`Entered: July 1, 2024
`
`Trials@uspto.gov
`571-272-7822
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`________________
`
`LG ELECTRONICS, INC.,
`Petitioner,
`
`v.
`
`CONSTELLATION DESIGNS, LLC,
`Patent Owner.
`________________
`
`IPR2023-00319
`Patent 10,693,700
`________________
`
`Record of Oral Hearing
`Held: April 17, 2024
`________________
`
`
`
`Before BRENT M. DOUGAL, MICHAEL T. CYGAN, AND
`SCOTT RAEVSKY, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`
`
`IPR2023-00319
`Patent 10,693,700
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`USMAN KHAN, Ph.D.
`W. KARL RENNER
`JEREMY MONALDO
`PATRICK DARNO
`Fish & Richardson, P.C.
`1000 Maine Ave SW Ste 1000
`Washington, DC 20024
`(202) 626-6383 (Khan)
`khan@fr.com
`axf-ptab@fr.com
`jjm@fr.com
`darno@fr.com
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`WILLIAM MEUNIER, ESQ.
`MICHAEL RENAUD, ESQ.
`KEVIN C. AMENDT, ESQ.
`Mintz, LLP
`One Financial Center
`Boston, MA 02111
`617-348-1845 (Meunier)
`WAMeunier@mintz.com
`mtrenaud@mintz.com
`keamendt@mintz.com
`
`
`
`
`The above-entitled matter came on for hearing Wednesday, April 17,
`
`2024, commencing at 9:20 a.m. MST, at the Denver, Colorado, USPTO
`Regional Office/via Video-conference.
`
`
`2
`
`
`
`IPR2023-00319
`Patent 10,693,700
`
`
`P R O C E E D I N G S
`- - - - -
`
`9:20 a.m.
`JUDGE DOUGAL: Good morning. So, first of all, we apologize for
`
`the delay. I'm informed there was some mix-up on the court reporter.
`Generally, we would have the court reporter right here in person. We
`apologize that that did not happen today. But we have a court reporter
`online.
`
`Because of that hiccup, we will need to be a little more cognizant of
`the fact the court reporter is not here in person. Try to maybe talk a little
`slower. And that also means that they don't have, or I'm sure they're
`probably getting them, the demonstratives and the other things, but they may
`not be as up-to-speed as our usual court reporters. So, just let's keep that in
`mind.
`So, I'll do our official start now.
`
`Good morning. I'm Judge Dougal. I have with me Judge Raevsky,
`
`and then, on the screen here is Judge Cygan. This is IPR 2023-00319
`between LG Electronics, Incorporated, and Constellation Designs, LLC.
`
`I hope everybody's in the right place.
`
`Okay. So, let's start with appearances. So, yes, Petitioner on this side
`and Patent Owner. So, who do we have for Petitioner?
`
`(Off-microphone comments.)
`
`JUDGE DOUGAL: Great.
`
`Patent Owner?
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`3
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`IPR2023-00319
`Patent 10,693,700
`
`MR. MEUNIER: Good morning, Your Honors. My name is Bill
`
`Meunier from Mintz. I'm here with Kevin Amendt, also from Mintz, on
`behalf of Constellation Designs, the Patent Owner.
`
`Thank you.
`
`JUDGE DOUGAL: All right. I'm getting the message that maybe
`your microphone wasn't on. So, I don't think they heard you. So, do you
`want to repeat?
`
`MR. KHAN: Good morning, Your Honors. My name is Usman
`Khan. I'm here with my colleagues Jonathan Driesslein, Jeremy Monaldo,
`and Karl Renner from Fish & Richardson.
`
`JUDGE DOUGAL: Thank you. And that was Petitioner.
`
`Okay. Obviously, as we just experienced, Judge Cygan and our court
`reporter will hear you if you speak into the microphone. Generally
`speaking, you're going to be at the podium right at the microphone. So, it
`won't be an issue. But just make sure that it's on.
`
`As you likely know, even though we have the demonstratives here, we
`also have them at our computers and have reviewed them. We have the
`entire file. But make sure to say the slide number that you're on, so
`everyone can follow along, and it makes for a better transcript.
`
`Let's see, we've got 45 minutes of time for each side today. Petitioner
`will start first. All right. Because we had this 15-20 minute delay, I want to
`confirm whether anyone has any problems with flights or other time
`constraints that might cause a problem.
`
`Petitioner?
`
`MR. KHAN: No concerns from Petitioner.
`
`JUDGE DOUGAL: Okay. Patent Owner?
`
`4
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`IPR2023-00319
`Patent 10,693,700
`
`MR. MEUNIER: Since we're starting now, I think we're okay.
`
`JUDGE DOUGAL: Okay, okay. Good. Okay. Good. Let's see.
`
`With objections, we ask that you not object during the presentation of the
`other party's argument. Please save those until your time.
`
`Petitioner, would you reserve any time for rebuttal?
`
`MR. KHAN: I'd like to request 15 minutes, please.
`
`JUDGE DOUGAL: Okay. Great. It's already set up.
`
`And, Patent Owner, you can decide at your time, but do you have a
`general idea of what you're thinking right now? Do you want a rebuttal?
`
`MR. MEUNIER: Right now, I probably will just save 5 minutes,
`Your Honors.
`
`JUDGE DOUGAL: Okay. Okay. Any questions?
`
`(No audible response.)
`
`JUDGE DOUGAL: I just wanted to confirm that everything's
`working out with the court reporter. But I guess, for our people online, I
`don't know if the court reporter can talk to us or message us, but I just want
`to confirm that he or she can hear us and is recording everything and not
`having any issues.
`
`(Off-the-record comments with the court reporter.)
`
`JUDGE DOUGAL: I guess we'll wait a few minutes. You're free to
`speak among yourselves, or whatever.
`
`(Pause.)
`
`JUDGE DOUGAL: All right. Let's go ahead.
`
`Yes, Petitioner, do you want to come up to the podium and get ready?
` And when you're ready, let me know. I do have a couple of questions to get
`things started with. All right.
`
`5
`
`
`
`(Pause.)
`JUDGE DOUGAL: People are typing messages.
`Okay, I think we're good. I think we're good. So, we'll go ahead and
`
`IPR2023-00319
`Patent 10,693,700
`
`
`
`
`start.
`One thing that I don't know we've really talked about yet in the
`
`briefing, I just want to clarify, is the question of, who has the burden to
`establish written description support, or the lack thereof, and what that
`burden is?
`
`So, I'll start off with those questions, and we can proceed whenever
`you are ready.
`
`
`
`DR. USMAN: Okay.
`
`JUDGE DOUGAL: And obviously, if that’s not a question that
`you’re prepared to talk about your colleagues in the back can think about it
`and look to the papers and we’ll talk about it later too.
`
`DR. USMAN: Thank you, Your Honors. If it may please the Board,
`before I begin can I just check in if my voice is clear? Being heard by
`everyone?
`
`(No audible response.)
`
`DR. USMAN: Thank you. Your Honors, I’ll start with maybe
`addressing your question, and then proceeding with some talking points.
`
`If we can turn to Slide 5 please. In the briefing we mentioned a few
`cases. This is the Indivior v. Reddy’s Labs case. What this case specifies is
`that the written description requirement requires a statement of the
`invention. And that the disclosure must be clear and unambiguous.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`6
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`IPR2023-00319
`Patent 10,693,700
`
`In fact, the, yes. The patent must demonstrate that the inventors had
`
`possession of the invention. It must demonstrate that. Now, it is generally
`upon the Petitioner to show that there is a lack of written description
`support. And once Petitioner has offered arguments then it is on the Patent
`Owner to rebut those arguments.
`
`And what we see in this case, and also in the Lockwood case, is that
`the requirement is not one of conjecture. That a inventor may have
`possessed this. It is that there must be clear description of the claimed
`features being supported.
`
`And as noted here, it is denying invitation for hunting, for a hunting
`license or a search expedition where a reader is required to go through
`different parts of the disclosure and patch together the claimed invention. It
`should be clear, it should be unambiguous. And I hope to show you today in
`our discussion that that is simply not the case here. There is no written
`description support.
`
`If I can continue? Did I answer your question?
`
`(No audible response.)
`
`DR. USMAN: Thank you. Slide 2 please. Your Honor’s, today I
`would like to address the following issues. These issues generally relate to
`whether or not the 777 patent provides adequate written description support,
`as we just discussed, for Claims 5, 15 and 25. I’ll generally be discussing
`Claim 5, but the arguments apply to Claims15 and 25 as well.
`
`In today’s presentation I will address the different theories that Patent
`Owner has provided for alleged written description support of the challenged
`claims. The first theory is that the description in 777 patent of optimized 1D
`constellations that are then orthogonalized to form 2D constellations
`
`7
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`IPR2023-00319
`Patent 10,693,700
`
`allegedly provide support for Claim 5. We’ll call this the orthogonalization
`theory. And this is addressed in Issue 1.
`
`The second theory is that the 777 patents disclosure of optimizing
`over each degree of (indiscernible) for each dimension, provides sufficient
`support. The second theory is addressed in Issue 2.
`
`The third issue is related to a requirement in the claims for a non, for
`each of the non-uniform multidimensional constellations to have a greater
`parallel decoding capacity than other constellations at the same SNR. We
`will discuss how both experts agree that Claim 5, how they agree that Claim
`5 would have been implemented and how these disclosures simply miss
`them with this 777 patent.
`
`Then to Issue 4. We will discuss some of the serious problems in
`Patent Owner’s expert declaration and why it should be given less weight
`relative to Petitioner’s evidence.
`
`In addressing these issues, Your Honors, it will become apparent that
`Patent Owner’s arguments rely on the omission of key clear terms, such as
`multidimensional in their analysis and their notes. And that the Patent
`Owner relies on an assumption that the capacity characteristics of a one-
`dimensional constellation, and of multidimensional constellations, are the
`same. This is simply not true.
`
`Moreover, the record reflects that the Patent Owner has not provided
`any evidence of the assumption being true, whereas the Petitioner has
`provided various pieces of evidence, including the testimony of both experts,
`explaining why this is (indiscernible) correct.
`
`Can Your Honors please turn to Slide 8? Before we dig into the
`issues, I thought it would be very helpful to revisit the claim language
`
`8
`
`
`
`IPR2023-00319
`Patent 10,693,700
`
`because it’s critical here. Claim 5 depends from Claim 1. It generally
`recites various limitations related to a communication system that has a
`receiver, and this receiver is using a non-uniform multidimensional symbol
`constellation pairs. And that these plurality of non-uniform
`multidimensional symbol constellations must have the same number of
`points.
`
`Claim 5, as you can see here, further specifies that in the plurality of
`different non-uniform multidimensional constellations, which according to
`Claim 1 are the same number of points, each one of these non-uniform
`multidimensional symbol constellations must be capable of providing a
`greater parallel decoding capacity that is specific to SNR, that is significant
`to this issue, than other constellations at the same SNR.
`
`Let’s briefly summarize what is required by Claim 5. One, non-
`uniform multidimensional symbol constellations. And I emphasize
`multidimensional because as you will see, Patent Owner often glosses over
`this requirement.
`
`Two, that each of these constellations has a greater parallel decoding
`capacity of particular SNR compared to all the other constellations at the
`same SNR. As we go to show you today, none of these features are
`described in the 777 patent.
`
`If Your Honors could turn to Slide 12 please. In the POPR the Patent
`Owner had argued that Figures 11B, 13B, 15B and 17B allegedly support
`Claim 5. As noted in Slide 12, Your Honor is pointing to Patent Owner’s
`statements and preliminarily agreed with Patent Owner. Note however, that
`in all of these statements the word multidimensional is missing. This was
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`9
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`IPR2023-00319
`Patent 10,693,700
`
`not a coincidence. This is because they use 11B, 13B, 15B and 17B, are all
`related to one dimensional constellations.
`
`Patent Owner programmatically, and incorrectly, pointed to one-
`dimensional constellations as allegedly providing support for claim capacity
`characteristics of multidimensional constellations. As (indiscernible) are
`required in Claim 5.
`
`Now the record has since developed, and plenty of evidence is now at
`the disposal of Your Honors. Petitioner believes this evidence clearly
`establishes the error in Patent Owner’s theories of written description
`support. And why these challenge claims are ultimately not entitled to
`priority in the 777 patent.
`
`And to briefly recap, pre-institution in response to the petition Patent
`Owner offered two theories of why Claim 5 is supported. Again, these two
`theories are covered in Issue 1, Issue 2. It’s also noteworthy that Patent
`Owner offered only attorney arguments in the Patent Owner response, they
`did not collaborate their positions by any expert evidence at the time of the
`Patent Owner response.
`
`And if Your Honors would please turn to Slide 16. Patent Owner’s
`theory is derived from the disclosure shown on the right slide here, Slide 16.
` The passage explains that when you orthogonalize a 1D constellation, such
`a PAM-8 constellation, one can construct a 2D constellation, such as a
`QAM-64 constellation, as shown in Figure 21 on the left.
`
`Which by the way, is the only figure in the patent that depicts a 2D
`constellation. PO, the Patent Owner, then combines this disclosure with
`Figures 11B, 13B, 15B, 17B. Again, a hunting expedition. They combine
`this disclosure with some other disclosure in the patent and say, well, these
`
`10
`
`
`
`IPR2023-00319
`Patent 10,693,700
`
`figures describe optimized 1D PAM constellations. And so when you
`orthogonalize them into 2D constellation, voila, you have optimized 2D
`constellations.
`
` JUDGE DOUGAL: So, this is Judge Dougal. How are we supposed
`to deal with the statements in your petition and your own expert testimony
`when, for example, you said at Paragraph 135, constellations of Figures 11B,
`13B, 15B, 17B are one-dimensional PAM constellations, however, the 777
`patent explains that one-dimensional geometric PAM constellations can be
`optimized into multidimensional QAM constellations?
`Provides a cite to the 777 patent. So it seems to be that your expert
`admitted that, consistent with what we found in the DI, that the 777 patent
`teaches optimizing these PAM constellations, well his words are, optimize
`into multidimensional QAM constellations.
`DR. USMAN: Thank you, Your Honor, for giving me the opportunity
`to clarify that. So the optimized in that sentence refers to the PAM
`constellations, not the multidimensional constellations. I think what he was
`trying to say there was that you can optimize the PAM constellations and
`four multidimensional constellations. That disclosure is there. We don’t
`dispute that. That disclosure is there.
`But to say that the multidimensional constellations are themselves
`optimized is not the case. And I think we tried Dr. Hochwald in subsequent,
`in their second declaration and tried to clarify that.
`And I think we will discuss the statement where he tries to make that
`clear. That yes, you can have optimized PAM constellations, you can use
`the prior optimized PAM constellation to create multidimensional
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`11
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`IPR2023-00319
`Patent 10,693,700
`
`constellations. But that does not mean the multidimensional constellations
`themselves are optimized.
`Why is that important? Because when you look at Claim 5, Claim 5
`actually requires that we are dealing with the parallel decoding, the parallel
`decoding capacity, and the SNRs, all the multidimensional constellations not
`those of 1D constellations. And in subsequent evidence I’ll prove to you, we
`have third-party evidence as well that suggests that when you have 1D
`constellations, it actually doesn’t suggest it explicitly shows, you have 1D
`constellations that are optimized to form 2D constellations, they will look
`different than what a 2D constellation optimized in multidimension looks
`like.
`
`JUDGE CYGAN: Counsel, this is Judge Cygan. Can you explain
`what it means in the description on your slide that says the capacity
`characteristics of the two-dimensional are identical to the, to those of the
`one-dimensional PAM?
`Why should we not draw from that that, you know, the teachings
`from, or the description from 13B, which is the PAM, which we relied on
`earlier in the institution decision, why aren’t those same characteristics being
`drawnfrom the two-dimensional based on that statement?
`DR. USMAN: Thank you very much. That was actually my next
`talking point. So that sentence is essentially what Patent Owner hangs its
`hat on. It says look, it says, the capacity characteristics of QAM and PAM
`are identical.
`What they didn’t pay attention to, and what is critical, and what we
`will show you in subsequent slides, are the last five words of that sentence.
`Says, on a per dimension basis.
`
`12
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`IPR2023-00319
`Patent 10,693,700
`
`
`So what it’s saying is, when it looked at a QAM in one dimension on
`per dimension basis, it has the same capacity characteristics. Yes, it has the
`same capacity characteristics because you’re looking in one dimension. And
`we’ll show you that in subsequent slides.
`If Your Honors could, and this actually is, this is the fundamental flaw
`in their theory. And we’ll show you why it’s not true. If Your Honors could
`turn to Slide 13 please.
`So on Slide 13 Dr. Hochwald, he explains that the disclosure in the
`777 patent provides information about the capacity characteristics in one-
`dimension, not the capacity characteristics of multidimensional
`constellations. He explains that when 2D QAM constellations are formed by
`or orthogonalizing 1D PAM constellations, the capacity characteristics on a
`per dimension basis, 1D, that’s 1D, may be the same. But the capacity
`characteristics over multidimensional constellations are not necessarily the
`same.
`
`They explained this, and this is shown actually better in Slide 14. If
`Your Honors could turn to Slide 14. You will notice that this phrase, on a
`per dimension basis, is found throughout the specification. Here is showing
`you various different snapshots of the disclosure in the 777 patent. Each one
`of them says, on a per dimension basis.
`Why is that important? Because the patent, and this is reflected in the
`figures, their patent is all about PAMs. And they have the throw away
`statement about, yes, you can generate QAMs and PAMs. The capacity
`characteristics that are discussed in the 777 patent are related to one-
`dimensional characteristics.
`JUDGE RAEVSKY: Counsel --
`
`13
`
`
`
`IPR2023-00319
`Patent 10,693,700
`
`
`DR. USMAN: If Your Honors could turn to Slide 17.
`JUDGE RAEVSKY: Counsel, before we go there. This is Judge
`Raevsky. Is Dr. Hochwald’s testimony supported by citations other than
`Fuentes or is that the sole basis upon which he bases his statement in Slide
`13?
`
`DR. USMAN: I think Dr. Hochwald, I’m sorry, and Slide 13 you
`
`said?
`
`JUDGE RAEVSKY: Yes.
`DR. USMAN: Give me one second please.
`(Pause.)
`DR. USMAN: So I can double check this particular point. My
`understanding is, Dr. Hochwald’s position, regarding this particular
`understanding, is based on a few different, first of all, he has decades of
`experience in this field. And he has indicated, and in fact we will show you
`that the SNR and how well we put in capacities. In fact, even the patent
`itself says that the (indiscernible) the spacing between constellation
`(indiscernible) is different from multidimensional and one-dimensional
`constellations.
`And I think we have a slide that points to them. I’m happy to give
`that slide number too. So I think as (indiscernible) is not just based on the
`Fuentes reference, it’s based on experience, based on what the patient
`features. And in addition is (indiscernible.)
`JUDGE RAEVSKY: Thank you.
`DR. USMAN: You’re welcome.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`14
`
`
`
`IPR2023-00319
`Patent 10,693,700
`
`
`JUDGE CYGAN: Counsel, this is Judge Cygan again. Before we
`leave Slide 13, can you talk a little bit about why Claim 5 requires it to not
`be on a third-dimension basis?
`DR. USMAN: Yes. In fact, if we can turn to Slide, uh, go down
`please. That shows Claim 5. Yes.
`Your Honor, as we’ll notice here that --
`JUDGE CYGAN: I’m sorry, which slide is that?
`DR. USMAN: Sorry. This is Slide 8.
`
`JUDGE CYGAN: Slide 8? Thank you.
`
`DR. USMAN: You’ll notice here that the words, on a per dimension
`
`basis, are missing. You will notice here that the word one-dimension is
`missing. What you see here repeated is the word multidimensional. And at
`the end you can also clearly see that it says, the multidimensional symbol
`constellations at the same SNR.
`
`The claim is directed to the capacity characteristics of
`multidimensional constellations. Those characteristics are not the same in
`one dimensional constellations.
`
`JUDGE CYGAN: Just to clarify that. If multidimensional is
`clarifying the single constellation, it’s not characterizing the characteristics
`you’re getting from them, correct?
`
`Claim 5 requires a greater capacity I’m not sure that I see where it
`requires the capacity to be not on a one, not on a per dimension basis. Can
`you talk about that a bit?
`
`DR. USMAN: Yes. Because what Claim 5 is saying is that you’re
`comparing the parallel decoding capacities of each of the plurality on
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`15
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`IPR2023-00319
`Patent 10,693,700
`
`multiple, each of the different non-uniform multidimensional symbol
`constellations with other multidimensional symbol constellations.
`
`The comparison is of a capacity characteristic of multidimensional
`constellations. And this capacity characteristic is not the same in 1D as it is
`in 2D.
`
`JUDGE CYGAN: Well, let me explain a little bit how my thinking is
`going. And I’ll give you a chance to weigh in on this. So if we’re starting
`with the PAM-8 constellation, that certainly seems to meet it for one-
`dimensional. Showing that it’s different for different values.
`
`Then we have the teaching that the capacity characteristics are
`identical. Which means that on a per dimension basis it’s going to meet the
`claim limitation. Why isn’t it sufficient for the claim that on a per
`dimension basis one of those, each of those different multidimensional
`constellations is capable of providing a parallel, a greater parallel decoding
`capacity on a per dimension basis at a specific SNR? Why can’t that be on a
`per dimension basis is my question?
`
`DR. USMAN: Because I think there is two issues with that. The first
`is that we’re inserting the words on a per dimension basis, which are not
`found in the claim. As we showed in that last slide, Slide 14, that phrase is
`littered throughout the specification, but it was not part of the claim. Which
`means they intentionally did not put it in part of the claim. Right?
`
`And secondly, why is it important? It’s because the very thing that
`you’re referring to, the parallel decoding capacity and the SNR, that is
`different for 1D than it is for 2D. And so what, essentially in order to arrive
`at that place where you’re saying, you have to do two things. First you have
`to insert these words into the claims. And then you’re essentially comparing
`
`16
`
`
`
`IPR2023-00319
`Patent 10,693,700
`
`a different entity. You’re comparing characteristics that are different than
`what have been actually claimed in Claim 5.
`
`JUDGE CYGAN: So in your first point are you advancing the scope
`of written description argument that it might satisfy it in a per dimension
`basis for that portion of it, but on a more generic meaning of it, it doesn’t
`satisfy it on a more generic reading of it?
`
`DR. USMAN: No, Your Honor. What I’m saying is the claim does
`not recite on a per dimension basis. What it recites is parallel decoding
`capacities of multidimensional constellations. And it also says, at the SNR
`of multidimensional constellations.
`
`How can we compare how will decoding capacity in one dimension
`with SNRs in multidimensions? That doesn’t make sense. And so, if we
`start inserting words and start assuming this might be what is meant by the
`claims, we end up in no man’s land essentially. That’s not what the claim is
`about.
`
`And maybe with the limited time I would like to jump ahead to a
`figure that perhaps shows this very clearly.
`
`(Pause.)
`
`DR. USMAN: Yes. If Your Honors could turn to Slide 29. Slide 29,
`this is the Fuentes reference. This is not something Dr. Hochwald came up,
`this is a third-party reference.
`
`What we see in the center is a QAM constellation, 64-QAM
`constellation, that has been orthogonalized in one dimension. And then,
`sorry. That has been optimized in one dimension and then orthogonalized to
`create a 2D constellation. We see that in the center.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`17
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`IPR2023-00319
`Patent 10,693,700
`
`And you can see the spacing in 1D, in the two dimensions being
`
`essentially the same. Right?
`
`What you see on the right side is a QAM constellation that has been
`optimized in two dimensions. And the Fuentes reference explains what's the
`difference. But when you optimize in two dimensions you can relax the
`spacing requirements. That allows more flexibility and better optimization.
`
`Fuentes reference even explains that you’d almost never go with what
`you see in the middle because the SNR and the capacity factors, when
`optimized in 2D, are significantly better than what you see in 1D.
`
`JUDGE RAEVSKY: Counsel, this is Judge Raevsky. Where does
`the claim require optimizing in multiple dimension?
`
`DR. USMAN: So, Your Honors, Petitioner does not believe that the
`claim requires optimization. The reason we’re talking about this is because
`Patent Owner presented these theories as written description support.
`In their Patent Owner response they said, there are two theories here
`that provide support. One, orthogonalization. Two, which unfortunately we
`haven’t got to yet, is optimizing over multidimensional constellations. I’m
`sorry, optimizing over multiple dimensions each (indiscernible.)
`In our Petitioner reply we even go on to address the problems with
`these, with the optimization theories, right? With this orthogonalization of
`1D to 2D. Relying on the fundamental assumption that the characteristics of
`1D and 2D are the same. They’re not. The Fuentes reference shows you
`that they’re not the same.
`The claim is directed to LDPC, SNRs, are multidimensional
`constellation, not one. Even their own expert, when we discussed this figure
`they said, yes, we actually refer to what you see in the middle, we refer to it
`
`18
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`IPR2023-00319
`Patent 10,693,700
`
`as a 1D constellation. That’s why in the caption you actually see it’s being
`referred to as 1D. Why? Because the capacity characteristics performance
`are essentially the same as 1D. That’s not a multidimensional constellation.
`In the second theory is, in the second theory there is ten lines of
`description. And the ten lines talk about how complicated it is. This, when
`you optimize, when you go from one dimension to two dimensions the
`complexity just, you know, it’s exponential.
`When we asked their own expert they said, and this is, we can show
`you Slide 26. If you can turn to Slide 26. We asked the expert, well, you
`know, can you explain to us what this additional complexity said? Well it’s
`going to take a lot of time, right? It’s very difficult to explain.
`None of that disclosure is part of the 777 patent. So what we see, the
`reason we’re talking about optimization is because these were the theories
`presented by Patent Owner. They’re not correct, they’re not right. They’re
`wrong. We’re not saying optimization is required, but they’re pointing to
`the support, we’re trying to explain that support isn’t sufficient, it misses the
`point.
`
`They’re going to try to say that, well, we’re reading optimization to
`the claims. No we’re not. What we’re saying is, actually we’re saying the
`opposite. Because optimization is not sufficient. Like the description of
`optimization that you provided here, that doesn’t teach a person, that does
`not show a person of ordinary skill in the art that you have possession of
`this.
`
`What it’s requiring you to do is to try to, and one other point here.
`The experts, the Patent Owner’s expert, and we show this in slide, yes, Slide
`37. Patent Owner’s expert, the POSITA standard that he uses to provide his
`
`19
`
`
`
`IPR2023-00319
`Patent 10,693,700
`
`statements was out of a PhD. Right? That’s not the POSITA standard that’s
`used in these proceedings.
`Why? Why? Well it’s because they’re actually requiring a much
`higher level of expertise and trying to insert that knowledge into the patent
`because it’s clearly not there. And then saying, yes, a POSITA would have
`understood this. No, a POSITA wouldn’t have understood that. Right?
`What you’re trying to do is trying to make up for a lack of written
`description support that’s just not there. And let’s just say --
`JUDGE CYGAN: Counsel, this is Judge Cygan.
`DR. USMAN: I’m sorry?
`JUDGE CYGAN: Counsel, this is Judge Cygan.
`DR. USMAN: Yes.
`JUDGE CYGAN: I have a question for you on your slide. That’s
`your assertion of ordinary skill in the art, correct?
`DR. USMAN: On the top left side is the --
`JUDGE CYGAN: Right.
`DR. USMAN: -- standard that has been agreed upon in the institution
`decision, was presented in the petition.
`JUDGE CYGAN: And doesn’t that standard say that superior
`education can compensate?
`DR. USMAN: Yes, it does.
`JUDGE CYGAN: So doesn’t that sort of bring into play PhD level
`education?
`DR. USMAN: It’s possible, Your Honors. But what that standard
`does not do, for instance, when you’re looking at the standard used by Patent
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`20
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`IPR2023-00319
`Patent 10,693,700
`
`Owner is it doesn’t show you what a persons with a master’s degree would
`have understood, right?
`What essentially their, their re-definition is of a person of
`extraordinary skill in the art, not ordinary. And mind you, they did not
`present any definition of their own. They did not --
`JUDGE CYGAN: Honestly I don’t think that’s being reflected on
`your slide. You’re showing us your assertion, which was agreed to, that you
`could have superior education, and then you have the testimony. Or at least
`the discussion from the surreply, which is a PhD and possibly some years
`after PhD. That see