`571-272-7822
`
`
`Paper No. 57
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`__________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`__________
`
`LSI CORPORATION and AVAGO TECHNOLOGIES U.S., INC.,
`Petitioners,
`
`v.
`
`REGENTS OF THE UNIVERSITY OF MINNESOTA,
`Patent Owner.
`
`__________
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`__________
`
`Record of Oral Hearing
`Held: January 19, 2021
`__________
`
`Before JENNIFER S. BISK, ROBERT J. WEINSCHENK, and
`CHARLES J. BOUDREAU, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONERS:
`
`
`KRISTOPHER REED, ESQ.
`EDWARD MAYLE, ESQ.
`Kilpatrick Townsend & Stockton LLP
`1400 Wewetta Street, Suite 600
`Denver, CO 80202
`303-405-8536 (Reed)
`303-607-3368 (Mayle)
`kreed@kilpatricktownsend.com
`tmayle@kilpatricktownsend.com
`
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`PATRICK McELHINNY, ESQ.
`MARK KNEDEISEN, ESQ.
`K&L Gates LLP
`210 Sixth Avenue
`Pittsburgh, PA 15222
`412-355-6334 (McElhinny)
`patrick.mcelhinny@klgates.com
`mark.knedeisen@klgates.com
`
`
`
`
`The above-entitled matter came on for hearing on Tuesday, January 19,
`
`2021, commencing at 1:00 p.m. EST, by video/by telephone.
`
`
`
`
`
`
`
`2
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`
`
`P-R-O-C-E-E-D-I-N-G-S
`
`1:00 p.m.
`JUDGE BISK: This is a hearing for IPR 2017-01068. I have a
`couple of administrative announcements before we get going.
`First, before we begin the proceedings, we'd like to have the
`attorneys spell their names for the court reporter. I'll have you do that in
`just a second. And then, after the hearing, if you could stay on the line for
`a few minutes to see if the court reporter has any questions for you, so we
`get a clear transcript.
`The second announcement is that I believe there's a public line in
`this hearing today, and I know there is some confidential information in the
`record. So, I just want the attorneys to make sure that you don't talk about
`anything here that cannot be made public.
`And then, the third thing is that we have Judge Weinschenk with us
`on video and we have Judge Boudreau who's joining us by the phone
`because he's having some power issues at his location.
`All right. So, could I have the Petitioners' attorney spell your name
`for the court reporter, please?
`MR. MAYLE: Good afternoon. This is Edward Mayle,
`M-A-Y-L-E, for Petitioners. On the line, also, is Kristopher Reed, lead
`counsel. Last name, R-E-E-D, and Kristopher is with a "K."
`JUDGE BISK: Okay. Thank you.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`3
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`And Patent Owner's Attorney?
`MR. McELHINNY: This is Pat McElhinny, M-C-E-L-H-I-N-N-Y,
`of K&L Gates. And with me is my partner, Mark Knedeisen,
`K-N-E-D-E-I-S-E-N.
`JUDGE BISK: Okay. Thank you.
`So, I believe we gave each party an hour of time. And, Petitioner,
`you can -- both parties, actually, can save some of their time for rebuttal, if
`they like.
`So, Petitioner, how much time, if any, do you want to save for
`rebuttal?
`MR. MAYLE: I'd like to save at least 20 minutes.
`JUDGE BISK: Twenty minutes? Okay. I will do my best to
`watch the time and give you a signal if you're coming close. I sometimes
`get wrapped up and forget. So, you might want to keep a note on the time
`yourself.
`Okay. Whenever you're ready.
`MR. MAYLE: Good afternoon, and may it please the Board, my
`name is Ed Mayle. I, along with Lead Counsel Kristopher Reed, represent
`Petitioners.
`What we're asking is that the Board find that Claims 13, 14, and 17
`of the '601 patent be found anticipated by Okada and by Tsang, T-S-A-N-G.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`4
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`We'll start with claim construction. If you look at slide 3, we have
`our demonstratives. Claim 13 is there. The only disputed claim in this
`IPR is the word "transition," which is in that claim. And as shown on slide
`4, the University is asking the Board to settle Claim 13 and this word
`"transition" with multiple magnetic limitation. The University argues that
`"transition" means a reversal in the magnetic orientation of adjacent bit
`regions along the recording track of a magnetic recording medium, but that
`is not the customary and ordinary meaning of "transition." And neither the
`intrinsic --
`JUDGE BISK: I'm sorry, can you say what the plain and ordinary
`meaning is?
`MR. MAYLE: Well, we know that it at least has to cover any
`binary system. So, it at least has to cover optical, magnetic, or any other
`binary system that you could conceive of. And it is a --
`JUDGE BISK: So, what about Patent Owner's alternative
`construction?
`MR. MAYLE: So, the alternative -- excuse me. I didn't know
`you were -- go ahead.
`JUDGE BISK: Oh, I'm sorry. I was just going to say that in your
`briefing you say that it's not helpful, but, to me, that construction, I'm not
`exactly sure what the problem with it is.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`5
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`MR. MAYLE: Well, first and foremost, the person of ordinary
`skill in the art would not be looking to the Webster's Dictionary to
`understand this patent. Setting that aside, it doesn't add any clarity other
`than the word "transition" that's already in the claim. For example, it
`introduces, the dictionary introduces this concept of a change in state or
`stage, and that could be ambiguous. You know, what is a state in this
`context? What is a stage? So, it seems to be not helpful.
`JUDGE BISK: Okay. So, what do you think a transition is? Is
`it a change between a 0 and a 1? And does it have anything to do with the
`physical medium? I guess I'm a little confused, I'll have to admit right at
`the beginning here, of how transition works in conjunction with the NRZ
`inverted encoding.
`MR. MAYLE: Okay.
`JUDGE BISK: And it seems like the parties are kind of talking
`around each other on that. So, just to give you a little background of what I
`think I understand, I think I understand that NRZ inverted encoding, in that
`scheme, a 1 symbolizes a transition. Okay. Is that your understanding?
`So, if we're talking about transition, I think it's going to depend on whether
`we're talking about a code word that's been encoded with NRZI or not.
`Am I right about that?
`MR. MAYLE: Right. Mostly. At the code word level, where in
`a real device there would be an encoder which is operating at the logical
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`6
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`level, at the level of 1s and 0s. And that's what, for example, Claim
`17 -- Claim 13 talks about binary sequences and imposing the constraints on
`the binary sequences. Claim 17 is the dependent claim that says NRZ, and
`it specifically says transitions from 0 to 1 and from 1 to 0. So, in NRZ,
`right, it's 0 to 1 and 1 to 0. In NRZI, a 1 would be a transition.
`But Claim 13 goes on and talks about the recorded waveform, and
`that's where the physical aspect comes in. At the end of the day of the
`method, there has to be a physical record being made on a disk --
`JUDGE BISK: Right.
`MR. MAYLE: -- but the transitions are at the binary level in the
`claims.
`JUDGE BISK: Okay. So, I know we're talking about claim
`construction, but can we transition for a minute to Okada and what it
`teaches?
`MR. MAYLE: Yes.
`JUDGE BISK: Because I'm a little confused by your argument on
`Okada. Basically, I'm confused about the reply brief. So, for instance,
`you have this Exhibit 1011 where you, basically, summarize the tables of
`Okada, and the first column are the input, the 8-bit inputs. The second
`column is what's in the tables of Okada. And then, I'm confused about
`what the third column is supposed to be. I thought, prior to reading your
`reply brief, that the third column was the 13-bit data from the tables encoded
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`7
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`using NRZI, but your reply brief seems to say it's the opposite. And I'm a
`little confused.
`MR. MAYLE: So, can we just go to my slide 12?
`JUDGE BISK: Yes.
`MR. MAYLE: And I have a couple of slides to walk you through
`
`that.
`
`JUDGE BISK: Okay.
`MR. MAYLE: Slide 12, that's from our reply brief. That's an
`annotated version of Okada's figure 6. Now, to start, I'll walk through that
`point and just make it up. Allow me to get to that point.
`JUDGE BISK: Okay.
`MR. MAYLE: On the left side of the figure, you'll see that 8-bit
`data words are coming in where it says, "Record Information." That's red
`on the slide.
`JUDGE BISK: Right.
`MR. MAYLE: Yes, that's (audio interference). And then, there's
`an 8 to 13 converter. That's No. 10, shown in blue.
`JUDGE BISK: Yes.
`MR. MAYLE: That's the part that's imposing the j and k constraint.
`Okay? We're not to the encoded waveform yet. That encoder is what is
`imposing the j and k constraints.
`JUDGE BISK: Okay.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`8
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`MR. MAYLE: And it's making 13-bit strings, which are the middle
`column for Exhibit 1011.
`JUDGE BISK: But how come here you say that they are 13-bit
`NRZI strings?
`MR. MAYLE: Right. So, to go from -- I'll get to that. So,
`they're in NRZI now. The NRZI modulator is what happens next. It's
`shown as figure 6 in Okada.
`JUDGE BISK: Okay.
`MR. MAYLE: And that is the portion that takes the data out of
`NRZI into NRZ.
`JUDGE BISK: So, I'm confused by that. It seems to me that
`Okada says the NRZI modulator puts it in NRZI format. So, if you look at
`column 4 of Okada, and then, the paragraph that's on 13 through 16, it says
`that, "Data after conversion is supplied to an NRZI modulator to become
`NRZI data." And you seem to be saying the opposite, that what comes into
`the modulator is NRZI data and what comes out is NRZ data.
`JUDGE BISK: Yes, it's NRZ data, which all the experts agreed,
`even the University's expert, Dr. McLaughlin, he had as an original opinion
`that Okada does not suppose (audio interference). But, on cross-
`examination, he reversed himself. That's in the record and we have it on a
`slide that --
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`9
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`JUDGE BISK: Yes, I saw his statement there, but it still doesn't
`make me understand how something that -- first of all, how is the converted
`information an NRZI string? Where does it become an NRZI string? And
`then, how does it come out of the NRZI modulator as NRZ? I'm not
`understanding that.
`MR. MAYLE: All right. There's a couple of examples of that
`where they make this very clear. If you look to column 9 in the Okada
`patent --
`JUDGE BISK: Okay.
`MR. MAYLE: -- the bottom paragraph on the left --
`JUDGE BISK: Okay.
`MR. MAYLE: -- they show you there's a bit string, it says after 8
`to 13 conversions. This is line 52.
`JUDGE BISK: Line what, 52?
`MR. MAYLE: 52.
`JUDGE BISK: Okay.
`MR. MAYLE: They say, "After 8 to 13 conversion." So, that
`means in box 10 on our slide 12. "After 8 to 13 conversions, there's a
`certain bit string." "When this data is subjected to the NRZI modulator,"
`which is box 6, "downstream event, it becomes" -- and it shows you. And
`if you compare those two, you can see that in the first bit string coming out
`of the encoder, wherever there's a 1, it will flip the bit from the previous bit.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`10
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`JUDGE BISK: Yes.
`MR. MAYLE: So, that means that the first string is NRZI. So,
`the output of box 10 is NRZI. Otherwise, this example would make no
`sense. That's the only way it makes sense.
`JUDGE BISK: It looks to me like it's the opposite. It says, after 8
`to 13 bit conversion, you have 00101. Okay? And then, when it's
`subjected to NRZI modulation -- that's box 6 -- it becomes -- and now the
`bits are flipping on a 1. So, there's a 00, then, where there's a 1, there's a bit
`flip. Then, there's -- oh, I see what you're saying.
`MR. MAYLE: Yes. And Professor McLaughlin agreed in cross
`that, if this is the NRZ format, that you look to the University briefs. They
`agreed that like a 1 here would be laser on; the 0 is laser off. So,
`everyone agrees. Both experts and all the parties are agreeing that --
`JUDGE BISK: So, can we look at your Exhibit 1011 --
`MR. MAYLE: Yes.
`JUDGE BISK: -- where you have those tables?
`MR. MAYLE: Yes.
`JUDGE BISK: Okay. Now let me get to it here.
`So, the first column is the 8-bit number that goes in. That's the red
`
`box.
`
`
`
`MR. MAYLE: Yes. Yes.
`
`11
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`JUDGE BISK: Then, the next one is the 13-bit encoded data, and
`then, your third column, you say it's NRZI recorded data, but, actually, now
`what you're saying is this is NRZ data?
`MR. MAYLE: Yes. So, that label introduced a lot of confusion.
`JUDGE BISK: Yes.
`MR. MAYLE: If you look to Professor Soljanin's declaration -- so,
`she made this exhibit and put that label up there, which doesn't specify
`(audio interference) recording. In her declaration that goes along with it,
`she always said the third column was after the NRZI mod. And the
`University in its surreply -- I think on page 17 -- also called it "after." So,
`that label should really be "after NRZI modulator." And so, after box 6 in
`the Okada slide, on my slide 12, that's what's taking it out of NRZI before it
`gets recorded. So now, it --
`JUDGE BISK: Okay. So now that I understand that, now let me
`ask you about what a transition is here. So, you're saying that a transition
`is when there's a change from a 0 to a 1 or a 1 to a 0?
`MR. MAYLE: In that third column that gets recorded, that's
`correct. You've got it. Yes.
`JUDGE BISK: Okay. So, then, can you explain, in your petition
`where you talk about Okada -- hang on. So, I'm not sure I'm going to be
`able -- here it is. It's on page 24 of your petition. You say that, all I see is
`you're saying, perform data conversion "before NRZI modulation in such a
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`12
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`way that 1 will not appear three or more times in a row." So, you're saying
`that, before it's converted from NRZI to NRZ, that it's in NRZI, so 1 will not
`appear more than three times in a row because 1 means a transition?
`MR. MAYLE: So, before the modulator, 1 is a transition.
`JUDGE BISK: Yes.
`MR. MAYLE: That point, I want to address this, page 24.
`JUDGE BISK: Okay.
`MR. MAYLE: If you look back at page 22 in the petition, this is
`addressing Claim 1, not Claim 13.
`JUDGE BISK: Right.
`MR. MAYLE: Claim 1, indeed, has an extra limitation that Claim
`13 does not.
`JUDGE BISK: But, when you go to Claim 13, all you say is, all
`you do is rely on Claim 1, right?
`MR. MAYLE: Yes, but, if I may, Claim 1 required, also, that the j
`constraints -- it's at the bottom of that page 22 -- "facilitate the reduction of a
`probability for a detection error in said receiver means" --
`JUDGE BISK: Okay.
`MR. MAYLE: That limitation is not in Claim 13. Now, in the
`District Court, the University tried to read it in, but they didn't try to here.
`But, on Okada, the way Okada does that -- and again, it's only Claim 1 -- the
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`13
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`way Okada detects errors had to do with this three 1s in a row. Those are
`not the transition. That's just the way Okada did Claim element one.
`JUDGE BISK: Okay.
`MR. MAYLE: And so, that got propagated through. And so,
`that's true; that's what Okada discloses, but that's not what's at issue in Claim
`13.
`
`JUDGE BISK: Okay. So, you're saying that a transition is a
`change between a 0 and a 1 and a 1 and a 0 on the recorded waveform?
`MR. MAYLE: Yes, in Okada, that's correct.
`JUDGE BISK: Okay.
`MR. MAYLE: Correct.
`JUDGE BISK: Okay. I'm sorry about that. I just was very
`confused and I needed to get that cleared up before I could understand your
`argument. You can now go ahead with your agenda.
`MR. MAYLE: Okay. I want to truncate the claim construction
`argument a little bit in the interest of time.
`JUDGE BISK: Okay.
`MR. MAYLE: If you just look at the claim language itself, there's
`nothing in Claim 13 about magnetic or anything like that. In Claim 17,
`which depends from it, it specifies exactly what we're saying; it flips from 0
`to 1 and from 1 to 0. And you see on our slide 5 the University's experts
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`14
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`testified that those would not be transitions. And so, it's a dependent claim.
`So, they've got to be wrong on it.
`That specification, the Phillips case teaches that there's two times
`when the specification can depart from the customary and ordinary meaning.
`One is when there's a special definition, in other words, a (audio
`interference). The University doesn't even argue that. It's not there.
`And the second is when there's a clear and unambiguous disavowal
`of the claim scope in the specification. It's not there, and the University
`actually resists that. They're not arguing that. They say you don't need to
`show that. They are not even arguing that. So, there's nothing in the
`specification that says, in that progression that you can only use magnetic.
`In fact, as shown on slide 6, you can see where the magnetic is
`summarized. It says that invention applies to devices "such as magnetic."
`But earlier in the patent it also mentioned optical recording.
`And on slide 7, I have two examples of the customary and ordinary
`meaning of "transitions." On the left, you see a patent from Dr. Immink,
`who's the creator of the compact disc. And he says transitions can be used
`for "a magnetic or optical medium."
`On the left of page 7, you see Dr. McLaughlin's own patent, and he,
`himself, used the word "transitions" before the time of the invention to refer
`to optical recording channels. So, the customary and ordinary meaning has
`to at least include these optical binary systems.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`15
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`And on slide 8, I don't want to get into the confidential material here,
`but you see a slew of exhibits where everyone at the University before this
`IPR -- the inventors, both the inventors, their patent attorneys, the executives
`at the University, their licensing expert, and others -- everyone knew that
`Claim 13 covers optical. And so, it's pretty apparent what the University is
`trying to do here now. They're trying to save the claims from the Okada
`patent by asking that you rewrite them, which we hope that you do not do.
`On slides 9 and 10, there was also some potential other confusion in
`the record on how to count the transitions. There's actually three ways in
`the record to do it, starting with slide 9. If you read the literal language of
`Claim 17, it says that the transitions go from 0 to 1 and from 1 to 0. So,
`kind of a round trip. That can be one. That's one way to do it.
`And Professor Soljanin, our expert, testified in her deposition that
`you can count one-way trips, just 0 to 1s, or you could count 0s, or you can
`sum them and do both. So, you'll arrive at a different number to describe
`the same thing.
`JUDGE BISK: So, I'm sorry, which of these different counting
`methods are you using or is your expert using?
`MR. MAYLE: Our expert testified that she used 0 to 1 only.
`JUDGE BISK: Okay. So, that would be the second one here?
`MR. MAYLE: Correct. So, an example on the next slide, 10, you
`see 01010.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`16
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`JUDGE BISK: Yes.
`MR. MAYLE: If you do it the way, you know, if you read Claim
`17 literally, and you think it's hard to go up and down the mountain twice, 0,
`1, and 0 is 1, and then, that 1 and 0, that's 2. Then, you can only do the 0s
`to 1s. Again, you're looking at the same thing; there's different ways to
`describe the same thing.
`The second and third methods, which is what Dr. Soljanin used,
`come up with 2 with the same bits.
`The University, basically, double-counts. They look at every hop
`in either direction, and then, they count them all and sum them up, and they
`get to four transitions there.
`So, hopefully, that clears it up. But this case doesn't depend
`upon -- you know, all three of those ways are arguably acceptable. But the
`outcome of this case will not depend upon that, and I'll get to that and I'll
`show that Okada has anticipate under any convention. Because, remember,
`Claim 13, the independent claim as no upper bound. It's just a letter "j"; j
`consecutive transition can be as high as you want to go. As long as it's
`(audio interference), it's anticipated. Both references have that.
`JUDGE BISK: It's only Claim 14 that has the upper bound, right?
`MR. MAYLE: Yes, and 17, which depends on 14.
`JUDGE BISK: Okay.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`17
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`MR. MAYLE: It's less than 10. It doesn't show -- there's no
`dispute about Tsang; there's only dispute about Okada. But, as we'll show,
`there's less than 10.
`But let's get to Okada. We've done a lot of it already on slide 12.
`Now let me go back to the converter part, No. 10. And Okada instructs that
`column 3, lines 16 to 63, that the 8 to 13 converter 10 -- and I'm
`quoting -- "The 8 to 13 converter 10 performs data conversion to expand the
`8-record information to 13-bit data according to one of the following tools."
`That's in Rule 1 and Rule 2.
`Now Okada indicates that it's up to the user to decide which of those
`two rules to follow. But picking one of those two rules is not the end of the
`process of imposing j and k constraints on the 13-bit sequences. This is
`very important to understand. So, I just will say it again. Just picking one
`of the two rules in the abstract is not the end of the Okada process of how
`Okada imposes constraints.
`JUDGE BISK: So, I have a question about this, the Rule 1 and
`Rule 2. So, Patent Owner says that both are required, even though you
`only use one at a time. And it's shown, in the tables, they use Rule 1 for,
`you know, all of the input up to, I guess, B8, and then, from B8 on, they use
`Rule 2. But the entire algorithm, I guess is what you're saying, requires
`both rules to be used. And you're saying that the two rules can be used
`independently on any number? And you could use Rule 1 for everything or
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`18
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`you could use Rule 2 for everything, and have a complete algorithm? Is
`that your position?
`MR. MAYLE: Yes. Yes, but I'll point you to the petition at page
`23 first --
`JUDGE BISK: Okay.
`MR. MAYLE: -- just so you can see what our petition really was.
`There was no new argument here. You see the petition under that figure 5,
`page 23. It says -- and I'm taking this right out of the patent -- "The 8 to 13
`bit converter 10 expands 8-bit input record information to 13-bit data
`according to either one of the two rules." And then, we say at the last
`column -- oh, sorry -- the last paragraph, it says, Rule 1 and Rule 2 of
`Okada, "each imposes," blah, blah, blah. So, they each anticipate either
`one. So, first off, that's our petition.
`JUDGE BISK: Right.
`MR. MAYLE: Now, in Okada, Okada -- I'll point you to -- as I
`was saying, you don't just pick one of the rules. If you look to Okada, at
`the top of column 4, that top right of column 4,- you see it says -- and I don't
`have the line number -- but it says, "Since the 8 to 13 converter 10 has no
`output rule to an input rule, a user needs to arbitrarily prepare output patterns
`that follow a selected rule." So, let me unpack that.
`So, if you have 8 bits coming in, that's two of these possibilities.
`When you have 13-bit code words coming out, you have many, many more
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`19
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`potentials. That's 8,192 to the 13th power potential code words. And so,
`the user, first, picks one of those rules, and then, the user, as it instructs in
`column 4, arbitrarily creates a set of 13-bit code words that follow the rule
`that was selected. And then, you have to map them back, and that's what
`those tables do. Now you have to assign a code word to an input word.
`It's this two-step process that imposes the j and k constraints that's
`described in Claim 13. You first select the rule, and then, you create the
`subset of the 13-bit values that comply with the rule and map them back to
`the inputs. And Okada gives the several examples.
`Before I go into that, let me just talk about the claim. If you look at
`Claim 13, which is what we're talking about here, you know, it's a simple
`method for encoding an unspecified number of n-bit data where it's to an
`unspecified number of n-bit code words.
`So, the University seems to be arguing that there needs to be two of
`the n code words. In other words, one for every input. That's Claim 18,
`which is not at issue here.
`Another dependent claim expressly requires a different, you know, a
`further limitation on the method, where there's a mapping of all to the n
`power data words to the n-ended code words. That's not Claim 13. So,
`first off, their whole argument is the wrong claim.
`So, let me just continue on with Okada. Okada is giving us several
`examples of how this can be done. For Rule 1, there are seven examples,
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`20
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`tables 1 through 7. For Rule 2, Okada provides two examples in tables 8
`and 9. Each table reflects the creation of a given set of arbitrary 13-bit
`encoded sequences that comply with either Rule 1 or Rule 2, and then,
`assigning those code words back to the inputs. That is how Okada imposes
`the constraints. Again, it's not just the selection of a rule -- the rule by
`itself doesn't do it -- but also the creation of a specific set of rule-compliant
`13-bit encoded sequences that map back to the inputs.
`Now we talked about the NRZI modulator. So, again, that's what
`takes the data from NRZI format and converts it to NRZ. And I pointed
`you to the specification of some examples where that was done.
`If you look at slide 13, we have an example here. In table 8 there
`are 31 8-bit data words that can be potentially input into the 8 to 13
`converter. To correspond with those 31 8-bit data words, Okada selected in
`the table 31 13-bit encoded sequences. They all comply with Rule 2.
`And the selection of these particular 31 Rule 2-compliant 13-bit encoded
`sequences out of the hundreds of other Rule 2-compliance 13-bit sequences
`that could have been picked, that is the thing that imposes the j constraint
`and the k constraint that's emerging from the encoded waveform from the
`converter, according to table 8.
`The same thing happens with the 8 to 13 converter programmed in
`compliance with Rule 2. And for an example, in table 9, there's nine
`unique 8-bit code words given there.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`21
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`JUDGE BISK: And could I ask you to jump a little bit to the k
`constraint? So, Patent Owner makes an argument that Rule 1 allows for 13
`0s in a row, which would, then, lead to, basically, no k constraint.
`MR. MAYLE: Yes.
`JUDGE BISK: But you said that the tables don't show any -- so, is
`your argument that it could, but you just don't choose that?
`MR. MAYLE: Yes.
`JUDGE BISK: Okay.
`MR. MAYLE: Yes. The anticipating disclosure does not show
`that. Theoretically, only if you interpret -- I think Rule 1 says there has to
`be an even number of 0s. If you interpret 0 as an even number, which I
`think is not a fair interpretation, but give them that for argument's sake, that
`means there can be 0, I guess --
`JUDGE BISK: I'm not sure that Rule 1 says anything -- Rule 1 just
`says "at least one 0" and an even number of 1s, of consecutive 1s.
`MR. MAYLE: Right. Right, right. So, they're saying one 0 and,
`then, 12 more 0s.
`JUDGE BISK: Yes.
`MR. MAYLE: They're saying an even number of 1s to 0.
`JUDGE BISK: Oh, oh, I see what you're saying.
`MR. MAYLE: Yes.
`JUDGE BISK: Okay. Now I understand.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`22
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`MR. MAYLE: Right.
`JUDGE BISK: Thank you. Yes.
`MR. MAYLE: That seems to be absurd, and I see you laughing at
`it. But, even if it wasn't absurd, that's not what Okada uses. Again, the
`point I was trying to make is you don't -- Okada rules aren't taken in the
`abstract. You have to actually assign the code words, and then, map them
`back.
`
`And one other point on that. In patent law, as long as one
`embodiment anticipates the claim, the claim is dead. It doesn't matter that
`they could come up with some strained hypotheticals that might not
`anticipate, according to their argument; the claim is still invalid. It's a red
`herring.
`JUDGE WEINSCHENK: Mr. Mayle, I have a question.
`MR. MAYLE: Yes, sir?
`JUDGE WEINSCHENK: Is there a specific value of j and k that
`results from these rules in Okada?
`MR. MAYLE: Not from the rules. You would have to look in the
`tables and count the 1s and 0s. You have to count a concrete set of
`rules -- sorry, not rules -- code words. You can't just look at the rule,
`because, as we just demonstrated, you could come up with something with
`13 0s that wouldn't work. So, what we look at is the embodiment --
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`23
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`JUDGE WEINSCHENK: Can you give me an example from one
`of the tables and tell me what the j and k values are in that example?
`MR. MAYLE: Sure. In our briefs in tables 8 and 9, we found that
`the worst-case scenario for the j constraint, there was a code word -- it's in
`our reply brief; I'd have to look it up, which one. There was a code word
`that ends in four transitions if you double count 0101s. If you counted the
`way the University --
`JUDGE WEINSCHENK: Mr. Mayle, it's helpful for me to see
`something. So, maybe can we use the example that you have on your slide
`13 here and go through that, and tell me what the j and k values will be for
`that one?
`MR. MAYLE: Right. So, these are only code words, a single
`report. What you have to do is look at the sequences and how you combine
`them.
`
`So, let me find the reply brief. There's a better one.
`If you look at page 16 of our reply brief, we address tables 8 and 9
`on page 16. And we've said that the data word FC in table 9, if you look at
`table 9 of Exhibit 11, it has four transitions from 0 to 1, from 1 to 0.
`JUDGE WEINSCHENK: Hold on, Mr. Mayle. You've got to
`slow down for a second for me.
`So, we're at table 9?
`MR. MAYLE: And it's 1011.
`
`
`
`24
`
`
`
`IPR2017-01068
`Patent 5,859,601 B2
`
`
`
`JUDGE WEINSCHENK: Okay. I'm at table 9 in Exhibit 1011.
`What am I looking at now?
`MR. MAYLE: If you look at the two code words FC and F8 --
`JUDGE WEINSCHENK: Okay.
`MR. MAYLE: So, that's the worst-case scenario for j. One of
`them ends in four transitions and one of them begins with four transitions, if
`you count transitions.
`JUDGE WEINSCHENK: How do I know where there's a
`transition here?
`MR. MAYLE: So, 0 to 1 flips or 1 to 0 flips are t