberattacks, as 5G is more dependent on software than previous communication systems.  Citing studies that have shown the potential impact of cyber crimes could be as high as €5t, or ~6% of global GDP, such fears pushed the EU toward the 2019 Network & Information System Cooperation Group’s initiatives to prevent ‘hostile state actors’ from ‘privileged access’, applying pressure to a vendor, or by invoking legal requirements.  With few hard (legally binding) and many soft (non-binding communication) laws, the EU built a framework for controlling cybersecurity.  Of course the ‘hostile state actors’ mentioned above refer to China and more specifically Huawei (pvt) and ZTE (000063.CH), both of whom were and remain the object of continuing US campaigns relating to the security of their telecommunication equipment.
The report also indicates that the cost of 5G deployment across all EU member states through 2025 is estimated to be between €281b and €391b, with the bulk of such investments being made by EU member nations themselves.  Between 2014 and 2020 5G development was supported with €4b, coming from both the EU budget and the European Investment Bank, which has provided loans of €2.5b for nine 5G projects in 5 member states., along with the Recovery & Resilience Facility, which will add another €724b in grants and loans for such future projects.  With all of this available capital one might expect the EU to be a leader in 5G rollouts however the report goes on to point out that under the 2016 EU 5G Action Plan member states were to have launched early 5G networks by the end of 2018, full commercial 5G services in at least one major city by the end of 2020, and uninterrupted 5G coverage in urban areas and main transport paths by 2025, adding (last year) that all populated areas should be covered by 2030.
By the end of 2020 23 of the 27 member states had achieved the goal of 5G access in at least one major city and by the end of 2021 that number had increased to 25 of the 27, however a recent study cited in the report indicated that only 11 member states are expected to meet the ‘uninterrupted coverage’ goal by 2025, and that 5G coverage (including both EU and non-EU countries) in Europe will reach only 35% in 2025, while the US is expected to reach 51% by that date, with no definition for quality of service.  The report indicates that a lack of clarity as to what ‘fully commercial’ services might mean, citing the difference between Luxembourg City, where 5G is available on only a few streets to Helsinki, where almost the whole territory is covered, is a stumbling block that continues to limit actual deployments, but on an overall basis the report goes further as to which members might be able to meet plan goals and who might not.

​Only 14 of the 27 member states have included some of the EU objectives indicated above in their 5G national strategies and only 3 have included all of the objectives, but more to the point, of the three 5G bands that have been selected for 5G (low – 700MHz.; Mid – 3.6GHz.; High – 26GHz.) member states were supposed to make the low band spectrum available to carriers by 6/30/2020, with the mid and high bands by 12/31/2020.  However by the end of 2020 only 13 member states had assigned the low band, 17 states had assigned the mid band, and 4 states had assigned the high band, totaling less than 40% of the available bandwidth.  By the end of October last year that had only increased to 53%. 
In some cases, particularly for the high band, carrier demand was weak, but conflicts with spectrum use across borders of non-EU countries are a problem, as some of those countries use the 5G low band for TV broadcast, and negotiations for the release of that spectrum vary considerably from country to country.  But the issues relating to the 2019 China security threat are also an impediment to 5G rollouts.  While the idea of a uniform and broad 5G security directive across the EU was the objective of the 2019 plan, as Chinese 5G vendors became a ‘national security threat’, each country had their own take on the problem and implementation, which limits any coordinated effort across EU member states, some of whom had already decided on key vendors when the problem became a focal point. 
The report goes further citing interviews with representatives of member states who considered the criteria for vendor selection ‘open to interpretation’, with 11 of member states national regulatory authorities voicing similar concerns, and a number of EU countries have included Huawei in their 5G equipment purchasing plans.  As of the end of 2020 40% of network equipment vendors in EU states were non-EU sanctioned vendors and 60% of core 5G equipment was from non-EU vendors, while 64% of EU state 4G customers were using Chinese communication equipment and the cost to ‘rip and replace’ Chinese equipment purchased since 2016 has been estimated to be €3b.
All in the report did not paint the coordinated picture of 5G deployment across EU member states that some might have hoped for, with only some of that weakness attributed to COBID-19.  The report suggests that by the end of this year a common definition of 5G quality of service (speed & latency) should be developed, and all member states should include the original and updated 2016 goal plans in their 5G national strategies, while continuing to negotiate with non-members as to the allocation of spectrum.  But the real problem still seems to be security, and as none of the 5G member state rules are legally binding there seems to be no way to coordinate a uniform security plan, especially one that can come up with real replacement cost estimates and a way for those with imbedded Chinese equipment to be compensated for its replacement. 
While we in the US are concerned over the 5G spectrum conflicting with aircraft altimeters, private enterprise controls much of 5G’s rollout across the US, and while each state and city might have rules that modify those plans a bit, it seems to be nothing like the difficulties facing EU countries and their EU and non-EU neighbors and while the report’s suggestions are lucid and convincing, implementing them is really the problem. 

On 11/19/21 we noted that both AT&T (T) and Verizon (VZ) agreed to defer the launch of 5G services on C-Band spectrum (3.7 to 3.98 GHz), a portion of the 5G spectrum that they acquired during the last auction, until early January, the 2nd time such a postponement had been made.  It seems that both companies have once again bowed to pressure from the FAA and are ‘temporarily deferring 5G rollout’ in locations around airports.  This 3rd postponement comes as the FAA and a number of airlines feel that 6G towers near runways could cause interference problems with altimeters used on aircraft and could cause reading to be incorrect or not work at all.  This differs from testing results done by the FCC, who approved the use and auction of the C-Band, which raised almost $70b.
While the US government has intervened (again) to smooth things over between the affected parties, the question remains as to why things have gotten to this point given the fact that for almost two years telecom carriers, airport administrators, airlines, and US regulators were unable to come up with a plan that would satisfy both safety concerns and the use of 5G C-Band.  Making it more embarrassing is the fact that the C-Band is being used in ~40 countries for 5G, with no reported altimeter problems, and the FAA, along with altimeter suppliers, have been working to determine which altimeters are reliable under 5G and which need to be retrofitted or replaced for months.
Carriers are not without some blame in this situation, having known that this was a potential issue when they were bidding on C-Band spectrum, and the FAA proposal for a 5G buffer zone (~1 mile) only affects ~50 airports across the US, where over 5,200 public and 14,700 private airports operate.  But there are other factors, such as the higher power 5G transmitters used in the US and the larger  buffer zones used in other countries (France has a 96 second buffer zone, while the FAA proposal is for a 20 second buffer) that are still being regaled by carriers.
According to statements made by the FAA on January 13 - 16, aircraft that are operating with untested altimeters ort those that are in need of retrofitting or replacement will be unable to perform low-visibility landings where 5G is deployed, although they have cleared ~45% of the US commercial fleet for low-visibility landings at ‘many’ airports.  In particular two altimeter models that were installed in a wide variety of  Boeing (BA) and Airbus (AIR.FP) aircraft have been approved, which opens runways on up to 48 of the 88 airports most directly affected by the 5G interference question.
All in, now that the controversy has made its way to CNN and other news media and a few foreign airlines were said to be planning to limit flights to the US if the 2nd 5G deferment was not extended, traveler interviews and aviation specialists will get their day in the sun until carriers figure they have waited long enough for the FAA to finish its evaluation and for the industry to adapt its equipment to the ever-changing technology environment and press for an end to the moratorium or final regulations, including upgrading or replacing altimeters that are affected..
Altimeters come in 4 types.  The first is based on barometric pressure and temperature.  These would not be affected by 5G.  The second are radio altimeters that measures the time it takes for a pulsed signal to be reflected from the surface.  As the pulse time is very short, separate transmit and receive antennas are needed.  We believe these could be affected by 5G signals.  The third type are GNSS (Global Navigation Satellite Systems) that use radio waves that are bounced off of satellites rather than the ground.  We believe these types could also be affected by 5G signals.  He fourth type are laser altimeters that use laser light instead of radio waves to measure distance similar to the way TOF works for short distances.  The most common of the 4 types is a barometric altimeter.
​

The Global Mobile Suppliers Association is a trade organization that promotes the use of 3GPP (3rd Generation Partnership Project) standard protocols for mobile telecommunications.  Such standards cover those related to 2G, 3G, 4G (LTE) and 5G and the organizations executive members include Apple (AAPL), Ericsson (ERIC), Intel (INTC) Qualcomm (QCOM), Huawei (pvt), Samsung (005930.KS), Nokia (NOK), and ZTE (000063.CH) along with thousands of members and the association also amasses global data on a variety of topics related to the current state of mobile communication.  Based on that accumulated data we have put together a sort of 5G year-end summary that points toward the state of 5G today.

• 481 operators in 144 countries/territories have invested in 5G to date.  That is up 16.7% y/y.
• Of those 481 operators, 189 in 74 countries have launched one or more 5G services, up 40% y/y.
• There are 81 operators offering residential or business 5G FWA (Fixed Wireless Access) services, up 84% y/y.
• 32% of FWA service providers offer speed-only based rates, while 36% offer volume-only based fees, with 9% offering both plans.
• FWA quoted speeds range from 50Mbps to 4.2Gbps, with just over half between 250Mbps and 1.0 Gbps.
• 207 5G FWA CPE (Customer Premise Equipment) devices have been announced (up 92% y/y) with 61.8% for indoor use.
• Private network deployments (664 total) are primarily LTE, but currently 5G private networks hold a 20% share, with combined 4G/5G adding another 5%.
• The total count of announced 5G devices is now over 1,200 with 832 commercial devices
• 5G form factors continue to increase (22 currently), with phones making up the largest share (~50%) (See Fig. 1)
• 596 5G phone models have been announced with 530 commercially available from 40 vendors, with 8 additional vendors with announced models.
• There are 5 suppliers of 5G chipsets and 47 commercially available 5G mobile processors/platforms (up 88% y/y) and 17 commercially available 5G modems, up 70% y/y.
• 55 5G Chipsets support sub-6 GHz spectrum (60.4%), while 36 support mmWave (39.6%)
• 82.1% of all announced 5G devices support sub-6 while 12.4% of announced devices support mmWave.

All in, despite any issues surrounding the mobile phone business, it has been a good year for 5G overall, with almost every metric expanding on a y/y basis.  While Sub-6 is the dominant spectrum, with a slower data speed than mmWave, overall 5G speeds continue to improve and the auctioning of C-band spectrum should continue that trend in 2022.  5G deployments globally have been delayed somewhat due to the COVID-19 pandemic, and we have little trust in coverage maps, but as deployment expands in 2022, mobile users will see less 5G drop and 5G FWA will continue to expand as a business or residential alternative to fiber and cable.  Phones with 5G capabilities continue to expand penetration in mid and lower priced tiers and consumers seem to have accepted that 5G capability is the obvious choice if you live in any urban or suburban area.  5G device suppliers are so common that Huawei, the supplier with the greatest 5G device share, has just over 5% of device share, and Samsung, one of the biggest proponents of 5G in the early years, remains under 5% in terms of device share (Fig. 2).

5G device growth (i.e. – the number of new devices with 5G capability) slowed in November, similar to last year, albeit at a slower pace than in 2020.  Device growth had been averaging 7.4% through September and has seen declines in October and November, which is beginning to look like a seasonal slowdown as new 5G devices have already been released for the holidays.  Device growth, while slower in November remains considerably above the trend line (Fig. 3).  5G smartphone growth also slowed a bit in November but the spike seen in September is another indicator that the 5G space has matured enough to see seasonality.  As the data for 5G growth is meaningless for 2019, 2020 is the only data points on which we rely for growth in December. 
Based on last year’s data, we would expect December 5G device growth to be in the range of 5.0% to 5.5%. and while overall 5G growth will slow a bit more in 2022 as penetration rates increase, we expect a weak January and stronger 5G device growth in February and March, a result of Mobile World Congress (2/28/22).  While overall device growth will slow in 2022, we expect penetration to continue to rise as 5G networks in a number of countries will begin in earnest in 2022, and continued expansion in the US and China will take place.  C Band (3.7 – 3.989GHz) auctions and assignments continue in this the most popular mid-band 5G spectrum with only a few European countries still in the planning stage, and about half of South American countries having already assigned the band.  The buildout of 5G systems in these countries is up to both the government or private carriers so the use of such spectrum is not a given, but such spectrum assignments pave the way for 5G usage when the towers are built.