UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`APPLE, INC.,
`Petitioner,
`
`v.
`
`CALIFORNIA INSTITUTE OF TECHNOLOGY,
`Patent Owner.
`____________
`
`Case IPR2017-00210; Case IPR2017-00219
`(Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`____________
`
`Record of Oral Hearing
` Held: April 19, 2018
`____________
`
`
`
`Before KEN B. BARRETT, TREVOR M. JEFFERSON and
`JOHN A. HUDALLA, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`

`

`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`
`.
`
`
`RICHARD GOLDENBERG, ESQUIRE
`Wilmer Cutler Pickering Hale & Dorr, LLP
`60 State Street
`Boston, MA 02109
`
`MICHAEL H. SMITH, ESQUIRE
`Wilmer Cutler Pickering Hale & Dorr, LLP
`1875 Pennsylvania Avenue, N.W.
`Washington, D.C. 20006
`
`
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`
`
`
`
`
`
`MICHAEL ROSATO, ESQUIRE
`Wilson Sonsini
`
`1875 Pennsylvania Avenue, N.W.
`Washington, D.C. 20006
`
`
`The above-entitled matter came on for hearing on Thursday, April 19,
`2018, at 9:30 a.m., at the U.S. Patent and Trademark Office, Madison
`Building East, 600 Delany Street, Alexandria, Virginia, before Walter
`Murphy, Notary Public.
`
`
`
`
`
`2
`
`

`

`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`P R O C E E D I N G S
`JUDGE JEFFERSON: Good morning. You can be seated. Good
`morning, I am Judge Jefferson and with me are Judge Hudalla and Judge
`Barrett. This is a trial in hearing IPR2017-00210, 00219 and 00297 which
`has been consolidated with 2017-00423. At the outset we received the
`jointly filed objections and that's paper 58 in the 297 case, paper 66 in the
`210 case and paper 65 in the 219 case. Those were objections against the
`demonstrative exhibits and we overruled those objections, they're denied.
`Demonstratives are not evidence as we've said repeatedly. Other cases we
`are capable of certainly distinguishing evidence in the record from
`demonstrative exhibits and arguments, and we're capable of distinguishing
`arguments that are newly presented, both at oral argument or in the papers
`that have come before us and our final decisions will take that into account.
`You are certainly free to argue those things before us today.
`So we'll get started. Our first, we'll have a consolidated hearing. This
`record will be filed in all three cases. The first two we will hear will be
`IPR2017-00210 and IPR2017-00219. They both concern U.S. patent No.
`7,116,710 which is owned by California Institute of Technology and of
`course Petitioner in this case is Apple, Inc.
`Depending on the timing of when these cases finish we will take a
`break and then move after that break we'll move into the last hour with
`respect to the 297. At this time I'm going to ask counsel to introduce
`yourselves at the lectern since that's where the microphones are and please
`introduce whoever's with you at counsel's table and of course anyone who's
`in the gallery. Start with Petitioner.
`
`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
`
`

`

`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`MR. GOLDENBERG: Good morning, Your Honor. Richard
`Goldenberg, lead counsel for Petitioner Apple. With me here is Michael
`Smith, back-up counsel. We also have James Dowd, back-up counsel,
`Kevin Chan, back-up counsel, David Cavanaugh, also from Wilmer Hale,
`and then from Broadcom, Mark Blake, one of the real parties of interest.
`JUDGE JEFFERSON: Thank you, sir. And from the Patent Owner
`please.
`MR. ROSATO: Good morning, Your Honor. Mike Rosato on behalf
`of CalTech, Patent Owner. I have with me at counsel table Quincy Lu from
`my law firm, Matt Argenti, Rick Torczon, and Todd Briggs. Thank you.
`JUDGE JEFFERSON: Nice to put faces with names we've heard on
`more than a few calls. Before we begin we remind the parties that this
`hearing is open to the public and a full transcript that will become part of the
`record. We don't have open motions to seal so if you're going to get into
`anything that you consider to be protected information put the onus on both
`parties and that also includes whether your opposing counsel is talking to let
`us know so we can handle the record appropriately.
`As you know from the trial order, each party has 30 minutes to present
`their argument in each case. The Petitioner has the burden to show
`unpatentability so we'll start with you, then move to the Patent Owner, and
`Petitioner may reserve rebuttal time. In order to keep this hearing focused
`on the merits, I think you've both been with us before, there'll be no
`objections or speaking objections during the presentation. You'll certainly
`have time to respond during your speaking time. Petitioner, you have 30
`minutes. Would you like to reserve some time for rebuttal?
`
`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
`
`
`
`4
`
`

`

`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`MR. GOLDENBERG: Yes, Your Honor. I'll probably make a
`decision exactly how much to reserve on the fly, but on the order of eight
`minutes is what I'd like to reserve.
`JUDGE JEFFERSON: Certainly. We'll try to give you some
`warning. The light should work but if not, I have a timer as well and we will
`get started that way. Just a brief warning, as you can hear my voice is a little
`weak. To the extent you can't hear me or something we have able judges
`who can take over if they need to but please feel free to tell me if you don't
`understand what I'm saying. So Petitioner you may begin when you're
`ready.
`MR. GOLDENBERG: Thank you, Your Honor. Again, my name is
`Richard Goldenberg and I'm pleased to be here to present Petitioner's oral
`argument. We're here today to try and answer any questions that the Board
`may have over any of the remaining issues in this IPR. Over the course of
`this IPR many issues have prompted phone calls with the Board and we
`appreciate the attention that the Board has provided in those calls and the
`guidance that we've received, those calls have resolved many of the issues
`but of course there are still live issues and we'd like to have the opportunity
`to answer any questions that may remain.
`On slide 2, we have a brief roadmap of the order that I plan to address
`the issues, but if the Board would prefer to hear things in a different order or
`different issues that are not addressed in the slide, I'm happy to try and
`address that. Just let me know. But if there are no questions at this time I'll
`proceed with a brief overview of the invalidity of these claims, and we can
`move to slide 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
`
`
`
`5
`
`

`

`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`The message here really is that although many issues have been raised
`throughout the course of this IPR, it remains a simple case. These claims are
`obvious over simple and straightforward reasons. Some of the claims are
`anticipated by the Frey reference. The remainder are rendered obvious over
`combinations involving Frey where the lead reference is the Divsalar
`reference, Frey is secondary and then in some cases there is a tertiary
`reference, the Luby97 reference.
`If we jump to slide 6, this is a comparison of the lead Divsalar
`reference to the code in the patent and the point here really is that Divsalar
`and the patent both have the same structure in their codes. They both have
`an outer coder on the left marked "202" for the patent, rate 1 over q
`repetition for Divsalar. Both of those outer coders are repeaters. In both
`cases the repeater is followed by a permuter labelled "P" here which is also
`an interleaver and then again in both cases downstream of the permuter is an
`inner coder "206" for the patent, the rate 1, 1 over 1 + D block in Divsalar
`which is an accumulator. So both those have repetition followed by
`interleaving followed by accumulation.
`The key difference is that in Divsalar, the repetition is regular. Every
`bit is repeated q times, whatever q is. It could be every bit as repeated, say,
`three times, whereas in the patent bits are repeated different numbers of
`times so some information bits say repeated three times, other information
`bits repeated four times. That irregularity is the feature that's missing from
`Divsalar, but that teaching of irregularity is provided by the Frey reference
`that we see here on slide 7. Frey is all about irregular codes and when
`combined with Divsalar, the irregularity of Frey invalidates the claims.
`
`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
`
`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`The irregularity of Frey can be seen graphically on slide 8. Here the
`bits on the bottom, the open circles on the -- the flow of data here is from
`bottom to top. In some of the other figures it had a different flow, but here
`it's bottom to top. The circles on the bottom represent information bits. If
`we start with the bottom figure those blocks labeled Rep 2, Rep 3, those are
`repeaters. The Rep 2 block repeats its bits two times, Rep. 3 repeats its
`block three times, and so on. There is the irregularity in a particular
`irregular repetition of Frey, and that can be seen graphically in the top figure
`as well where again the bottom open circles represent information bits.
`Some of those information bits have four edges emanating from them,
`meaning those bits are repeated four times. Other of the open circles have
`two edges emanating from them, meaning that those bits are repeated two
`times. So Frey says this in the text. It shows it graphically. It is performing
`irregular repetition and, again, when added to Divsalar that results in an
`invalidating combination, and as we --
`JUDGE JEFFERSON: Counsel, let's go back to the --
`MR. GOLDENBERG: Yes, Your Honor.
`JUDGE JEFFERSON: -- the prior slide. I mean clearly the parties,
`the key dispute seems to be that they say Frey doesn't support what you just
`said. It might be irregular but it's not just irregular information bits, and it's
`not the same kind of coding because this is -- let me try to get this right --
`Frey is a non-systemic and the target is systemic, in the combination of
`systems. Can you explore that difference because I understand what you're
`saying but clearly opposing counsel sees it differently.
`MR. GOLDENBERG: Yes. This is an issue that's disputed by the
`parties certainly and I don't believe that there's a dispute about the issue of
`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
`
`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`systematic, so in a systematic code the codeword will include information
`bits and parity bits.
`JUDGE JEFFERSON: The codeword being the output of the --
`MR. GOLDENBERG: The codeword being the output, yes, Your
`Honor. So systematic has both a non-systematic code. It can be only parity
`bits. The information bits need not be directly represented there. But in
`Frey I don't believe it's disputed that Frey is systematic. The issue that the
`Patent Owner has raised is whether the open circles on the bottom here
`actually represent information bits and in figure 2 of Frey, that's the figure
`on the bottom of slide 8, the figure 2 caption refers to codeword bits and so
`the Patent Owner says well, those bits are codeword bits, we can't tell
`whether they're information bits or not. They could all be parity bits, and we
`of course believe that that's incorrect. The open circles on the bottom do
`indeed represent information bits and there's a number of reasons for that.
`The Patent Owner actually admitted that this itself in the POPR. In
`the POPR at page 20 and 21, Frey explains that in both of these figures the
`bottom circles represent information bits. Also back to the notion of a
`codeword in a systematic code including both information and parity the bits
`on the bottom of the figure 2 caption says codeword bits, but the information
`bits are codeword bits. In a systematic code the information bits are part of
`the codeword and that's what Frey was referring to in the caption.
`The other piece that I would mention is that in our reply and our reply
`declaration -- and the reply declaration was written by Brendan Frey, the
`author of this reference -- Brendan Frey explains in his reply declaration that
`indeed the open circles at the bottom are referring to information bits. And
`while we're on the subject of the Frey declaration I should mention that the
`8
`
`
`
`

`

`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
`
`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`Patent Owner chose not to depose Dr. Frey. He was available. The Patent
`Owner never requested to do that deposition and as the record stands, the
`reply declaration of Dr. Frey and several other declarants all stand
`unchallenged in the record.
`JUDGE JEFFERSON: We do have a declaration and a deposition
`from Dr. Davis who -- my understanding is Patent Owner is saying that Dr.
`Davis agrees with them that Frey is showing non-systemic -- where basically
`these are basically parity bits and not information bits. Am I understanding
`that correctly?
`MR. GOLDENBERG: I don't think so, Your Honor.
`JUDGE JEFFERSON: Okay.
`MR. GOLDENBERG: I don't think that there's a dispute that the Frey
`reference is systematic and maybe I could ask my co-counsel to find the best
`cite in Frey or our petitions, but I don't believe the systematic issue is
`disputed. And you're correct, there was a deposition of Dr. Davis, the
`original declaration who filed a declaration in support of our petition. He
`was deposed. He filed a second short declaration explaining why he was
`unavailable to help with the replies and the Patent Owner chose not to
`depose Dr. Davis on his sworn testimony explaining that he was unavailable.
`We filed that declaration after the Patent Owner had raised the charge that
`Davis was somehow running and hiding, and --
`JUDGE JEFFERSON: Well, accepting that he's not available, we
`have two declarants and we have Dr. Davis and Dr. Frey and my
`understanding is the Patent Owner is saying Dr. Davis has made some
`admissions that agree with them and their position, which is that you can't
`take the irregularity as described in Frey and apply it to Divsalar because of
`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
`
`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`these technical difficulties that they are different types of encoding, if you
`want to follow up.
`MR. GOLDENBERG: Yes.
`JUDGE JEFFERSON: So I mean the idea is how are we supposed to
`manage the distinctions between Dr. Davis and Dr. Frey, assuming there are
`differences or are there no differences, in your opinion?
`MR. GOLDENBERG: I don't believe there are any differences and
`there are no differences in particular on this issue. Professor Davis never
`said that the open circles on the bottom here in the Frey figures are parity
`bits. I don't believe there's any testimony to that effect, and Dr. Frey in his
`unchallenged testimony of course restates that as well. Any further
`questions on that, Your Honor?
`JUDGE JEFFERSON: No, we are getting to the point. Let's move to
`motivations combined or at least some understanding of motivations
`combined. Patent Owner's argument seems to be that the phrase you can't
`take the wholesale (phonetic) the concept in Frey and apply it to Divsalar or
`Frey doesn't give you the pure motivation to combine to make the
`information bits in Divsalar irregular.
`MR GOLDENBERG: The Patent Owner makes a similar argument to
`that about the Luby reference which is relevant to the next hearing, and also
`MacKay which is relevant to the third hearing. I don't believe that the Patent
`Owner has argued that Frey's bits are not information bits. There is an issue
`of motivation to combine. The motivation to combine, as we see on slide 9,
`simply comes in the form of Frey explaining that his irregular code provides
`superior results. The Patent Owner has challenged that Frey's superiority
`isn't true in all cases and we would accept that if irregularity is not a
`10
`
`
`
`

`

`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
`
`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`panacea, it's not that adding irregularity will automatically make any code
`better. The exciting thing about Frey though is he showed a result if
`incorporating irregularity into a turbocode and it did provide improved
`results. In fact, you don't need to go further than the abstract.
`The abstract of Frey explains that the reason he's writing this paper is
`that because he obtains a coding gain in the irregular code. One of ordinary
`skill would have been motivated by the improved performance provided by
`irregularity taught both in Frey and other references, they would have
`motivated to want to want to take advantage of that irregularity in other
`codes such Divsalar's, and if we move back just to the combination for a
`moment I don't have Frey and Divsalar on the same figure – actually let's
`see, maybe. I'm going to try just jump ahead in the slides because there
`actually may be. Yes, so I'm on slide 12 here. These show the structure of
`the codes. Frey has his, in the figure on the left is figure 1 from Frey which
`is a Tanner graph from the Frey reference. The figure on the right is a
`Tanner graph depiction of Divsalar, and doing this here with the Tanner
`graphs we could look at it in terms of the block diagram of Divsalar as well.
`The point is that both of these codes have the same structure and they have
`the same structure as the patents too. They begin with repetition. That's
`seen by the lines emanating from the -- so I mentioned earlier there's a
`rotation of orientation. In the figure on the left of Frey the data flow is from
`bottom to top. In the Divsalar figure the flow is from left to right. In both
`cases we have blue boxes around the information bits and Frey's repetition
`of the information bits is indicated by the lines emanating those open circles
`at the bottom. Divsalar's repetition is indicated by the edges emanating from
`those nodes labeled U. In Divsalar the information bits are labeled U. In
`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
`
`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`both codes, there's repetition followed by permuting followed by
`accumulation and you can see that same Divsalar structure back in in block
`diagram on slide 6. Again, it's repetition followed by permutation followed
`by accumulation
`JUDGE HUDALLA: How do you respond to Patent Owner's
`argument that, you know, in the reply brief you brought in a whole host of
`new experimental evidence? In your slide there it says it's simple but some
`of that experimental evidence is coming in the reply. I mean it leads us to
`believe that maybe it's not so simple. You had to go through some pretty
`heavy experimentation, come up with some results that maybe shore up your
`idea that this is something that could be done simply.
`MR. GOLDENBERG: Yes, Your Honor. So the Patent Owner has
`made that argument certainly. Our position is that the petition and the
`opening Davis declaration laid out is a straightforward obviousness case.
`Frey teaches irregularity, teaches that better regularity improves the code
`and one of ordinary skill would have wanted to obtain that betterment by
`using irregularity in other codes such as Divsalar's.
`JUDGE HUDALLA: Well, again, don't both Dr. Davis and Dr.
`Mitzenmacher give this idea that hey, this is kind of an unpredictable field
`here. You can't just stick in a code and expect it's going to work simply as
`your slide says.
`MR. GOLDENBERG: So, well there's a few issues here. Let me try
`to take them one at a time, Your Honor. It's undisputed that it's difficult to
`develop a new code or modify an old one and predict in advance
`mathematically how well that code is going to perform. Experimentation is
`what's routine for those of ordinary skill in the art in this field. That's what
`12
`
`
`
`

`

`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`those of ordinary skill do. They test a code by, and test a new idea, by
`experimenting with it. One of ordinary skill reading Frey and Divsalar
`would have been motivated to do those experiments.
`Now none of these claims require any sort of performance. Nothing
`in the claim would require an optimally performing code, even a better code.
`Once one of ordinary skill in the art has made the combination of applying
`Frey's regulator to Divsalar, those claims are invalid. That combination
`meets the claims.
`JUDGE HUDALLA: So your contention is that there doesn't have to
`be a reasonable expectation of success from these things or --
`MR. GOLDENBERG: No, no, Your Honor. So we still need a
`motivation for one of ordinary skill to have to have made the combination in
`the first place. The motivation was attempting to obtain the improved
`performance that Frey holds out the crumbs for. That's the motivation.
`As far as the reasonable expectation of success, another issue the
`Patent Owner has raised, the reasonable expectation of success is shown by
`the fact that the combination would have been a simple one. These codes
`have the same structure of modifying Divsalar's repeaters. They repeat some
`bits different numbers of times instead of repeating them all at the same
`time. That's an easy modification. Once that modification is made, all the
`claim elements are met and reasonable expectation of success doesn't require
`anything more. The key is that one of skill in the art would have been
`motivated to make the combination in the first place.
`JUDGE HUDALLA: Can I ask you a question about that as well?
`MR. GOLDENBERG: Yes, Your honor.
`
`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
`
`
`
`13
`
`

`

`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
`
`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`JUDGE HUDALLA: I mean we've talked a bit about this idea that
`are you making an obvious-to-try argument? There's that sense in your
`petition and, you know, are you trying to put it under that kind of case law or
`is this something different that you're putting forward here?
`MR. GOLDENBERG: I don't think it's quite obvious-to-try, Your
`Honor. I haven't thought about it in those terms. I think it's -- the
`combination of Frey and Divsalar fits within the realm of what those of
`ordinary skill in the art typically do in this field. So in actual practice, one
`of ordinary skill may have tried one modification of Divsalar and tested it,
`and then tried another. That was just what's routine in this field.
`Back to Your Honor's question about the experimentation. So we
`made this argument in our petition and declaration there was a
`straightforward combination. The Patent Owner then argued that one of
`skill in the art actually would have been dissuaded from using Frey's
`irregularity in Divsalar. They would not have been so encouraged. That
`was an argument we didn't expect. Again, you don't have to get past the
`abstract of Frey to see the reason he's writing this paper is to tout the
`performance gains of adding irregularity to the turbocodes provided.
`It's a strange reading to read Frey and think that one of ordinary skill
`would be discouraged from using irregularity. They made that argument, we
`didn't anticipate it. The answer to that argument was in the experimental
`results that Dr. Frey provided in his reply declaration. Once we saw that
`argument that we didn't expect, he did the experiments to show both that the
`combination was a simple one and that the combination actually does
`provide improved benefits as one of ordinary skill would have expected, and
`back to the unchallenged nature of the Frey declaration, if the Patent Owner
`14
`
`
`
`

`

`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
`
`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`had issues with those experiments, issues with Dr. Frey's explanation, they
`had a perfect opportunity to examine the author of the Frey reference but
`they chose not to depose him. I think it's an inescapable conclusion that the
`reason the Patent Owner did not depose Dr. Frey or Dr. Davis on his second
`declaration is that they knew they were not going to like their testimony.
`Instead of confronting the substance of our reply explanations and the
`explanations in Dr. Frey's declaration, they had resorted to procedure
`attempting to exclude those declarations and of course we feel that's
`improper.
`JUDGE HUDALLA: I would say, counsel, we understand the tension
`there between their decision not to depose Dr. Frey and maybe making a
`litigation decision in this case. We understand both sides of that issue, so
`maybe you can focus more on some of the meat of the case. I didn't mean to
`take you off the track.
`JUDGE JEFFERSON: Well let me -- there's something he said that
`sparked a thought in me which is you keep using the word it's a simple
`substitution. Is it fair that it's from Frey itself that doing this is making
`choices about what components to use and how to make this work? Is it
`easy to take some thought and experimentation as you said, I mean doesn't
`that lead away from a reasonable expectation of success if you were trying
`something with the intention that you're not sure what the outcome is going
`to be?
`MR. GOLDENBERG: I don't think so, Your Honor. So Frey does
`use the words not trivial and other words to that effect. What Frey is saying
`is that to develop an optimally performing code that can take some work, but
`to obtain some improvement or to attempt to obtain some improvement,
`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
`
`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`those things were obvious and in fact not difficult. Now sure, it might have
`some extra work to find the optimally performing, the perfectly performing
`modification of Divsalar, that could take some work, but these claims don't
`require that. They don't -- any degree of irregularity if Divsalar repeats all
`the information that's q times, repeating one bit q plus one times is enough to
`meet these claim limitations. Now one of skill in the art probably would
`have done something different. In the Frey experiment he repeated some
`bits three times, other bits seven times and compared that to another code
`where all bits were repeated five times so they all have the same code rate.
`It was an apples-to-apples comparison, and in that example as the
`unchallenged testimony of Dr. Frey shows, there is an improvement which is
`what one of skill in the art would have expected.
`JUDGE BARRETT: Before you move on I want to address
`something, make sure I understand your position. You mentioned that you
`were, if I understand this correctly, you were surprised by the teach away
`argument?
`MR. GOLDENBERG: Yes.
`JUDGE BARRETT: So my understanding of teach away is that the
`reference itself discourages the combination and if I understood you
`correctly, your responsive evidence is to show that the combination would
`indeed work. That to me doesn't seem to address whether the reference
`discourages the combination in the first place. So did I misunderstand your
`position?
`MR. GOLDENBERG: Well I think, Your Honor, the teaching away
`argument is answered in our opening papers. Frey is all about touting the
`performance gains that can be achieved by adding irregularity. That alone
`16
`
`
`
`

`

`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`was sufficient to motivate one of ordinary skill to make the combination.
`The experimental results was like icing on the cake, if you will. It's
`confirming the expectation that was set forth in our opening papers. Does
`that answer the question?
`JUDGE BARRETT: Not exactly. At the end of the day we're going
`to have to decide whether all this evidence is appropriate rebuttal,
`appropriate reply, or not and now I seem to understand you to say well, we
`hit it head on in the petition -- we actually addressed in the petition. Having
`an expert come in later and say well, say, it would have been easy doesn't
`really tell me whether the person of ordinary skill in the art would have
`known that ahead of time, would have been discouraged or not.
`MR. GOLDENBERG: Yes, I understand, Your Honor, and just to go
`back to one point just before we lose it, Dr. Mitzenmacher's declaration at
`paragraph 90 says that Frey is a systematic code, so I think that does resolve
`the issue. There's not a dispute on that one.
`But to Your Honor's question, I think that that answer is the
`straightforward one, that Frey alone with many other references taught the
`benefits of modifying regular codes to be irregular. That's a clear incentive
`for one of ordinary skill to experiment, to modify a code like Divsalar's to
`make it irregular. The experimental results in the reply declaration, they're
`really just confirmatory.
`JUDGE BARRETT: Okay.
`JUDGE JEFFERSON: We're down to about seven and a half minutes
`left, and I want you to turn to the antedating issue specifically.
`MR. GOLDENBERG: Yes, Your Honor.
`
`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
`
`Cases IPR2017-00210; IPR2017-00219 (Patent 7,116,710 B1)
`Case IPR2017-00297 (Patent 7,916,781 B2)
`
`
`JUDGE JEFFERSON: Let's start with the publication date of Frey
`because I think one of my problems with the evidence introduced by
`Petitioner is that at best you show that Frey was shipped to University of
`Illinois Urbana-Champaign, happens to be my alma mater, on February 20th
`of 2000 I believe, and that's it. Am I missing something?
`MR. GOLDENBERG: Well, so there are two points to the Frey prior
`art status. The first is as I've got there on slide 18. Frey's been available in
`the Cornell University library since March 20, 2000. That's before any
`priority date as claimed by the patents. That March 20 date is sufficient to
`establish the prior art status of Frey and that's substantiated by the testimony
`here from the Cornell University librarian that's quoted on slide 18. We did
`provide some additional evidence about an even earlier date for Frey than
`March 20, but I think it's more important to focus on the Patent Owner's
`failure to establish an earlier date of invention.
`So in the Patent Owner response they provided some software files
`evidence that had been written by one of the inventors, Dr. Jin, and some
`testimony attempting to establish an earlier date. That attempt fails and as
`we see here on slide -- so even with the March 20 date Frey is prior art. On
`slide 19 there's a table here showing some of the software files that the
`Patent Owner has cited to and the Patent Owner has said that the dates in
`black are the dates for those files and that's what shows their earlier date of
`inve