`December 17, 2015
`
`trials@uspto.gov
`571-272-7822
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`ERICSSON INC. and
`TELEFONAKTIEBOLAGET LM ERICSSON,
`Petitioner,
`
`v.
`
`INTELLECTUAL VENTURES II LLC,
`Patent Owner.
`____________
`
`Case IPR2014-01185
`Patent 7,269,127 B2
`____________
`
`Held: October 21, 2015
`____________
`
`
`
`
`
`BEFORE: JUSTIN BUSCH, PETER P. CHEN, and J. JOHN
`LEE, Administrative Patent Judges.
`
`The above-entitled matter came on for hearing on Wednesday,
`October 21, 2015, commencing at 2:00 p.m., at the U.S. Patent
`and Trademark Office, 600 Dulany Street, 9th Floor, Hearing
`Room D, Alexandria, Virginia.
`
`
`
`
`
`Case IPR2014-01185
`Patent 7,269,127 B2
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`
`J. ANDREW LOWES, ESQ.
`
`CLINT WILKINS, ESQ.
`
`Haynes and Boone, LLP
`
`2505 North Plano Road, Suite 4000
`
`Richardson, Texas 75082-4101
`
`
`
`
`
`
`
`
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`and
`
`JENNIFER WELLS, ESQ.
`Ericsson
`6300 Legacy Drive
`Plano, Texas 75024
`
`LORI A. GORDON, ESQ.
`STEVEN W. PETER, Ph.D., ESQ.
`Sterne Kessler Goldstein Fox
`1100 New York Avenue, N.W.
`Washington, DC 20005
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`P R O C E E D I N G S
`
` 2
`
`
`
`
`
`1
`
`
`
`Case IPR2014-01185
`Patent 7,269,127 B2
`
`
`- - - - -
`JUDGE LEE: Good afternoon. This is the oral hearing
`in case IPR2014-01185, concerning U.S. Patent Number
`7,269,127. In a moment, I will ask the parties to identify
`themselves for the record, but first, I am Judge John Lee, we have
`Judges Chen and Busch attending from the Silicon Valley and
`Detroit offices, respectively. Because of that, I would just like to
`remind the parties, please speak into the microphone at all times
`so that Judges Chen and Busch can hear what you are saying, and
`when you are referring to documents, be they papers or exhibits
`or demonstratives, please refer to things by number, paper
`number, slide number, et cetera, so that they can follow along.
`So, let's start with counsel for Petitioner, please identify
`yourselves for the record.
`MR. LOWES: Yes, Your Honor, I'm Andrew Lowes,
`lead counsel for Petitioner.
`JUDGE LEE: Could you just speak into the
`microphone, please.
`MR. LOWES: I'm sorry, Your Honor, I'm Andrew
`Lowes, lead counsel for Petitioner, Ericsson. With me at counsel
`table is Clint Wilkins, backup counsel. Also in attendance today
`is Jennifer Wells, who is in-house counsel for Ericsson.
`JUDGE LEE: Thank you, Mr. Lowes.
`For Patent Owner?
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`
`MS. GORDON: Yes, Your Honor, I'm Lori Gordon,
`from the law firm of Sterne Kessler Goldstein Fox and I'm going
`to be arguing today on behalf of the Patent Owner, Intellectual
`Ventures, and with me at counsel table is Steve Peters, also from
`Sterne Kessler.
`JUDGE LEE: Thank you. Before we begin arguments,
`does either party have any preliminary matters they wish to raise?
`Mr. Lowes?
`MR. LOWES: No, Your Honor, other than the
`projector currently is not working, if we could have a moment to
`try to make that connection.
`JUDGE LEE: I'll give you that before we begin.
`Anything from Patent Owner?
`MS. GORDON: No, Your Honor. Thank you.
`JUDGE LEE: Thank you. All right, Mr. Lowes, both
`parties have 45 minutes for their arguments. Petitioner, you may
`reserve some time for rebuttal, if you'd like; are you going to
`reserve some time?
`MR. LOWES: Yes, Your Honor, I would like to use 35
`minutes for my primary presentation, reserve 10 minutes for
`rebuttal.
`JUDGE LEE: That would be fine. And take whatever
`time you need to get your equipment hooked up.
`MR. LOWES: Thank you. Okay, I believe we're ready
`to go with the AV equipment.
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`
`JUDGE LEE: All right, Mr. Lowes, you may begin
`when you're ready.
`MR. LOWES: Thank you, Your Honor. As I indicated,
`I am lead counsel for Petitioner, Ericsson, and this is, for the
`record, IPR2014-01185, and today I'll be referring to
`demonstrative exhibits marked as Ericsson or ERIC 1037.
`So, in terms of my presentation today, I first want to
`give a little background about the '127 patent, which is the patent
`involved in the IPR, as well as then moving to the claim itself,
`and going through the claim, and some claim construction issues,
`followed by an application of the prior art to the claim.
`Turning now to slide 2 of Exhibit 1037, these are the
`claims that are involved in the IPR, for which trial is instituted.
`Claims 1 through 3 and 5, trial was instituted on a combination of
`Schmidl and Arslan, as demonstrating that the claims were
`unpatentable. For claims 4 and 6 through 10, it's the combination
`of Schmidl, Arslan and Kim that demonstrate that the claims are
`unpatentable. And for claim 17, it's Schmidl, Arslan, Kim and
`Heiskala. Of the claims in the trial, only claim 1 is in
`independent form.
`Slide 3, please. The '127 patent, which is Exhibit 101 in
`our exhibits, is directed to synchronization of a data frame
`between a transmitter and a receiver, as set forth in the
`background of that patent, it was known to use training symbols
`in the data frame for synchronization, and also to use pilot
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`symbols in that data frame for synchronization. The patent,
`rather than focusing on the pilot symbols that are part of the data
`structure, it's focused on providing efficient preambles, so it's
`trying to make the preamble structure of the data frame as
`efficient as possible to make the synchronization quick.
`And you can see here, on slide 3, I've copied in a
`portion of the background, where it indicates that there is a need
`for an efficient preamble structure that provides time and
`frequency synchronization, and then with respect to summarizing
`what the invention is about, the present invention provides a
`system for providing efficient preamble structures.
`So, here, the patent is recognizing that it wants to
`generate efficient preamble structures. A portion of figure 6 from
`the '127 patent is illustrated at the bottom of slide 3. You can see
`there that on the left-hand portion, it's showing the preamble
`structure 70, and to the right-hand side is the data structure 72.
`Within that preamble structure, it's got a first training symbol
`that's called the enhanced training symbol, and then it's followed
`by another training symbol as well as can be followed by other
`overhead items in the frame. So, here, again, the patent is about
`this preamble structure, and really it's about generating this
`enhanced training symbol.
`Can I go to slide 4, please. Turning now to the claim,
`on slide 4 we have reproduced claim 1. You can see that the
`claim is about a transmitter of a communication system, 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
`
` 6
`
`
`
`
`
`
`
`Case IPR2014-01185
`Patent 7,269,127 B2
`
`transmitter comprising an encoder, and then it also refers to a
`modulator and at least one transmit antenna. And as to the
`modulator, that's where the elements of the preamble are
`described. So, it talks about midway through the claim, a
`preamble structure and a data structure, the preamble structure
`comprising at least one training symbol and an enhanced training
`symbol. And then what's transmitted, it's discussing transmitting
`the enhanced training symbol in a single symbol.
`Now, these elements were disclosed by the primary
`reference, the Schmidl reference, in our filing, and in the Patent
`Owner's filings, they're really unrebutted that those elements were
`disclosed with respect to the modulator and the at least one
`transmit antenna. So, the dispute is about the encoder element of
`the claim. And that element requires an encoder having a pilot/
`training symbol inserter, the pilot/training symbol inserter
`configured to insert pilot symbols into data blocks and then to
`combine training symbols with data blocks.
`You'll notice that the word "training" symbol here for
`the encoder is used with -- in an encoder element. The same
`claim language is used with respect to the modulator, at the
`bottom there at line 63, where it says, "at least one training
`symbol." So, we have the same term being used twice in the
`claims, and as we'll discuss, those actually have different
`meanings.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
` 7
`
`
`
`
`
`
`
`Case IPR2014-01185
`Patent 7,269,127 B2
`
`
`Next slide, slide 4, of Exhibit 1037. Here's a
`combination of figures 2 and 3 from the '127 patent, and this is
`coming from the Patent Owner's response, and it's indicating that
`we have different domains described both within the patent, as
`well as set forth in the claim. And the domain depends on
`whether you're in the frequency domain to the left of the [I]DFT,
`so here in the modulator in figure 3, the [I]DFT component
`transforms the output of the encoder into the time domain, so
`everything to the right of the [I]DFT is in the time domain, and
`everything to the left is in the frequency domain.
`So, going back, if you would, please, go back to slide 4,
`with respect to the claim, the training symbol referred to in the
`encode element would be in the frequency domain, because that's
`the domain that it's operating in, while the training symbol
`referred to with respect to the modulator component, it says that
`each modulator outputting a frame structure comprising, among
`other things, a training symbol. That's going to be in the time
`domain. So, just to be clear, there is some domain switching
`going on and it creates some confusion by the way the patent is
`worded.
`If you would go, please, to slide 6. So, with respect to
`the encoder element, the parties have different views on what it
`means to insert pilot symbols into data blocks, and whether that
`insertion could encompass embodiments that result in separate
`OFDM pilot symbols in the time domain, or whether the patent --
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`the claim actually excludes such embodiments. So, it's our
`position that, indeed, if you insert pilot symbols into data blocks
`in the frequency domain, you can, in fact, end up with OFDM
`pilot symbols in the time domain. It's within the scope of the
`claim, it encompasses it, it's at least one embodiment that satisfies
`the claim.
`The Patent Owner has taken the position that pilot
`symbols are not continuous sections of symbols, and therefore
`cannot generate pilot symbol in the OFDM time domain, and
`also, in their response, they said pilot symbols are not placed on
`every subchannel. As we'll discuss, that's not correct, as well as
`their expert during cross examination agreed with us that there's
`no limitation that says you can't place pilot symbols on every
`subchannel.
`Next slide, please.
`JUDGE CHEN: Counsel?
`MR. LOWES: Yes, Judge?
`JUDGE CHEN: What is Petitioner's proposed
`construction for "pilot symbol" and where is the support for that
`in the specification of the '127 patent?
`MR. LOWES: We initially proposed that it be a -- a
`symbol in the data structure for synchronization, and the Board's
`institution decision indicated that no construction was necessary,
`and we think since the prior art utilizes the term "pilot symbols,"
`that, indeed, no construction is necessary.
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`
`In terms of the support for our understanding of what's
`in the -- disclosed in the patent, I would definitely get to that, and
`I could certainly jump to that now if you'd like, or it's coming up
`in just a couple of slides.
`JUDGE CHEN: Okay. Proceed.
`MR. LOWES: Okay, thank you.
`So, next slide, slide 7. Here, here is displayed from the
`Patent Owner's response, in the upper left-hand corner, Hartogs
`figure F, which Dr. Hartogs is Patent Owner's expert and he is
`illustrating the checkerboard effect where the hashed or dashed
`symbols are pilots that have been inserted in the data blocks and
`the data blocks run left to right horizontally, so that first data
`block has four pilot symbols inserted in it.
`In his declaration, he said that the figure G below that
`would not satisfy the claim, as it has data blocks and then a
`complete and filled block of pilots, and then more data blocks.
`During cross examination, we asked him, here you can
`see we've highlighted his testimony on the right, "is there a limit
`to how many pilots could be adjacent to one another in a
`particular data block the way you've represented it here?" His
`response, "only -- I'd say the only limitations here are put up by
`the cleverness of the implementer. Obviously, if you get to the
`point where you have the entire block filled with pilots, then it
`really has just become another training symbol and you probably
`have enough information to just reinitialize your transmission."
`
` 10
`
`
`
`
`
`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
`
`
`
`Case IPR2014-01185
`Patent 7,269,127 B2
`
`
`So, even Dr. Hartogs recognized there's no limitation to
`how many pilot symbols you can put in the data blocks.
`JUDGE BUSCH: Counsel, in those figures F and G,
`are you saying that each row is one symbol, one data symbol?
`MR. LOWES: No, Your Honor, each row is one data
`
`block.
`
`Chen --
`
`JUDGE BUSCH: Data block.
`MR. LOWES: Each of the squares is a symbol.
`JUDGE BUSCH: Okay.
`MR. LOWES: Next slide, please, slide 8. And, Judge
`
`JUDGE CHEN: Counsel?
`MR. LOWES: Yes, sir?
`JUDGE CHEN: Before we get to that, so is it
`Petitioner's position that the testimony at deposition of
`Dr. Hartogs undercuts to an extent figures F and G, which he
`created for purposes of this IPR?
`MR. LOWES: It certainly undercuts his explanation of
`figure G, and figure F is maybe a possible figure, but certainly the
`claims are not limited to an embodiment like figure F. There's
`nothing in the claims that designates where those pilot symbols
`have to be and that's where his testimony undercuts it, because he
`recognizes that the entire data block, running left to right, all six
`of those symbols could be filled with pilot symbols.
`Anything further?
`
` 11
`
`
`
`
`
`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
`
`
`
`Case IPR2014-01185
`Patent 7,269,127 B2
`
`
`JUDGE CHEN: Please continue. Thank you.
`MR. LOWES: Moving to slide 8 of Exhibit 1037, this
`is from Petitioner's expert, Dr. Haas, he is showing his
`understanding from the patent of how the patent operates to insert
`pilot and training symbols into the data blocks, and ultimately,
`the net result in the time domain.
`So, beginning at the top of the drawing, you have a
`series of data blocks entering the encoder. You also have a series
`of pilot and training symbols on the right-hand side, they're
`shaded blue. The encoder then inserts those, the training
`symbols, in the front of the data, and then the pilot symbols are
`inserted into the data blocks and in this illustration, it fills the data
`block. Those are then fed into the [I]DFT, where they're
`converted to the time domain, which outputs a series of time
`domain blocks. A cyclic prefix is added and ultimately you end
`up with the time domain frame shown in figure A, which is
`essentially a portion of figure 6 of the '127 patent actually
`showing the pilot symbols.
`We'll go further into kind of what supports this, but this
`is our position --
`JUDGE BUSCH: And, counsel, in this figure, in this
`figure, each one of the squares, again, is one symbol and each
`group that's separated off is a separate block is the position here?
`MR. LOWES: Yes, Your Honor.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
` 12
`
`
`
`
`
`
`
`Case IPR2014-01185
`Patent 7,269,127 B2
`
`
`JUDGE BUSCH: So, there are three symbols in each of
`these little sections labeled "pilot symbols?"
`MR. LOWES: Yes. Yes, Your Honor.
`JUDGE BUSCH: Okay.
`JUDGE CHEN: Counsel, why are the recited pilot
`symbols not limited to the frequency domain as the Patent Owner
`alleges?
`MR. LOWES: Certainly there is a number of reasons,
`one of which there is an express disclosure in the patent itself,
`through incorporation by reference, that the pilot symbol will be a
`known OFDM symbol. OFDM symbols are in the time domain,
`that's exactly what you have illustrated here where you have a
`time domain block with a cyclic prefix added to it. In addition,
`the patent describes that pilot blocks are transmitted with the data
`blocks, exactly as shown in this figure A that's been annotated, in
`the upper portion there shown in green it has a data block, and
`then next to that you have a pilot block, each of those separated
`by a guard symbol, the cyclic prefix.
`I believe some of that will become apparent in the next
`couple of slides, if you'll indulge me.
`Next slide, please, slide 9. This is portions of the
`specification that support Dr. Haas' understanding of the patent
`and the disclosure, as well as inform us about the scope of the
`claims.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
` 13
`
`
`
`
`
`
`
`Case IPR2014-01185
`Patent 7,269,127 B2
`
`
`Next slide, please. Referring to slide 10 of Exhibit
`1037, we have here from the patent, the '127 patent, that the
`patent itself incorporates an earlier provisional of the inventor's,
`which we refer to as the Mody provisional, Exhibit 1035. That
`was incorporated by reference into the '127 patent. Patent
`Owner's expert in reaching his conclusion that it needs to be that
`checkerboard pattern or a sprinkled pattern into the data blocks
`did not consider the earlier work that was incorporated by
`reference, which we have here his testimony, "Did you review
`those documents?
`"I'm sorry, I had not looked at that second provisional."
`If you would advance to the next slide, please. So,
`what's in that Mody provisional is recognition by the inventors,
`again, incorporated into the '127, that pilot in the form of known
`OFDM symbols can be sent for at least Q symbol periods. So,
`the inventors themselves are recognizing that the pilots in the
`time domain are actually going to be known OFDM symbols.
`They're not composite with data, and as Dr. Haas recognizes in
`his testimony, since the pilot symbols, which we have reproduced
`on the slide 11, since the pilot symbols are known OFDM
`symbols, this implies that they are known to the receiver, so the
`transmitter/receiver can synchronize, do not contain any user data
`in the frequency domain and contain pilot symbols in the
`frequency domain. That would make it a known symbol, the
`receiver is going to get what it expects.
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`
`And then, finally, the Mody provisional discloses
`separate OFDM pilot symbols in the time domain. Again, going
`back to Dr. Haas' figure, that's what's represented in figure A.
`JUDGE CHEN: Counsel?
`MR. LOWES: Yes, sir?
`JUDGE CHEN: So, the Mody provisional that is
`Ericsson 1035, what's the serial number on that one? Is that the
`60/322, 786?
`MR. LOWES: Yes, Your Honor.
`JUDGE CHEN: And it's Petitioner's position that this
`testimony of Dr. Hartogs was that he did not review 60/322,786,
`but did review 60/322,785?
`MR. LOWES: Your Honor, he testified that he did not
`review 60/322,786, I don't believe there is any testimony whether
`he reviewed the first provisional.
`JUDGE CHEN: Okay. Continue, please.
`MR. LOWES: So, moving on to slide 12, this is a
`portion of the patent that's just been reproduced, describing a pilot
`symbol. It's a relatively large portion so we have broken it up
`into two slides, B-1 and B-2.
`Move to slide 13, please. Here, you can --
`JUDGE CHEN: Well, counsel, the specification does
`follow the claim language about the insertion of the pilot symbols
`into the data blocks. So, I'll await your presentation on Schmidl
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
` 15
`
`
`
`
`
`
`
`Case IPR2014-01185
`Patent 7,269,127 B2
`
`and Arslan and how they address the insertion into data blocks,
`but please continue.
`MR. LOWES: Yes, Your Honor. So, next slide, slide
`13, please. Here, it's a portion of the specification, the term "pilot
`blocks" as used in this description refers to symbols provided by
`the pilot/training symbol inserter which are periodically
`inserted -- are inserted periodically into the data blocks.
`Typically pilot symbols can be inserted at any point into the data
`blocks. So, there's no limitation here that you have to have them
`spaced from one another or where they appear in the data blocks.
`The next slide, slide 14 of Exhibit 1037. With respect
`to this portion, which is from column 7 of the patent, lines 40 to
`45, it recognizes that pilot blocks are typically transmitted with
`the data blocks to calibrate, i.e. synchronize, the receiver to the
`transmitter on the small scale. Here again, the training symbols
`disclosed in the '127 patent are for the large-scale synchronization
`at the beginning of a transmission, where you -- the receiver and
`the transmitter haven't necessarily been in communication and
`they need to do a large-scale synchronization.
`As Dr. Haas recognizes here, the pilot symbols are for
`the smaller scale synchronization, and he says, "whereas for a
`periodic calibration, after the initial synchronization, which is
`accomplished by the training symbols, as few as one pilot symbol
`in the time domain (from multiple pilot symbols in the frequency
`domain) may be used for periodic calibration." So, this is also
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`disclosing that with respect to what is transmitted, not what the
`encoder generates, but what is transmitted in the time domain,
`that the pilot blocks themselves are actually transmitted with the
`data blocks. So, that's also from the specification.
`Next slide, please.
`JUDGE CHEN: You mean the pilot symbols are
`transmitted?
`MR. LOWES: Well, the pilot blocks are transmitted. If
`we went back to slide -- if we go back to slide 8, please. Here at
`the bottom of slide 8, showing figure A, you can see that a pilot
`symbol in the time domain is made up of a pilot block, as well as
`a guard G, a cyclic prefix, just like a data symbol is made up of a
`data block, as well as a cyclic prefix guard G, and similarly the
`training symbols in the time domain are made up of training
`blocks, as well as a guard G.
`JUDGE CHEN: Okay, thank you.
`MR. LOWES: So, if we could go to slide 15, please.
`Slide 15 of Exhibit 1037. So, here we have from the patent, the
`closest example we have of, you know, what does the frame
`structure look like in the time domain? What's represented here
`on slide 15 is a portion of figure 6 showing just one antenna from
`the multiple antenna system. Here we've already talked about
`this, it's got the preamble structure 70 on the left, followed by the
`data structure on the right. In terms of the description of what it
`is in the time domain, the patent says, although omitted from
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`figure 6 for simplicity, "pilot symbols may also be intermittently
`inserted into the data symbols by the pilot/training symbol
`inserter 46, as discussed above."
`So, the patent itself is recognizing that pilot symbols
`would be in the time domain frame structure illustrated in figure
`6. It's simply a question of, well, how would you illustrate those?
`They're missing, obviously they're in the claims, but they chose
`not to illustrate them, but how does one understand how you
`would illustrate those in the patent?
`If we can go to the next slide. We put that question to
`Dr. Hartogs, Patent Owner's expert, during his deposition, and we
`asked him, "Well, the patent states that the pilot symbols are
`omitted from figure 6 for simplicity. How would one represent
`pilot symbols in figure 6 if one wished to do so?"
`His answer: "I don't know."
`We said, "Could you represent a pilot symbol in figure
`
`6?
`
`"Nothing -- certainly, nothing comes to mind."
`So, Dr. Hartogs was not able to demonstrate how one
`would illustrate pilot symbols in the data structure of figure 6.
`Next slide, please. Dr. Haas, Petitioner's expert, also
`addressed this issue, and we have his testimony here concerning
`that quotation from the patent. He points to, in response to
`questions during cross examination, toward the bottom here, he
`says, "If you take figure 6 -- more precisely, one frame of figure
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`6, which is number 68, one of the 68s, and you modify it to -- not
`to omit the pilot symbols as figure 6 does, then the result would
`be something similar to my bottom of figure A." So, he's
`referring to his figure A.
`Slide 8, please, of Exhibit 1037. Here you can see a
`comparison between on the bottom Dr. Haas' figure A, the bottom
`is his figure A, and the existing portion of figure 6 from the '127
`patent. And the only --
`JUDGE BUSCH: Counsel, I need to clarify something.
`MR. LOWES: Yes, sir?
`JUDGE BUSCH: You mentioned earlier that there's
`two different symbols and I guess I want to make sure I
`understand what you're saying. So, the pilot symbols shown, I
`guess higher up on that figure A, where there were three symbols
`that made up a block in the frequency domain, those were pilot
`symbols in the frequency domain that made up a block, and now
`a pilot symbol in the time domain is what you're alleging here in
`the bottom of this figure?
`MR. LOWES: Yes, Your Honor.
`JUDGE BUSCH: Okay. Thank you.
`MR. LOWES: You're welcome.
`And that comes about, and I believe that is our next
`slide. Yeah, we could actually flip to the next slide, slide 19, it's
`just another representation here. It comes about because the pilot
`symbols in the frequency domain are then provided to the [I]DFT,
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`which converts them into the time domain, and turns them into a
`pilot block. Then a cyclic prefix is added and you end up with
`what's shown in the bottom, figure A.
`Would you please go back to slide 18. So, here, the
`difference is that in Dr. Haas' version, he's added an extra data
`symbol, as well as he's illustrating two pilot symbols, again, this
`is in the time domain. So, these are OFDM pilot symbols, just
`like the provisional incorporated by reference says, "pilot
`symbols of known OFDM symbols would be transmitted," that's
`exactly what's shown here in the bottom portion of figure A on
`slide 18.
`And what makes up a pilot symbol? It's the pilot block,
`which has been transformed from pilot symbols by the [I]DFT,
`and then what's added to that is the G, which is a cyclic prefix,
`and here you have them interspersed in the data symbols.
`Next --
`JUDGE BUSCH: And again, just to make sure I
`understand your argument, your argument is that if it's a known
`OFDM symbol, if it were intermixed with the data and became
`part of the data symbol or data block, I guess if you're ignoring
`the cyclic prefix, that that wouldn't be a known symbol because
`it's intermixed with the data?
`MR. LOWES: Correct. If you have data inside the
`pilot symbol then it's no longer a known symbol and it can't be
`used for synchronization.
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`
`Next slide. We've already gone through this, I'm happy
`to answer more questions, but I think we'll keep moving, if you
`would.
`
`So, turning now back to claim 1, claim 1, element 1.1, is
`the primary element in dispute. The remaining elements are
`disclosed by Schmidl, including Schmidl discloses two different
`training symbols, one having equal halves, which satisfies the
`enhanced training symbol, as well as it has the antenna for
`transmitting it, and additional features.
`So, the element in contention is "an encoder having a
`pilot/training symbol inserter, the pilot/training symbol inserter
`configured to insert pilot symbols into data blocks and to
`combine training symbols with data blocks." And as we'll show,
`the combination of Schmidl and it's encoder that inserts training
`symbols would be modified to achieve the benefits of having
`pilots in the data symbols as taught by Arslan.
`Next slide, please, turning to slide 21. Here's figure 1
`from the Schmidl reference. You can see on the far left side, the
`baseband data bits are coming into the encoder, that's just the data
`coming from some user device. The encoder 14 then bundles that
`data and adds and inserts training symbols to the data to create
`frames of data. Each of those bundles is then sent from the
`encoder in parallel format to the IFFT, which converts it into the
`time domain. The DSP adds a cyclic prefix and then ultimately
`it's transmitted via the antenna on the far right.
`
` 21
`
`
`
`
`
`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
`
`
`
`Case IPR2014-01185
`Patent 7,269,127 B2
`
`
`JUDGE CHEN: Counsel, how does the encoder in
`Schmidl accomplish the inserting function that's recited in the
`claim?
`
`MR. LOWES: It accomplishes it via controlling the
`inputs on the various subcarriers. It's described, I believe, in
`column 12 of the patent. It talks about how it inserts -- each of
`the subchannels can be controlled either for the enhanced training
`symbol version, the odd numbered subcarriers are held to zero,
`and the even ones, a pseudo-noise code is applied to those, within
`a single symbol interval. So, it's actually controlling the inputs
`for each symbol interval, and then those inputs are -- on the
`various subcarriers are passed to the IFFT for conversion into the
`time domain.
`Moving to slide 22. Here is the output of the Schmidl
`transmitter, in the time domain, figure 6, which is also similar to
`figure 6 of the '127 patent. Here you've got the two training
`symbols. In time period one you have a training symbol, time
`period two, also, and then three through M are data symbols, and
`as we'll discuss, these are OFDM training symbols, so they have a
`training block as well as cyclic prefix.
`Figure 6 illustrates that the placement of the training
`symbols within a data frame, you have two in the front of the data
`frame, that it describes that training symbols 134 and 136 are
`preferably placed at the beginning of the data frame, as shown in
`the figure. It also recognizes, while that's a preferred
`
`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 IPR2014-01185
`Patent 7,269,127 B2
`
`embodiment, so that the second OFDM training symbol
`immediately follows the first OFDM training symbol 34.
`Schmidl also recognizes that although the two symbols need not
`be placed next to each other, doing so minimizes the effect of
`time variation of the channel, but it's recognizing that those
`symbols themselves could be spaced from each other and could
`be separated by data if desired.
`Next slide, please. Moving to slide 23, this is just
`adding the detail from f