Your browser is not supported. Please update it.

14 May 2015

Ericsson’s Apple suit: the cellular patents lock-in proves tough to break

The shape of the mobile phone has changed enormously since the early days, yet the pattern of patents power has not. There have been many efforts to address that, from patent pools to antitrust probes, but none has had a significant effect. The companies which develop and patent the 3GPP-approved modem technologies still hold the balance of power. There was a chance to change that with the move to LTE, but it failed – but ‘5G’ may provide another opportunity for those wanting a more open platform.

In the meantime, the litigation between Ericsson and Apple, which the former stoked up this week by extending it to Europe, is a highly significant testing ground, epitomizing the battle between the old and new kings of the mobile industry. The outcome of such fights, which also include Ericsson’s action against Xiaomi, will help decide whether 5G will evolve with a new set of assumptions about IPR in mobile devices – an outcome which will be particularly important, since it will also affect a huge variety of other products which will include a cellular modem in future, within the internet of things (IoT).

At the heart of Ericsson’s escalating litigation with Apple is the issue of whether, in the era of the smartphone, so much power should still lie with the owners of standards-essential patents (SEP) in the 3GPP modem and its associated technologies. Apple and many others argue that other, non-standardized elements of modern devices have greater value and differentiation, yet those with SEP can not only charge high fees for their IPR, but have a significant commercial advantage over rivals with no patents to trade. For instance, when Nokia was still making handsets, it could leverage its huge store of SEP to sign cross-licensing deals with other patent holders which might be cost-neutral, while a rival with similar devices but no SEP would have a far higher cost base because of the royalty fees.

Of course, the technologies behind these patents take huge sums and resources to develop (this debate is not about trolls, which have a different kind of power and are under attack by both traditionalists like Ericsson and mobile upstarts like Google). But Apple argues that, in the case of the technologies which are most important to the mobile device of 2015, there are few opportunities to generate significant licensing revenues nor to influence the shape of future platforms and standards. Many of the most crucial standard aspects of the mobile platform, such as codecs, are managed under open source, royalty-free or patent pool procedures, but every attempt to introduce these to the core areas of cellular technology have been squashed by the weight of the traditional players.

In other words, the 3GPP and the standards it controls command a disproportionate share of patent revenues and technology power in relation to their modern-day position in the overall smartphone landscape.

A teardown of a cellphone in the late 1990s would have revealed little beyond the modem, RF and antenna plus some basic embedded software, and the battery, SIM and keypad. Among those core components, the modem commanded by far the highest complexity, development cost and value. By contrast, a 2015 smartphone has a huge variety of additional elements, some of them more expensive and more complex to develop than the modem itself. The touchscreen and its surrounding components; processors for applications, graphics, image and other functions; a huge range of sensors; connectivity chips from WiFi to NFC; complex MIMO antennas and advanced power management facilities; not to mention a high level operating system and other software.

So why, the Apple camp argue, does the 3GPP still shape the industry, to the benefit of companies which have invested in underlying cellular technology, but not those which have created the modern smartphone by developing user interfaces or touchscreens?

When it comes to the technologies inherent in cellular communications, the traditional vendors still hold the upper hand, and were successful in carrying their 2G and 3G dominance over into 4G. Even though LTE is based on a whole new OFDM air interface, which could have ushered in new vendors and new patent regimes, in fact Qualcomm, Ericsson and others worked hard to develop their own technologies in this area. There was some shift in the premier league of SEP leaders – more presence for ZTE and Huawei, less for some Japanese firms and InterDigital – but the Qualcomm/Nokia/Ericsson triumvirate held firm.

The WiMAX community, whose technology was also OFDM-based, tried to introduce a different approach, with more transparent, centralized licensing processes similar to those in WiFi. That failed partly because of the mighty fightback by the 3GPP leaders – Ericsson, Qualcomm and others.

Of course there were other issues which destroyed WiMAX’s bid to be the mainstream 4G technology rather than 3GPP-driven LTE. But its sad story does raise questions about how far new competitors and new technologies can thrive in the mobile world – questions which also lurk behind the bloodbath in the cellular modem market, most recently highlighted by Nvidia’s confirmation that it would exit that space and stop trying to compete with Qualcomm. That was not all about IPR power – Nvidia had strong soft modem technology, acquired with Icera, but lacked Qualcomm’s long experience in the difficult art of integrating many chips into a low power SoC, so its own all-in-one processor/modem offering came too little too late.

But there is continuing unease, reflected in a rising tide of probes into Qualcomm’s business practices, that market leadership and IPR leadership in the cellular technology world is too closely tied together, disadvantaging new entrants. Nno market benefits from high barriers to entry. It is noticeable that when WiMAX was having its moment, it attracted a large number of innovative start-ups in the device baseband space, partly because it was a non-3GPP technology with more open approaches to patents and licensing. Sequans, Altair, Beceem, Comsys, Wavesat and others battled alongside Intel and Fujitsu. The technology may have failed to gain significant traction, but it offered a glimpse of a different standards landscape (it is notable that the few surviving LTE modem independents, such as Sequans and Altair, came from a WiMAX start, and have been skilful in targeting niches where Qualcomm hardly goes).

Hence the importance of the Ericsson suits against Apple and Xiaomi, both very successful smartphone vendors, both lacking in significant cellular IPR.

This week, Ericsson turned up the pressure on the iPhone maker, suing it in three European markets and claiming that it expects special treatment when it comes to licensing fees for SEP. Apple currently has no licence for its antagonist’s technologies, since their former deal expired early this year.

The escalation of the dispute will bring back bad memories of Apple’s last major battle over SEP, with Nokia, which ended in a settlement that was favorable to the Finnish company. That highlighted the fact that, while Apple has had some successes in patent infringement actions against Android players, notably Samsung, those have related to non-standardized elements such as displays, user interfaces and device design. In fundamental mobile IPR, it is a minnow.

Even those with significant SEP, like Samsung, find it hard to fight against Ericsson. One of the Swedish firm’s first major legal attacks of this kind was against Samsung. It accused the Korean firm of violating 24 patents, while its opponent claimed it was asking for unreasonable royalties which did not comply with Frand rules. But the case was settled, in January 2014, in Samsung’s favor, adding about $500m to Ericsson’s bottom line.

Apple aims to tip the balance of power in its own direction, which would significantly reduce its costs and increase its overall influence. In complaints made against Ericsson in a California court in January, Apple argued that the price of modern electronic devices is driven by the technologies in which it has power – the operating system, touchscreen interface and so on – which are unique to each design, not by standards-based elements, notably the modem. It said Ericsson “seeks to exploit its patents to take the value of these cutting-edge Apple innovations”.

The Swedish gian, has filed suit in Germany, the Netherlands and the UK. It claims it wants Apple to license its technologies on the same Frand (fair, reasonable and non-discriminatory) terms as everyone else, but that the US firm has refused to do so, since their previous contract expired in January.

Apple, in various counterclaims, has been alleging that Ericsson uses its dominant IPR position to demand “excessive” royalties, accusing the company of “abusive licensing practices”. It is also disputing that some of the patents in question are essential to LTE standards at all. It has also passed on Ericsson’s proposal to refer the argument over Frand terms to independent arbitration in Texas, an offer which has now expired, helping to spark the new actions.

It seems that the new European actions, as well as opening new fronts (and possibly bringing the European Commission into play in future) are specifically addressing Apple’s claims that key Ericsson patents are not SEP.

Gustav Brismark, VP of patent strategy at Ericsson, said in an interview: “Our Frand terms [with Apple] are in line with the other some 100 licensees using our technology. Our policy is to be fair and treat everybody the same in the market. We simply demand a fair royalty.”

Ericsson has taken the high moral ground with its offer to take not just this spat, but its entire Frand licensing framework, to arbitration, and so hope to obtain objective endorsement of its approach, at a time when many international regulatory and standards bodies are rethinking how Frand terms are defined. The European Commission, for instance, has been investigating the way that companies such as Samsung license their SEP, and whether this could be construed as giving the patent holders an unfair competitive advantage.

When the licensing deal for 2G and LTE patents expired early this year, Ericsson accused Apple of infringing on seven of its essential patents, while its opponent filed suit against Ericsson in California, denying any infringement and challenging the claims of being fundamental to LTE standards at all. However, Ericsson points out that the seven assets at issue in that case are a “small subset” of the total and “would not in any way resolve the wider dispute covering hundreds of patents”. The fights cover 2G and LTE standards as well as non-standardized technologies related to cellular communications and wireless modems.

Further legal actions have followed from Ericsson, including the demand for arbitration, several infringement lawsuits in Texas, and complaints at the US ITC (International Trade Commission). The ITC hearings are scheduled for late 2015 or early 2016. Now Ericsson has broadened its attack to its home front with the European suits.

Though Ericsson is not commenting on the sums involved, Reuters calculates that, if the firm wins, Apple would owe between $240m and $725m a year, based on forecast levels of handset sales and royalties per phone. However, any settlement could be affected by broader industry debates about per-device charging – Qualcomm, in particular, is under pressure to shift to charging licensing fees based on the price of the chip, not the whole smartphone. Apple has been in the vanguard of that campaign, which is already influencing patent renewal deals across the industry, and is harnessing the same arguments against Ericsson. The Swedish firm continues to maintain that royalties should be based on “the end prices of entire LTE devices”.

“Apple continues to profit from Ericsson’s technology without having a valid licence in place,” Kasim Alfalahi, Ericsson’s chief intellectual property officer, said in a statement. “Our technology is used in many features and functionality of today’s communication devices. We are confident the courts in Germany, the UK and the Netherlands will be able to help us resolve this matter in a fair manner.”

Ericsson has one of the wireless industry’s largest IPR portfolios, with more than 37,000 granted patents, and about the same number of pending applications worldwide. It says it has about 100 licensing deals, covering most wireless vendors and including major cross-licensing deals with rivals such as Nokia. Like Nokia (now freed of accusations of bundling by having no handset business), it aims to wield this asset more actively to generate money and influence, and more broadly, to keep the landscape of mobile SEP unchanged as the industry heads to 5G.

This makes the case against Xiaomi an important one too, especially as it also involves, the position of MediaTek, one of the few credible rivals to Qualcomm left in the cellular modem and SoC world. The targeted Xiaomi devices are those based on MediaTek processors, since these are not covered by the licences which would come with a Qualcomm chip. Nothing better sums up the disadvantage that a non-SEP owner, however successful in technology terms, has against a major IPR powerhouse.

In December, the rapidly growing Chinese handset maker saw some of its products subject to a temporary sales ban in India, when the Delhi High Court ordered a halt in import, advertising, manufacture or sale until the case could be resolved. The specific models were not named in the order, which just referred to products that “include the AMR, 3G and EDGE technology/devices/apparatus as patented” by Ericsson, but the he main devices which Xiaomi sells in India are the Mi3 and Redmi smartphones and the Redmi Note phablet, using MediaTek chips.

The precise details remain shadowy, but it seems that this is one of many cases in which patents rights in SoCs are complex. It seems that MediaTek has either not licensed the Ericsson patents in question, or has licensed them but without the rights to sub-license them to its customers – which would mean Xiaomi was not covered by MediaTek’s deal. Had it used Qualcomm chips, patent experts say, it would have been protected because Qualcomm’s extensive cross-licensing agreements with Ericsson would extend to its chip customers too.

The dispute shows the knotty issues of SEP licensing starting to affect the competitive landscape in emerging economies, not just the US and Europe, where most of the major court cases revolving around this type of IPR have dragged on in recent years. These issues will be important in markets like India too, because their scale and future growth will make them highly attractive to hard-pressed handset makers, but they impose very low cost bases on vendors, so the level of IPR fees will be critical to margins.

The Xiaomi case also points to MediaTek’s disadvantage – as it closes the technology gap with Qualcomm, it still lacks the SEP which is so useful to gain real market power. That is the concern raised by the exit of Nvidia, Broadcom, ST-Ericsson, Texas Instruments, Freescale and others from the handset modem game. So far, consolidation has not weakened innovation (nobody could accuse Qualcomm of sitting on its laurels in R&D terms), and price wars persist, even with just a few suppliers, because of the huge buying power, and daunting price pressures, of the major handset customers – another rapidly consolidating business. But the cellular modem and its 3GPP standards remain the foundation stone of mobile standards, and the companies which develop and control this area of technology also wield considerable power across the whole industry.

MediaTek is becoming an increasingly powerful threat to Qualcomm in its heartland, as it moves rapidly up from its traditional low-cost market into 4G superchips, sometimes managing to pip the US firm to the post with a new development. That happened last week when it launched the industry’s first commercial 10-core application processor, the Helio X20 (see separate item).

The company also promised that its new octacore offerings, due to ship in the second half of this year, will help boost its lackluster sales (up just 3.3% year-on-year in its first quarter, to NT$47.5bn/$1.55bn). It is looking for a 50% jump in shipments in the second half, based partly on the eight-core products, and if it delivers that, it would be a sharp contrast with Qualcomm’s warning that its full year revenues and profits would be below its previous guidance.

The question is, whether it can really level the playing field with Qualcomm, even with its hugely advanced 10-core monster, without more patents?

There are two developments which may disrupt the status quo at last, and neither of them concerns US law courts. One is the increasing IPR power of Chinese vendors such as ZTE – and Xiaomi itself, which says it applied for over 2,300 patents in 2014 and is aiming to have “tens of thousands of patents” within a decade, partly to protect itself against actions like Ericsson’s. Companies like ZTE and Huawei have the R&D resources and industry clout to ensure their technologies take their place in the SEP pantheon for new 3GPP standards, as well as in other platforms like those governed by the IEEE.
It is unclear, so far, whether the Chinese majors will shore up the existing system, as long as they can benefit from it, or will push for a new approach that might weaken the hold of their rivals from the west.

And less immediately, ‘5G’ is likely to see a far wider proliferation of technologies in the mobile and wireless landscape than 4G, or the smartphone revolution did – multiple air interfaces perhaps, including new ones that might not emanate from the 3GPP community; and a huge variety of other connectivity, software and device chip technologies. That will fundamentally influence the pattern of power in mobile IPR and in the baseband space – and will decide whether there will be the opportunity for a new modem standard, potentially with a new WiMAX-like array of innovators, to shake up the industry.

It seems unlikely that the old guard will be able to hang on to the same level of power in that fragmented world – but for Apple, Xiaomi and others, the aim will be to break down the old walls long before the next decade, and 5G, dawns.

Korea and EC join forces in Qualcomm probe:

Qualcomm may have settled with China’s authorities over investigations of its business practices, but it still faces probes in other regions. Now South Korea’s Fair Trade Commission (FTC) and the European Commission (EC) say they will share the progress of their respective examinations, and the steps they plan to take, in a sign that pressure is mounting on the US giant.

According to ZDnet sources, the FTC is close to a decision that Qualcomm’s patent licensing practices do, indeed, violate Korean antitrust law. And that may influence Europe – FTC officials met with Cecilio Madero Villarejo, deputy director-general for antitrust at the EC, recently to discuss how competition laws could be applied in a cooperative way when it comes to standards-essential patents (SEP) and whether these are licensed according to Frand (fair reasonable and non-discriminatory) guidelines.

“The Qualcomm case can affect many countries globally, and we plan to cooperate with the EU and the US and other developed countries’ authorities in handling the case,” an FTC spokesman said.