For those in the business of mobile networks, “Open RAN” is the “next big thing”. In the old world of 4G networks, the key interface between radios and baseband processors was in principle an open standard, but in practice completely proprietary, because of ‘vendor-specific’ data fields and formats. Consequently, signing a supply contract with Nokia, Ericsson, or Huawei was more than a little like applying for a marriage license, and divorce was expensive. You couldn’t change radio vendors, for example, without changing everything else.

Today, the ORAN Alliance publishes specifications for key interfaces, and consequently many new players are attempting to break into the previously closed market for RAN gear, like radios and the software and servers that will accomplish baseband processing.

It promises to be a lucrative business.

Truth be told, there are two serious obstacles. One is surmountable. The other, less so.

The first obstacle is that the most important standard, for “fronthaul”, is still incomplete, despite being a published spec. It’s not plug-and-play. According to a number of reports, vendors who want to ‘interoperate’ must first work out agreements on some of those data fields and formats, which may in fact still be changing.

The good news here is that all players, and in particular the members of the ORAN Alliance, are more or less committed to finishing the standard and agreeing on certification testing. It’s probably just a matter of time before this happens — although ‘time’ may mean a year, or two, or longer.

There is a more serious obstacle, one that seems to go to the heart of the business case for open RAN, or at least for new players.

The standards behind 4G and 5G mobile network technology are developed and published by the members of 3GPP, members who are permitted to take out patents covering said technology, under a sort of gentlemen’s agreement to license their patents under “fair, reasonable, and non-discriminatory” terms, the infamous ‘FRAND’ — infamous because the meaning of those fine words is left to the judgment of thousands of courts across the globe.

They are called “Standard Essential Patents” — SEP, if you like acronyms. A member of a standards body like 3GPP contributes work to the common effort, and declares part of that work essential to the standard, but their own intellectual property, meaning there is no way to implement the standard without infringing on the patent.

The claim of course may be subject to debate, i.e. litigation. But this can be an enormously expensive crap-shoot.

More to the point, if you are a company intending to market virtualized RAN software implementing real standards, there may be hundreds, if not thousands, of SEPs to worry about.

This is not new. The old guard, namely Nokia, Ericsson, and Huawei, are used to dealing with this. Each has its own portfolio of SEPs and bargains with the others to cross-license.

New players don’t have portfolios of their own, to speak of. Cross-licensing is not an option.

In theory, a new player has to work out which of thousands of patents must be licensed, and negotiate with the big OEMs, and also with up to a half-dozen pool administrators who have themselves collected patents over the years. And remember, mobile networks have to support 4G as well as 5G, so there are many years’ of patents at issue.

Or maybe a network operator might do it for them. There is some logic to this. Carriers like ATT have many established relationships as well as patents of their own, and might be very well positioned to negotiate blanket licenses to cover x number of radios, say. This would give them a great deal of flexibility, to deal with new market entrants without worrying about whether they will be forced to replace gear that is tied up in patent litigation.

Whatever happens, this is something the industry is going to have to deal with. Yet, at the moment, it appears not much is happening. At least in mobile networks. In some areas, like cellular vehicle connectivity, aka connected cars or C-V2X, litigation is proceeding apace.

Essential patents can not be overlooked. The open RAN industry will have to pay the piper, one way or another. Sooner or later.