UNITED STATES PATENT AND TRADEMARK OFFICE
`______________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`______________
`
`SAMSUNG ELECTRONICS AMERICA, INC., AND SAMSUNG
`ELECTRONICS CO. LTD., AND APPLE INC. AND LG ELECTRONICS,
`INC.,
`Petitioner,
`
`v.
`
`FASTVDO LLC,
`Patent Owner.
`______________
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`______________
`
`Record of Oral Hearing
`Held: August 24, 2017
`______________
`
`
`
`
`
`Before KARL D. EASTHOM, JEFFREY S. SMITH, and PATRICK M.
`BOUCHER, Administrative Patent Judges.
`
`
`
`
`
`
`______________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`______________
`
`SAMSUNG ELECTRONICS AMERICA, INC., AND SAMSUNG
`ELECTRONICS CO. LTD., AND APPLE INC. AND LG ELECTRONICS,
`INC.,
`Petitioner,
`
`v.
`
`FASTVDO LLC,
`Patent Owner.
`______________
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`______________
`
`Record of Oral Hearing
`Held: August 24, 2017
`______________
`
`
`
`
`
`Before KARL D. EASTHOM, JEFFREY S. SMITH, and PATRICK M.
`BOUCHER, Administrative Patent Judges.
`
`
`
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONERS:
`
` DAVID L. FEHRMAN, ESQUIRE
` RYAN MALLOY, ESQUIRE
` Morrison Foerster LLP
` Marunouchi 1-Chome
` 29th Floor 5-1
` Chiyoda-Ku, Tokyo 100-6529
` (81)3-3214-6522
`
`ON BEHALF OF THE PATENT OWNER:
`
` WAYNE HELGE, ESQUIRE
` WALTER D. DAVIS, JR., ESQUIRE
` Davidson Berquist Jackson & Gowdey, LLP
` 8300 Greensboro Drive
` Suite 500
` McLean, Virginia 22102
` 571.765.7708
`
`ALSO PRESENT:
`
` Cyndi Wheeler - APPLE
`
`
`
`
`The above-entitled matter come on for hearing on Thursday, August
`
`24, 2017, at the U.S. Patent and Trademark Office, 600 Dulany Street,
`Alexandria, Virginia in Courtroom A, at 11:17 a.m.
`
`
`
`2
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
`
`
`
`P R O C E E D I N G S
`- - - - -
` JUDGE SMITH: Good morning. Welcome back,
` everyone.
` We're here in this matter of inter partes review
` 2016-01203.
` This is the matter of Apple and LG Electronics as
` Petitioner versus FastVDO as Patent Owner.
` I'd like to start by getting appearances of
` counsel.
` Who do we have here on behalf of the Petitioner?
` MR. FEHRMAN: Good morning, your Honor. David
` Fehrman of Morrison & Forester. I'm joined by Ryan Malloy of
` Morrison & Forester and Cyndi Wheeler from Apple.
` JUDGE SMITH: Thank you.
` Who do we have on behalf of Patent Owner?
` MR. HELGE: Good morning, Your Honor. Wayne Helge
` for Patent Owner, FastVDO, from Davidson, Berquist, Jackson &
` Gowdey. And I have with me backup counsel, Walter Davis.
` JUDGE SMITH: Thank you.
` I want to go over a few administrative details.
` The hearing order indicated that both sides will have 30
` minutes of argument.
` Petitioner, you'll present your case first. Patent
` Owner will be allowed to respond to Petitioner.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`3
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` Petitioner, if you wish, you may reserve time for
` rebuttal. Do you wish to do so?
` MR. FEHRMAN: Yes, Your Honor. Five minutes.
` JUDGE SMITH: Five minutes for rebuttal. So you'll
` have 25 minutes to present your case.
` You may begin when ready.
` MR. FEHRMAN: Thank you, your Honor.
` I'd like to start. I know that the court was very
` aware of the patent and its contents, but I'd like to do a
` quick review of the background because I think it's very
` important to set up the understanding here of those of skill
` in the art.
` So the state of the art of the prior art is
` described in the '482 patent in fairly clear detail. And
` various things that are relevant to the claims here --
` JUDGE SMITH: If you could, Counsel, when you
` reference the slides, please --
` MR. FEHRMAN: Yes, Your Honor.
` JUDGE SMITH: -- indicate the slide number.
` MR. FEHRMAN: So we're on Slide 4 now. And that's
` the general indication that transformation, quantization, and
` subsequent encoding were known in the art, looking at Column
` 2, lines 11-14 of the patent. We'll go into a little more
` detail. Transforming image data was known as indicated in
` Slide 5, 2, 23 to 32 of the patent. And quantizing was known
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`4
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` as we indicate in Slide 6.
` These are all excerpts from the background of the
` patent.
` And entropy coding was known in the art. And, in
` addition, specifically to the patent here and the claims, as
` we indicate on Slide 8 that's in the detailed description,
` split field entropy coding was also known in the art. And
` that's a reference to the Fiala article. As known to those
` skilled in the art, a variety of those codes can be separated
` into prefix and suffix. And that's indicated at Column 16 of
` the '482 patent.
` And another thing that's quite important is the
` observation and the background. As we indicate on Slide 9,
` that the occurrence of bit errors during a storage and
` transmission, so either one of them, of compressed data was
` known in the art. And this is indicated at Column 4, line 66
` to Column 5, line 5.
` And the transmission and storage here is basically
` grouped and indicates that both of these are referred to as a
` channel. And both of those in a channel are subject to
` corruption.
` And another observation in the background is that
` applying unequal error protection during storage and
` transmission was known in the art. And that's to provide
` more efficient error protection by not having to apply 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
`
`5
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` same amount of error protection to different portions of the
` data. And that's indicated on Slide 10.
` The '482 method and apparatus I think the board is
` very familiar with. That's starting on Slide 11 I think
` going through 18. So it indicates that -- at Column 13, that
` there's a split field code providing information in the
` prefix about the length of the suffix. And the suffix is
` represented by the original data that's indicated at Column
` 13, lines 51 to 58. And the term "split field code" is one
` that was coined in the patent, apparently.
` Onto Slide 13, in the disclosed coding scheme, the
` suffix is afforded a lower error of protection than the
` prefix because the prefix is more critical to keep protected
` because it can result in errors that will propagate. And as
` was indicated and discussed in the earlier hearing, it may be
` that the suffix fields are not provided with any protection
` whatsoever. They're lower protected than the prefix all the
` way down to no protection whatsoever. That's indicated on
` Slide 13.
` And in the patent at 16, 15 to 22, the suffix
` fields may be channel encoded with a lower level of error
` protection or may not be channel encoded, thereby providing
` no error of protection.
` And the specification describes this application of
` error protection to split field codes as being the novel
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`6
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` aspect of the claimed invention. And that's indicated in
` Slide 14, Column 16, 28 to 35, just a statement that the
` proposed codes have not been previously separated in order to
` provide error resiliency as provided by the method and
` apparatus of the invention.
` The coding scheme is indicated basically in
` Figure 1 and shown here on Slide 15. And it's -- that
` includes a suffix generator, prefix generator, and then the
` application of unequal error protection to those two portions
` of the data. And then the data is applied to either -- sent
` to either a transmitter for transmission or to a storage
` medium.
` And throughout the patent, there's really no
` distinction made between transmission or storage and the
` effect of channel errors on either one of those. They're
` treated essentially the same. We just list exemplary Claim
` 1, Slide 16 and 3, Slide 17.
` Just to review, on the institution decision, we
` have two grounds applicable here. The first ground is
` various other claims based upon obviousness over Kato, the
` same reference that was dealt with in the previous hearing.
` And the second ground is obviousness based upon the
` Fiala article that was identified in the patent and an
` article to Fazel and patent '622 to Fazel indicated on Slide
` 18.
`
`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
`
`7
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` Let me first address the obviousness in view of
` Kato. We have four main areas of argument. We'll address
` them one at a time. We indicate the four in Slide 21.
` And we highlight the first one on Slide 22. And
` that's with respect to a first set of Claims 1-3, 7-9, 13,
` 14, 23, and 24. And the Patent Owner argues that it was not
` obvious to modify Kato's prefix to specify the length of the
` suffix more directly.
` In terms of the disclosure of Kato, it discloses --
` very clearly discloses split field coding in terms of the
` division of code word into a first portion PI and a second
` portion RI. And that's indicated at Slide 24.
` The first portions and second portions are
` separated. And this is the portion of the specification at
` 31, 51 to 62 and 31, 65 to 67. And that they apply unequal
` error protection to these different portions of the split
` field code. And that's indicated in terms of the -- the
` split field code is shown in Figure 7, which we have on Slide
` 25. And the code is contained in a data store region.
` The data store region has first portions P sub I on
` the left, second portions R sub I on the right. And in this
` particular embodiment, that's one section of a data store
` region that would have continuous addresses. And the data is
` written from the left to the right for P sub I and from the
` ending edge on the right to the left for R sub I. And in
`
`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
`
`8
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` this description, these can both be variable length codes, so
` it's indicated the blocks here, P1 to PN may have different
` lengths and R1 to RN may also have different lengths.
` And the encoder -- encoding apparatus is
` illustrated in Figure 6A that we show here on Slide 26. And
` that shows the ECC encode operations will provide the unequal
` error protection at 603 after the data has been split into
` the two portions to form a split field code.
` And the split field code is contained in the RAM
` 617, which is a big issue here in terms of whether that's the
` the only place that this data store region exists or if it
` exists in other places as well, or it would be obvious to be
` in other places as well.
` In Kato, the code length -- we're on Slide 27 now.
` Kato discloses that the prefix PI provides L, the entire
` length of the code, rather than a direct indication of L2,
` the length of the suffix.
` In our view, that's essentially the only thing
` lacking from Kato. And our position it would have been -- is
` that it would have been obvious that the code could
` directly -- the prefix could directly indicate the length of
` the suffix rather than length -- indicating the length of the
` overall code.
` And we refer here on column -- on Slide 27 to
` Column 24, lines 48 to 45 [sic] which just indicates 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
`
`9
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` overall code length, L, is indicated in the prefix, PI.
` And onto Slide 28. Dr. Lippman explained why it
` was obvious to modify PI to specify L2 instead of L. And he
` basically says, in the end, in order to operate on the
` decoding side, you need to have the value of L2. And whether
` you get that by looking at L and then deriving that or
` providing it directly, the result is the same because you
` have to have that L2 value in order to be able to perform
` your decoding.
` And because of that, Dr. Lippman concludes a person
` of ordinary skill in the art would have found it obvious to
` modify Kato by coding PI to contain the value of L2 directly,
` that there's essentially no difference in the end result of
` obtaining L2 to facilitate the decoding operation.
` And onto Slide 29. The Patent Owner provided no
` expert testimony contradicting Dr. Lippman's testimony.
` Instead, their entire argument here is based upon one
` position, and that's that contrary to Petitioners' argument,
` Kato does not disclose using L2 in decoding PI and RI at the
` receiver. And that's our key issue here. We believe that
` that argument is wrong. And if it is wrong, the Patent Owner
` has no argument that it's not obvious to employ L2 directly
` as opposed to the overall length L.
` Their argument is that because Kato does not
` disclose using L2 during the decoding process, our argument
`
`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
`
`10
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` fails at the beginning, at the outset. But as we indicate on
` Slide 30, it's very clear that Kato does employ L2 in the
` decoding process. And we have here a number of excerpts from
` the discussion of the decoding operation, and those excerpts
` are essentially all from Columns 29 and 30. And it's Column
` 29, line 26, to Column 30, line 9, is basically the operation
` of the decoder with respect to its use of L2 in performing
` decoding. It needs to extract L2 and L1 in order to properly
` decode the data.
` So the three statements we have here are explicit
` statements that L2 is obtained. So we have L2 can be
` recovered at this stage. The data VI is recovered using the
` data L2. In the registered 632 and the word length data L2
` for the second portion is executed by subtraction unit. And
` it's set into the register to update the register 632. It's
` very apparent that L2 is in fact used in the decoding
` process.
` JUDGE SMITH: I think Petitioners' point -- I mean,
` Patent Owner's point, if I understand it correctly, that the
` part of Kato that you're referencing now, the Columns 29
` through 30, disclose what you have up there, but it also
` discloses not using L, using L max instead of L.
` MR. FEHRMAN: Well, that's based upon a
` misunderstanding of the overall operation of the decoder,
` Your Honor. The L max is only to properly align the data
`
`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
`
`11
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` with the RAM that is used during the decoding process.
` And if I could switch to the Elmo briefly. Is
` there -- is there somebody controlling the Elmo?
` MR. HELGE: He just walked out.
` MR. FEHRMAN: I cannot switch to that.
` If you refer to Column -- to Figure 6B of the
` patent, the Kato patent, it indicates the RAM 637, which gets
` the data VI and VI prime applied to it. And just for
` aligning the overall data with the input to the RAM
` correctly, it does use the maximum area; but in order to read
` out the data correctly from the data store region, which has
` been transmitted by the transmitter and applied with error
` correction code with the unequal error protection, so that
` transmitted data, the data store region is in the
` transmission or it could be in storage.
` In order to properly read that out, the two
` registers, 632 and 631, have to be set to coincidence with
` the proper beginning of each of the words of the data store
` region as indicated in Figure 7.
` So if we go back to Slide 25, the register 632,
` when each word is read out, it starts with the ending edge
` and reads out to the end of that word, but then it needs to
` be reset. And it's only reset by retrieving the value of L2,
` which is what this entire discussion is about. It retrieves
` the value of the L2 for the next data word, so it knows how
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`12
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` many to count down to set the starting point for that
` register for reading the next word.
` So L2 and equally L1 has to be used to read out the
` data from the proper point, otherwise, you're not going to
` read out the words that you're decoding.
` So both in respect to Figure 6B and the decoding
` operation and referencing it to the data store region in
` Figure 7, it is abundantly clear that L2 is used and directly
` used and required to be used. And that's indicated. Again,
` it's explicitly stated in Slide 30 at three different places
` and the overall discussion of that section makes it very
` clear how the operation is obtained to retrieve the proper
` words and read them in the right order to then decode them.
` So onto Slide 31, which is the second issue that
` we're dealing with, which is the issue of storing unequal
` error protected data. So Kato discloses -- clearly discloses
` storage of unequal error protected data. The Patent Owner's
` essential argument is that you can't look past the RAM of
` 617, but that's completely inconsistent with Kato and
` inconsistent with the patent itself in terms of what a data
` channel is and what can be considered a store.
` If this data store region is data that is
` transmitted, it's transmitted with error correction applied
` to it in Kato, and it's not restricted to just existing in a
` temporary fashion in the RAM 617 before the error protection
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`13
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` is applied to it.
` JUDGE SMITH: Let me ask you about that. What you
` have up there -- the sentence that you have up there ends
` with, "For example, a given former area of the data store
` region."
` Doesn't that support the contention that what Kato
` is talking about here is what was previously stored in the
` data store region, that phrase "former area of the data store
` region"?
` MR. FEHRMAN: I think the former area means the
` front region. So this is --
` JUDGE SMITH: I see.
` You don't think it means previous? The word
` "former" there, you think does not mean previous, it means
` front?
` MR. FEHRMAN: I don't think so, Your Honor. I
` think it's talking about applying the unequal error
` protection code selectively to the two selections of this
` data.
` JUDGE SMITH: Within the data store region itself?
` MR. FEHRMAN: Yes. Well, it's added onto this data
` store region, the error correction codes, and then the data
` store region is transmitted along with the error correction
` codes and received at the decoder.
` JUDGE SMITH: Yeah, I think it's -- as far as I
`
`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
`
`14
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` understand, everybody agrees they're transmitted together.
` MR. FEHRMAN: Right. So this only illustrates the
` region without the core pieces.
` JUDGE SMITH: Right. So it's given former area of
` the data store region, does that mean that the first region
` with the PI, within the data store region or --
` MR. FEHRMAN: That's my understanding, Your Honor.
` JUDGE SMITH: -- does that mean former in the sense
` of the data's already been read out from the memory?
` MR. FEHRMAN: I think it may work either way, Your
` Honor.
` JUDGE SMITH: I see.
` MR. FEHRMAN: The first area is what's given the
` higher level of protection, the prefix portion.
` JUDGE SMITH: And whether it's within the physical
` memory or not is ambiguous?
` MR. FEHRMAN: Well, the physical memory 617
` contains precisely just what's shown in Figure 7, which is
` the data store region prior to ECC encoding. And then that
` data store region is subjected to the ECC encoding and
` transmitted.
` JUDGE SMITH: Is that what that says? Does it say
` the data store --
` MR. FEHRMAN: No, this is the unequal error
` protection.
`
`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
`
`15
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` JUDGE SMITH: Where does it say the data store
` region in memory 617 is subjected to the unequal error
` protection?
` MR. FEHRMAN: Well, the Patent Owner's argument is
` that the data store region is restricted to memory 617. And
` our position is that's not the case at all.
` First, in terms of it is an obviousness issue, not
` an anticipation. Issue.
` And second, as we've show on the next slide,
` Slide 33, the Kato patent itself refers to that data region
` -- data store region doesn't have to be a physical data
` region, it can be a packet transmission, it can be a data
` region in the case of the disc-shaped recording medium.
` Both of those -- certainly the packet transmission
` is outside of the RAM 617. That's after the ECC encoding,
` modulation, and transmission. And that's -- substantially
` our issue is that the storage here can occur, it would be
` obvious to form storage after the modulation and after the
` ECC correction is applied.
` The Patent Owner is arbitrarily restricting any
` storage of the data store region to the RAM 617. So this is
` just from the patent itself, but --
` JUDGE SMITH: But that still begs the question --
` no, I mean, I think data is stored in the medium. Kato
` discloses 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
`
`16
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` The question is, does Kato disclose the error
` corrected data -- could you go back to the previous slide.
` MR. FEHRMAN: Sure.
` JUDGE SMITH: So, you know, again, the data store
` region could be the physical medium --
` MR. FEHRMAN: Yes.
` JUDGE SMITH: -- you know, as you pointed out in
` Column 33.
` MR. FEHRMAN: It could be a physical medium.
` JUDGE SMITH: But this basic question is this error
` correction applied to data within that region or does it only
` apply after its read from the data store region?
` MR. FEHRMAN: Well, that region by itself, when
` it's in 617, it does not have error correction applied to it,
` yet, it's following the ECC 603.
` JUDGE SMITH: So what evidence -- when you say its
` obvious to store it after providing error correction, what
` evidence do you --
` MR. FEHRMAN: Well, one evidence is that it's
` talking about in the next slide, Slide 33, a packet
` transmission, that the RAM 617 has nothing to do with a
` packet transmission. The transmission may be that the output
` 605 of the encoder. And then it also equates this and says
` that would apply equally to a disc-shaped recording medium.
` It doesn't restrict that to, well, that's only
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`17
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` inside at the RAM 617. And so that's the patent itself.
` And also, and as we recall, the background of the
` patent talks about channels and storage and transmission and
` doesn't make any distinction between them. They're both
` subjected to errors potentially.
` And that's what Dr. Lippman testified to as well,
` that the reason that the data is compressed is so that it can
` be more efficiently stored or transmitted. And the issues
` faced are the same in each case.
` And so in terms of our obviousness position, it's
` that it's obvious that the data at the end of the encoder can
` be stored in the error protected form or it can be
` transmitted, which is exactly what's shown in the patent
` itself.
` And we've never taken the position that the error
` -- that the data in the RAM 617 is error protected. That,
` obviously, occurs at the ECC encode box. And --
` JUDGE SMITH: You have about a minute left.
` JUDGE EASTHOM: Is there any evidence -- or does
` your expert discuss the contention you're making about the
` packet being downstream of the error correction code, the
` packet formation?
` MR. FEHRMAN: That specifically, Your Honor, I
` don't think so; but he's talking about in general here and
` other places in his declaration that the reason for
`
`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
`
`18
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` compression is to facilitate more efficient storage and/or
` transmission.
` In general, things are often stored and then
` transmitted. It would be stored in an error protected form
` and then later transmitted.
` JUDGE EASTHOM: Okay.
` JUDGE SMITH: In Figure 6B of Kato, when this
` transmission with the unequal error protection is received,
` is it stored anywhere along the way before being read out of
` the decoder?
` MR. FEHRMAN: Well, I would imagine that it's
` stored to separate data in the ECC decode, but we have not
` identified things that don't have a specific statement of
` storage. So the ECC decode would occur first at 621 and then
` at 635 would be, again, a data store region that's shown in
` Figure 7 without the ECC data applied to it. And that's --
` that data requires L2 to enable read-out of the data -- L1
` and L2 to enable those to be read-out and properly decoded.
` JUDGE SMITH: Okay. Thank you.
` MR. FEHRMAN: Thank you, Your Honor.
` JUDGE SMITH: Patent Owner, you have 30 minutes.
` You may begin when ready.
` MR. HELGE: May it please the board. Thank you,
` Your Honor.
` Your Honor, I would like to pick up on something
`
`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
`
`19
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` that Petitioner just talked about. We had seen a number of
` different slides on the -- and, Your Honors, just for
` clarification, we have no demonstratives again. I'll be
` referring to parts of the record, parts of Apple's slides as
` well -- in particular, parts of Apple's slides.
` Your Honor, something that Petitioner has said a
` few times was that Dr. Lippman's paragraph 80 of his
` declaration talks about the reason for compression is for
` more efficient storage or transmission. The question I have
` for you and ask you to think about as you're preparing your
` final written decision is what does that have to do with
` error correction encoding? It has nothing to do to do with
` error correction encoding.
` Compression is about entropy encoding. Compression
` is about reducing the size of your data that comes in, for
` example, into the encoding circuit 602 for storage in the
` RAM.
` Error correction encoding is the opposite of
` compression. Dr. Lippman's statement about compression, the
` purpose of it being for more efficient storage or
` transmission, has nothing to do with the teachings or
` applying the teachings of Kato to then lead you to store
` error correction encoded data. They do not follow.
` JUDGE SMITH: Well, let me just ask about this
` because Petitioners' point, if I understand Petitioner
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`20
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` correctly, what Petitioner is saying is, because you have
` this variable length encoding with these split fields, the
` reason why you need to add the additional error correction to
` the prefix field is because if you don't, then you lose
` synchronization, you lose the whole thing. So while the
` additional error correction adds bits instead of compresses
` bits, it does so to make the decoding of the compressed bits
` possible. That's why you would do it to this compressed
` split field encoded data.
` MR. HELGE: Your Honor, I think what you're getting
` at is the idea of the fundamental nature why do we apply
` error correction codes at all. And I think what you're
` getting at is exactly what's shown in Kato 6A, which is we're
` going to apply entropy coding, we're going to store it in
` RAM, and then we're going to put it to the ECC encoder before
` it's transmitted.
` And why do we do it before it's transmitted,
` because that's where the errors could occur. And that's what
` Kato discloses.
` But in terms of Dr. Lippman's statement in
` paragraph 80, and I encourage you to --
` JUDGE SMITH: Could they -- I'm sorry, just real
` quick.
` MR. HELGE: Certainly, Your Honor.
` JUDGE SMITH: Could they also occur if they're
`
`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
`
`21
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` stored in an electronic memory?
` MR. HELGE: Your Honor, our background talks about
` the possibility of noisy storage. And what I would
` distinguish here is the concept of noisy storage versus what
` Kato discloses as RAM.
` There's no reason to apply error correction encodes
` to Kato's RAM. It's not classified a noisy. Kato doesn't
` talk about it being noisy.
` Petitioners' just admitted that you don't apply
` error correction encoding to the data that's stored in the
` RAM. So I think just for the purposes of distinguishing the
` two, if there was a recognition that storage is noisy, I
` think our background talks about that as a concept. That is
` not applicable here in Kato. That's not a situation that
` we're faced with.
` Now, what I mentioned with Dr. Lippman's paragraph
` 80, if you read Dr. Lippman's -- well, a number of times
` throughout his declaration, he has very conclusory statements
` and paragraph 80 is one of those, in particular, where he
` says here, I'm going to quote, "Given Kato's teaching that
` the first portions PI should be subjected to higher error
` protection than the second portions RI, one of ordinary skill
` in the art would have recognized the importance of storing PI
` in a higher error protected data block of a storage medium
` than the RI."
`
`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
`
`22
`
`
`
`
`Case IPR2016-01179
`Case IPR2016-01203
`Patent 5,850,482
`
` What that statement doesn't deal with or what it
` really assumes going in is that you're going to store it in
` the first place. In terms -- what he's really dealing with
` there in paragraph 80 is the idea of the different data
` blocks. He's not talking about the idea of storage. And as
` the Petitioner just said, they don't point to anything
` downstream -- well, I want to be clear about that -- they
` don't point to the ECC encoder. They don't point to the
` modulator in Figure 6A.
` Dr. Lippman has a number of these types of very
` conclusory statements, and I want to point this out because
` these are pervasive throughout his declaration.
` He continuously makes assumptions in a manner the
` Federal Circuit has recently spoken to in The personal web
` technologies case, the In re: NuVasive case, these are
` techniques that the Federal Circuit says this is not enough.
` It's not simply whether a person of ordinary skill in the art
` could have seen these two refere
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
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.
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.
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
