[00:00:00] Speaker 03:
1726, Cisco Systems v. Remote. Ms. Oliver, please proceed.

[00:00:08] Speaker 00:
Good morning, and may it please the Court. Angela Oliver on behalf of Cisco. This appeal involves three separate IPR proceedings covering two related patents. I'll refer to these cases by their IPR numbers, the 575 IPR, the 576, and the 1283. There are two overarching issues for this Court.

[00:00:28] Speaker 00:
First, the board focused on the wrong part of the prior art reference, Tara Shook, when it rejected Cisco's combination because the prior art taught analog predistortion, and the board failed to address evidence that undermined that very determination. Second, in the 576 and 1283 IPRs, the ones involving the deltas limitations, the board again failed to properly consider the combination Cisco proposed. For each case, the court should vacate and remand.

[00:00:56] Speaker 00:
As to the first issue... the board focused on the wrong portion of the Tereshook reference. The board should have focused on Tereshook's lookup table, not Tereshook's feedback loop. The sole basis for the board's decision in the 575 IPR was that Tereshook's feedback loop uses analog predistortion, and the patent claims exclude the use of analog predistortion based on some prosecution history. But the problem is Cisco never relied on that feedback loop in the Tereshook reference. That was not part of our combination.

[00:01:27] Speaker 00:
In the combination, almost all the limitations were taught by Roberts, and we just used one piece of the Terashuk reference, the lookup table with the digital-to-digital mapping from Terashuk. That was the only thing the board should have been considering from the Terashuk reference.

[00:01:41] Speaker 00:
Now, as part of the reasoning, the board said that Cisco could not rely on portions of the reference without looking at the rest of it. They called it a piecemeal analysis. But that's a legal error. Obviousness does not require bodily incorporation. It does not require us to use the entirety of a reference. A reference has to be considered for all that it teaches. We knew this from this court's case law.

[00:02:02] Speaker 01:
The claim here is digital, right, and terrorism is analog.

[00:02:08] Speaker 00:
That's not exactly correct. So the... The portion of Tereshuk that we relied on, the lookup table, that piece is digital to digital. And there's actually no dispute about that, that Tereshuk discloses a digital to digital mapping. The Red Brief concedes this at page 39. So that piece is digital to digital, and that's what we relied on. Now, the analog portion... is in the feedback loop of TerraShook. It's about how you generate the mapping that goes into that lookup table.

[00:02:39] Speaker 00:
That does have an analog component, and that's what the board got focused on. The board said that was analog predistortion. The claims exclude analog predistortion, so therefore you can't use this reference. But again, we didn't rely on that feedback loop. We just needed the board to look at the digital-to-digital lookup table. And so that is sufficient basis to vacate the board's decision process and remand to focus just on that relevant piece. Now, also on this issue, there's a separate problem with the board's analysis, is that even if you kind of go into the reasoning and say, does Terashuk only disclose analog pre-distortion, Cisco presented evidence in its reply brief that there is a different way to do this mapping in Terashuk that does not involve that feedback loop at all.

[00:03:24] Speaker 00:
This is the calculated in advance teaching from Terashuk. It's at appendix... 2271, and it might be helpful just to look at this briefly.

[00:03:35] Speaker 00:
It's at Appendix 2271.

[00:03:36] Speaker 00:
It's Volume 2, Column 7. Tereshuk specifically says, the mapping implemented by the linearizer can be calculated in advance and loaded into the linearizer for use. Alternatively, it may be advantageous to periodically recalculate the mapping. Periodically recalculating, that's the feedback loop. So there were two disclosures of how you do this mapping. One had to do with analog, and one was not. It was calculated in advance. The board expressly refused to consider that evidence at Appendix 26 because it said we did not provide an expert declaration.

[00:04:12] Speaker 00:
That, of course, is not required.

[00:04:13] Speaker 03:
Can you show me where they expressly refused?

[00:04:21] Speaker 00:
Yes, it's at Appendix 26, Your Honor.

[00:04:26] Speaker 00:
And here is the first full paragraph.

[00:04:30] Speaker 00:
It says, Petitioner's reply includes positions not supported by reply declaration.

[00:04:36] Speaker 00:
Petitioner does not support the contention that this could be used with or without a feedback loop. This is the discussion of the calculated in advance portion.

[00:04:45] Speaker 00:
And the board says petitioner does not, excuse me, petitioner neither supports this with the testimony of Dr. Blumenthal or the disclosures of Robert's terror shook or any other reference. But as you can see from the next line in the board's opinion, what we cited was exactly this portion of terror shook. Column 7, lines 15 through 16. So we did cite directly to the reference, and that cannot be conclusory if we've cited that. This is not bare attorney argument.

[00:05:16] Speaker 03:
What they say is that it hasn't supported your contention, so they've read Tara Shook differently than you've argued it. It doesn't say that they rejected Tara Shook or refusing to look at it. It also doesn't say that they're kicking you just because you didn't have a declaration. They're saying you didn't convince them with either a declaration or portions of the references you pointed to. They expressly cite the portions of the reference you point to, so it's not that they ignored them or didn't review them.

[00:05:45] Speaker 00:
Your Honor, I think that would be reading into the board's opinion. They've said, we don't have a reply. This is just attorney argument. and we haven't supported the disclosures.

[00:05:54] Speaker 03:
And there's just... And they cite your petition at the exact pages where you make the argument, and they cite Exhibit 1... See there? Petition 43 to 45. They cite Exhibit 1006, the columns of the reference. That's the exact columns of reference you all cited to them. So it's not the case that the board disregarded the evidence that you put towards it. It cited it expressly and said, this evidence is incandescent. But what... I don't understand what... That...

[00:06:22] Speaker 03:
Maybe you want me to disagree with their fact-finding, but I don't see that I can dispute that they made one.

[00:06:28] Speaker 00:
Your Honor, we just respectfully disagree with how this paragraph should be read. There's no substantive analysis of the calculated in advance portion of this reference. They do not say that it is or is not analog predistortion. The entire premise of the board's decision is that Tereshuk only does analog predistortion. So without some sort of clear finding that calculated in advance is also analog predistortion, I have a hard time understanding how this can be fully vetted by the board's analysis here.

[00:06:59] Speaker 00:
And this was the crux of the entire opinion, that this was the only thing Tereshuk teaches.

[00:07:05] Speaker 00:
Your Honor, before I turn to the next issue, I'll just briefly note that on this issue, this analog predistortion discussion, it came up in all three IPRs. It was dispositive in the 575 IPR. It is not dispositive in the other IPRs because, as we'll talk about next, this was just not the issue in those cases. In those cases, Cisco relied on the background knowledge of a skilled artisan as informed by the right reference to show the disputed limitations at issue, to show how to do the mapping.

[00:07:35] Speaker 00:
So how TerraShook does its mapping, which is what we've discussed here, is not relevant to the next two IPRs. So with that, I'll turn to the next two IPRs, the 576 and the 1283 IPRs. These are the Delta's limitations, and here the board did not adequately address Cisco's proposed combination for these limitations. Cisco's analysis was based on the background knowledge of a skilled artisan as evidenced by the right reference to show how an artisan would have known to create a mapping that compensates for nonlinearities in an MZ modulator using the reference right as the key analysis.

[00:08:16] Speaker 00:
You can see this. I think it's helpful to visualize some of this. If the court would mind turning to Appendix 3102.

[00:08:25] Speaker 00:
This is in volume two.

[00:08:28] Speaker 00:
And there's essentially a graph here that shows how this compensation is done. This is a basic mathematical type analysis. So at appendix 3102, you can see there's a blue line, a blue sinusoidal curve. That's the natural response of the device, the MZ modulator. There's an orange straight line, which is what we're trying to get to. We're trying to linearize the curve. And there's a green dotted line that shows the compensation curve that is taught by the right reference as explained by our expert.

[00:08:59] Speaker 00:
And this green compensation curve is exactly how you create this map. The inputs are on the x-axis, the outputs are on the y. So that is your input to output in the digital mapping that gives you the mapping and the contents of that lookup table that we're looking at for these limitations. Now you can see on the next page at appendix 3103 that this is how the deltas work. So this is a subset of a larger chart. But this subset has four rows in it, and there are four input values with four corresponding output values.

[00:09:35] Speaker 00:
And then you can see the three deltas, or the differences, between those output values. And the claims here are simply asking, as the input values increase or decrease, do the deltas increase or decrease? That's what we've shown in this table on Appendix 3103. And so this was the crux of Cisco's analysis, but that is not at all what the board addressed. Instead, the board focused on a completely separate issue and said that these curves that our expert went through based on the right reference said that this actually has to reflect the terror-shock compensation techniques.

[00:10:11] Speaker 00:
And that caused the board to focus on an unrelated issue about whether Tereshook could compensate for nonlinearities without also addressing logical mismatches. All of this was really a sideshow because Cisco did not rely on Tereshook to show how to fill the contents of that table. This was all based on the right reference and our experts' discussion of the math of how you would fill that table. So the board's discussion of Tereshook in this context was simply off-base. and the board should have addressed our arguments based on right and our expert testimony.

[00:10:44] Speaker 00:
In particular, the board only addressed right in one sentence. It did not go through any of the analysis that we've discussed here, these charts, these graphs that we presented in our petition and our expert's testimony. The board just did not go through any of that. And so we, and had the board done that, it, I think, would have had a much different analysis in its opinion. So we are simply asking that the board... have an opportunity on remand to consider all of this evidence, focus the analysis on what Cisco actually presented with respect to how to show these DILTAs limitations.

[00:11:17] Speaker 00:
It's the right reference, and it's our expert's testimony explaining that. This was never based on Tara Shook, and that is what the board focused on.

[00:11:25] Speaker 03:
Okay, thank you, Ms. Oliver.

[00:11:29] Speaker 00:
Thank you.

[00:11:53] Speaker 04:
Good morning. May it please the court. Denise Demore on behalf of Remote at Tel Aviv University.

[00:11:59] Speaker 04:
Judge Moore, Your Honor, I believe that you had it exactly right, not just with regard to the reply evidence, but in this case, we have three final written decisions in which the board fully considered all of Cisco's positions, recounted the positions of all of the parties, And then in each of the final written decisions articulated their reasons why they disagreed with regard to what Cisco presented in their petitions. There is nothing in the board, there's nothing that they did not consider.

[00:12:34] Speaker 04:
In fact, Cisco's entire appeal is premised solely on the notion that they didn't consider something. But in the final written decisions in this case and in the petitions for rehearing, which Cisco did go back to the board and ask them to rehear, they considered the arguments. They just didn't credit them. And as a result, the board correctly found that Cisco failed to meet its burden in these IPRs, and the claims were improperly affirmed.

[00:13:02] Speaker 04:
If we get to even Cisco's first argument about the fact that in the context of the based-on claims, which is the 575 IPR that they talked about, in fact, the board did consider everything that Cisco presented.

[00:13:22] Speaker 04:
As Cisco pointed out in its briefs, the references should be considered for all that they teach, not just ignore the rest of the reference. And there is nothing in Tarachuk, and if you look at the final written decision, Cisco claims that the board ignored the couple of columns that they cited in the reply. But in fact, in the final written decision at appendix pages 23 through 29, I apologize, They consider everything in column 6 and column 7 and column 8. The board has reviewed TerraChuck and for the reply evidence says, sorry, that doesn't support your position.

[00:13:57] Speaker 04:
But otherwise, as to if we even just look at 23, I mean, Cisco is now saying we were only looking at TerraChuck for the linearizer. But on page 23 alone, the board says that they – That in order, the baseline claim that is the subject of our red brief, that independent claim one recites altering the linearity of an optical response to the modulator, the linear compensation is performed in the claim converter.

[00:14:29] Speaker 04:
And in Terachuk and in Roberts, it happens before the converter. Whether it's analog, whether it's digital, it happens before the converter. So there is no output of the converter that is, that meets the requirements of the claim. And everything that Cisco points to in both Tarachuk and Roberts happens before the converter that they assert meets the limitation. And in fact, Cisco didn't even address this portion of the decision in its brief.

[00:15:02] Speaker 04:
It claims an abuse of discretion. It's not clear in what way they claim the board abused its discretion, since it considered everything that was there. That's not even addressed in the brief either.

[00:15:14] Speaker 04:
And with regard to the second argument, with regard to the Delta's limitation, the Board was very clear in its analysis there as well, and essentially, at the end of the day, determined that Mr. Blumenthal's analysis was not representative of anything in Tarachuk, and that Frankly, the way in which it was presented lacked credibility and made credibility determinations about who presented this properly.

[00:15:45] Speaker 04:
In fact, if we look at...

[00:15:55] Speaker 04:
They said, you know, we've looked at what Mr. Blumenthal did, and, in fact, that is not representative of what's taught in Terichuk, and so Cisco failed to meet its burden. It preserved that factual question specifically for trial, and ultimately Cisco did not meet its burden.

[00:16:14] Speaker 04:
The reasoning in both opinions is sound. Cisco hasn't pointed to anything that they didn't consider, so I'm not sure what the abuse of discretion even could be. And the IPRs should be upheld. Anything further?

[00:16:37] Speaker 02:
I don't think so. I see that there's an ex parte proceeding going on and an infringement in the district court. Yeah. Are those still pending or progress is to be made?

[00:16:49] Speaker 04:
So there are two district court cases. One in Texas is stayed pending the ex parte reexams. There were 10 total on the three patents in Texas. Nine of them have been resolved. All claims have been affirmed. So there's one more left. There's an appeal hearing on that in May. And in the district court, there's a district court case pending in Delaware as well. And Judge Williams issued a claim construction as part of summary judgment and found non-infringement.

[00:17:21] Speaker 04:
So the appeal for that is just starting. So that's where we are on that one. The process is just starting for that. The decision was in November.

[00:17:32] Speaker 04:
But I think this is the end of all of the IPRs with one EPR left to go.

[00:17:38] Speaker 04:
as a result. And to this point, all claims have been affirmed.

[00:17:43] Speaker 04:
Thank you.

[00:17:49] Speaker 03:
Thank you. Ms. Oliver, you have some rebuttal time.

[00:17:54] Speaker 00:
Your Honor, just one quick point on each IPR here. So particularly on the 575 IPR, at Appendix 26, the Board does not give any substantive analysis of the calculated in advance disclosure in Tara Shook. Now, this is particularly important here because this was the crux of the Board's entire decision in this case. The entire premise was the finding that Tara Shook only discloses analog predistortion. We presented evidence for the contrary. While the Board did cite it at Appendix 26, there is no substantive analysis of whether that teaches analog predistortion.

[00:18:28] Speaker 00:
Agencies are required to provide a reasoned basis for their decisions, and given the importance of this particular issue, we would have expected a bit more discussion of this particular evidence from the board before rejecting that evidence in a fairly conclusory fashion. And the same is true with respect to the next IPRs. We simply suggest that we presented extensive discussion of evidence that was just nowhere in the board's opinion. Our experts' discussion of the right reference and expert testimony explaining how the mapping would be created and how a skilled artisan would have known, based on background mathematical principles, that this is how this mapping would have been done, and it would have been easy to compensate for the nonlinearities.

[00:19:09] Speaker 00:
And so we would just ask that the court vacate and remand these cases to the board so the board can consider all the evidence that Cisco presented.

[00:19:16] Speaker 03:
Thank you. This case is taken under submission.