Your browser is not supported. Please update it.

13 November 2014

BT DSL patent appeal turns case further in Assia’s favor

British Telecom has lost an appeal court revisit to its patents dispute with US broadband management specialist Assia, damages are yet to be set, but this round was a serious reversal for BT, since the appeal court added another patent breach to the first court room result.

Our first heads up came from the legal practice Wragge Lawrence Graham & Co which acted for Assia in the case and then we had an email exchange with Assia and read the verdict.

In December last year the first British Court found BT guilty saying that BT’s Rate Adaptive Management Box (RAMBo), which monitors and controls its DSL lines, infringes upon one Assia patent

Assia’s DSL Expresse software is already used to manage well over 70 million broadband connections across five continents and this relies on the technology in two European patents numbered 1,869,790 and 2,259,495. The 790 patent was seen by the original court as both valid and infringed upon, while the 495 patent was ruled valid but not infringed because it was deemed to be obvious. The Appeal court has now reversed that issue with respect of the 495 patent, so pushed the case forwards with 2 core patent violations, instead of one.

Patent royalties have not really played a major part in the development of the fixed broadband eco-system. In DSL broadband John Cioffi, founder of both Amati Communications, which won the first AT&T DSL contract, and since then Assia, is widely acknowledged as one of the main inventors with 100 patents to its credit; while in cable broadband, the R&D collective CableLabs makes all the running, but does this on a royalty free basis.

Cioffi doesn’t need the money, Amati was acquired for $395 million (Cioffi owned 10%) by Texas Instruments, the day after Amati won the market defining AT&T DSL deal. Prior to that there were many forms of DSL, most of them proprietary and this contract drove the technology into standards, which has created the market as it is today.

DSL reaches the majority of 800 million broadband homes today and even the broadband ambitions of cable companies were triggered by that initial AT&T deal.

Interestingly Alcatel has told us that it uses a different approach to managing its DSL lines than Assia, not varying the power or wavelengths in order to get around cross-talk, but much of the BT approach relied on this, and this seems to be core to Assia’s method

These patents go to the heart of Assia’s current product range, DSL Expresse, a set of software tools which behave like an expert system, suggesting ways to improve throughput on fixed DSL broadband lines.

These and other patents underpin ideas that will come into their own even more so, once vectoring begins to be installed in earnest.

There are a number of incumbent telcos who are trying to build their own DSL management technology and Assia has picked BT as the first operator to bring to heel with a legal action. BT wanted to go it alone with its own in-house technology, but this relied on information gleaned from in-house trials and also from publicly available works, some of which include those written by Ken Kerpez, now the Principal Systems Engineer at Assia, which are in the public domain. Assia has always asserted that when it did consulting work with British Telecom way back in 2006, some ideas were incorporated into BT’s thinking on the subject. It is loath to say it copied Assia’s ideas, but instead says it “tried to,” suggesting the copies aren’t as good as the original technology, but still breach patents.

BT lost the case in December and has now lost the appeal, in worst shape and we don’t yet know if it will move to the UK’s Supreme Court (which we understand has taken over from the House of Lords), or if it will submit to the appeal court verdict and pay up. There is no date set as yet for the hearing to set the damages, likely to be January at the earliest, but BT is in a difficult position given that several million DSL lines (about 2.7 million Infinity lines) are already being controlled using its infringing technology, and it may have to simply pay damages as well as license the technology. Given that these particular patents are not essential patents contained in a standard, unlike those used in Vectoring, the amount that a court may award in damages could be considerable, given that FRAND terms have never been applied to these.

Assia has said in the past that it attempted for years to work with BT and that by refusing to take a license or admit infringement, BT was effectively blocking access to the entire UK market for Assia. That could be true in most countries in Europe.

Assia’s General Counsel Noah Mesel said “the Court’s findings are highly relevant to our discussions with other operators who we believe also implement broadband management systems that read to Assia’s extensive patent portfolio.”

It looks likely that this result will form a test case that will be used extensively in Assia’s future negotiations and litigation with network operators and manufacturers.

BT is supplied by Fujitsu and Huawei in DSLAMs and multiservice access nodes (MSANs), so those could also supply management software, and we understand that Huawei has such a product, but it is not used by BT and anyway Huawei is not shown as a partner on the Assia site and may not have a license to these patents either (but then again may not use them).

In the past BT has said that it could make minor changes to its programming to avoid the issue entirely and that the court’s decision will have no material effect on the operation of BT’s networks. We think that legal workarounds on patents are usually covered by what tend to be very broadly worded patents, and we expect that BT has nowhere else to go but to license the patents.