[00:00:00] Speaker 06:
Our final case for today is 22-1493, Sisvel v. Sierra Wireless.

[00:00:05] Speaker 06:
Mr. Carraza, please proceed.

[00:00:08] Speaker 01:
Good morning.

[00:00:08] Speaker 01:
May it please the court?

[00:00:09] Speaker 01:
Robert Carraza on behalf of Appellant Sisvel International.

[00:00:13] Speaker 01:
First, I would just like to say thank you for the court's flexibility in allowing me to appear remotely by video conference while I'm ill.

[00:00:21] Speaker 01:
I do apologize in advance if I cough at any point and interrupt the argument.

[00:00:25] Speaker 01:
There are multiple issues on appeal, and I'm happy to answer questions on any.

[00:00:29] Speaker 01:
I would like to address two in detail, specifically the board's reasoning on combining and puncturing being taught in Chen.

[00:00:39] Speaker 01:
I would like to start with combining first, where the board's reasoning highlights the oversimplified approach that led to a lack of adequate support or explanation.

[00:00:50] Speaker 01:
Turn into page 37 of the appendix.

[00:00:54] Speaker 01:
That's where the board first addresses the post-institution argument and evidence.

[00:00:59] Speaker ?:
at the first full paragraph above the block or at the bottom.

[00:01:02] Speaker 01:
The board simply says that the evidence and the arguments that are submitted by assistable post-institution were unavailing.

[00:01:10] Speaker 01:
As we explained in our briefing, that doesn't comport the APA standards.

[00:01:14] Speaker 01:
But what I hope to do is provide a little clarity in understanding how the board came to that error.

[00:01:20] Speaker ?:
And I think the clearest way is to look at the next paragraph, where the board said instead, having considered the entirety of the record,

[00:01:27] Speaker ?:
a record, excuse me, we are persuading as we initially set forth in the institution decision.

[00:01:34] Speaker 01:
The board then cites two specific portions of Chen.

[00:01:40] Speaker 01:
Both of those have to do with accumulation.

[00:01:42] Speaker 01:
And that's a key term here, accumulation.

[00:01:46] Speaker 01:
First one is accumulating energy.

[00:01:48] Speaker 01:
And the second, which is on Apex 38, is that accumulation can be accomplished with an arithmetic logic unit.

[00:01:55] Speaker 01:
That term accumulation is key here because that's how the petitioners, the cross-content, framed this dispute.

[00:02:03] Speaker 01:
And if you look at Apex 26, for example, the first full sentence at the top of the page, the board recounts how the petitioner contended that Chen describes the process of accumulating the code symbols from the transmitted and retransmitted code blocks as interleaving, not the less precise term combining.

[00:02:24] Speaker 01:
In other words, the petitioners were telling the board that interleaving in the language of Chen is accumulating.

[00:02:34] Speaker 01:
And therefore, as the board also explained on page 36 of their opinion, turning back to that main section, in the middle of that paragraph, the board recounted again that argument, that accumulating different code symbols, interleaving, is what Chen calls.

[00:02:48] Speaker 01:
Instead of the word used combining in the 561 pattern,

[00:02:54] Speaker 01:
And so essentially what the board had here was a lexicographical paradigm that they were looking at these terms combining and interleaving through.

[00:03:04] Speaker 01:
And that led the board to its reasoning at the bottom of page 36, again, related to the accumulate aspect, where the board said, in the second body of the Chen, interleaving is the exact, sorry, combining in second, sorry, interleaving the second body of the Chen

[00:03:25] Speaker 01:
is the exact same thing as combining in the 561 bet.

[00:03:28] Speaker 06:
Mr. Garrison, can I ask a question?

[00:03:32] Speaker 06:
The board clearly found that interleaving and combining are the same thing in Chen.

[00:03:40] Speaker 06:
They found it in the institution decision, and then they found it again in the final written decision.

[00:03:46] Speaker 06:
Now, surely they could have gone into more detail on why that was the case.

[00:03:51] Speaker 06:
But what's wrong with it?

[00:03:54] Speaker 06:
I mean, how did they err here?

[00:03:57] Speaker 06:
It's not like this was new or unexpected.

[00:03:59] Speaker 06:
They made the same finding in the institution decision.

[00:04:03] Speaker 01:
They erred, Your Honor, by not addressing the evidence that was submitted post-institution by Sissible.

[00:04:09] Speaker 01:
And for example, that includes Mr. Bates' explanation on page 35 and 36 of the appendix that the board recounted, where Mr. Bates explained that interweaving

[00:04:19] Speaker 01:
is a term of art, and so is combining, and that they are not the same thing.

[00:04:24] Speaker 01:
And so what the board said instead, they focused on this idea of accumulation.

[00:04:28] Speaker 06:
But it says, OK, wait.

[00:04:29] Speaker 06:
But it actually says, even having considered Pat Noner's arguments and his response in certify, and having considered Mr. Bates's declaration of deposition testimony, and the entire record at trial, we find the argument unavailing.

[00:04:44] Speaker 06:
And it goes on to say, as we initially set forth, and it explains why,

[00:04:48] Speaker 06:
They think from Chen they are the same.

[00:04:51] Speaker 06:
So it doesn't seem to me that the board failed to address your argument.

[00:04:56] Speaker 06:
It seems like it mentioned it and rejected it.

[00:04:59] Speaker 06:
Now, you might have preferred a greater degree of explanation as to why it was rejecting the argument.

[00:05:06] Speaker 06:
That's often the case.

[00:05:08] Speaker 06:
But how did they err?

[00:05:10] Speaker 01:
I think the need for explanation and understanding of the evidence that was presented

[00:05:17] Speaker 01:
needs to be more than simply saying it's unavailing because it addresses the issues precisely.

[00:05:22] Speaker 01:
And indeed, if you look at those two references that the board uses at the bottom of 37 to 38, for example, there was confusion at the board because of this idea of accumulation.

[00:05:33] Speaker 06:
Let me understand something.

[00:05:35] Speaker 06:
Is the argument you're making today and in your brief a sufficiency argument, an APA challenge, the board failed to address our arguments, or is your

[00:05:47] Speaker 06:
argument that based on what the board said it got it wrong, meaning a merits related.

[00:05:52] Speaker 06:
Are you making an APA challenge or a merits challenge?

[00:05:56] Speaker 01:
It's an APA challenge that we presented in the brief, Your Honor.

[00:05:58] Speaker 01:
And the APA challenge was that the board simply said that our evidence post-institution was unavailable.

[00:06:05] Speaker 01:
There was no other explanation.

[00:06:06] Speaker 01:
What I hope to do today is explain why that's critical, because the board was confused and therefore had to address that evidence beyond just the procedural requirement to do so.

[00:06:17] Speaker 01:
Take, for example, the accumulation evidence that the board cited on page 37 to 38.

[00:06:22] Speaker 01:
The first one is about accumulation of energy.

[00:06:26] Speaker 01:
We know that that applies to both types of accumulation.

[00:06:30] Speaker 01:
And we can find that on 1221 of the appendix.

[00:06:32] Speaker 01:
That's discussion of energy.

[00:06:34] Speaker 01:
But critically, if you look at the next piece of evidence that the board cites, that accumulation can be accomplished with an arithmetic logic here, that is the second piece of the only two potential

[00:06:46] Speaker 01:
grounds of evidence that the board cited to say that the evidence that we presented was unavailing.

[00:06:53] Speaker 01:
That accumulation can be accomplished with an arithmetic logic unit.

[00:06:57] Speaker 01:
Critically, that's found on page 18 of Chen, which is 1219 of the record.

[00:07:04] Speaker 01:
That's in lines 37.

[00:07:06] Speaker 04:
Mr. Garza, what about the citations, the column 14 of Chen, which the board cited and are really talking about

[00:07:17] Speaker 04:
accumulating code symbol packets.

[00:07:20] Speaker 04:
And then the board makes the finding okay.

[00:07:25] Speaker 04:
It discloses accumulating packets and combining packets.

[00:07:28] Speaker 04:
That's combining the code symbols you're getting from the original transmission with whatever code symbol

[00:07:33] Speaker 04:
you're getting from the retransmission.

[00:07:36] Speaker 04:
And Chen calls that interleaving, but the board makes the fact-finding.

[00:07:41] Speaker 04:
That's the same thing that the claimed invention is doing when the claimed invention calls for combining code symbols from the original transmission with code symbols from the retransmission.

[00:07:53] Speaker 01:
So perhaps the best way to answer your question, Judge Chen, is if I back up just for a moment and try to explain simply what accumulating in Chen is.

[00:08:00] Speaker 01:
And there's no dispute about it.

[00:08:02] Speaker 01:
If you look at the claims in Chen at 1223, the claims decoding method with the last step is accumulating.

[00:08:08] Speaker 01:
Claim four, dependent claim in Chen on 1223, explains that accumulating can be combining.

[00:08:15] Speaker 01:
Dependent claim seven explains that accumulating can be interleaving.

[00:08:19] Speaker 01:
And so combining and interleaving are two forms of accumulating in Chen and in DR itself.

[00:08:26] Speaker 01:
And critically, what the board did is they mixed them together, essentially saying accumulating and, or sorry,

[00:08:32] Speaker 01:
combining and interweaving are both forms of accumulating, and therefore they're equivalent to each other.

[00:08:38] Speaker 01:
And apple and orange might both be fruit, but that doesn't mean, as a rational logical matter, they are equivalent to each other.

[00:08:44] Speaker 01:
And the board's error is especially clear in the recitation of the accumulation can be accomplished with the arithmetic logic unit.

[00:08:51] Speaker 01:
That's reliance on page 1219 of the record.

[00:08:54] Speaker 01:
That arithmetic logic unit is only used in the exemplary embodiment, which across the balance have disavowed reliance on.

[00:09:02] Speaker 01:
And that's crystal clear in that page that it's talking about combining.

[00:09:06] Speaker 01:
So what the board is simply saying is accumulating, collecting, can be done with an arithmetic logic unit, but that's wrong.

[00:09:14] Speaker 04:
So if you just accept the fact that perhaps the court would look at the references to column 14 about Chen's alternative embodiment

[00:09:28] Speaker 04:
and the board is relying at least in part on that portion of Chen discussing the accumulation of code symbols from the original transmission and the retransmissioning.

[00:09:39] Speaker 04:
And I don't want to say the word combining, but at least uniting them somehow.

[00:09:44] Speaker 04:
What is your response to the board's at least partial reliance on that passage in Chen?

[00:09:54] Speaker 04:
Why is that not combining as the term combining is used in the claimed invention?

[00:09:59] Speaker 01:
I would say that's error because what the board misunderstood, and what the paradigm was set up for them, and what they were assuming was that accumulation in Chen meant interleaving.

[00:10:10] Speaker 01:
But it's very clear that accumulation in Chen could be combined or interleaving.

[00:10:15] Speaker 01:
And on page 14 of Chen, that alternative embodiment is clearly using interleaving.

[00:10:20] Speaker 01:
That's the importance, the criticality of Mr. Bates' testimony, where he explained that in the art, interleaving is a well-known thing.

[00:10:27] Speaker 01:
It's error correction.

[00:10:29] Speaker 01:
where you take data spread across rows and you create columns.

[00:10:32] Speaker 01:
And then you permutate that information across the rows so that you reduce the rate of the error.

[00:10:38] Speaker 01:
That does, as your honor said, include some form of kind of joining, or as the board put it, collecting the symbols together.

[00:10:46] Speaker 01:
But it's important that the claims here, they don't claim interlink.

[00:10:50] Speaker 01:
They claim combining, which is known to be that simple arithmetic combination

[00:10:55] Speaker 01:
And indeed, the board recognized that as they cited to the arithmetic logic unit for support.

[00:11:01] Speaker 01:
But they made that mistake of citing the arithmetic logic unit because that's only used in CHED in the exemplary environment, which across opponents have disavowed reliance on.

[00:11:12] Speaker 01:
And I don't necessarily fault the board for it because it was confusing the record below.

[00:11:16] Speaker 01:
As we explained to the board in our server apply at Apex 566 and 571 in the record,

[00:11:24] Speaker 01:
The petitioners changed theories mid-straining.

[00:11:28] Speaker 01:
They disavowed reliance on the exemplary embodiment and this ALU.

[00:11:33] Speaker 01:
Their expert had relied on the ALU, had relied on this combiner, the 640.

[00:11:38] Speaker 01:
But then in the middle of the idea, they said, we don't need that.

[00:11:41] Speaker 01:
We're not going to rely on it.

[00:11:43] Speaker 01:
And so that, I believe, is the source of the board's confusion here.

[00:11:46] Speaker 01:
They're citing evidence from the exemplary embodiment, whereas they shouldn't have been doing it because that's not part of the invalidity approach.

[00:11:55] Speaker 01:
Indeed, I think the board's error here in kind of taking an overgeneralized view and saying apples and oranges are both fruit, and therefore an apple is an orange, can also be applied to the board's finding on puncturing itself.

[00:12:08] Speaker 01:
And so what the board essentially said was selectively transmitting adjust the code rate.

[00:12:15] Speaker 01:
There's no doubt about that.

[00:12:17] Speaker 01:
In Chen, it also teaches that puncturing adjusts the code rate.

[00:12:22] Speaker 01:
And the problem here is that the claims of the 561 patent don't claim selective transmit.

[00:12:28] Speaker 01:
They claim puncture, which is a specific type of adjusting the code rate.

[00:12:34] Speaker 01:
So in that sense, the 561 here is claiming an apple.

[00:12:37] Speaker 01:
Just because an apple might also be a fruit doesn't mean that it's an orange.

[00:12:41] Speaker 01:
And that's what the board reasoned, is that they both have this kind of idea of reducing the code rate.

[00:12:47] Speaker 01:
And I think the clarity there is found on page 14 of Chen, 12, 15 of the record.

[00:12:52] Speaker 01:
where Chen explains that if you selectively transmit, you have two generators, and it creates a code rate of one half or one two.

[00:13:00] Speaker 06:
Mr. Garza, you're well into your rebuttal time.

[00:13:01] Speaker 06:
Did you want to save it?

[00:13:04] Speaker 01:
I do.

[00:13:04] Speaker 01:
I will quickly wrap up, Your Honor.

[00:13:06] Speaker 01:
I appreciate that.

[00:13:07] Speaker 01:
In the next paragraph, though, they say that the puncturing the code rates, the same generators, if you're puncturing, have a code rate of one two.

[00:13:14] Speaker 01:
If it's punctured, it's three four.

[00:13:16] Speaker 01:
So it's very clear that Chen even makes clear that selectively transmitting

[00:13:21] Speaker 01:
is a different way of changing your code rate than puncturing.

[00:13:25] Speaker 01:
And that's the board's error there, the same assumption, the high-level assumption they made with combining.

[00:13:30] Speaker 01:
And with that, I would like to reserve the remainder of my rebuttal time to answer some questions.

[00:13:35] Speaker 06:
That sounds good.

[00:13:35] Speaker 06:
Thank you very much, Mr. De La Rosa.

[00:13:38] Speaker 05:
Thank you.

[00:13:42] Speaker 05:
Thank you.

[00:13:43] Speaker 05:
Please, the court.

[00:13:45] Speaker 05:
The parties have obviously raised a number of issues on appeal.

[00:13:47] Speaker 05:
my colleague has spent quite a bit of time on their issues on appeal so I want to make sure I briefly address those but I will do my best to move quickly so we have time to address cross-appellants issues as well.

[00:13:57] Speaker 05:
So turning first to combining and interleaving.

[00:14:02] Speaker 05:
Council has presented some new arguments based on the claims of Chen that haven't been presented below or in my understanding.

[00:14:08] Speaker 04:
Did the board make a mistake at 837, 838 when it seemed to be relying on the idea of accumulating energy as being

[00:14:17] Speaker 04:
the matching up with the combining?

[00:14:21] Speaker 05:
Your Honor, I believe the board had it spot on on Appendix 36 when they addressed, just as you did as well, page 14 of Chen at 1215, where it discloses exactly the same type of accumulation that we're talking about here.

[00:14:34] Speaker 05:
And I can hear from Chen in particular, where it describes the code symbols for the retransmitted packets.

[00:14:42] Speaker 05:
And this is at 1215, line 9, or 8.

[00:14:48] Speaker 05:
are interleaved not combined with the corresponding code symbols from prior transmissions.

[00:14:54] Speaker 05:
So what they're talking about and they're providing in great detail in the particular embodiment

[00:14:59] Speaker 05:
that we have consistently, and I respectfully disagree, have ever changed course on.

[00:15:03] Speaker 05:
We have consistently always relied on this embodiment, presented on 1215 for anticipating the claims of the 561 patent, describes taking one set of code symbols from the original transmission, a different, keyword different set of code symbols from the retransmission, and mushing them together.

[00:15:21] Speaker 05:
Use whatever term you want, whether it's combining, interleaving, accumulating, collating.

[00:15:26] Speaker 05:
I think I've seen them all across the briefing.

[00:15:29] Speaker 05:
bringing those two different packets together.

[00:15:31] Speaker 05:
And that's exactly the analysis that the board provided at 36 when it talks about Chen disclosing exactly that in the connection with the particular embodiment that the Cross Appellants were relying on.

[00:15:44] Speaker 04:
So what do you think was going on at 837, 838 where they're talking about accumulating energy?

[00:15:51] Speaker 05:
I think the board was trying to refer to and

[00:15:54] Speaker 04:
That's a different embodiment than the embodiment you're relying on.

[00:15:57] Speaker 05:
Right.

[00:15:57] Speaker 05:
I think the board had intended to refer to the accumulation description at page 18 of Chen, which is 1219 of the appendix, where it did explain that accumulation can mean these different things.

[00:16:08] Speaker 05:
But there's no doubt that when we're talking about the particular embodiment that we've consistently relied on, it is this second form where you've got the different code symbols from the transmission and retransmission being brought together.

[00:16:21] Speaker 05:
and that the Board's analysis of that is detailed and does describe it here in the final written decision on 36.

[00:16:30] Speaker 05:
And I'd also point out that the Board's analysis is recorded more than just by one page.

[00:16:35] Speaker 05:
The Board has a helpful explanation in Appendix 26 and then walks through its analysis at 35 to 38.

[00:16:40] Speaker 05:
And so while the Board might mention

[00:16:42] Speaker 05:
the other form of accumulation.

[00:16:45] Speaker 05:
It in no way undercuts the thorough analysis it did of the actual form of these symbols being brought together that we relied on below.

[00:16:53] Speaker 04:
Is puncturing and selective transmission different from each other?

[00:17:00] Speaker 05:
Not at all.

[00:17:00] Speaker 05:
And we provided below a very detailed 200 plus page declaration walking through the state of the art.

[00:17:08] Speaker 05:
Puncturing is anything that removes

[00:17:10] Speaker 05:
coded data symbols.

[00:17:12] Speaker 05:
You've got the channel coding, which is adding symbols, and puncturing is tailoring it down.

[00:17:16] Speaker 05:
Whether you're taking away a little or none, that's puncturing, and that is what is being done here.

[00:17:20] Speaker 04:
Right, but there's nothing in Chen that actually states we're going to generate all the code symbols, and then we're going to delete two of the four code symbols for each data bit.

[00:17:34] Speaker 05:
I respectfully disagree that there's nothing in here to a posita that says that.

[00:17:37] Speaker 04:
Well, I'm just saying, expressly says that to a common reader like myself.

[00:17:43] Speaker 04:
I didn't see a sentence that says that.

[00:17:46] Speaker 05:
I understand the point you're making, but I would refer to Dr. Kakeyas' detailed explanation in his declaration that walks through how a convolutional encoder works.

[00:17:56] Speaker 05:
And by virtue of it being a convolutional encoder, a posita would understand it is necessarily always generating this

[00:18:03] Speaker 05:
one-fourth coded data block, always.

[00:18:06] Speaker 05:
And so if you're saying that you have a one-half coded data block coming out of it, it means that you have cut or punctured, and you've removed some of the coded data symbols to reduce down to that one-half coded data block.

[00:18:19] Speaker 05:
And so I appreciate the point you're making, Your Honor, but I respectfully disagree that to a Posita, and there was ample evidence below demonstrating how a Posita would have understood Chen's disclosure of a convolutional encoder

[00:18:30] Speaker 05:
to convey this already punctured coded data block.

[00:18:37] Speaker 04:
Means for detecting?

[00:18:38] Speaker 05:
Means for detecting.

[00:18:40] Speaker 05:
Happy to jump there.

[00:18:41] Speaker 05:
Thank you, Your Honor.

[00:18:42] Speaker 05:
So on indefiniteness, as you're all aware, the board declined to address the unpatentability of independent claims five and 10.

[00:18:49] Speaker 05:
Now, we had provided a detailed chart in the materials at appendices 967 to 969 that shows that these claims are substantively identical to claim one, which is already found to be invalid as obvious over single reference obviousness.

[00:19:10] Speaker 05:
In view of the fact that the mirror identical claims, with the only differentiation being that instead of claiming a need for retransmission, simply a means for detecting, that merely restating that shouldn't allow these claims to survive the IPR process.

[00:19:30] Speaker 05:
the parties and the board.

[00:19:32] Speaker 04:
We still have to go through the 112F exercise.

[00:19:35] Speaker 04:
We do, we do.

[00:19:35] Speaker 04:
Finding corresponding structure.

[00:19:37] Speaker 05:
Happy to jump straight into that.

[00:19:39] Speaker 05:
And so we believe that we presented below enough evidence that there was at least some corresponding structure, enough corresponding structure that the court could have proceeded through with its infidelity analysis under 102-103.

[00:19:50] Speaker 02:
So you want to remand?

[00:19:52] Speaker 05:
Yes, Your Honor.

[00:19:53] Speaker 02:
That's an intel case in the rest, right?

[00:19:55] Speaker 05:
Exactly.

[00:19:56] Speaker 02:
So was that an issue on which you had a burden to raise that issue?

[00:20:02] Speaker 02:
And you didn't argue to the board that it had this duty?

[00:20:08] Speaker 05:
Well, we argued below that the board had enough issues.

[00:20:10] Speaker 02:
So the first question under the intel is whether or not it's impossible, right?

[00:20:17] Speaker 02:
Once you found that there was no structure or so,

[00:20:22] Speaker 02:
claim fails under means plus function, can it still be looked at for 103?

[00:20:26] Speaker 02:
So we don't know the answer to the impossibility question.

[00:20:31] Speaker 05:
I would say, Your Honor, that the board's decision to the contrary suggests that it is very much possible.

[00:20:35] Speaker 02:
Suggested.

[00:20:36] Speaker 02:
That's not the same.

[00:20:38] Speaker 02:
No, they did not.

[00:20:39] Speaker 02:
Your remand would be for two purposes.

[00:20:41] Speaker 02:
One, first for the board to decide if it's impossible, and then if it's not impossible to do the trick.

[00:20:47] Speaker 05:
Correct, Your Honor.

[00:20:48] Speaker 02:
OK, so my question is, did you have a burden to raise that?

[00:20:53] Speaker 02:
Or did you waive that argument?

[00:20:56] Speaker 02:
You didn't ask the board for that when you were down there.

[00:20:59] Speaker 05:
So I had reviewed, based on the court's instruction in Intel v. Qualcomm, that the board has already been challenged with this task of identifying whether or not it was possible.

[00:21:09] Speaker 05:
I guess I would see that as the board's responsibility.

[00:21:12] Speaker 05:
But I will say, petitioners did provide the answer.

[00:21:15] Speaker 02:
There is an independent obligation of the board.

[00:21:17] Speaker 02:
It's not an issue that you have to raise.

[00:21:20] Speaker 02:
That would be my, yes, Your Honor.

[00:21:22] Speaker 02:
I'm sorry, coming out of SAS.

[00:21:23] Speaker 05:
Yes, Your Honor.

[00:21:27] Speaker 05:
So I will go back to the sufficiency of what we did demonstrate.

[00:21:31] Speaker 05:
So whether or not it provides the adequate corresponding structure, we have and were able to, relying on the 561 specification, point to some corresponding structure.

[00:21:42] Speaker 05:
that arc that is clearly associated with the means for detecting 224.

[00:21:47] Speaker 04:
Are you relying just on the three letters arc in the box for means 224 in figure two?

[00:21:54] Speaker 04:
Or are you really relying on column two where it says hybrid FEC slash arc?

[00:22:02] Speaker 05:
I would argue it's the combination of both.

[00:22:04] Speaker 04:
I don't know what that means, the combination of both.

[00:22:08] Speaker 04:
The background sections already says ARQ in there.

[00:22:12] Speaker 04:
So how, in what way are you relying on both?

[00:22:17] Speaker 05:
Well, so you're exactly right.

[00:22:19] Speaker 05:
Those are the two spots in the specification.

[00:22:21] Speaker 05:
So the arc in that box is certainly clearly one-for-one associated in the means for detecting 224, but read in the context of the entire specification, we believe it was understood that the claimed invention of the 561 patent was directed to GSM technology and therefore would have therefore also understood

[00:22:42] Speaker 05:
it to disclose the use of that hybrid arc that was discussed in the background section.

[00:22:47] Speaker 05:
We could also have a conversation about whether or not that same posita would have already known that without having to read that basic knowledge in the background as well.

[00:22:55] Speaker 04:
Just to confirm, GSM already existed.

[00:22:59] Speaker 04:
That was in the background.

[00:23:00] Speaker 04:
Absolutely, Your Honor.

[00:23:01] Speaker 04:
Not the particular version we're fighting over here, but GSM was already in there.

[00:23:04] Speaker 04:
Absolutely.

[00:23:05] Speaker 04:
A skilled artisan, when they see ARC, and then they see in the background hybrid FEC slash ARC, they would understand that the means for detecting contemplated by this inventor was hybrid FEC slash ARC?

[00:23:18] Speaker 05:
Absolutely, which would include FEC and the CRC algorithm.

[00:23:22] Speaker 05:
and would have understood that's how you would implement a means for detecting.

[00:23:25] Speaker 05:
Now, I'm not going to stand here today and try my darnedest to persuade you that this claim isn't indefinite.

[00:23:30] Speaker 05:
I suspect it is.

[00:23:31] Speaker 05:
The question in my mind is, is there enough structure here that the board could have and it was possible for the board?

[00:23:37] Speaker 04:
And I guess what you're saying is when you identify by name a known existing software protocol,

[00:23:48] Speaker 04:
that's enough to identify the algorithm as corresponding structure for a means plus function limitation.

[00:23:55] Speaker 04:
You don't need to go further and actually explain the known details and operational steps of which that identified protocol comprises.

[00:24:08] Speaker 04:
Is that the argument?

[00:24:09] Speaker 05:
I would say it's case specific.

[00:24:11] Speaker 05:
And here where we're talking about a known mandatory specification, where it would have been known,

[00:24:17] Speaker 05:
I think this is the case where there may be an argument made that disclosure of just the GSM protocol with ARC should be sufficient, but certainly not.

[00:24:26] Speaker 05:
I'm not attempting in any way to argue that's always the case.

[00:24:29] Speaker 02:
In your argument, are you led to a specific algorithm or to a POSA who would know, supposedly, what algorithm to write?

[00:24:39] Speaker 05:
It's the latter, Your Honor.

[00:24:40] Speaker 02:
The POSITA would know, having looked at... And is there more than one possible algorithm that could be written?

[00:24:46] Speaker 05:
I suspect that there would be different ways to write the software for running the CRC check and the forward error correction, Your Honor, yes.

[00:24:55] Speaker 02:
So when you're identifying the structure in a maintenance function claim, the idea is to identify a structure specifically so that one can find equivalence thereof, correct?

[00:25:05] Speaker 05:
Correct, Your Honor.

[00:25:05] Speaker 02:
That's the whole idea.

[00:25:07] Speaker 02:
So if the possibility was that you could be led to any number of actual algorithms,

[00:25:16] Speaker 02:
How are you going to know which one is the equivalent of the many that might have been picked?

[00:25:23] Speaker 05:
Well, Your Honor, you don't have to persuade me.

[00:25:24] Speaker 05:
The claim is indefinite.

[00:25:26] Speaker 05:
I agree with you on that.

[00:25:27] Speaker 02:
No, no, no.

[00:25:27] Speaker 02:
I'm trying to get it in terms of whether or not there were.

[00:25:30] Speaker 02:
The question here was, in these claims, there's a requirement for solving.

[00:25:35] Speaker 02:
You've got to write an algorithm, right?

[00:25:37] Speaker 00:
Right.

[00:25:38] Speaker 02:
And it really sort of means give us a specific algorithm so then we can decide for equivalent's purpose whether somebody found

[00:25:45] Speaker 02:
an equivalent because your patentee wants to be able to reach equivalents.

[00:25:50] Speaker 02:
But the patentee shouldn't be able to reach equivalents of equivalents of equivalents because I have never established the ground point, right?

[00:25:56] Speaker 05:
Agreed, Your Honor.

[00:25:57] Speaker 02:
But wouldn't that suggest that it's insufficient for purposes of disclosing the algorithm simply to disclose a place where people can write algorithms and they can write more than one?

[00:26:12] Speaker 05:
So I agree with the point you're making, Your Honor.

[00:26:14] Speaker 05:
And for cross-appellant's purposes, our point is merely that there was enough disclosure here that a postita would have had some understanding of this.

[00:26:21] Speaker 02:
Well, I hear you almost to be saying you're willing to wave off the argument as to whether or not this was a satisfactory mens plus function claim, because you believe the duty is sufficient, and you want to do it on a 102, 103, because it doesn't do you any good.

[00:26:34] Speaker 02:
to have a funk as a means by something?

[00:26:37] Speaker 05:
Yes.

[00:26:38] Speaker 05:
In fact, that's the opening I was hoping for, to point to the interests that the court has identified that are at issue here, the interests of the parties.

[00:26:45] Speaker 02:
If we agree with you on sending it back, we could say we don't need to decide.

[00:26:49] Speaker 05:
Exactly.

[00:26:50] Speaker 05:
Although I can't help but take a moment, if I can, to point out that Chen does disclose this exact same thing to the extent

[00:26:56] Speaker 05:
that the 561 patent discloses that means you're detecting the truth.

[00:27:00] Speaker 06:
But the weird thing is, it's such a strange posture for this to be in.

[00:27:04] Speaker 06:
You have the board basically saying it can't find structure, and so they can't technically hold the claims indefinite, but they have effectively done that.

[00:27:15] Speaker 06:
And then I have you, who want the claims to be invalid, arguing to me that there is in fact sufficient structure so that they should have reached the prior art.

[00:27:23] Speaker 06:
issues.

[00:27:23] Speaker 06:
That sound about right?

[00:27:24] Speaker 05:
You are exactly right, Your Honor.

[00:27:26] Speaker 05:
It's a fine line.

[00:27:26] Speaker 05:
I'm trying to walk here.

[00:27:27] Speaker 05:
But there is an interest to the parties, to the court, to the judicial process.

[00:27:30] Speaker 06:
So if there's a district court litigation, there's still a couple people.

[00:27:33] Speaker 05:
Exactly.

[00:27:34] Speaker 05:
Exactly, Your Honor.

[00:27:34] Speaker 05:
And there still remains the ability for patent owner to continue to assert these claims against others because they have technically survived the IPR process.

[00:27:42] Speaker 04:
I guess Judge Clevenger raises an interesting point, which is how definite does the identified

[00:27:52] Speaker 04:
software protocol have to be in order to satisfy the corresponding structure requirement for 112f, 1126.

[00:28:03] Speaker 04:
You know, if another claim had said means for analyzing data, means for combining data, means for transmitting data, and then the specification said a processor, a conventional processor will

[00:28:22] Speaker 04:
perform those functions of combining, analyzing, transmitting data.

[00:28:29] Speaker 04:
We would probably all say that's good enough.

[00:28:32] Speaker 04:
But we also all know that there are like 10,000 different processors out there.

[00:28:38] Speaker 04:
And no one processor is truly identical.

[00:28:42] Speaker 04:
They have different little features attached to each of these 10,000 processors.

[00:28:48] Speaker 04:
But we wouldn't say, holy smokes, we didn't really

[00:28:51] Speaker 04:
go out of your way and give us a truly developed detailed picture of which particular processor among the class of processors you are saying is your corresponding structure.

[00:29:05] Speaker 04:
So I'm trying to figure out now we go back to your case in this software world where things are just more abstract and things can go in a lot of different directions and then

[00:29:16] Speaker 04:
as your expert seemed to suggest.

[00:29:18] Speaker 06:
Is there a question in here, Joseph?

[00:29:21] Speaker 04:
Yeah.

[00:29:22] Speaker 06:
I'm just curious.

[00:29:22] Speaker 06:
I'm waiting for it.

[00:29:23] Speaker 04:
You shouldn't be laughing at that.

[00:29:28] Speaker 04:
You can laugh if you want, I guess.

[00:29:30] Speaker 04:
My point is that if this FEC-ARQ thing is slightly indeterminate, like maybe there's some core traits that are

[00:29:39] Speaker 04:
common throughout all the different versions of the protocol.

[00:29:42] Speaker 04:
But there's a lot of different versions of doing it.

[00:29:45] Speaker 04:
Is that going to destroy the ability to say, oh, this is a sufficient corresponding structure despite this degree of indeterminacy?

[00:29:55] Speaker 06:
Well, it looks like your time is up.

[00:29:56] Speaker 06:
So thank you for your time.

[00:29:57] Speaker 06:
I'm kidding.

[00:29:58] Speaker 06:
I'm kidding.

[00:29:58] Speaker 06:
Go ahead and answer if possible.

[00:30:02] Speaker 05:
I appreciate the line you're drawing, and I guess I'm being a little over simplistic here, but I keep coming back to... This is very nuanced.

[00:30:09] Speaker 05:
It is very nuanced, but at core, whatever it is in the 561 patent, there's a one-for-one correlation in Chen.

[00:30:18] Speaker 05:
And so I guess I don't think it's maybe necessary to delve in and figure out where that line can be drawn.

[00:30:24] Speaker 05:
I appreciate that, Your Honor, you want to draw that line.

[00:30:28] Speaker 05:
I'm not sure where it would be.

[00:30:29] Speaker 05:
I just know that

[00:30:31] Speaker 05:
It is one for one in Chen disclosed.

[00:30:34] Speaker 05:
And therefore, these claims should be emerging from the IPR process.

[00:30:40] Speaker 05:
OK.

[00:30:40] Speaker 06:
Thank you, counsel.

[00:30:42] Speaker 06:
Mr. Gayarza has some rebuttal time.

[00:30:44] Speaker 06:
How much time is left?

[00:30:49] Speaker 06:
One minute.

[00:30:49] Speaker 06:
And we went over by two minutes with her.

[00:30:51] Speaker 06:
So give him four minutes on the clock, please.

[00:30:55] Speaker 01:
Thank you, Your Honors.

[00:30:56] Speaker 01:
I appreciate it.

[00:30:58] Speaker 01:
I'm short on time.

[00:30:58] Speaker 01:
I don't necessarily want to open a can of worms for cross rebuttal, but I do want to just quickly address the means for detecting and the procedural problem that I see when I picked up the briefs here.

[00:31:09] Speaker 01:
The board on Apex 59 said that the means for detecting and the lack of proven structure there was an alternative basis, an additional basis.

[00:31:18] Speaker 01:
But if you look at page 17 of the board's opinion, what the board described the lack of explanation to be,

[00:31:24] Speaker 01:
was petitioner has not shown that the 561 patent presents an algorithm for how the error detection code detection error.

[00:31:33] Speaker 01:
Now that's been the focus of the yellow brief and that's also been the focus of the gray brief, a gray of 24.

[00:31:37] Speaker 01:
They cite Apex 937 and Apex 986-89 as evidence to support that.

[00:31:45] Speaker 01:
The problem is that the board didn't stop there.

[00:31:47] Speaker 01:
The board said,

[00:31:49] Speaker 01:
Did the petitioner explain the circumstances under which the error correcting code cannot correct errors and what constitutes sufficient certainty?

[00:31:59] Speaker 01:
That alternative function, that additional function has never been addressed and it's not addressed by their expert nor was it addressed in the briefing itself.

[00:32:07] Speaker 01:
That's a straightforward basis for affirmation by this court.

[00:32:10] Speaker 06:
I have a procedural question.

[00:32:12] Speaker 06:
Mr. Yarza, I don't understand.

[00:32:14] Speaker 06:
The board effectively found that it couldn't identify an algorithm for a 112-6 limitation in this patent.

[00:32:23] Speaker 06:
They basically rejected the FEC ARC combination.

[00:32:28] Speaker 06:
Is that right?

[00:32:28] Speaker 01:
I believe so, yes.

[00:32:30] Speaker 06:
Wouldn't that render these two claims indefinite?

[00:32:34] Speaker 01:
On the record that was presented,

[00:32:37] Speaker 01:
preventing the board from doing it.

[00:32:39] Speaker 01:
It was the across-the-field firm, and they did not present the evidence to show that structure.

[00:32:46] Speaker 01:
It was not our burden to come forward with our own expert to explain how those algorithms work and where they could work.

[00:32:53] Speaker 01:
I believe if presented in trial, we could do that.

[00:32:56] Speaker 01:
But that wasn't our burden to do that here.

[00:32:58] Speaker 01:
And indeed, more fundamentally, the board didn't just say it was an algorithm for detecting errors in a code, the means

[00:33:06] Speaker 01:
for detecting a need for retransmission also encompasses, as the board found, the fact that the error correcting code cannot correct errors occurring on the channel with sufficient certainty.

[00:33:18] Speaker 01:
In other words, it detects the code, or it detects it and it could fix it, but it couldn't fix it.

[00:33:24] Speaker 06:
I'm just so baffled.

[00:33:26] Speaker 06:
Why do you want us to leave in place a decision by the board that on this record there was insufficient disclosure of an algorithm to satisfy the structural requirements of 112.6?

[00:33:40] Speaker 06:
I don't understand.

[00:33:42] Speaker 01:
Because we do not believe that their expert actually sufficiently showed the structure.

[00:33:47] Speaker 01:
Could that expert have?

[00:33:49] Speaker 01:
I believe they could have, but they did not.

[00:33:51] Speaker 01:
That was a deficiency in their case that they bore the burden.

[00:33:54] Speaker 04:
Would it be a waste of my time to ask you what is the corresponding structure for your means for detecting claim?

[00:34:00] Speaker 01:
I'm not an expert in the field, Your Honor, but I'm sure an expert at trial could explain that with adequate opportunity.

[00:34:07] Speaker 01:
But it wasn't our burden before the board.

[00:34:09] Speaker 01:
But it was their burden on appeal to appeal not only the error detection code, but the board's further finding that they failed to explain the circumstances under which the error correcting code cannot correct errors and will constitute sufficient certainty.

[00:34:24] Speaker 01:
That was not addressed by their expert at all in what I can see that they're citing to this court on apex 937, 986.

[00:34:31] Speaker 01:
So that deficiency is clear.

[00:34:33] Speaker 01:
Turning quickly into remaining time that I have, I would like to address your question, Judge Chen, on the combined limitation on 12-15 of the record, where in the Chen's description of the alternative embodiment, it never describes interleaving as anything except accumulation.

[00:34:51] Speaker 01:
And that's the criticality of Mr. Bates' testimony here.

[00:34:54] Speaker 01:
He said, in this field, interleaving is a special type of accumulating.

[00:34:59] Speaker 01:
And it's not the combining that's claimed in the patent.

[00:35:01] Speaker 01:
And he explained that.

[00:35:03] Speaker 01:
And essentially, what the board did is they said, we disagree on page 36.

[00:35:08] Speaker 01:
And they did so based on the words that were being used.

[00:35:11] Speaker 01:
In other words, Chen was not using the word interleaving as understood in the art.

[00:35:16] Speaker 01:
And the patent wasn't using the word combining as understood in the art.

[00:35:20] Speaker 01:
And they never explained how that could possibly be.

[00:35:24] Speaker 01:
Combined also with their erroneous reliance on 37 to 38 on ALUs, which did come up from their expert at 1091 and 1114, their expert relied on that also, led the board to that error.

[00:35:38] Speaker 01:
The board is clearly wrong to rely on ALUs as support for disclosing accumulation being interweaving in the Chen reference, unless there is

[00:35:56] Speaker 01:
and validate and affirm the remain.

[00:35:58] Speaker 06:
Okay, thank you, Mr. Gajarza.

[00:36:00] Speaker 06:
Ms.

[00:36:00] Speaker 06:
Merrill, I'll give you one minute limited to responding to what Mr. Gajarza had to say about the cross-appeal.

[00:36:08] Speaker 05:
Thank you, Your Honor.

[00:36:11] Speaker 05:
Thank you.

[00:36:11] Speaker 05:
I will turn back to the court's decision in Intel v. Qualcomm 21F4801

[00:36:18] Speaker 05:
In that case, the patent challenger had agreed that the claims were indefinite, but the court had found that the board still had an obligation to determine if there was sufficient structure.

[00:36:27] Speaker 05:
So while I appreciate my friend's position that it was somehow a failing on the part of cross-appellants, the board had an obligation to do its best to find sufficient structure.

[00:36:36] Speaker 05:
And not only that, based on the structure that it was able to find, and we did present and the board did make some findings as to structure,

[00:36:44] Speaker 05:
then determine whether or not it was possible to assess something.

[00:36:47] Speaker 04:
What about what your opposing counsel said about A17, about how there is a failure to find an algorithm for detecting an error, and in addition, an algorithm for figuring out whether it's possible to or not possible to correct any errors?

[00:37:06] Speaker 05:
I believe this is a new argument, Your Honor.

[00:37:07] Speaker 05:
And so we're talking about appendix page 913, you said?

[00:37:10] Speaker 03:
Page 17 of the board decision.

[00:37:12] Speaker 05:
Oh, 17 of the board's decision.

[00:37:19] Speaker 03:
post-poll sentence of A-17.

[00:37:26] Speaker 03:
Petitioner is not shown.

[00:37:29] Speaker 05:
Petitioner is not shown.

[00:37:30] Speaker 05:
So I respectfully disagree.

[00:37:34] Speaker 05:
We've got extensive testimony from Dr. Kakaus, both in his first declaration and the second declaration, that explain how the posita would understand, and that's the court's test, having presented some evidence, which is the ARQ, as Your Honor has already pointed out,

[00:37:49] Speaker 05:
So then the board is then able to look to export testimony as to what Episcida would have understood from it.

[00:37:55] Speaker 05:
And Dr. Kakeyas did provide detailed explanation in his declaration as to what that CRC algorithm is, as understood by Episcida.

[00:38:04] Speaker 05:
Okay.

[00:38:06] Speaker 05:
Thank you, counsel.

[00:38:07] Speaker 06:
This case is taken under submission.