Obscure Prior Art Less Likely to be Combined

Google LLC v Sonos, Inc 2022 FC 1116 Zinn J

            2,545,150 / Adaptive Echo and Noise Control

This brief decision is notable for providing further clarification of how obscure prior art is to be treated in the context of an obviousness attack.

Various “virtual assistants” such as Alexa or Google Assistant, respond to voice commands that may be picked up by a range of smart appliances, such as Sonos’s well-known home audio speakers. Google’s 150 patent relates to a system of echo cancellation and noise suppression in which the order of echo cancellation and noise suppression is adaptively determined based on an amount of noise in the received signal [86]. Google brought this action alleging that Sonos’s smart speakers infringed. In defence, Sonos disputed infringement and alleged invalidity based on obviousness.

As is often the case, much turned on claim construction. A preliminary issue was whether recourse to the disclosure is always permitted or is permitted only when the claims are ambiguous. Biogen v Pharmascience 2022 FCA 143, released only a week before, settled that recourse to the disclosure is always permitted. While Zinn J would not have had the benefit of Biogen v Pharmascience, he nonetheless came to a consistent conclusion after reviewing the recent FC caselaw [42]–[44].

A key point on the facts was whether “echo cancellation” is a subset of “noise suppression” or whether the two are mutually exclusive [47]. Zinn J concluded that the two were distinct, in an analysis that turned primarily on a purposive reading of the patent as a whole, rather than on the expert evidence: see [50], noting that the two must be distinct to give meaning to Claim 7; and [51], noting that the two are treated as distinct throughout the disclosure. Zinn J’s construction of other contentious claim elements turned largely on the expert evidence.

Zinn J held that Sonos did not directly infringe because the allegedly infringing features were disabled on delivery [65], so that infringement could only be established by inducement. But he also had no difficulty in concluding that if the user infringes, that infringement will have been induced by Sonos [67]–[69], so infringement simply reduced to whether the user would infringe [70]. He answered this question in the negative, in an analysis that turned entirely on the evidence [81].

The obviousness inquiry was originally taken to be an essentially factual inquiry as to whether the invention would have been obvious to a person skilled in the relevant art. From this, it follows directly that the state of the art relevant to an obviousness attack does not comprise the entire body of publicly available information, but only that information which would actually be available to the skilled person. When the non-obviousness requirement was codified in s 28.3, it provided that obviousness is to be assessed in light of information disclosed “in such a manner that the information became available to the public in Canada or elsewhere.” On its face, this states that the entire body of public information constitutes the state of the art. The question then was whether codification was intended to change the law on this point, and if so, what the specific effect of the change was. In Hospira 2020 FCA 30, discussed here, the FCA held that “it is an error to exclude from consideration prior art that was available to the public at the relevant date simply because it would not have been located in a reasonably diligent search” but at the same time “The likelihood that a prior art reference would not have been located by a [person skilled in the art] may be relevant to consideration of step 4 of the obviousness analysis (whether differences between the state of the art and the inventive concept constitute steps which would have been obvious to the PSA) in that the uninventive PSA might not have thought to combine that prior art reference with other prior art to make the claimed invention” [86]. As I noted in my post on Hospira, this suggests that there can be no mosaicing of prior art that is not available in a reasonably diligent search. However, the FCA’s comments in Hospira were brief, as the Court remanded the issue for reconsideration.

The issue of obscure prior art arose in this case because a key element of prior art was the US patent 5,668,871. The 871 patent was very clearly obscure prior art—which is to say, prior art that would not have been found in a reasonably diligent search; it was not known to either expert prior to the litigation, nor was it found by either expert during their preparation, “despite their personal familiarity with the field and despite [Sonos’s expert] having conducted a diligent search” [92].

Google argued that the 871 patent was not eligible prior art for the purposes of the obviousness attack for that reason. Zinn J rejected this argument on the clear authority of the above-quoted paragraph 86 of Hospira [96].

The question then is how the obscure prior art may be used. Zinn J pointed out that “the difficulty of locating a document is a matter that may be considered at the final step,” relying on the same passage from Hospira [97].

[98] The question thus becomes whether, given the obscure nature of the 871 Patent, the uninventive POSITA might have thought to combine the 871 Patent with other prior art to make the claimed invention.

On the facts, Zinn J held that “given the difficulty in locating the 871 Patent, that the POSITA would not have been led directly and without difficulty to combine these references” [105], and he consequently rejected Sonos’ argument that the 150 Patent was obvious based on “any combination involving the 871 Patent” [106]. I must say that it is not clear to me exactly which references in addition to the 871 patent Zinn J was referring to as “these references” but I take it the other references were part of the prior art, but not part of the common general knowledge.

Google had relied on Kane J’s statement in Teva v Pharmascience 2020 FC 1158 [796], to the effect that “the prior art relied on by Pharmascience would not all have been found by the POSITA,” affd 2022 FCA 2 (Locke JA) [32] noting that Kane J “was apparently concerned that, given the difficulty in locating certain prior art, the PSA would not have been led directly and without difficulty to combine these references.” Zinn J considered that Locke JA was not contradicting his own prior statement in Hospira but was simply saying that “the difficulty of locating a document is a matter that may be considered at the final step” [97]. I think this is right, considering Kane J’s statement at [801] that “I agree with Teva that Pharmascience has not explained how or why the POSITA would regard these pieces of prior art together in a mosaic to lead to the invention,” and moreover “even if the POSITA was handed this mosaic, it does not lead them directly to the subject matter of the claims.” So, I don’t see any inconsistency.

Zinn J’s decision is consistent with Fothergill J’s decision in dTechs 2021 FC 190 (see here). However, it is possible that it diverges from Manson J’s decision in Biogen v Taro 2020 FC 621; in my post on Manson J’s decision I suggested that his analysis was notable for making a mosaic of obscure prior art. Manson J’s decision holding the patent at issue invalid for obviousness was just affirmed by the FCA in Biogen v Pharmascience 2022 FCA 143 [143]–[171]. However, the FCA decision did not address this issue; the main ground of appeal was that Manson J had failed to adopt the mantle of the POSITA, essentially by failing to recognize that the skilled person in the field would be highly skeptical of any new alternative treatments that were not supported by double-blind, placebo-controlled trials [144], [148]. So far as I can see, the issue of mosaicing of obscure prior art was simply not addressed at all. I think the most likely explanation is that I was simply wrong in thinking that the prior art relied on by Manson J was obscure prior art, which would explain why the issue was not raised on appeal. In that case, there is in fact no divergence between these decisions on this issue. In any event, even if Manson J did combine obscure prior art, the fact that the FCA affirmed on obviousness cannot be taken as holding that it is permissible to combine obscure prior art, given that the issue was not discussed. If the FCA had intended any comment on that point, it would surely have discussed Hospira.