`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY 25, 2015
`
`Divsalar'’s repeat irregular, one way would be to
`
`partition the information bits into subblocks and
`
`repeat the bits in each block a different number of
`
`times, right?
`
`A.
`
`One of the many ways of adding
`
`irregularity would be to add irregularity in a way
`
`of changing the number of times things are repeated.
`
`Q.
`
`Okay.
`
`MR. GLASS: And just objection to previous
`
`question as calls for a legal conclusion.
`
`from Frey '99 of partitioning bits into subblocks where I repeat each
`
`Go ahead.
`
`BY MR. DOWD:
`
`Q.
`
`And what's shown in Figure 2 of Frey is
`
`that you've got a group of bits that you have
`
`partitioned into subblocks Fl, F2, F3,
`
`through FD,
`
`and you repeat the bits in each block a different
`
`number of times, correct?
`
`MR. GLASS:
`
`Same objection.
`
`And outside
`
`the scope.
`
`THE WITNESS:
`
`It shows nodes that have
`
`different repetitions -- whatever exactly that means
`
`in this paper ~~ attached to them.
`
`BY MR. DOWD:
`
`QO.
`
`So if I
`
`took the concept
`
`il
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`Apple vs. Caltech
`IPR2017-00701
`Apple 1140
`
`Apple vs. Caltech
`IPR2017-00701
`Apple 1140
`
`
`
`subblock a different number of times and I apply
`
`that to the repeater of Divsalar Figure 3,
`
`the
`
`result would be an irregular repeat, correct?
`
`MR. GLASS: Objection. Vague. Calls for
`
`a legal conclusion.
`
`THE WITNESS:
`
`No, it's false.
`
`MR. GLASS: Calls for a legal conclusion.
`
`Incomplete hypothetical.
`
`Go ahead.
`
`BY MR. DOWD:
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY 25, 2015
`
`if
`
`Q.
`
`A.
`
`Q.
`
`SO --
`
`It's false.
`
`So are you saying that if I
`
`take the input
`
`block to the repeater in Figure 3, divide that into
`
`subblocks and repeat
`
`the bits of each subblock
`
`different numbers of times,
`
`that's not an irregular
`
`repeat;
`
`is that your testimony?
`
`A.
`
`That is not what 18 written --
`
`MR. GLASS:
`
`Same objections.
`
`THE WITNESS:
`
`That is not what is written
`
`in Figure 2. What is written in Figure 2 is that
`
`you take the code word bits, which is something
`
`entirely different.
`
`BY MR, DOWD:
`
`Q.
`
`Well,
`
`try my question.
`
`My question is,
`
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`286 |
`
`I
`
`take the input bits to the repeater in Figure 3,
`
`divide them into subblocks, and repeat each subblock
`
`a different number of times,
`
`then I have an
`
`irregular repetition, correct?
`
`MR. GLASS:
`
`Same objections. Asked and
`
`answered.
`
`THE WITNESS: What you're asking me is if
`
`I
`
`take a repeater accumulated code and make it
`
`irregular,
`
`is it irregular? Yes.
`
`BY MR. DOWD:
`
`Q.
`
`Okay.
`
`And so taking a set of input bits,
`
`dividing that into subblocks and repeating each
`
`subblock a different number of times,
`
`that to you is
`
`the definition of an irregular code?
`
`MR. GLASS: Calls for a legal conclusion.
`
`Outside the scope.
`
`THE WITNESS:
`
`I have no opinion on that.
`
`BY MR. DOWD:
`
`Q.
`
`Now,
`
`is it your position that Dr. Frey and
`
`Dr. Divsalar were in different groups that didn't
`
`talk te each other?
`
`A.
`
`This would be best posed to them.
`
`I have
`
`
`
`
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`So to the best of your knowledge, 25
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`absolutely no idea.
`
`Q.
`
`Okay.
`
`Dr. Frey and Dr. Divsalar may well have talked to
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY 25, 2015
`
`each other?
`
`I could speculate, but
`
`I prefer not to.
`
`222:
`
`2223
`
`Well, we don't have to speculate.
`
`O5:
`
`22:
`
`MR. DOWD: Let's mark as Exhibit 22, a
`
`O5:
`
`22:
`
`copy of a document Caltech24021, it's an e-mail
`
`from
`
`O5:
`
`22:
`
`Dr. Frey to Dr. Divsalar.
`
`05
`
`222:
`
`(Urbanke Exhibit 22 was marked for
`
`222:
`
`identification and attached to the
`
`05
`
`222:
`
`transcript.)
`
`BY MR. DOWD:
`
`Do you have Exhibit 22?
`
`Yes.
`
`This is an e-mail that Dr. Frey sent
`
`to
`
`. Divsalar in December of
`
`'99, right?
`
`The date reads:
`
`12/8/1999.
`
`11
`
`12
`
`13
`
`14
`
`| 08
`
`222:
`
`53
`
`OS:
`
`22:
`
`53
`
`O05:
`
`22
`
`253
`
`222:
`
`54
`
`122:
`
`54
`
`222:
`
`57
`
`Right? That is a sentence that appears; that's
`
`"Dr. Divsalar, have you had a chance
`
`to look through the Allerton paper?"
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`A.
`
`Q.
`
`And he says ~- he references the irregular
`
`turbo codes work that he's been doing, right?
`
`A.
`
`He -- in -- in there it's written
`
`"irregular turbo codes" is two of the words that
`
`appear in the e-mail;
`
`that is correct.
`
`Q.
`
`And he asks:
`
`223:
`
`O5
`
`2:23:
`
`09
`
`O5:
`
`23:
`
`14
`
`05
`
`223:
`
`18
`
`223:
`
`22
`
`723
`
`224
`
`223:
`
`28
`
`223
`
`234
`
`05
`
`223
`
`236
`
`O5
`
`223:
`
`39
`
`05
`
`223:
`
`41
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`correct.
`
`Q.
`
`Exhibit 21 is the irregular turbo codes
`
`paper from the 1999 Allerton conference, correct?
`
`A.
`
`Whether he refers to the turbo code paper,
`
`I have no idea.
`
`It doesn't say specifically.
`
`It says ~~ he mentions: Have you looked
`
`through the Allerton paper? Which paper he refers
`
`to,
`
`I have no idea.
`
`QO.
`
`My question was, Exhibit 21 is the
`
`irregular turbo codes paper Dr. Frey presented at
`
`Allerton in 1999, correct?
`
`A.
`
`Reference 21 is a code -- a paper
`
`entitled:
`
`"Trregular Turbo Codes," which was
`
`published or which was presented presumably at the
`
`Allerton conference in 1999.
`
`Q.
`
`Okay.
`
`And if we return to Exhibit 22, he
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`paper we've been talking about so far today, right?
`
`"Regardless,
`
`it would be interesting
`
`to extend the work that you and Bob have
`
`done to the case of irregular turbo
`
`codes."
`
`Have I read that correctly?
`
`Yes.
`
`If we go to Exhibit 6, that's the Divsalar
`
`A.
`
`QO.
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`Yes.
`
`And two of the authors on that paper are
`
`Divsalar and Bob McEliece, right?
`
`That is correct.
`
`And so at this time Divsalar and McEliece
`
`A.
`
`Q.
`
`were working together on RA codes, correct?
`
`A.
`
`Their main motivation was to extend what
`
`They were working on a problem that resulted in the RA codes that we have in Exhibit 6,
`
`THE REPORTER:
`
`T’m sorry. Start over.
`
`THE WITNESS: Their main motivation was to
`
`extend what was called the interleaver gain
`
`conjecture,
`
`I believe,
`
`to turbo codes.
`
`And they
`
`succeeded in the RA paper to do that for the very
`
`specific case of RA codes.
`
`BY MR. DOWD:
`
`Okay.
`
`So --
`
`That is what their main work was at that
`
`So my question was, at that time
`
`. Divsalar and Dr. McEliece, what
`
`they were
`
`working on together was RA codes, right?
`
`A.
`
`They were working on the weight
`
`distribution problem.
`
`Q.
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`224;
`
`57
`
`224;
`
`58
`
`2:25:
`
`O01
`
`225:
`
`05
`
`225:
`
`09
`
`225:
`
`13
`
`225:
`
`16
`
`2:25:
`
`18
`
`125:
`
`18
`
`2:25:
`
`20
`
`225:
`
`22
`
`225:
`
`27
`
`225:
`
`30
`
`225:
`
`36
`
`725;
`
`39
`
`225:
`
`40
`
`225:
`
`40
`
`225:
`
`44
`
`2:25:
`
`44
`
`225:
`
`47
`
`2:25:
`
`52
`
`225:
`
`53
`
`225:
`
`56
`
`225:
`
`56
`
`2:25:
`
`58
`
`
`
`MR. GLASS: Objection. Asked and
`
`answered.
`
`THE WITNESS:
`
`As it's written even in the
`
`paper,
`
`I believe,
`
`their motivation for looking at it
`
`and their main work for doing it was to solve the
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`definition, right?
`
`interleaver gain component conjecture.
`
`BY MR. DOWD:
`
`Q.
`
`A.
`
`Okay.
`
`By doing this they looked at a specific
`
`case or in order to accomplish it of RA -~- of RA
`
`codes. And, hence,
`
`I would consider that what
`
`the
`
`main concern at that point was -- was to prove or to
`
`establish the validity of this interleaver gain
`
`conjecture to various forms of turbo codes.
`
`QO.
`
`Now, we said earlier that RA codes are a
`
`form of turbo; do you recall that testimony?
`
`A.
`
`Q.
`
`That is correct.
`
`The Frey irregular turbo codes paper from
`
`1999, Exhibit 21,
`
`is about making turbo codes
`
`irregular, right?
`
`A.
`
`Q.
`
`That is correct.
`
`And if I make the repetition of the RA
`
`code in Figure 3 irregular,
`
`I have an IRA code by
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`particular situations as there were Luby '$7 and
`
`MR. GLASS: Objection. Vague.
`
`THE WITNESS:
`
`So, you know, without giving
`
`a legal opinion here, one way of making --
`
`accomplishing I -- IRA codes is of -- by definition
`
`introducing irregular repeats.
`
`BY MR. DOWD:
`
`Q.
`
`Okay.
`
`Now,
`
`I
`
`take it that if we return to
`
`Exhibit
`
`2 you'd never seen -~- oh, withdrawn, sorry.
`
`I'm sticking with the e-mail exhibit 22.
`
`A.
`
`.
`
`Right.
`
`So I
`
`take it that before you were retained
`
`for this case, you had never seen the e-mail that
`
`we've marked as Exhibit 22?
`
`A.
`
`Q.
`
`I'm pretty sure no.
`
`Okay.
`
`So let's see if we can agree on
`
`some kind of basic points.
`
`We can agree that making an LDPC code
`
`irregular improved performance over a regular LDPC
`
`code, right?
`
`A.
`
`We can agree that --
`
`MR. GLASS: Objection.
`
`Incomplete
`
`hypothetical.
`
`Go ahead.
`
`THE WITNESS: We can agree that in
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`292 |
`
`Luby '98 and Richardson '99, each of them with very
`
`specific restrictions,
`
`some form of irregularity
`
`improved the performance.
`
`The restrictions in the
`
`Luby '37 paper were for the binary racial channel
`
`for particularly cascaded codes.
`
`The restrictions
`
`in this Luby '98 paper were that there were a
`
`particular way of doing the decoding, which is
`
`neither equal to message passing nor equal
`
`to the
`
`flipping algorithm.
`
`And in the case of LDPC codes in the ~-
`
`in
`
`the Richardson '99 paper this was strictly concerned
`
`with Gallager type but
`
`irregular codes.
`
`BY MR. DOWD:
`
`on
`
`So if we go back to Exhibit 17, Luby '98,
`
`at page 925,
`
`the statement there is:
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`performance than regular graphs only as
`
`"Tt is shown that using irregular
`
`graphs yields codes with much better
`
`performance than reqular graphs."
`
`Have I read that quote correctly?
`
`A.
`
`In the context of the quotes,
`
`they're
`
`consistent.
`
`Oo.
`
`And in Luby '98,
`
`they don't say:
`
`"Tt's shown that using irregular
`
`graphs yields codes with much better
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`compared to a flipping algorithm when
`
`you're decoding," do they?
`
`MR. GLASS: Objection. Vague. Asked and
`
`answered.
`
`THE WITNESS: That's the only possible
`
`conclusion you can draw because that's the only
`
`thing they show in the paper.
`
`BY MR. DOWD:
`
`Q.
`
`Well, and maybe that's the only conclusion
`
`you can draw, but --
`
`A.
`
`It's the only conclusion someone, you
`
`know, versed in the art could draw.
`
`Q.
`
`Okay. Sir,
`
`in the statement where they
`
`explain why it's better performance,
`
`there's no
`
`mention here on this page of flipping,
`
`is there?
`
`MR. GLASS: Objection. Asked and answered
`
`
`
`
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`You
`
`several times now.
`
`Go ahead.
`
`THE WITNESS:
`
`As
`
`I mentioned, it's not
`
`customary that every single time when you talk about
`
`something that you would repeat all possible
`
`restrictions that you're considering.
`
`Typically in the abstract you would say,
`
`we're considering, you know,
`
`a certain type of code.
`
`We're considering certain type of decoder.
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`would not repeat that at every single sentence.
`
`05
`
`2:31:
`
`56
`
`BY MR. DOWD:
`
`05
`
`232:
`
`00
`
`Q.
`
`Right.
`
`In the abstract, can you point out
`
`232:
`
`00
`
`where it says that this is about a flipping decoder
`
`05
`
`232:
`
`03
`
`in the abstract?
`
`232:
`
`06
`
`A.
`
`Here it's in the abstract, but it's very
`
`05
`
`232:
`
`06
`
`clearly explained in the paper that that decoding
`
`05
`
`232:
`
`06
`
`does not send a message passing decoder, but --
`
`O05:
`
`32:
`
`06
`
`THE REPORTER: Wait. You're going to have
`
`232:
`
`06
`
`to start that answer over.
`
`05:
`
`32:
`
`13
`
`11
`
`12
`
`13
`
`THE WITNESS:
`
`In the paper it is clearly
`
`232:
`
`13
`
`explained that that is not a standard message
`
`05
`
`232:
`
`15
`
`pattern. We couldn't -- but
`
`the Gallager algorithm,
`
`05
`
`232:
`
`20
`
`2:32:
`
`23
`
`so why don't you guys try again.
`
`as we had discussed beforehand, and it is followed
`
`05
`
`by a flipping algorithm.
`
`BY MR. DOWD:
`
`Qo.
`
`Oh,
`
`so now it's not
`
`important enough to
`
`make it into the abstract, but it's still --
`
`MR. GLASS: Objection. Argumentative and
`
`| 05
`
`232;
`
`25
`
`05:
`
`32
`
`225
`
`05:
`
`32:
`
`27
`
`232:
`
`29
`
`232:
`
`32
`
`2:32:
`
`34
`
`232
`
`134
`
`232
`
`234
`
`| 05
`
`0s
`
`05
`
`232:
`
`34
`
`O5
`
`2:32:
`
`34
`
`O05:
`
`32:
`
`39
`
`(Overlapping speakers.)
`
`Wait. Wait. There was not
`
`even a complete question there, and I don't know
`
`what your response was,
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`BY MR. DOWD:
`
`Q.
`
`Sir,
`
`I am correct that the abstract of
`
`Luby nineteen ninety -- 1998 does not say anything
`
`about flipping decoders, right?
`
`MR, GLASS: Objection. Asked and
`
`answered.
`
`THE WITNESS:
`
`It talks about the Gallager
`
`algorithm.
`
`BY MR. DOWD:
`
`Q.
`
`A.
`
`Okay.
`
`Sorry, sorry,
`
`let me read this again.
`
`It talks about for which our decoding
`
`algorithm.
`
`It doesn't specify exactly what that is
`
`in the abstract because the abstract doesn't specify
`
`let's turn to
`
`every single technical detail but it doesn't talk
`about a message passing algorithm, but it talks
`
`about our --
`
`THE REPORTER: Wait. Wait.
`
`Slow down.
`
`"l..but it doesn't talk" ~-
`
`Start there.
`
`THE WITNESS:
`
`It doesn't talk about
`
`the
`
`message passing algorithm, but it talks about our
`
`decoding algorithm.
`
`BY MR. DOWD:
`
`Q.
`
`And if we go to Luby '97,
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`mWNH
`
`11
`
`12
`
`13
`
`14
`
`That would be exhibit?
`
`MR. GLASS:
`
`9.
`
`MR. DOWD:
`
`93.
`
`THE WITNESS:
`
`9.
`
`Page, sorry?
`
`. DOWD:
`
`05
`
`2:33:
`
`47]
`
`233:
`
`49
`
`233:
`
`55
`
`OS:
`
`33
`
`258
`
`05
`
`2:34:
`
`Ol
`
`| 08
`
`234:
`
`03
`
`Last page, 944. Actually, second to the
`
`234:
`
`03
`
`Yes.
`
`O05:
`
`34:
`
`09
`
`05
`
`2:34:
`
`10
`
`The acknowledgement section they describe,
`
`05
`
`734:
`
`10
`
`05
`
`234:
`
`14
`
`"Strong evidence that irregular degree
`
`234;
`
`14
`
`sequences are better than regular degree
`
`05
`
`234:
`
`17
`
`sequences.”
`
`| 05
`
`234:
`
`19
`
`24
`
`there's not technical description what "strong" means.
`
`regular degree sequences", closed quote.
`
`Correct?
`
`A.
`
`That's some acknowledgement.
`
`I don't --
`
`It doesn't mean what the decoder is.
`
`It
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`Do you see that?
`
`A.
`
`You're talking about
`
`the last paragraph
`
`before the acknowledgement?
`
`Q.
`
`I'm talking about
`
`in the acknowledgement
`
`they say: There has been, quote, "Strong evidence
`
`that irregular degree sequences are better than
`
`fos
`
`O5:
`
`34:
`
`05
`
`234:
`
`24
`
`O05:
`
`34:
`
`33
`
`234:
`
`34
`
`:34:
`
`37
`
`734:
`
`42
`
`05
`
`734:
`
`45
`
`05
`
`234;
`
`4’]
`
`O05:
`
`34:
`
`51
`
`05
`
`234:
`
`53
`
`05
`
`234:
`
`56
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY 25, 2015
`
`doesn't mean what
`
`type of irregularity.
`
`It doesn't
`
`234:
`
`59
`
`say about, you know, what exactly the channel
`
`they
`
`235:
`
`02
`
`are talking about,
`
`so it would be impossible to make
`
`235:
`
`04
`
`297 |
`
`a qualified statement about that.
`
`Q.
`
`¥3,
`
`it doesn't mention any of those
`
`things, right?
`
`A.
`
`Tt doesn't mention what
`
`they are talking
`
`about.
`
`Q.
`
`All it says is: There's strong evidence
`
`that, quote, "irregular degree sequences are better
`
`than regular degree sequences.”
`
`A.
`
`If they have found strong evidence that
`
`in general,
`
`they would have
`
`ii
`
`12
`
`13
`
`235:
`
`07
`
`235:
`
`09
`
`235:
`
`12
`
`235:
`
`17
`
`2:35;
`
`19
`
`235:
`
`19
`
`235:
`
`21
`
`235:
`
`24
`
`2:35:
`
`26
`
`:35:
`
`29
`
`
`
`to Page 1826, do you see there are results on that page?
`
`proves such a thing,
`
`published it at some point.
`
`Q.
`
`Now, we can agree that making turbo codes
`
`irregular improve their performance over regular
`
`turbo codes, right?
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`235:
`
`32
`
`235:
`
`37
`
`O05:
`
`741
`
`O05:
`
`:44
`
`> 46
`
`247
`
`OS:
`
`247
`
`O5:
`
`249
`
`O05:
`
`253
`
`05:
`
`254
`
`O05:
`
`205
`
`05:
`
`716
`
`MR. GLASS: Objection. Vague.
`
`Incomplete
`
`| OS:
`
`hypothetical.
`
`THE WITNESS:
`
`Some particular way of
`
`improving might help;
`
`some other ways might hurt.
`
`BY MR. DOWD:
`
`Q.
`
`Well,
`
`if you'd turn in Exhibit 21,
`
`the
`
`Frey '99 paper,
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`18 -- sorry,
`
`18 --
`
`26.
`
`Thank you.
`
`Do you see there are results shown on that
`
`page, under the heading 5?
`
`A.
`
`Q.
`
`Figure 4 we're talking about?
`
`I'm saying -- Y3, under the heading 5
`
`is ~- that's the heading that says:
`
`"Results."
`
`A.
`
`Q.
`
`Oh,
`
`I see in that section.
`
`And the second paragraph says:
`
`"The
`
`irregular turbo code clearly performs better than
`
`the regular turbo code for bit error rates,” and
`
`then it lists them?
`
`A.
`
`Yes.
`
`So some specific irregular turbo
`
`code performs in their experiment better.
`
`Q.
`
`Okay.
`
`So we can agree that, at least in
`
`Luby '99,
`
`the regular turbo code outperformed the
`
`regular turbo code, right?
`
`Luby doesn't talk about
`
`turbo codes,
`
`I
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`--
`
`I'm sorry.
`
`Let me restate that.
`
`We can agree that in Frey '99,
`
`the
`
`irregular turbo code performed 0.15-dB better than
`
`the regular turbo code, right?
`
`The particular
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`MR. GLASS: Objection. Vague.
`
`THE WITNESS:
`
`The particular irregular
`
`turbo code that they're -- that they're referring to
`
`has a better performance, yes.
`
`BY MR. DOWD:
`
`Q.
`
`Now,
`
`if a person of ordinary skill had
`
`read Divsalar and Luby or Divsalar and Richardson or
`
`Divsalar and Frey and wanted to make Divsalar's
`
`repeat irregular,
`
`they could have done so, right?
`
`MR. GLASS: Objection.
`
`Incomplete
`
`hypothetical.
`
`Compound.
`
`THE WITNESS:
`
`In my analysis or my
`
`expertise,
`
`I was asked in particular to refer to
`
`Luby '97, Luby '98 and Richardson '99, and that's
`
`what
`
`I have done.
`
`BY MR. DOWD:
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY 25, 2015
`
`hypothetical. THE WITNESS:
`
`Q.
`
`Okay.
`
`So if a person of skill read
`
`Divsalar and Luby '97 and wanted to make Divsalar's
`
`repeater irregular they would have known how to do
`
`so, right?
`
`No.
`
`Okay. What about Richardson and Divsalar?
`
`MR. GLASS: Objection. Vague.
`
`Incomplete
`
`As
`
`I have opinioned in my
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`expert report,
`
`there were many reasons why the
`
`| 05
`
`2:38:
`
`26
`
`invention of turbo codes was something novel and
`
`O05
`
`738
`
`:31
`
`something surprising.
`
`Number one, RA codes ~~ sorry,
`
`irregular
`
`RA codes were semi novel and surprising.
`
`Number one, at that point
`
`in time, RA --
`
`RA codes were routinely represented in an entirely
`
`238:
`
`36
`
`238:
`
`38
`
`238:
`
`41
`
`738:
`
`45
`
`238:
`
`49
`
`different way.
`
`So there were no notions of variable
`
`238:
`
`55
`
`nodes or check nodes or anything like that.
`
`BY MR. DOWD:
`
`239;
`
`00
`
`05
`
`239:
`
`03
`
`Q.
`
`I understand.
`
`I -- I'm really not asking
`
`239:
`
`04
`
`that question.
`
`My question was a very specific
`
`239:
`
`05
`
`05
`
`239:
`
`09
`
`11
`
`12
`
`13
`
`these things mean. What does it mean to say to make
`
`THE WITNESS:
`
`I don't know what any of
`
`a repeater irregular What notions would that have
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`question.
`
`If a person of skill in '99 read Divsalar,
`
`read the Richardson '99 paper and decided that
`
`239:
`
`09
`
`O05:
`
`39:
`
`15
`
`they'd like to make Divsalar's repeater irregular,
`
`| 05:
`
`220
`
`they would have the technical ability to do that,
`
`right?
`
`A.
`
`I don't know what --
`
`MR. GLASS: Objection. Objection. Vague.
`
`Incomplete hypothetical.
`
`Go ahead.
`
`O05:
`
`225
`
`OS:
`
`225
`
`05:
`
`228
`
`O05:
`
`228
`
`05:
`
`230
`
`O05:
`
`231
`
`OS:
`
`232
`
`O5:
`
`233
`
`O05:
`
`236
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`carried over the Luby paper which talks about
`
`specific class of codes which are LDPC codes?
`
`So unless you tell me exactly what
`
`representation you have in mind and what notions
`
`would have carried over,
`
`I don't see how to combine
`
`those.
`
`BY MR. DOWD:
`
`Q.
`
`Okay.
`
`Now,
`
`in your report you talk about
`
`the prosecution that led to the patents; do you
`
`recall that?
`
`A.
`
`I did a very cursory summary to the best
`
`of my knowledge.
`
`I'm not an expert.
`
`I'm not sure
`
`if I got this all right, but
`
`I tried to, a little
`
`bit,
`
`for my benefit,
`
`to summarize.
`
`Q.
`
`Okay.
`
`And if we turn,
`
`for example,
`
`to
`
`Paragraph 87,
`
`there's a discussion about how the
`
`patents claim priority to a provisional application
`
`filed on May 18, 2000.
`
`Do you see that?
`
`Right.
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`transcript.)
`
`MR. DOWD: Let's mark as Exhibit 23 a copy
`
`of the provisional application.
`
`(Urbanke Exhibit 23 was marked for
`
`identification and attached to the
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`il
`
`12
`
`13
`
`BY MR. DOWD:
`
`:
`
`A.
`
`.
`
`.
`
`Do you have Exhibit 23?
`
`Yes.
`
`Do you recognize it?
`
`You know,
`
`some of the figures look
`
`familiar, but
`
`I must say I have browsed through this
`
`very, very quickly, and so, you know, with so many
`
`41:
`
`17
`
`41:
`
`17
`
`41:
`
`18
`
`241:
`
`19
`
`4l:
`
`29
`
`4l:
`
`32
`
`4l:
`
`35
`
`documents to review,
`
`I cannot say with certainty
`
`241:
`
`40
`
`what all these documents are.
`
`Q.
`
`So you can't tell me whether you've read
`
`Exhibit 23 before?
`
`05:
`
`4l:
`
`43
`
`4l:
`
`45
`
`4l:
`
`49
`
`A.
`
`Oh, I've flipped through the history,
`
`OS:
`
`4i:
`
`49
`
`that's for sure, but, you know, how exactly these
`
`4i:
`
`52
`
`55
`
` 05:
`
`pages looked like and what exactly it contained,
`
`don't recall.
`
`Q.
`
`All right.
`
`My question is a simpler one.
`
`Is Exhibit 23 the May 18, 2000 provisional
`
`application that you are referencing in Paragraph 87
`
`41:
`
`42;
`
`00
`
`42:
`
`00
`
`42
`
`203
`
`42:
`
`07
`
`42:
`
`10
`
`42:
`
`16
`
`42:
`
`19
`
`42:
`
`24
`
`42:
`
`27
`
`42:
`
`31
`
`42:
`
`33
`
`: O5;
`
`O5:
`
`O5:
`
`O5:
`
`OS:
`
`O5:
`
`O05:
`
`05:
`
`O05:
`
`OS:
`
`of your report?
`
`A.
`
`T would not know for sure. Clearly,
`
`writing this particular part,
`
`I had legal counsel,
`
`and to the best of my knowledge,
`
`I tried to
`
`summarize. But it's possible that I might have
`
`gotten some of the facts not exactly correct.
`
`not a lawyer.
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`303 |
`
`Q.
`
`Okay.
`
`So this part
`
`6 of your report,
`
`the
`
`| 05:
`
`42
`
`235
`
`summary of the patents-in-suits, who wrote that, you
`
`O05:
`
`42:
`
`37
`
`or the Caltech's lawyers?
`
`242:
`
`42
`
`A.
`
`I went over the file history very quickly
`
`42:
`
`45
`
`and had legal assistance trying to sort out and, you
`
`O05:
`
`42;
`
`48
`
`know, explain to me what, you know,
`
`some very
`
`OS:
`
`42:
`
`53
`
`standard notions were. And so with help of legal
`
`O5:
`
`42:
`
`56
`
`counsel,
`
`this was written.
`
`Q.
`
`So let me see if I understand.
`
`You -- you had explained to you what
`
`happened in the file histories of the four
`
`patents-in-suits;
`
`is that correct?
`
`OS:
`
`43:
`
`01
`
`O05:
`
`43
`
`:03
`
`O5:
`
`43;
`
`07
`
`05:
`
`43:
`
`10
`
`O05:
`
`43:
`
`14
`
`MR. GLASS: Objection. Mischaracterizes
`
`05:
`
`43:
`
`16
`
`43;
`
`16
`
`ll
`
`12
`
`13
`
`of this, where this section appears, who actually typed this up; did you type this up?
`
`questions.
`
`I ask many of the questions.
`
`I got some
`
`/ 05:
`
`43;
`
`19
`
`explanations. Whether or not
`
`indeed I fully
`
`understood these,
`
`is not clear to me.
`
`I'm not a
`
`lawyer.
`
`BY MR. DOWD:
`
`Okay.
`
`That's not my main expertise.
`
`And in terms of paragraphs 85 through 119
`
`05:
`
`43:
`
`22
`
`O5:
`
`43:
`
`25
`
`O05:
`
`43:
`
`28
`
`O05:
`
`43:
`
`28
`
`05:
`
`43;
`
`28
`
`OS:
`
`43
`
`229
`
`05:
`
`43:
`
`31
`
`O5:
`
`43:
`
`39
`
`05:
`
`43:
`
`42
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`the testimony.
`
`THE WITNESS: No.
`
`I simply had a lot of
`
`05:
`
`O5:
`
`43;
`
`18
`
`
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY 25, 2015
`
`This is about 60 pages. MR. GLASS: Objection. Objection.
`
`locations?
`
`A.
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`A.
`
`Many of these things might not have been
`
`that I
`
`typed up, but were in discussions.
`
`Q.
`
`A.
`
`Okay.
`
`I don't recall exactly which parts exactly
`
`I
`
`typed up, but
`
`I clearly had legal counsel.
`
`Q.
`
`Okay.
`
`So let's return to Exhibit 23,
`
`the
`
`provisional application. And I'd like you to take a
`
`moment and just look through the -- the slide deck
`
`that is attached, which begins on Caltech Page 6584
`
`and runs through 6616, and just let me know when
`
`you're done.
`
`A.
`
`What was the last page, 661 -- I don't
`
`remember.
`
`on
`
`It's the last page of the document,
`
`so
`
`just to the last slide there on Page 6616.
`
`A.
`
`So there's several slide decks;
`
`is that
`
`correct?
`
`Q.
`
`There should be two.
`
`Just let me know
`
`when you've gotten to the end,
`
`A.
`
`Qo.
`
`I've reached the end.
`
`Okay.
`
`Can you tell me whether the
`
`provisional application discusses indexing memory
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`Outside the scope.
`
`THE WITNESS:
`
`I don't know exactly even
`
`what
`
`the definition of -- of the term that they have
`
`used would be.
`
`It would be impossible for me now to
`
`tell.
`
`BY MR. DOWD:
`
`Q.
`
`Okay.
`
`So in your review sitting here now,
`
`did you see in anything that discussed memory
`
`locations?
`
`MR. GLASS:
`
`Same objection. Outside the
`
`Same objection. Outside the
`
`252
`
`253
`
`255
`
`758
`
`202
`
`202
`
`202
`
`705
`
`:07
`
`708
`
`209
`
`209
`
`:11
`
`714
`
`216
`
`720
`
`223
`
`728
`
`231
`
`2:32
`
`232
`
`THE WITNESS:
`
`These are slides. Many of
`
`the things, you know, might not be in there, might
`
`be in there.
`
`I don't know exactly what
`
`the
`
`particular term means in the realm that you are
`
`talking about.
`
`BY MR. DOWD:
`
`Q.
`
`A.
`
`Q.
`
`Okay.
`
`I have no idea.
`
`I'm just asking, you just read through the
`
`slides; we took several minutes to do it.
`
`Can you
`
`point me to any place that discusses an indexed
`
`memory location?
`
`MR. GLASS:
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`BY MR. DOWD:
`
`Q.
`
`"Yes" or "no"?
`
`MR. GLASS:
`
`-- of the expert report.
`
`THE WITNESS:
`
`I don't even know what the
`
`term “index memory location" means for you.
`
`BY MR. DOWD:
`
`:52:
`
`32
`
`05
`
`252:
`
`34
`
`52
`
`234
`
`O05:
`
`52:
`
`35
`
`: 05
`
`252:
`
`37
`
`05
`
`252:
`
`37
`
`QO.
`
`Have you heard the term "index memory
`
`05
`
`252:
`
`40
`
`location"?
`
`That means many things to different
`
`Does it have meaning to you?
`
`| 05
`
`252:
`
`44
`
`05
`
`252:
`
`44
`
`05
`
`252:
`
`44
`
`O05:
`
`52:
`
`45
`
`Potentially, but it's not something that I
`
`O05:
`
`52:
`
`45
`
`have thought about.
`
`05
`
`152:
`
`48
`
`ll
`
`12
`
`13
`
`(Overlapping speakers.) THE REPORTER: Wait.
`
`What does "index memory location" mean to
`
`252:
`
`48
`
`I don't know in this particular thing.
`
`I
`
`O35:
`
`523
`
`51
`
`— 05
`
`252:
`
`51
`
`have not been asked to opinion on that.
`
`It's not my
`
`05:
`
`52:
`
`53
`
`reaim of expertise.
`
`Q.
`
`And am I correct that in reviewing the
`
`Slides just now you didn't see anything about
`
`index
`
`memory locations?
`
`MR. GLASS: Objection. Outside the scope
`
`05
`
`05
`
`05
`
`252:
`
`56
`
`252:
`
`56
`
`752:
`
`58
`
`253:
`
`03
`
`753:
`
`OS
`
`253:
`
`06
`
`2:53:
`
`06
`
`253:
`
`06
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`MR. GLASS:
`
`-—- the expert report.
`
`Go
`
`THE WITNESS:
`
`IT didn't look for
`
`253:
`
`07
`
`253:
`
`O07
`
`253:
`
`07
`
`any particular instance of that.
`
`I wasn't even
`
`05:
`
`53:
`
`12
`
`aware of that term --
`
`253:
`
`12
`
`THE REPORTER: Repeat your answer and slow
`
`53:
`
`12
`
`753:
`
`13
`
`THE WITNESS:
`
`I was not
`
`in particular
`
`O05:
`
`53:
`
`13
`
`aware of the term until you told me.
`
`I think that
`
`O05:
`
`53:
`
`15
`
`would be a particular term to be looking out for.
`
`05
`
`253:
`
`19
`
`BY MR. DOWD:
`
`| 05
`
`253:
`
`21
`
`Q.
`
`Okay.
`
`Take a minute and just tell me,
`
`is
`
`05
`
`253:
`
`21
`
`there any discussion in here of an index memory
`
`05
`
`253:
`
`25
`
`11
`
`12
`
`13
`
`MR. GLASS: And objection. Calls -- THE WITNESS:
`
`location?
`
`| 05
`
`253:
`
`28
`
`MR. GLASS:
`
`Same objection. Outside the
`
`Q5
`
`253;
`
`28
`
`THE WITNESS:
`
`Can you give me a definition
`
`of what that means.
`
`BY MR. DOWD:
`
`| 05;
`
`229
`
`05%!
`
`229
`
`05:
`
`230
`
`OS:
`
`732
`
`Q.
`
`You -- your understanding, as the person
`
`05:
`
`232
`
`of skill in this art,
`
`is better than mine;
`
`so using
`
`whatever understanding of that term you have.
`
`A.
`
`I don't have any particular --~
`
`
`
`understanding.
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`O5:
`
`235
`
`Q5:
`
`238
`
`05;
`
`240
`
`O05:
`
`240
`
`05:
`
`241
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`MR. GLASS: And objection. Calls for a
`
`308 |
`
`legal conclusion.
`
`BY MR. DOWD:
`
`of the expert report.
`
`Q.
`
`A.
`
`You don't know what a memory location is?
`
`I have some understanding what a memory
`
`location might be. Whether or not that has any
`
`legal bearing or any opinion on what a memory
`
`location means in that case,
`
`I have no idea.
`
`Qo.
`
`What is your understanding of what a
`
`memory location is?
`
`A.
`
`It might simply be a place in memory,
`
`perhaps physical, perhaps abstract.
`
`Do you know what an index is?
`
`I know what an index could be.
`
`Do you know -- what is your understanding
`
`of an index with respect to memory?
`
`A.
`
`I don't know.
`
`An index could be
`
`important,
`
`so perhaps we're talking about
`
`important.
`
`I have no idea.
`
`Q.
`
`Okay.
`
`So with that understanding of what
`
`a memory is and what an index is, do you see any
`
`discussion of a memory location or an index in the
`
`provisional application?
`
`MR. GLASS: Objection. Outside the scope
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`li
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`309 |
`
`THE WITNESS: This is about 50 pages.
`
`If
`
`that's a question,
`
`I would take, you know, quite a
`
`long of time to look at that in detail.
`
`Now within
`
`a few minutes,
`
`it would be impossible to answer that
`
`question.
`
`BY MR. DOWD:
`
`Q.
`
`Well,
`
`take a few minutes and just if you
`
`could see --
`
`(Overlapping speakers.)
`
`THE REPORTER: Wait. Wait. Wait.
`
`You
`
`cannot interrupt him, please.
`
`THE WITNESS:
`
`Sorry.
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`a few minutes on the fly. That's not --
`
`THE REPORTER:
`
`Can I get a clean question,
`
`MR. DOWD:
`
`Sure.
`
`Q.
`
`I'd like you to take a few minutes, go
`
`back through the slides and tell me if you can
`
`identify anything in there that is discussing memory
`
`locations or indexing.
`
`MR. GLASS:
`
`Same objection. Outside the
`
`scope of the expert report.
`
`THE WITNESS: You're asking right now to
`
`do work that typically an expert would take many
`
`hours or perhaps days to do within a few seconds or
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`BY MR. DOWD:
`
`Well --
`
`-~- a reasonable request.
`
`-- whether it's reasonable or not, please
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`is that correct? MR. GLASS: Calls for a legal analysis.
`
`MR. GLASS:
`
`Same objection. Outside --
`
`outside the scope of the expert report.
`
`THE WITNESS:
`
`I would not even know
`
`exactly how I would look for whatever you want me --
`
`I don't know exactly the definition you have in
`
`mind.
`
`I would have no idea how to get started with
`
`this task.
`
`MR. GLASS: Objection. Calls for a legal
`
`analysis.
`
`BY MR. DOWD:
`
`Q.
`
`So if all I ask you is, can you identify
`
`for me a set of memory locations or a corresponding
`
`index in a provisional patent application, you would
`
`have no ability to understand what you're supposed
`
`to look for?
`
`MR, GLASS: Mischaracterizes the
`
`testimony. Outside the scope.
`
`BY MR. DOWD:
`
`QO.
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Go ahead.
`
`THE WITNESS:
`
`I would need to know exactly
`
`what you have been looking for.
`
`I would then take a
`
`considerable amount of time to ponder that question.
`
`I would have to look at these documents in detail,
`
`and after a considerable amount of time, which might
`
`be hours, which might be days,
`
`I might be able to
`
`perform an opinion. But it's not something that
`
`can -- that can be done within a certain amount of
`
`minutes.
`
`BY MR. DOWD:
`
`VIDEOTAPED DEPOSITION OF RUDIGER L. URBANKE
`CONDUCTED ON WEDNESDAY, FEBRUARY25, 2015
`
`transcript.)
`
`Q.
`
`So I
`
`take it, sitting here right now, you
`
`can't tell me one way or the other whether a memory
`
`location or a corresponding index is anywhere
`
`disclosed in this Exhibit 23?
`
`MR. GLASS:
`
`Same objections.
`
`And
`
`mischaracterizes the testimony.
`
`THE WITNESS: Without a détailed analysis,
`
`this would be