Last week, patent pool firm MPEG LA and frequent patent challenger Unified Patents announced a patent pool for 3GPP infrastructure named Alium. I have no idea why they chose the Latin word for garlic as the name for a telecommunications patent pool. If the pool smells from anything, it's from the apparent agenda of high-volume implementers seeking to devalue patents.
As I noted three months ago, there are various reasons for which Open RAN is certain to increase SEP licensing costs. At the time, I was not taking a position on the bottom-line desirability of Open RAN, or the O-RAN software interfaces and specification. My focus was narrowly on IP licensing and litigation issues. I later reported on growing unease in Europe over the geopolitical dimension of the push for adoption of O-RAN, and on the security risks identified by a German government agency. It's a multi-faceted topic. With deep-pocketed organizations beating the drums for Open RAN, the potentialy intriguing aspects of the approach have received far more publicity than the caveats.
According to its initial announcement, the Alium pool purports to "provid[e] a solution to the uncertainty and risk posed by the more than 100,000 essential patents already self-declared." The same press release describes patent uncertainty as "threaten[ing] to slow adoption" of Open RAN. But a pool can't eliminate uncertainty as patent holders have no obligation to join it. They can stay outside the pool and pursue their agenda. In that case, however, Unified Patents comes into play, with a "Patent Quality program to build the integrity of the Open RAN ecosystem by discouraging the assertion of invalid patents." Post-grant challenges to patents are Unified's business. They make a lot of noise every time they get a patent invalidated; but there are many cases in which their challenges fail, which serves to harden certain patent portfolios. Companies that hold many patents are sometimes fine to lose half a dozen patents or more if only one of them emerges stronger than ever. Whether the Unified glass is half-full or half-empty is in the eye of the beholder.
An Alium pool license is a bundle: as an implementer, you get a license to all "essential" patents held by a list of licensors--which has not been released yet, but we'll know more after the turn of the year--and additionally you get Unified's service of challenging other patents (not the ones in the pool, obviously). This is a unique combination. All other patent pools that I know are neutral with respect to patents outside a given pool.
By becoming a joint venture partner in a patent pool firm, Unified knowingly and willingly runs into a conflict of interest: there may patent holders whose patents Unified would like to challenge (and those patents don't have to be telecommunications infrastructure patents), but those patent holders may already be, or may be courted to become, Alium licensors.
Unified's other contribution relates to the utilization of AI to determine the share each licensor receives out of the pool income (after deduction of a pool fee and Unified's fee for challenging external patents). However, as the Alium website explains, essentiality determinations will not be left to AI: each licensor will have to prove in a traditional way (i.e., subject to manual review) that it holds at least one essential patent.
The pool focuses on patents potentially infringed by a Radio Unit (RU) as defined by 3GPP. This is different from baseband communications between a base station and an end-user device.
The licensed field of use is "Radio Units (RUs) compliant with O-RAN Alliance’s adopted reference designs for Picocells and Outdoor Micro Cells (low PHY layer functionality per 7-2x baseband split)."
While no licensor has been named, an announcement of the names of some contributors is expected to come in early 2022. What has already been announced is the rate: it starts at $25 per unit and goes down to $10 for all units sold after the first 200K. It's highly unorthodox to determine a rate while it is still unclear whose patents will be in the pool...
Unless there will only be a tiny number of patents in the Alium pool, that rate is extremely low compared to, for instance, Justice Birss's top-down royalty determination in Unwired Planet. For the time being, there is every indication that the Alium pool is an implementer-driven pool. Presumably the key licensors will be companies whose primary objective is to bring down licensing costs on the implementer side. Alium's initial presentation talks about "establish[ing] a FRAND rate," and in this case, the plan appears to be a sub-FRAND rate, hoping that courts and policy makers will be swayed by a pool rate. I believe major patent holders will know how to explain to a judge or a jury that Alium is too agenda-driven to be indicative of the fair and reasonable market value of patents in this space.
Let's see what Alium will announce early next year. While pools can create transactional efficiencies, this pool is designed to serve those who may have to defend themselves against infringement litigation brought by companies that don't contribute patents to that particular pool.
MPEG LA's most famous pool, the AVC/H.264 video codec pool, definitely played a key role in the success story of that codec standard. Certain significant patent holders such as Nokia and Motorola preferred not to participate, as they felt they could get a fair value only outside the pool, but MPEG LA's licensors had to bring litigation against a number of companies over the years in order to get them to take a license. It would be an oversimplification to say that MPEG LA was an implementers' pool--but it was certainly designed to help large-scale implementers reduce licensing costs by means of an annual royalty cap. While an annual royalty cap may appear generous to some, it's questionable from an innovation and competition policy perspective: it massively reduced the per-unit licensing cost of the largest companies in the industry, giving them a competitive advantage over smaller rivals. Alium does not appear to have a cap, but its royalty rate goes down with volume, to an extent that far exceeds what is justified by administrative economies of scale.
Share with other professionals via LinkedIn:
