throbber
UNITED STATES DISTRICT COURT
`CENTRAL DISTRICT OF CALIFORNIA - WESTERN DIVISION
`HONORABLE MARIANA R. PFAELZER, U.S. DISTRICT JUDGE
`
`1
`
`THE CALIFORNIA INSTITUTE OF TECHNOLOGY, )
`)
`Plaintiff/Counter-Defendant, )
`)
`Case No.
`)
`) CV 13-7245 MRP (JEMx)
`)
`)
`)
`
`vs.
`HUGHES COMMUNICATIONS, INC.; HUGHES
`NETWORK SYSTEMS, LLC; DISH NETWORK
`CORPORATION; DISH NETWORK, LLC; and
`DISHNET SATELLITE BROADBAND, LLC,
`Defendants/Counter-Plaintiffs.)
`________________________________________)
`
`))
`
`REPORTER'S TRANSCRIPT OF
`CLAIMS CONSTRUCTION AND MOTION HEARING
`WEDNESDAY, JULY 9, 2014
`1:05 P.M.
`LOS ANGELES, CALIFORNIA
`
`________________________________________________________
`MYRA L. PONCE, CSR NO. 11544, RMR, CRR
`FEDERAL OFFICIAL COURT REPORTER
`312 NORTH SPRING STREET, ROOM 430
`LOS ANGELES, CALIFORNIA 90012
`(213) 894-2305
`
`UNITED STATES DISTRICT COURT
`
`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
`
`Hughes, Exh. 1026, p. 1
`
`

`

`APPEARANCES OF COUNSEL:
`
`2
`
`FOR THE PLAINTIFF/CROSS-DEFENDANT:
`QUINN, EMANUEL, URQUHART & SULLIVAN
`BY: JAMES R. ASPERGER
`Attorney at Law
`865 South Figueroa Street, 10th Floor
`Los Angeles, California 90017
`(213) 443-3000
`QUINN, EMANUEL, URQUHART & SULLIVAN
`BY: KEVIN P.B. JOHNSON
`BY: MARK TUNG, Ph.D.
`Attorneys at Law
`555 Twin Dolphin Drive, Fifth Floor
`Redwood Shores, California 94065
`(650) 801-5000
`QUINN, EMANUEL, URQUHART & SULLIVAN
`BY: EDWARD J. DeFRANCO
`BY: ERIC HUANG
`Attorneys at Law
`51 Madison Avenue, 22nd Floor
`New York, New York 10010
`(212) 849-7000
`
`FOR THE DEFENDANTS/CROSS-PLAINTIFFS:
`WILMER, CUTLER, PICKERING, HALE & DORR
`BY: JAMES M. DOWD
`BY: AARON S. THOMPSON
`Attorneys at Law
`350 South Grand Avenue, Suite 2100
`Los Angeles, California 90071
`(213) 443-5309
`WILMER, CUTLER, PICKERING, HALE & DORR
`BY: WILLIAM F. LEE
`BY: RICHARD GOLDENBERG
`Attorneys at Law
`60 State Street
`Boston, Massachusetts 02109
`(617) 526-6556
`
`UNITED STATES DISTRICT COURT
`
`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
`
`Hughes, Exh. 1026, p. 2
`
`

`

`3
`
`WEDNESDAY, JULY 9, 2014; 1:05 P.M.
`LOS ANGELES, CALIFORNIA
`-oOo-
`THE COURTROOM DEPUTY: Calling Item No. 1,
`CV 13-7245 MRP, The California Institute of Technology versus
`Hughes Communications, Inc., et al.
`Counsel, please state your appearances.
`MR. ASPERGER: Good afternoon, Your Honor. Jim
`Asperger appearing on behalf of CalTech. With us from CalTech
`are Jennifer Lum, Chantal D'Apuzzo, Hannah Dvorak-Carbone, and
`Jennifer Shockro.
`THE COURT: Fine. I say welcome to them all.
`MR. JOHNSON: Good afternoon, Your Honor. Kevin
`Johnson on behalf of CalTech.
`MR. DeFRANCO: And, Your Honor, Ed DeFranco also on
`behalf of CalTech.
`MR. TUNG: Mark Tung on behalf of CalTech.
`MR. HUANG: Eric Huang on behalf of CalTech.
`THE COURT: All right.
`MR. LEE: Good afternoon, Your Honor. Bill Lee from
`Wilmer Hale with Aaron Thompson and Richard Goldenberg and Jim
`Dowd for Hughes. And with us are Dean Manson and Kim Jessett
`from Hughes.
`THE COURT: All right. Please.
`That was -- that was very helpful yesterday.
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 3
`
`

`

`4
`
`MR. JOHNSON: Thank you, Your Honor. Again, Kevin
`Johnson. And we plan on starting, as we discussed yesterday,
`with -- by going back and forth and starting with the "repeat"
`term. And I'm going to begin by addressing Your Honor's
`question. And just so -- so we're all on the same page, of
`course there are -- and we've handed out slides, Your Honor,
`to -- to Your Honor as well as to the clerk and to the court
`reporter.
`
`THE COURT: All right.
`MR. JOHNSON: So there are four CalTech patents
`involved in this lawsuit. And I'm at Slide 2. And there are
`six terms that are the subject of discussion for today. And
`I'm going to start with the "repeating" term. And I'm going to
`answer Your Honor's question first, and then I'll discuss the
`competing constructions and the intrinsic evidence with respect
`to the "repeating" term.
`And, Your Honor, from the CalTech's side, the idea is that
`I will deal with the "repeat" term, the "transmission" term,
`and the "codeword" terms. And then my partner Mr. DeFranco
`will deal with the balance of the terms.
`THE COURT: The indefiniteness.
`MR. JOHNSON: That's right. The Tanner graph, the
`equation, and "combined."
`THE COURT: That's fine.
`MR. JOHNSON: So, Your Honor, the Court posed the
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 4
`
`

`

`5
`
`following question at Slide 5. "If the term 'repeat' were
`construed to require duplication of bits, would this
`construction necessarily exclude from the scope of the claims
`the LDGM embodiment described in column 3, lines 51 to 59, of
`U.S. Patent No. 7,116,710?"
`THE COURT: I raised this, frankly, yesterday.
`MR. JOHNSON: Right. Right. And the answer is that
`assuming that "duplication" means "re-use," the LDGM embodiment
`is not excluded. It would be covered by the definition of
`"repeating."
`THE COURT: I know that.
`MR. JOHNSON: However, if -- if Hughes contends that
`"duplication" means something other than "re-use," if it means
`copying bits, it's still going to be a disputed issue as to
`whether the LDGM is covered or not. And let me explain why.
`THE COURT: I said to you yesterday, you want to
`include in the meaning of the term that alternate embodiment.
`MR. JOHNSON: That's correct, Your Honor.
`THE COURT: Yes.
`MR. JOHNSON: And that's because a person skilled in
`the art reading the patent and the specification and the claims
`would understand that the term "repeat" has to cover everything
`that -- all of the embodiments that's disclosed in the patent.
`THE COURT: Quite so.
`MR. JOHNSON: And so it has to disclose and has to
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 5
`
`

`

`6
`
`cover both the notion of duplication, which I'll get into in a
`sec, but it also has to cover this alternate embodiment. And
`the alternate embodiment is, in fact, different. And a person
`of ordinary skill in the art would understand that it operates
`differently and it doesn't sequentially duplicate bits as --
`THE COURT: No.
`MR. JOHNSON: -- Hughes argues.
`THE COURT: Right.
`MR. JOHNSON: And it doesn't duplicate bits if by
`"duplication" Hughes means copied the bits. And we've seen a
`reference to copying the bits in Dr. Stark's declaration.
`And so -- and that's because -- we talked about it in the
`tutorial yesterday, the fact that when you have an LDGM, a
`low-density generator matrix, it engages in a mathematical
`process of linear algebra where the bits are being multiplied
`and re-used.
`THE COURT: Yes, I know that.
`MR. JOHNSON: Okay. So if we just go to Slide 6.
`And this is -- on the top of the slide I've pulled up the
`portion of the specification that Your Honor referred us to.
`The LDGM is an alternative embodiment that uses a low-density
`matrix as the coder.
`The purpose of the LDGM coder is to use a low-density
`sparse matrix to reduce the complexity of the calculations in
`generating the parity bits. And there's no dispute on either
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 6
`
`

`

`7
`
`side that that's what the purpose of an LDGM is. It uses a
`sparse matrix, which has a low number of 1's in it, to reduce
`the complexity of the calculations. And it uses linear algebra
`to perform this matrix multiplication, and it has to re-use the
`bits as part of that matrix multiplication.
`Now, we also know that the LDGM code has randomness
`already present in the structure, and that is right out of the
`patent specification. If we look at the portion of the spec
`that I've highlighted on this slide, it says right there
`that --
`
`THE COURT: Didn't I read this to you yesterday?
`MR. JOHNSON: I believe you did, Your Honor, yes.
`"Randomness already present in the structure" of the LDM -- of
`the LDGM.
`So yesterday, Your Honor, Hughes put up this slide in its
`tutorial. And there was a lot of discussion about Slide 44.
`And I've reproduced it at the top here.
`THE COURT: Yes. That's right.
`MR. JOHNSON: And they put this slide up,
`Your Honor. And if you read the portion that's bold, it says,
`"This is an example of a (very small) LDGM code that performs
`an irregular repeat."
`THE COURT: Yes.
`MR. JOHNSON: But we know when pressed and -- and
`asked of whether, in fact, this was an LDGM code,
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 7
`
`

`

`8
`
`Mr. Goldenberger -- Mr. Goldenberg yesterday admitted that this
`is not an LDGM code. And that's because -- and, Your Honor, if
`I may approach the screen from time to time.
`THE COURT: Yes. Anytime you want.
`MR. JOHNSON: Thank you, Your Honor.
`And that's because this LDGM code has five 1's -- two,
`three, four, five -- and five 0's. And there -- they've been
`manipulated so there is a 1 in each column and a 0 in each
`column.
`And Hughes admitted yesterday this isn't an LDGM code
`because LDGM codes are sparse, low-density, and they're random,
`as we just looked in the patent specification. A person of
`ordinary skill in the art would know that there's randomness
`associated with it. So that's not an LDGM matrix.
`And Your Honor may recognize on Slide 8, this is a matrix
`that is reproduced from Hughes's responsive brief at page 6 and
`it's also from Dr. Stark's, Hughes's expert, declaration at
`paragraph 49.
`And they say that this is an example generator matrix.
`But again, this is not a low-density, sparse generator matrix.
`It has no randomness. It's a structured matrix that's designed
`by Hughes to perform duplications. We see that the 1's are
`organized in a structured fashion there. And it's not sparse.
`The 1's are all clustered together. So this isn't an LDGM
`either.
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 8
`
`

`

`9
`
`And when we look at Slide 9, the top of this slide is,
`again, a reproduction from Hughes's tutorial yesterday,
`Slide 42.
`
`THE COURT: Yes, it is.
`MR. JOHNSON: They put this up to say in an LDGM
`code, the generator matrix is low-density, meaning that most of
`its entries are 0.
`Now, we agree that the size of this matrix is more
`accurate than the size of the matrix here that's two-by-five,
`with the five 1's and the five 0's. But going back to slide --
`my Slide 9, in this slide Hughes intentionally placed no more
`than a single 1 in each column. And that's intentional and
`it's important to distinguish so that -- they've been
`highlighted in blue here. And you see --
`THE COURT: I see them.
`MR. JOHNSON: And again, this is not random. It's
`structured in a way to show 1 -- a single 1 in each column.
`And that's -- that was done intentionally.
`When you go forward and you look at -- and, Your Honor,
`Hughes submitted a video tutorial as part of its tutorial prior
`to yesterday's in-person tutorial. And what we did was we
`pulled this slide, this video -- this is an image that was
`captured from their video. And this is my Slide 10.
`This is an LDGM code. And this shows how a person of
`ordinary skill in the art would understand an LDGM code, that
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 9
`
`

`

`10
`
`it can have one bit in a column but it can also have more than
`one bit in a column, as I've highlighted here. And again, the
`top part of this is from Hughes. And it shows more than one
`bit in a column and the code -- this LDGM code proves our
`point. It establishes that the bits are re-used but are not
`sequentially duplicated.
`And so a person of skill in the art would understand,
`reading the patent, that a person -- that an LDGM code can have
`more than one in a column. It can have more than one 1 in a
`column. It can have one 1 in a column, but it has to be able
`to anticipate and be able to have more than one 1 in the
`column. And the reason that that's important is because this
`slide that's up here, this LDGM code -- and I -- CalTech and, I
`think, Hughes would even agree to this, this repeats, it
`re-uses the bits, but it doesn't sequentially duplicate the
`bits.
`
`And so because the claims require the term "repeat,"
`"repeat" must include the notion that this LDGM code would be
`covered. And in covering this LDGM code, it means that the --
`there's no sequential duplication here. And if we multiply out
`this LDGM code by the information bits, we will see that bits
`are re-used but are not sequentially duplicated.
`If we move forward to Slide 11, you know, going -- again,
`this is a reproduction of Slide 44 from Hughes's tutorial.
`Their Slide 44 is incorrect for another reason. And -- and
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 10
`
`

`

`11
`
`that is when you -- when you look at the five -- when you look
`at the matrix again at the top, the contrived nature of
`Hughes's matrix is demonstrated by simply switching these two
`bits or any two bits.
`If you were to make this 1 a 0, as I've done here in red,
`and if you were to make this 0 a 1, as I've done here in red,
`switching those two bits eliminates the sequential duplication.
`And you can see that by -- if you follow out the
`multiplication --
`THE COURT: No, I understand that. I do.
`MR. JOHNSON: So the point is that a person of
`skill, reading the patent with the intrinsic evidence, would
`understand that the term "repeat" has to cover all of these
`situations. And -- and that's why as we move forward -- and
`we'll get into the intrinsic evidence in a second -- "repeat"
`on itself is actually a pretty good term for a person of skill
`to understand.
`THE COURT: Yes.
`MR. JOHNSON: And it doesn't -- and so I go back to,
`you know, the plain and ordinary meaning of it from our
`standpoint is "re-use," sequential duplication is -- adds
`ambiguity and imports additional limitations and really tries
`to limit these patents to one particular embodiment when we all
`know that there are alternate embodiments to disclose.
`THE COURT: I want you to know that what I am
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 11
`
`

`

`12
`
`concerned about -- I can see that point. But down the line,
`"re-use" is not clear and not more helpful. And "repeat," in
`its plain and ordinary meaning -- I'm not talking about to a
`layman -- but in its plain and ordinary meaning seems to me to
`be better.
`
`MR. JOHNSON: I mean, we -- we agree that "repeat,"
`the term "repeat," the plain and ordinary meaning is good
`enough. And that, you know, as we went through --
`THE COURT: It's got to be good enough. That's what
`everybody uses.
`MR. JOHNSON: Right. And so from our standpoint --
`we went down this path of trying to come up with synonyms. We
`know what it doesn't mean. It doesn't mean sequential, it
`doesn't mean duplication or copied. It has to be broader than
`that.
`
`And -- and so as we look through the intrinsic evidence
`and we looked at the extrinsic evidence to come up with a
`term --
`
`THE COURT: That, however, is not particularly
`desirable, from the standpoint of the Court, down the road.
`MR. JOHNSON: Well, maybe as we go through this a
`little bit more, there will be some other ideas. And, you
`know, from my standpoint --
`THE COURT: I don't -- what I'm asking you is: Why
`are we looking for an idea?
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 12
`
`

`

`13
`
`MR. JOHNSON: Well, because Hughes is trying to read
`in limitations.
`THE COURT: I understand. But we'll leave that for
`
`them.
`
`MR. JOHNSON: Right. So we would be fine just
`leaving the term as "repeat" and having it as the plain and
`ordinary meaning of "repeat," as long as it doesn't include
`"sequential duplication." This may fall into one of those
`cases where a negative limitation makes sense here because
`that's how Hughes is trying to read "repeat."
`We know what it isn't. It's not sequential duplication.
`It's not duplication.
`THE COURT: Go on.
`MR. JOHNSON: So, Your Honor, if we go to Slide 12,
`as I said, the LDGM multiplies the message word, which is a
`collection of k bits, with a generator matrix. The generator
`matrix is here in -- it's labeled "G" here. And the codeword
`is generated using linear algebra, special math. It's referred
`to as matrix multiplication.
`And the patent spec we talked about defines the LDGM as a
`sparse, low-density generator matrix. But it doesn't perform
`sequential duplication because it doesn't repeat the same bits.
`You know, a 1 doesn't get repeated four times as 1, 1, 1, 1.
`Instead, the repeat is accomplished by using vector
`information's -- vector information bits.
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 13
`
`

`

`14
`
`So here we see on this side -- and this is from our
`Exhibit J, the Rizzo article. This shows the vector on the
`right side. Yesterday we heard Hughes say that it -- most of
`the time they see it on the left. I've seen it most of the
`time on the right, but it doesn't much matter for purposes of
`discussing the -- the point is there's a process of multiplying
`what we see under x here by this matrix G --
`THE COURT: Yes.
`MR. JOHNSON: -- to generate the parity bits on the
`left. And that's what "re-use" means. You're re-using those
`bits in the multiplication of the matrix.
`And so if I look at Slide 13 -- and this is just intended
`to be, again, an example. This matrix here in the middle is
`not a low-density generator matrix. This is, instead, used
`just to explain what the linear algebra is.
`What we see is that you multiply each bit in the first row
`here of the matrix with the corresponding bit in the message
`word. So you take this 1 and you multiply it by this 1 in red,
`and that's shown down here. 1 times the 1 red. Then you take
`this 1 and you multiply it by this 0 in orange, and that's the
`1 times the 0 over here. Then you take this 1, multiply it by
`the blue 0. That's what we see here, 1 times the 0 in blue. 0
`times the green 1; that's what we see here. 0 times the blue
`1. And then, finally, 0 times the purple 1. And that is the
`process of linear algebra we learned in college when we were
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 14
`
`

`

`15
`
`studying engineering.
`And you end up, when you follow that multiplication out,
`with a code -- with basically a -- with parity bits of 1, 0, 1,
`1, 1. Each bit of this message word is re-used according to
`the 1's present in the generator matrix. And what that means
`is we see that this bit gets re-used because there's a 1 here.
`So --
`
`THE COURT: Yes. That's right.
`MR. JOHNSON: We see it in red. And we see -- so
`whenever you multiply the 1 in the matrix by this vector over
`here (indicating), you're re-using the bits and you're re-using
`them according to the 1's that are present in the generator
`matrix. And that's why when we go back -- that's why here in
`this example of an LDGM, when the -- when you have more than
`one 1 in a column and when the 1's aren't next to each other,
`adjacent to each other -- either right next to each other or on
`top of each other, you're not going to have sequential
`duplication.
`So the patent has to cover both of those ideas, both --
`when you have more than one 1 in a column and when you only
`have one 1 in a column. And that's how a person of skill would
`understand a low-density generator matrix to work. They're
`trying to -- the purpose of these low-density generator
`matrices are, again, to increase efficiency and -- because
`they're sparse and they're low-density.
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 15
`
`

`

`16
`
`THE COURT: Yes.
`MR. JOHNSON: Mr. Tung talked about the fact that
`you have -- want, you know, things to move quickly, low
`overhead. And that's how these things are used.
`So -- so let's move forward. Slide 13 we were on. And I
`want to show you also Slide 14 just to make a point that Hughes
`addressed yesterday. If you move the vector to the left-hand
`side --
`
`THE COURT: Yes.
`MR. JOHNSON: -- and you transpose them -- the
`matrix -- so this matrix in the middle here is the same as on
`the other page, it's 1, 1, 1, 0, 0, 0 in the top row, and now
`that's been put in the column. So it's been transposed, 1, 1,
`1, 0, 0, 0. You get the same parity bits, 1, 0, 1, 1, 1 on the
`right side, just as you did in the previous example.
`So moving the vectors and transposing the matrices gives
`you the same result. And my point simply here is that the
`patent has to cover both of these situations, whether the
`vector's on the left or on the right and whether the -- whether
`the -- the low-density generator matrix has 1's adjacent to it
`or otherwise.
`And so what we're saying is "repeat" covers the situation
`where -- where there may be sequential duplication, but it also
`has to cover the situation where there's not. And that's how a
`person of ordinary skill, reading the patent and the intrinsic
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 16
`
`

`

`17
`
`evidence, would understand the term --
`THE COURT: So now I'm back to the question again.
`Why is "repeat" -- "repeat" not clear?
`MR. JOHNSON: We think it is clear.
`THE COURT: Well, we can leave it at that.
`MR. JOHNSON: So just -- and I'll try to go through
`this relatively -- more quickly. But again, the distinction
`between the constructions -- CalTech's construction is re-used
`in forming a code. Defendant's construction is sequential
`duplication.
`The real -- the key issue here is does "repeat" require
`that the bits be sequentially duplicated? The answer is no.
`And it's instructive, as the case law says, to start with the
`relevant claim language.
`THE COURT: Let's -- let's try, without
`sequential, the word "duplication." Is that clear?
`MR. JOHNSON: I think, as I said at the beginning,
`"duplication" has some inherent ambiguity based upon how I
`think Hughes is reading it. So if "duplication" means you're
`going to re-use it, that we would be fine with.
`But if -- if, as I think Hughes is going to get up and say
`in a few minutes, that "duplication" requires some copying,
`which is in their expert declaration, they want to have the
`same notion that there has to be some copying of the bits. And
`that's not what the LDGM does, as I walked through.
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 17
`
`

`

`18
`
`The LDGM uses linear algebra to re-use the bits and
`multiply them one at a time to generate parity bits. And in
`some instances, they may be -- you may have bits that -- that
`are duplicated, but in others you won't.
`And the patent -- that's why the term "repeat" is, I
`think, clear because it tells a person of skill that it has to
`cover both of those --
`THE COURT: That is what I was asking you before.
`MR. JOHNSON: Understood.
`THE COURT: All right.
`MR. JOHNSON: So if I could, I'll just go through
`the -- the claim language, again, at 19. I'm on Slide 19. I
`think it's instructive to look at why the claim language
`supports CalTech's construction. If we look at Claim 13 of the
`'032 patents -- and these patents all obviously share the same
`disclosure -- but it shows that the LDGM is used to output a
`random sequence of repeats of the message bits.
`THE COURT: I read that to you yesterday, I think.
`MR. JOHNSON: And if this -- this part -- because
`the sequence is random here, comes right out of the claim, and
`there is no sequential duplication -- all right. If it were
`sequentially duplicated, there wouldn't be randomness. There's
`no interleaver in this particular claim. The randomness comes
`from the LDGM.
`THE COURT: Yes, it does.
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 18
`
`

`

`19
`
`MR. JOHNSON: And so Claim 13 uses the word "repeat"
`to mean "re-use." It confirms our construction.
`And when you look at Dependent Claim 7 of the '710 patent,
`it also confirms that Claim 1, which uses the term "first
`encoding including repeating the data elements," when we look
`at the dependent claims from that, we see that the first coder
`comprises a low-density generator matrix coder.
`So Claim 1, which has "repeating," and talks about the
`first coder has to cover the LDGM. So LDGM -- "repeat" has to
`be broader than "sequential duplication."
`And finally, with respect to Claim 15 of the '710 patent,
`again, it recites a coder that repeats said stream of bits
`irregularly and scrambles the repeated bits. And again,
`there's no interleaver here. And Dependent Claim 20 talks
`about -- includes the limitation of low-density generator
`matrix.
`So because the LDGM must be capable of scrambling the
`repeated bits, "repeat" has to be broader. It can't be
`"sequentially duplicated." And that's just from looking at the
`claims.
`If we -- if we look at the specification, it describes a
`specific -- I'm on Slide 22. It describes specific embodiment
`where repeating can include duplication. That's where part of
`the spec talks about n equals k -- sorry, n equals q times k.
`And we agree, that part talks about duplication.
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 19
`
`

`

`20
`
`THE COURT: Go back over that just one more time.
`MR. JOHNSON: Sure.
`So I'm going to talk about two parts of the specification.
`The first part is the part that Hughes relies on. And it
`talks -- it says, "In an embodiment" -- right? -- which is
`important because they're just describing one embodiment
`here -- it says, "the outer coder is a repeater that
`repeats" --
`
`THE COURT: Correct.
`MR. JOHNSON: -- "the k bits in a block a number of
`times q to produce a block with n bits, where n equals q times
`k."
`
`So that's a specific embodiment where repeating can
`include duplication. But -- and this is the important part --
`the specification also describes, as Your Honor pointed out, an
`alternate embodiment further down in Claim 3 that describes the
`LDGM that performs an irregular -- irregular repeat of the k
`bits in the block. And that means n does not equal q times k.
`That's how a person of ordinary skill in the art would
`understand it. I don't think there's any dispute with Hughes
`on that point.
`But this LDGM embodiment is an example of repeating that's
`not sequential duplication. And a person of skill reading
`these two passages together would understand that repeating has
`to include both sides. It includes the -- the sequential
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 20
`
`

`

`21
`
`duplication potentially, but it also has to include the LDGM
`which doesn't do sequential duplication. And so the term is
`properly construed as "re-use."
`Now, another part of this spec is also instructive. And
`I'm at Slide 23 here, and I'm referring to column 3, lines 56
`to 64. Because, again, the LDGM coder is described here and it
`says that "The interleaver may" --
`THE COURT: Yes.
`MR. JOHNSON: -- "be excluded" --
`THE COURT: Of course.
`MR. JOHNSON: -- "due to randomness." So that makes
`
`sense.
`And when you look at that and connect it to the Tanner
`graph in Figure 3, which is Slide 24, we see that the graph
`represents, as was described yesterday, "Describes
`relationships between the message bits and the parity bits in
`an IRA codeword that's generated according to the claims."
`And here, each line -- some of the examples are
`highlighted in yellow here. Each line that exits the
`information nodes shows a re-use of that message bit. And this
`random permutation box in the middle, which I've underlined
`here in red, represents how those random connections are
`ultimately repeated.
`THE COURT: Yes. I know that. The -- it would
`ordinarily on another situation have many lines in 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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 21
`
`

`

`22
`
`MR. JOHNSON: That's right. That's right,
`
`Your Honor.
`And so, again, Figure 3 is consistent with CalTech's
`construction. Nothing in this figure suggests with respect
`that these nodes are sequentially duplicated.
`And when you look at Hughes's example of the generator
`matrix, again, it's not an LDGM. One of the -- if we look at
`the extrinsic evidence that was cited at Docket No. 75-3 --
`this is Hughes's extrinsic evidence -- even that reference
`defines LDGM to be random and sparse. That's consistent with
`the patent. It's consistent with a person of ordinary skill's
`understanding.
`And that leads us to the next argument that Hughes
`discusses in its brief, which is that CalTech's construction
`somehow renders terms in the patent claims superfluous. So, in
`particular, Hughes asserts that CalTech's construction makes
`the interleaving step superfluous because it leaves for some
`randomness.
`And so I've put up Figure 1 here on Slide 26. And we see
`in the middle here that it refers to "interleaving the repeated
`data elements in the first encoded data block."
`Now, the LDGM has some randomness. We know that. A
`person of skill knows that. But that doesn't render
`interleaving -- the interleaving step as meaningless because
`some randomness can be followed by an interleaving step that
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 22
`
`

`

`23
`
`introduces additional randomness, further scrambles the bits up
`for even better error correction, potentially.
`And so each one -- our construction doesn't eliminate
`or -- or take away any of the meaning from any of these
`limitations. Each one of these limitations in the claims has
`meaning. And as you look forward into the -- as I said, in the
`specification, when Your Honor referenced this yesterday, the
`spec discloses "an interleaver may be excluded when using an
`LDGM."
`
`THE COURT: Yes. That's right.
`MR. JOHNSON: It doesn't --
`THE COURT: I read it to you.
`MR. JOHNSON: Right.
`And it doesn't say that the interleaver "must be
`excluded." It says it "may be excluded."
`So re-use of the bits, as the LDGM coder would do, does
`not eliminate the need for an interleaver. No claim term is
`rendered superfluous. They all have meaning with CalTech's
`construction.
`So what does Hughes do? Hughes sends us to extrinsic
`evidence. And they cite -- and this is Slide 28. They cite
`the eleventh definition from a common usage dictionary, and
`they ignore the earlier dictionary definitions that we believe
`are consistent with CalTech's proposed construction.
`Again, our construction is "re-use." That's consistent
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 23
`
`

`

`24
`
`with the first definition of "to say or do again." It's
`consistent with "to undergo again," which is Definition No. 5.
`And when you look at the article that Hughes relies upon
`from Professor McEliece, Slide 29, they somehow argue that
`Professor McEliece admitted in this article that "repeat" means
`"sequential duplication." This article doesn't say that. This
`article shows an example of repeating, doesn't reflect how a
`person of ordinary skill in the art would understand
`"repeating" as it's used in the claims.
`And, in fact, this is an article, refers several times to
`"for example," "in our example." He's not defining "repeat" in
`the context of the intrinsic evidence used in the four patents
`in this case.
`This patent -- and this -- and this portion of
`Dr. McEliece's book doesn't suggest that "sequential
`duplication" is the proper construction of "repeat."
`So when you compare the constructions, CalTech's
`construction is consistent with all the intrinsic evidence.
`It's the only construction that's consistent with all of the
`dependent claims. It's the only construction that embraces and
`covers all of the described embodiments, including the
`alternate LDGM embodiment. And defendant's narrow construction
`of "sequentially duplicating" imports limitations and isn't
`supported by the intrinsic evidence.
`THE COURT: All right.
`
`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
`
`UNITED STATES DISTRICT COURT
`
`Hughes, Exh. 1026, p. 24
`
`

`

`25
`
`MR. JOHNSON: With that, Your Honor, I'll turn it
`over to Hughes.
`THE COURT: Please.
`MR. LEE: Your Honor, Bill Lee for Hughes. And I
`will address the "repeat" limitation, Tanner graph, and
`"combined." And Mr. Dowd will address some remaining terms, if
`it's all right with Your Honor.
`THE COURT: Fine.
`MR. LEE: Your Honor, I'm goi

This document is available on Docket Alarm but you must sign up to view it.


Or .

Accessing this document will incur an additional charge of $.

After purchase, you can access this document again without charge.

Accept $ Charge
throbber

Still Working On It

This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.

Give it another minute or two to complete, and then try the refresh button.

throbber

A few More Minutes ... Still Working

It can take up to 5 minutes for us to download a document if the court servers are running slowly.

Thank you for your continued patience.

This document could not be displayed.

We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.

Your account does not support viewing this document.

You need a Paid Account to view this document. Click here to change your account type.

Your account does not support viewing this document.

Set your membership status to view this document.

With a Docket Alarm membership, you'll get a whole lot more, including:

  • Up-to-date information for this case.
  • Email alerts whenever there is an update.
  • Full text search for other cases.
  • Get email alerts whenever a new case matches your search.

Become a Member

One Moment Please

The filing “” is large (MB) and is being downloaded.

Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.

Your document is on its way!

If you do not receive the document in five minutes, contact support at support@docketalarm.com.

Sealed Document

We are unable to display this document, it may be under a court ordered seal.

If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.


Access Government Site

We are redirecting you
to a mobile optimized page.





Document Unreadable or Corrupt

Refresh this Document
Go to the Docket

We are unable to display this document.

Refresh this Document
Go to the Docket