`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`GOOGLE LLC,
`Petitioner,
`
`v.
`
`KONINKLIJKE PHILIPS N.V.,
`Patent Owner.
`
`Case IPR2017-00437
`Patent 6,772,114 B1
`____________
`
`Record of Oral Hearing
`Held: February 13, 2018
`____________
`
`
`
`
`Before KEVIN F. TURNER, ROBERT J. WEINSCHENK, and
`KAMRAN JIVANI, Administrative Patent Judges.
`
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`DAVID M. KRINSKY, ESQUIRE
`CHRISTOPHER SUAREZ, ESQUIRE
`Williams & Connolly, LLP
`725 Twelfth Street Northwest
`Washington, D.C. 20005
`(202) 434-5338
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`JUSTIN OLIVER, ESQUIRE
`SEAN WALSH, ESQUIRE
`Fitzpatrick, Cella, Harper & Scinto
`975 F Street Northwest
`Washington, D.C. 20004
`(202) 530-1010
`
`
`
`
`The above-entitled matter came on for hearing on Tuesday, February
`
`13, 2018, commencing at 1:00 p.m., at the U.S. Patent and Trademark
`Office, 600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`2
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`
`P R O C E E D I N G S
`- - - - -
`JUDGE WEINSCHENK: Good afternoon, everyone. This is a
`hearing for IPR2017-00437, Google, LLC versus Philips N.V. Let's start
`with appearances, who do we have for Petitioner? And please step up to the
`center podium when you introduce yourselves.
`MR. KRINSKY: Good afternoon, Your Honor,
`David Krinsky for Petitioner, Google. At counsel table with me is Mr.
`Suarez, and I'd also like to recognize Kevin Hardy and Christopher Geyer
`we're all from Williams and Connolly, LLP. Also here is John Colgan from
`Google.
`JUDGE WEINSCHENK: Great, thank you. And who do we have for
`Patent Owner?
`MR. OLIVER: Good afternoon, Your Honor.
`Justin Oliver of Fitzpatrick, Cella on behalf of the Patent Owner. With me
`at counsel table is Sean Walsh, also of Fitzpatrick, Cella.
`JUDGE WEINSCHENK: All right, thank you very much. Before we
`get started just a few housekeeping matters. As you can see Judge Turner
`and Judge Jivani are appearing remotely, so in order for them to be able to
`hear you whenever you're speaking please step up to the center podium and
`speak into the microphone. And when you're referring to any of your slides
`please use slide numbers so they can follow along. As you know from our
`order each side has 30 minutes to present their case. We'll start with
`Petitioner, just let us know if you'd like to reserve any time for rebuttal.
`MR. KRINSKY: Thank you, Your Honor, and I would like to reserve
`15 minutes 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
`26
`
`
`
`3
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`
`JUDGE WEINSCHENK: 15?
`MR. KRINSKY: 15, I mean if I go a little into that I can use less on
`rebuttal. Before we begin also would Your Honor like a copy of the slide
`deck in paper?
`JUDGE WEINSCHENK: Sure.
`MR. KRINSKY: May I approach?
`JUDGE WEINSCHENK: Yes.
`MR. KRINSKY: Just one?
`JUDGE WEINSCHENK: Yes, thank you. You may begin when
`you're ready.
`MR. KRINSKY: Thank you, Your Honor, may it please the Board.
`The Tucker patent application, Exhibit 1004, anticipates and renders obvious
`the instituted claims of the '114 patent in significant part because it is
`directed to the same type and structure of subband coder and decoder. If we
`could go directly -- and I apologize in advance for skipping around a little
`bit -- I'm going to jump to slide 4. I've put up on the screen a copy of Claim
`20, which is just the decoder side, but it's a representative claim for these
`purposes.
`And as you can see from Claim 20 there are really only three terms in
`dispute, and I would submit really only two fundamental issues in dispute.
`The question of whether Tucker teaches the required second coded signal
`within a high frequency range and whether Tucker teaches the required
`low-pass and high-pass filters. And I think it makes sense, given the sort of
`smoke and mirrors that I think we're likely to see from Philips, to begin with
`the filter terms. I think those are the ones that are perhaps the most
`confusing.
`
`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
`
`
`
`4
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`
`If we could jump to slide 21, slide 21 has a copy of figure 2 of Tucker,
`and I would submit I've highlighted here the portion of figure 2 -- this is
`Tucker's decoder -- I've highlighted the portion of figure 2 that relates to
`decoding the high band. In both Tucker and the '104 patent, of course,
`there's one portion which is directed to decoding -- encoding and decoding
`the upper band.
`Typically, 4-8 kilohertz signal, and another portion that's directed to
`encoding and decoding. In this case, decoding the 0-4 kilohertz signal, the
`lower band, the highlighted portion here being the upper band, and the
`question as to both filter terms is what does figure 2 mean when it says
`"interpolate" in box 32 that's in the lower band.
`JUDGE WEINSCHENK: Mr. Krinsky, you've told us what
`interpolate means, right? You said interpolate means upsampling followed
`by a low-pass filter, right?
`MR. KRINSKY: That's right, when interpolate doesn't have any
`further color, that's correct. That's what interpolate means in box 32, of
`course, that needs then to be reflected to the upper band in box 26. And I
`think the key point here is everyone agrees in the art that interpolate may or
`may not include a filter depending on the context. When you upsample that
`yields ghosting, that you wind up with two copies of the signal, essentially,
`that are mirror images of each other frequency --
`JUDGE WEINSCHENK: But you've told us that in the context of
`this Tucker reference interpolate means upsampling followed by a low-pass
`filter, right?
`MR. KRINSKY: That is correct. That is correct, and that's what it
`means in the art when you do filtering, again, when there's no further word
`
`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
`
`
`
`5
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`in the claim. Box 26 says interpolate and reflect to the upper band, and the
`question is what is the person of ordinary skill in the art interpret that phrase
`to mean. You have to wind up in the upper band. So, Dr. Quackenbush
`opined in his opening declaration that when you read interpolate and reflect
`to the upper band what that means is you actually use a reflection of the
`filter card, and so that encompasses upsampling and a high-pass filter.
`JUDGE WEINSCHENK: But that's not what Tucker actually teaches
`though, is it?
`MR. KRINSKY: That is what Tucker teaches.
`JUDGE WEINSCHENK: So, I think Patent Owner has pointed us to
`a portion of Tucker that says -- let me try to find it for you so that we're
`talking about the same thing.
`MR. KRINSKY: I think, sorry, Your Honor. I don't mean to be --
`JUDGE WEINSCHENK: No, it's all right.
`MR. KRINSKY: I think I may know where you're going. If we could
`bring up slide 13, is this the disclosure you're referring to?
`JUDGE WEINSCHENK: Nope, nice try.
`MR. KRINSKY: I won't guess then, I'm sorry.
`JUDGE WEINSCHENK: So, it's on page 14 of Tucker and I don't
`know if you have Exhibit 1004 in front of you or it's in a slide.
`MR. KRINSKY: I have it here. I don't know if I have it on a slide or
`not but I'll follow along here on paper.
`JUDGE WEINSCHENK: I'll let you get to page 14 and then --
`MR. KRINSKY: I'm there, Your Honor.
`JUDGE WEINSCHENK: We're talking about lines 21 to 23 and it
`says, the synthesized signal then passes to a processor, 26, which
`
`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
`
`
`
`6
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`interpolates the signal and reflexes it to the upper subband. But to me, the
`way I would read this is that, in box 26 we interpolate the signal which
`means upsampling and low-pass filtering, and then we reflect the signal that
`results from that to the upper band. Not that we're reflecting the filter itself
`but that we're reflecting the output of the low-pass filter.
`MR. KRINSKY: And I think that is one perfectly good way of
`reading that phrase and one perfectly good implementation of what Tucker is
`teaching. And if we could go back to slide 21 in box 26 and that, I think,
`goes to the nub of the dispute here in what Your Honor just posited. If you
`have as three conceptually separate steps upsampling followed by low-pass
`filter and followed by reflecting the combination of the low-pass filter and
`the reflection is a high-pass filter.
`JUDGE WEINSCHENK: Well, that's not exactly true though, is it,
`right? So, we have a low-pass filter, if you've got a signal coming into a
`low-pass filter it let's the low frequencies go through and it attenuates the
`high frequencies. So, we've lost everything in the high frequency band. We
`have information coming through in the low frequency band and then we're
`flipping up to the high band --
`MR. KRINSKY: Well, the key, Your Honor, is that in this context --
`JUDGE WEINSCHENK: Mr. Krinsky, let me finish the question.
`MR. KRINSKY: I apologize, I thought you were done.
`JUDGE WEINSCHENK: So, what we're doing there is we're
`attenuating the high frequency and letting the low frequency pass. Then
`later we're just flipping that information up to a different frequency band. If
`we have a high-pass filter we're allowing the high frequency signal to go
`through and attenuating low frequency. Although we may end up with a
`
`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
`
`
`
`7
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`high frequency signal in the end those signals may be different depending on
`how they do it, correct?
`MR. KRINSKY: In this context they're not different and that's the
`key. I would agree with you that in some other context, if you were to
`simply take a low-pass filter and reflect its output you might not get the
`same result as if you applied a high-pass filter. But in this context, because
`you have just up sampled the signal, that goes into the low-pass filter in what
`Your Honor just posited is a mirror image of itself around the center of the
`spectrum, around the 4 kilohertz point.
`So, you low-pass filter and reflect that output, the result is exactly the
`same as if in one step you attenuated only the lower frequencies. That is the
`functional point of box 26, and it's described in functional terms, to
`upsample the signal and then pull out the upper band. That is, by definition,
`what a high-pass filter does and it's what a high-pass -- what the box 26 is
`doing here. The parties agree -- excuse me. The parties agree, number one,
`that a filter may comprise other components.
`There's no requirement -- this is, as a practical matter, implemented as
`a processer in software. And, so whether it's one step or two steps is
`irrelevant to whether there's anticipation. And also in his deposition Dr.
`Johnson agreed that a high-pass filter may have a low-pass filter as a
`component. So, if you read the disclosure of Tucker to require a low-pass
`filter followed by a reflection step the combination of those two components
`in this context is a high-pass filter as Philips has defined the term and as
`defined in the art.
`JUDGE WEINSCHENK: I just want to back up for a moment,
`because I want to make sure I understand your position. Because I think
`
`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
`
`
`
`8
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`Patent Owner pointed out there might have been a little bit of flip-flopping,
`maybe intentionally or not intentionally, about what the position is here. Do
`you agree that what's disclosed in that box is a low-pass filter that's followed
`by reflecting the output of that low-pass filter to the upper band or do you
`think that it's a high-pass filter, that it's reflecting the low-pass filter to make
`a high-pass filter? Are we reflecting the output or are we reflecting the
`filter?
`MR. KRINSKY: I don't think it tells you which of those it is, I think
`is at the center of what Philips describes as a flip-flop. That's really an
`implementation detail. As Dr. Quackenbush originally conceived of it you
`would reflect the filter kernel. That's just, in his opinion, the person of
`ordinary skill in the art would most simply implement this, but where you
`could implement it the other way it would equally much be anticipation.
`So, if Tucker doesn't specify which of those approaches to take, but it
`doesn't need to in order to be able to practice the disclosure and either one
`anticipates. The choice for the Board is is it one of those two options; Dr.
`Quackenbush, or you have no filter at all which is what Dr. Johnson is
`positing which, in his own words, leads you to a bit of a mess in the output.
`JUDGE WEINSCHENK: Let's go back to one more thing here. You
`said before that if we have a low-pass filter and we reflect the output of it to
`the upper band the result would be the same in some circumstances and
`mainly in the characteristic of this reference. Is having the same output or
`same result enough for anticipation when the claims recites a specific
`structure? Here, we have the structure recited as a high-pass filter. The fact
`that we maybe get a similar result as a high-pass filter but do it with a
`low-pass filter, doesn't that distinguish it for purposes of anticipation?
`
`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
`
`
`
`9
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`
`MR. KRINSKY: I would submit that it does not, Your Honor.
`Because high-pass filter as the parties agree is defined functionally. It is
`something that passing through the upper band that substantially attenuated
`the lower band. That's exactly what this does regardless of the specifics of
`the limitation. And that's why I suggested the purported inconsistencies in
`Dr. Quackenbush's testimony really are beside the point. I mean, if I may
`use an analogy, it's a bit as if Tucker disclosed attach a shelf to a wall and
`Dr. Quackenbush opined, well, the person of ordinary skill would read that
`and naturally screw the shelf into the wall and then in deposition he's asked,
`well, couldn't you also glue the shelf and he said, gosh, I hadn't thought
`about that. And then later on reflection opined that, yes, you could also glue
`the shelf.
`JUDGE WEINSCHENK: But if you had a claim that recited screw
`the shelf to the wall and you glued it to the wall then it wouldn't meet that
`claim, right?
`MR. KRINSKY: If that were the claim but that's not the claim here.
`The claim here requires a high-pass filter which, again, the parties agree is
`defined functionally and what Tucker discloses is that 26 is a processor. The
`same is true in the '114 patent, the high-pass filter isn't some component you
`buy and pull out of a bin and solder or something and practice, they
`(inaudible) digital signal processing. So, whether the structure is
`subdivided into multiple subroutines, if you will, one of which you
`denominate a low-pass filter and then you denominate a reflection step, or
`whether those are combined into a reflection and a high-pass -- excuse me,
`I'm getting tongue tied -- a reflective low-pass filter and thereby a high-pass
`filter, it really is up to the implementer of Tucker.
`
`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
`
`
`
`10
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`
`It's not disclosed, it doesn't need to be disclosed because either way
`for the Board's present purposes there's anticipation. That is a high-pass
`filter however implemented. If Your Honors have further questions on
`filters I note I only have a couple of minutes left and I would like to address
`the '742 claims, unless there are further questions on the high-pass filter
`claims.
`JUDGE TURNER: Can I ask a quick question, and I'm game to give
`you more time if need be. I am and maybe my fellow panelists don't, but I'm
`game to give you a little bit more because I think this is a crucial question.
`If you can go with me to slide 23.
`MR. KRINSKY: Certainly, Your Honor.
`JUDGE TURNER: What's in the dotted line there in the center, does
`Petitioner agree that that's supposition?
`MR. KRINSKY: I don't think I'd agree that that's supposition. I think
`this is Dr. Quackenbush's illustration of exactly the hypothetical
`Judge Weinschenk posed.
`JUDGE TURNER: It's at least his interpretation, right?
`MR. KRINSKY: It's an interpretation --
`JUDGE TURNER: Because that's what the slide says, so unless the
`slide's lying it's his interpretation.
`MR. KRINSKY: Yes.
`JUDGE TURNER: That's not actually reflected in the disclosure. He
`saying that's what would be there, what's in the dotted line, where you have a
`high-pass filter in that slide, right?
`MR. KRINSKY: Well, the disclosure here is of interpolate and reflect
`to upper band, and the question for the Board is what does that mean. So,
`
`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
`
`
`
`11
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`this is one implementation that that could mean to the person of ordinary
`skill. The dotted line is -- reflects a one-step high-pass filter or it reflects a
`union of a low-pass filter and a reflection. But either way that is a
`component of what interpolate and reflect means to the person of ordinary
`skill in the art according to
`Dr. Quackenbush, and it's the only alternative as to what that could be under
`which Tucker achieves the goal that it says it achieves of reconstruction of
`the upper band.
`JUDGE TURNER: Okay. So, am I correct in assuming that if we
`don't accept Dr. Quackenbush's interpretation we don't find anticipation or
`am I incorrect?
`MR. KRINSKY: I mean I think the Board needs to find that
`interpolate and reflect mean something, so the question is what does it mean.
`And really only two options have been posited, one that includes a high-pass
`filter, however implemented, and one that does not. And the one that does
`not, which is Dr. Johnson's interpretation, simply doesn't work and as he
`testified the person of ordinary skill would understand that and would
`understand had to be wrong.
`So, I do think that if the Board concludes that this isn't a (inaudible)
`disclosure the Board can find obviousness on the same ground. It's clearly
`what, and even in Dr. Johnson's opinion, a person of ordinary skill would
`know needed to happen in order to make Tucker work. But I think the best
`legal framework for that is to say that interpolate and reflect to upper band in
`this context would be interpreted by the POSITA to mean interpolate -- or
`upsample and high-pass filter and therefore it's 102.
`
`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
`
`
`
`12
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`
`JUDGE TURNER: One last just quick question and I apologize for
`taking too much time, but even if we said that we had to find that this is an
`obvious variation, is that in the petition? Is that in part of the grounds that
`we've instituted, I mean, can we do that?
`MR. KRINSKY: I think Your Honor could do that given that it's an
`issue that's been sort of squarely presented to the Board by both sides. We
`addressed it at the petition as an anticipation but that's what we think it most
`properly is. Anticipation is through the eyes of the person of ordinary skill
`in the art, so does the Board have the power to address it as an 103 instead, I
`think it does.
`JUDGE TURNER: Right, but in terms of the 103 that was posited by
`Petitioner, that wasn't part of the 103, right?
`MR. KRINSKY: The way that we had teed up 103 was if the Board,
`for whatever reason, finds a component not to be expressly disclosed here
`it's certainly implicitly disclosed. So, I think it's fairly read within the scope
`with what we said in the petition, but Your Honor is correct that we didn't
`spell out at length a 103 analysis of a high-pass filter because in Dr.
`Quackenbush's opinion and our position is that one is there, one is disclosed.
`JUDGE TURNER: Okay. Thank you.
`JUDGE WEINSCHENK: Mr. Krinsky, you're cutting into your
`rebuttal time now. If you want to keep going we can always decide on
`rebuttal whether you need additional time and we can accommodate that.
`MR. KRINSKY: That would be fine as long as Your Honor has no
`problem with my resting on the briefs on second coded signal and perhaps
`addressing it later, if appropriate.
`
`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
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`
`JUDGE WEINSCHENK: Well, if you'd like to address it let's address
`it now and then we can always give both parties a little more time at the end
`if we feel it's necessary.
`MR. KRINSKY: Thank you, Your Honor, I appreciate that. I'll be
`very brief and jump to slide 6. Actually, lets, on second thought, let's jump
`back to that same figure, figure 21. The points I want to get across on
`second coded signal, just very quickly, there's been some confusion about
`what exactly it means for a signal to be within the high frequency range, and
`I just wanted to be very clear here, everyone agrees that the second coded
`signal is not a signal that itself has a frequency.
`It's a digital signal, it's ones and zeros that describes a signal having a
`particular frequency. And the key to the dispute on second coded signal
`within a high frequency range is that whether or not there is decimation,
`whether or not the signal as part of the encoding process is brought down
`first into the 0-4 kilohertz range before it's turned into LPC and gain
`coefficients. The shaping and amplitude of the signal that's described by
`those coefficients is the shaping having to do with the characteristics of the
`upper frequency band.
`JUDGE WEINSCHENK: So, you've taken the position that if you
`have a high frequency signal and you decimate it to lower the frequency that
`the shape, or amplitude, or other characteristics would still be the same.
`What's the evidence you have that you can point us to that supports that
`those things are the same? The characteristics are the same after
`decimation?
`MR. KRINSKY: Well, the relevant characteristics here are -- I guess
`my answer is twofold. One is Tucker, itself, teaches that this, that the signal
`
`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
`
`
`
`14
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`after decimation, after encoding into these coefficients represents and allows
`you to reconstruct the upper band signal, so it necessarily has the
`characteristics of that signal. But, second of all, Dr. Quackenbush explained
`at some length in his declarations how when you perform the decimation
`step you wind up with these -- the same shape of the signal just reflected,
`and so the coefficient is described -- in a sense they describe both signals --
`but fundamentally the point of those coefficients is to describe the
`characteristics of the upper band signal that's to be reconstructed.
`JUDGE WEINSCHENK: When you say that you have the same
`signal mirrored in a low frequency band in an upper frequency band, can
`you explain to me what that means?
`MR. KRINSKY: Sure.
`JUDGE WEINSCHENK: Because there's a lot of arguments about
`this mirroring effect and I want to be sure I understand that completely.
`MR. KRINSKY: Yes, let me see if I can illustrate it. This is -- the
`figure is sort of directed to a different purpose, but it'll illustrate the point if
`we go to slide 32. This is Dr. Johnson's interpretation -- this is Dr.
`Quackenbush’s simulation of what
`Dr. Johnson is saying happens when you just don't have filtering at all in the
`interpolation steps. And what you note about this signal is that it's
`symmetrical about the 4,000 frequency -- 4,000 kilohertz frequency. The
`right side and the left side have the same spikes in corresponding locations,
`it's like you folded the paper in half, that's what I mean by a mirroring
`signal.
`So, you can describe the shape of the left half or you can describe the
`shape of the right half, you're describing the same information in either case.
`
`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
`
`
`
`15
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`The only difference being are you going to the left or are you going to the
`right. With that, I'll reserve whatever is left of my rebuttal time.
`JUDGE WEINSCHENK: Okay.
`MR. OLIVER: Before I start, Your Honor, would you prefer a paper
`copy of our demonstratives?
`JUDGE WEINSCHENK: Sure, thank you.
`MR. OLIVER: Good afternoon, may it please the Board. This is a
`simple case of a petitioner failing to meet its burden, and I agree with the
`Petitioner in this case that the two most important features to talk about can
`be seen with respect to Claim 20, which can be seen on slide 3 of our
`demonstratives. I think there's been a lot of discussion of the high-pass filter
`issue, so I'll start my discussion of that. But I will tee up for both of these
`what the issues are, go to the high-pass filter, and then we can address the
`second pass -- the second decoder and second coded signal.
`Now, with respect to Claim 20, as Petitioner's counsel pointed out, we
`had a second decoder receiving a second coded signal within a high
`frequency range that is higher than the low frequency. Now, with respect to
`Tucker, the signal that is coded is simply not in the high frequency range,
`and for that reason we do not believe the burden has been met with showing
`that that is what is in Tucker, as I'll address in more detail.
`Similarly, the second issue that's of importance here is that the second
`decoder applies a high-pass filter and, simply put, Tucker does not describe
`a high-pass filter. Now, there are two problems with the arguments that
`have been presented here today by Petitioner. First, they're improper in that
`it's a complete flip-flop from the original petition. The original petition put
`forth the idea that there was this high-pass filter existing in there, and Dr.
`
`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
`
`
`
`16
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`Quackenbush swore up and down that no low-pass filter existed in Tucker's
`box 26.
`Once it was established that that position was unattainable there was a
`complete 180 degree flip, and now the position is that there is a low-pass
`filter in box 26, but we should understand that to be a high-pass filter. On
`the first point, the flip-flop is improper because it's changed the arguments.
`On the second point, the current position is simply incorrect, and I'd like to
`deal with that incorrection first and foremost to highlight a top level idea.
`JUDGE WEINSCHENK: Mr. Oliver, before you get there you're
`saying that their initial position was that there's a high-pass filter?
`MR. OLIVER: Yes, Your Honor.
`JUDGE WEINSCHENK: Maybe I'm a little bit confused, because I'm
`looking at their petition on page 39 and they say that Tucker teaches a
`reflected low-pass filter, i.e. a high-pass filter. So, to me that was their
`argument that they were saying, yes, it's a low-pass filter but it's a reflected
`one. So, I'm not sure I see flip-flop in the same way that you do and maybe
`you can point that out to me.
`MR. OLIVER: Certainly, Your Honor. The flip-flop is in that what
`does it mean to be a reflected low-pass filter that is a high-pass filter. We
`asked Dr. Quackenbush about this, and I would also point to Dr.
`Quackenbush's actual declaration at paragraph 151, Exhibit 1002. And what
`he testified to in this declaration was that that's what was going on in box 26
`and this is a quote here "would be understood by a POSITA to teach use of a
`high-pass filter during the interpolation process." That's what his testimony
`was in his initial declaration at paragraph 151, so that was our
`understanding, okay. He's saying that the interpolation process, instead of
`
`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
`
`
`
`17
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`using a low-pass filter uses a high-pass filter, and we establish that that is
`simply incorrect.
`JUDGE JIVANI: But, counsel, doesn't the sentence immediately
`following that in paragraph 151 read interpolation typically involves
`upsampling followed by application of a low-pass filter as I described
`above?
`MR. OLIVER: Yes, Your Honor, I completely agree that the idea is
`that interpolation typically involves -- interpolation can involve an
`upsampling followed by some type of filter, and the typical filter is a
`low-pass filter. I understand what he's saying in box 151 to say, but in this
`case it would be understood to be a high-pass filter because of the exact
`language he uses. Now, in any event --
`JUDGE JIVANI: And then at the bottom of that paragraph he
`continues, reflection to the high band would be the equivalent of upsampling
`followed by application of a high-pass filter instead. And so I think what
`you just told me is that you read that paragraph and those sentences to
`expressly require a high-pass filter?
`MR. OLIVER: That's our understanding based on the questioning and
`we'll go through some more slides showing that he was asked about this
`specifically and in his first deposition indicated that there would absolutely
`be a high-pass filter and there would be no low-pass filter in box 26, which
`is obviously the opposite of what is being argued right now. That being
`said, and I think I should address some of those contradictions. I'm sorry, I
`keep looking at the monitor when I should be looking at the camera, so I
`apologize for that Your Honor.
`JUDGE JIVANI: I'm used to it, don't worry about it.
`
`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
`
`
`
`18
`
`
`
`Case IPR2017 00437
`Patent 6,772,114 B1
`
`
`MR. OLIVER: I'm sure you are. But before I get to that I think that's
`an important issue. I think what we should first look at is what the current
`position is by the Petitioners, and I believe that's on slide 12 of our deck.
`So, we have now the current position is upsample, low-pass filter. Now, that
`would be what I believe they're arguing is the interpolation process, that
`interpolation is upsample, low-pass filter, and box 26 does require reflection.
`JUDGE WEINSCHENK: Mr. Oliver, is that a reasonable reading of
`what Tucker teaches, because I think there's at least some admission by your
`expert that interpolation is generally understood to be upsampling followed
`by a low-pass filter?
`MR. OLIVER: Certainly, Your Honor. The general or the most
`common way that interpolation is done is upsample followed by low-pass
`filter. Our argument is not that that is not a typical way to interpolate. Our
`argument was that, first of all, there was a contribution between the
`interpretation of interpolation in box 32 and 26. In one case saying it was
`low-pass filter and the other case high-pass filter.
`If they had made an obviousness argument that said a person of
`ordinary skill in the art knows that you can interpolate using high-pass or
`low-pass filters we think it would be obvious in this case to choose a
`low-pass filter, that would be a different ball game and we wouldn't be
`disputing. But putting that aside I certainly take the position that there could
`be an obvious argument, although, not presented in the petition that this is
`what box 26 should look like.
`Now, what our problem with this is is not only is it changed from the
`original position we believe, but what they have circled or squared off in the
`dotted line is not a high-pass filter. It is a low-pass filter and reflect. Now,
`
`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
`
`
`
`19
`
`
`
`Case IPR2017 00437
`Patent 6,