`571-272-7822
`
`
`
`
`
`
`
`IPR2014-00501, Paper No 47
`June 3, 2015
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`MOTOROLA MOBILITY LLC,
`Petitioner,
`
`v.
`
`INTELLECTUAL VENTURES I LLC,
`Patent Owner.
`____________
`
`Cases IPR2014-00501
`Patent 7,136,392
`____________
`
`Held: May 5, 2015
`____________
`
`
`
`BEFORE: MICHAEL W. KIM, PATRICK R. SCANLON, and
`KRISTINA M. KALAN, Administrative Patent Judges.
`
`
`
`
`
`The above-entitled matter came on for hearing on Tuesday, May 5,
`2015, commencing at 1:30 p.m., at the U.S. Patent and Trademark
`Office, 600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`JOHN C. ALEMANNI, ESQ.
`
`
`MITCHELL STOCKWELL, ESQ.
`
`
`Kilpatrick Townsend
`
`
`1001 West Fourth Street
`
`
`Winston-Salem, North Carolina 27101
`
`
`ON BEHALF OF PATENT OWNER:
`
`
`JOHN KING, ESQ.
`
`
`BRENTON R. BABCOCK, ESQ.
`
`
`TED CANNON, ESQ.
`
`
`Knobbe Martens
`
`
`2040 Main Street, 14th Floor
`
`
`Irvine, California 92614
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
` 2
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`
`
`
`
`
`
`
`
`
`
` P R O C E E D I N G S
`
`- - - - -
`
`JUDGE KIM: Good afternoon, everyone, please be seated.
`
`It looks like everyone is ready, so how much rebuttal time would you
`
`like to reserve?
`
`MR. STOCKWELL: Ten minutes, Your Honor.
`
`JUDGE: Great. You can begin when you're ready.
`
`MR. STOCKWELL: Good morning, Your Honor, or good
`
`afternoon. As Mr. Alemanni mentioned, I'm Mitch Stockwell, I
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`represent the Petitioner, Motorola Mobility. I want to start with just
`
`11
`
`reviewing what is disputed and what is not disputed. What's not
`
`12
`
`disputed is that the Ayyagari references that we rely on disclose all the
`
`13
`
`features of claims 1 and 3. It's also not disputed that motivation to
`
`14
`
`combine the Ayyagari White Paper with the IEEE 802.11-1999
`
`15
`
`standard exists that renders claim 10 obvious. So, the only issue as to
`
`16
`
`these claims would be whether the Ayyagari references are prior art,
`
`17
`
`not the substance of the references themselves.
`
`18
`
`With respect to the remaining claims, there are disputes,
`
`19
`
`there are several different disputes. We've grouped the claims into
`
`20
`
`four different groups, groups A through D, based on the limitation that
`
`21
`
`the Patent Owner contests as being absent in the Ayyagari references,
`
`22
`
`and I'm going to spend probably five minutes just giving a little bit of
`
`23
`
`an overview and then 10 minutes each on groups A and B, a few
`
`
`
`
`
` 3
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`minutes on the group C and D claims, and then on the last issue,
`
`whether the Ayyagari references are prior art, just a few minutes.
`
`I wanted to give you that overview because obviously to the
`
`extent that the Board has any questions, I think that's our principal job
`
`here is to answer your questions. Just in terms of the overview,
`
`obviously the IEEE 802.11-1999 standard, the WiFi standard, allowed
`
`for the carrier sense mechanism. That was the mechanism by which
`
`before you sent a data message, you would listen on the medium, find
`
`the idle period and then send the message if another station was
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`attempting to grab that transmission opportunity, and you got a
`
`11
`
`collision, it would run a backoff before attempting a retransmit.
`
`12
`
`What 802.11 did not have, and what Mr. Wentink described
`
`13
`
`in his declaration, and this is on slide 7 of Patent Owner's slides, for
`
`14
`
`the record, Exhibit 1021, as Mr. Wentink said in his declaration, in
`
`15
`
`attempting to submit evidence of conception, diligence and reduction
`
`16
`
`to practice, he was working with the IEEE Task Force and he said
`
`17
`
`specifically, the mission of the task force was to address quality of
`
`18
`
`service issues within the 802.11 standard, and the reason for that was
`
`19
`
`that at the time that he began his work on the task force, the 802.11
`
`20
`
`standard had stations that essentially had one queue, so there was no
`
`21
`
`way of prioritizing traffic and giving QoS, and in fact, this is also the
`
`22
`
`problem that Mr. Ayyagari and his co-authors were focused upon,
`
`23
`
`Exhibit 1005, at page 3, in explaining the approach, the Ayyagari
`
`24
`
`authors explained that a QoS guarantee for real-time application data
`
`25
`
`traffic is currently not available in the IEEE 802.11 standard.
`
`
`
`
`
` 4
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`So, that was the problem that both Mr. Wentink and his
`
`co-inventors and Mr. Ayyagari and his co-authors were each facing.
`
`And we think the approach that both groups took to the problem was
`
`the same.
`
`In the case of the '392 patent, there were essentially two
`
`approaches to the problem. The first approach was to add queues so
`
`that you would have multiple queues and thereby you could
`
`differentiate your traffic. You could have a queue for high-priority
`
`video traffic, and a separate queue for low-priority data traffic.
`
`10
`
`Once you had the queues, though, you then had to ensure
`
`11
`
`that you gave preferential access to the communications channel,
`
`12
`
`those transmission opportunities, to higher priority traffic. And the
`
`13
`
`mechanism by which, and this is on slide 9, Exhibit 1021, the
`
`14
`
`mechanism by which the '392 patent provided that preferential access
`
`15
`
`was to specify contention windows and define those contention
`
`16
`
`windows by reference to the priority level of the queue. And by
`
`17
`
`setting up the contention window timing for each queue, you could
`
`18
`
`ensure that higher priority messages received preferential access over
`
`19
`
`lower priority messages.
`
`20
`
`I want to jump to slide 11 and talk a little bit about what
`
`21
`
`Ayyagari disclosed in terms of that group's approach to the problem.
`
`22
`
`And it was the same. At the top of the slide, you can see that they
`
`23
`
`expressly discussed grouping packets with similar priority levels into
`
`24
`
`queues so that you would have multiple queues. That's also
`
`25
`
`referenced in the Ayyagari White Paper that you would have to
`
`
`
`
`
` 5
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`establish multiple queues. They would use the carrier sense
`
`mechanism, CSMA/CA, and I'm just going to refer to that as carrier
`
`sense rather than the algorithm, in the IEEE 802.11 to look for the
`
`transmission opportunity, and at the bottom of Exhibit 1021, 11, they
`
`discussed providing preferential access by manipulating the
`
`contention windows based on the priority level of the packet. So, as it
`
`says here, the time interval size delay, the CW, is a function of the
`
`priority level of the packet.
`
`So, that's the general background in terms of the two group's
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`approach to the IEEE problem, and it brings us really to our first
`
`11
`
`dispute, and this is the claim language on slide 12, this is claim 4. The
`
`12
`
`group A claims 4 through 6, 8, 13 through 15 and 19 through 21 have
`
`13
`
`similar limitations, and generally, what this limitation discusses is,
`
`14
`
`attempting to initially transmit a message data unit, and in essence, for
`
`15
`
`this particular scenario, the message data unit is coming out of a
`
`16
`
`low-priority queue. If you imagine the timers for the low and
`
`17
`
`high-priority queue expire at the same time and they're each trying to
`
`18
`
`grab the transmission opportunity, you want to defer to the
`
`19
`
`high-priority queue and as the claim limitation at the end says, you
`
`20
`
`treat the low-priority message as superseded and as if an unsuccessful
`
`21
`
`attempt to transmit the message had occurred, i.e., you back it off.
`
`22
`
`That's what the claim language says, and the specification
`
`23
`
`explains the claim language. Here at the top, I've got a cite on slide
`
`24
`
`13 from Exhibit 1001, this is the column 7 cite and the column 11 cite,
`
`25
`
`but just focused on the top cite here, so it talks about what happens
`
`
`
`
`
` 6
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`when there's a collision within the same station, it says the lower
`
`priority colliding queues behave as if there were an external collision
`
`on the wireless medium.
`
`So, that "as if" language in the claim is describing the
`
`behavior of the lower priority queue when there's this internal
`
`collision, and effectively what that behavior amounts to is the lower
`
`priority queue is superseded and delayed, as the next quote makes
`
`clear.
`
`So, that's the basic set-up for the group A claim limitations.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`Where are the disputes? Well, slide 14 summarizes that. The
`
`11
`
`Ayyagari references, according to the Patent Owner, don't teach this
`
`12
`
`claim limitation, and the Patent Owner's primary argument there
`
`13
`
`seems to be that the reference in Ayyagari to superseding the lower
`
`14
`
`priority queue and even if -- and then treating it as if backed off, is
`
`15
`
`only a possibility. We obviously disagree with that. We believe when
`
`16
`
`both queues contend for the transmission opportunity, the Ayyagari
`
`17
`
`'508 clearly describes superseding the lower priority queue and
`
`18
`
`backing it off.
`
`19
`
`So, this is the first substantive dispute between the parties.
`
`20
`
`If you go to slide 15, this is the language that the parties have focused
`
`21
`
`on and the parties' experts have focused on in Ayyagari, the second
`
`22
`
`bullet point. The first bullet point just refers to some other language
`
`23
`
`in the Ayyagari '508 patent that talks about what happens when an
`
`24
`
`external collision, that is, when two different stations try to grab the
`
`
`
`
`
` 7
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`same transmission opportunity, basically each of those stations are
`
`randomly backed off before they try to transmit again.
`
`The second bullet point, though, explains what happens if
`
`these queues within a station each try to grab the same transmission
`
`opportunity. So, it starts with a packet with higher priority than a
`
`packet waiting here in step 640, and that packet waiting is the lower
`
`priority packet, may be transmitted if the higher priority packet has a
`
`shorter wait time, even if it results in the superseded packet, that's the
`
`lower priority one, being forced into another backoff state.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`And as you can see, in slide 16, this is the language that
`
`11
`
`Dr. Roy explained teaches one of skill in the art the limitations of the
`
`12
`
`group A claims, the "attempting to transmit" limitations.
`
`13
`
`So, one of the things that I want to focus on is the Patent
`
`14
`
`Owner challenges this language and says it's insufficient to show a
`
`15
`
`disclosure of the "attempting to transmit as if" limitation. What I
`
`16
`
`would like to do is turn to the Patent Owner's response.
`
`17
`
`Ms. Gray, if you could pull that up, it's paper number 8.
`
`18
`
`And if we could go to page 46 in the PDF version of that, page 41 of
`
`19
`
`the actual document. If you could blow up claim 4. Thank you.
`
`20
`
`So, this is the Patent Owner's response, and it's the portion
`
`21
`
`of the response where the Patent Owner lays out a comparison of the
`
`22
`
`claim language to the exhibits to Mr. Wentink's declaration that was
`
`23
`
`submitted during the original prosecution to establish conception -- an
`
`24
`
`attempt to establish diligence and reduction to practice. And this
`
`25
`
`claim chart comparison was put together by Dr. Tewfik, who opined
`
`
`
`
`
` 8
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`that the documents showed conception, that is they would have
`
`identified to skilled persons how to build the image and to
`
`demonstrate possession by the inventor.
`
`And I want to focus on some of the language that Dr. Tewfik
`
`and Intellectual Ventures acknowledge necessarily disclose the
`
`limitations of claim 4, and if you look at the middle paragraph, I'll
`
`read it so that Your Honors who do not see the screen can hear this.
`
`It's a cite to Exhibit 1002, that's Mr. Wentink's declaration at 224,
`
`that's one of the attachments to the declaration. It says, "A collision
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`between local accesses are resolved locally. The higher priority frame
`
`11
`
`will be send" -- I think that means sent -- "first, while the lower
`
`12
`
`priority it collided with should act as if it is to defer on a Tx-Opp" --
`
`13
`
`that's a transmission opportunity -- "so would do another backoff."
`
`14
`
`Now, that language, the language about forcing the lower
`
`15
`
`priority message into another backoff is the identical disclosure that
`
`16
`
`we see in the Ayyagari '508 patent. Now, Patent Owner contends that
`
`17
`
`this does not matter, because the "even if" language in the Ayyagari
`
`18
`
`reference somehow indicates that Ayyagari's backoff of the lower
`
`19
`
`priority packet in favor of the higher priority packet merely indicates a
`
`20
`
`possibility, okay? And we take issue with that because what the "even
`
`21
`
`if" language indicates is what happens when a particular condition
`
`22
`
`occurs? Obviously there's -- when you're dealing with a multiple
`
`23
`
`queue station, some queues may be empty, and may not have
`
`24
`
`high-priority packets to transmit. You may not have low-priority
`
`25
`
`packets to transmit. There may be no internal collision.
`
`
`
`
`
` 9
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`If a queue is empty, and you're simply sending messages out
`
`of a single queue, the condition at which a collision will occur so that
`
`the lower priority packet will then be forced into a backoff state,
`
`doesn't happen, but when both queues are full, their timers expire or
`
`they're otherwise seeking to engage on the same transmission
`
`opportunity. The Ayyagari reference discloses that in that instance, in
`
`that condition, we are going to supercede the lower priority packet,
`
`send the higher priority packet, and push the lower priority packet into
`
`a backoff state.
`
`10
`
`That is not an indication of possibility, that is a description
`
`11
`
`of what happens under the conditions as noted when the queues
`
`12
`
`expire.
`
`13
`
`14
`
`15
`
`JUDGE KALAN: Counsel, can you hear me?
`
`MR. STOCKWELL: Yes?
`
`JUDGE KALAN: So, are you reading the "may be
`
`16
`
`transmitted" language as is transmitted effectively?
`
`17
`
`MR. STOCKWELL: I'm sorry, I'm not sure which "may be
`
`18
`
`transmitted" language you're referring to.
`
`19
`
`JUDGE KALAN: In the cited section on page 15 of your
`
`20
`
`slides.
`
`21
`
`22
`
`MR. STOCKWELL: If you could jump back to that.
`
`JUDGE KALAN: "A packet waiting during step 640 may
`
`23
`
`be transmitted." Are you saying that's not conditional or are you
`
`24
`
`reading it as is transmitted, even if it results in the superseded packet
`
`25
`
`being forced into another backoff state?
`
`
`
`
`
` 10
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`MR. STOCKWELL: Okay, so the "may be transmitted?"
`
`JUDGE KALAN: Um-hmm.
`
`MR. STOCKWELL: That's not conditional, that's
`
`explaining what happens when this condition occurs, in our view. So,
`
`a packet with higher priority, that's the packet that may be transmitted.
`
`JUDGE KALAN: But you're not going so far as to say it is
`
`transmitted, it's just talking about this condition?
`
`MR. STOCKWELL: Well, but that's exactly, it's not talking
`
`about it in a conditional sense, it's saying when this condition occurs,
`
`10
`
`then we're going to send the higher priority packet. That's the whole
`
`11
`
`reason for having higher priority packets. And, in fact, Patent Owner
`
`12
`
`does not urge that the "may be transmitted" here is what, in their
`
`13
`
`argument, creates this possibility. They urge that it's this "even if it
`
`14
`
`results" language. They say that's what suggests this is a mere
`
`15
`
`possibility.
`
`16
`
`Our explanation, and Dr. Roy's explanation is, no, you have
`
`17
`
`two queues. You know one queue is higher priority than the lower
`
`18
`
`queue. You know that sometimes those queues will engage for the
`
`19
`
`same transmission opportunity. When that happens, the higher
`
`20
`
`priority queue may be transmitted even if you end up putting the
`
`21
`
`lower priority queue into a backoff state.
`
`22
`
`So, what Dr. Roy explained is this segment here explains the
`
`23
`
`conditions under which that internal collision occurs, and what
`
`24
`
`happens. And in this case, the superseded packet is delayed, forced
`
`
`
`
`
` 11
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`into a backoff state, it behaves as if it had been subject to an earlier
`
`transmission.
`
`Did I answer your question?
`
`JUDGE KALAN: Yes, thank you.
`
`MR. STOCKWELL: I'm going to skip forward and deal
`
`with the next group of claims, this is the group B claims, because I
`
`think the other Ayyagari White Paper references are largely the same.
`
`The only thing I would like to say about the Ayyagari White Paper is,
`
`with respect to the group A claims, and this is on slide 17, Patent
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`Owner's position on this disclosure, which Dr. Roy also explained,
`
`11
`
`met the "attempting to transmit as if" limitation, is essentially that this
`
`12
`
`language in the second bullet point on slide 17, the first in/first out
`
`13
`
`priority queue, is talking about a single queue and therefore somehow
`
`14
`
`the Ayyagari White Paper cannot disclose this internal collision that
`
`15
`
`occurs and the delay of the packet. And we think that's absolutely
`
`16
`
`incorrect.
`
`17
`
`If you look at the Ayyagari White Paper, at page 7, it
`
`18
`
`expressly says, "define multiple FIFO buffers, one for each priority
`
`19
`
`level," bridging between pages 5 and 6 on Exhibit 1005. It talks about
`
`20
`
`"in order to minimize the complexity of the network, we suggest the
`
`21
`
`use of the non-gated head-of-line priority scheme using simple FIFO
`
`22
`
`priority queues."
`
`23
`
`So, what the disclosure and what Dr. Roy explained is,
`
`24
`
`clearly there are multiple queues in the Ayyagari White Paper, the
`
`25
`
`description of the first in/first out priority queue, and sending the
`
`
`
`
`
` 12
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`highest priority messages explains and teaches the skilled person the
`
`"attempting to transmit" limitation.
`
`I'm going to now talk a little bit about the group B claims,
`
`the dispute here deals with this limitation that's shown on slide 20, it's
`
`the means for sensing the communication medium for an opportunity
`
`to transmit message data units without interference. And we have
`
`essentially two disputes here. If we look at slide 21, one of them is a
`
`claim construction dispute. It's a straight-up what is the appropriate
`
`Section 112, 6 structure that informs the mean of the "means for"
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`language, and the other is a dispute as to whether or not the Ayyagari
`
`11
`
`references teach the means for sensing.
`
`12
`
`So, with respect to the first issue, if we go to slide 22 --
`
`13
`
`excuse me for a moment, I'm going to grab some water.
`
`14
`
`In the middle of the slide, we have set forth the algorithm
`
`15
`
`that the Board noted was the appropriate algorithm for the means for
`
`16
`
`sensing. It's the six-part algorithm outlined in the specification.
`
`17
`
`Leading up to this initial finding by the Board, the parties were
`
`18
`
`disputing the appropriate structure for means for sensing, largely the
`
`19
`
`parties agreed that the means for sensing included the transceiver in
`
`20
`
`terms of a structure for sensing the carrier mechanism, and a CPU,
`
`21
`
`then the debate was whether the CPU -- whether the specification
`
`22
`
`disclosed a precise algorithm for programming the general purpose
`
`23
`
`CPU. And this is the Board's finding as to what the algorithm is.
`
`24
`
`Now, the Patent Owner now contends that there is some
`
`25
`
`additional structure that should inform the means for sensing element,
`
`
`
`
`
` 13
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`specifically they contend that there are multiple schedulers shown in
`
`one of the figures that should be part of the structure.
`
`Well, that is certainly not within the Board's construction as
`
`articulated with respect to the algorithm. We went back and looked at
`
`each one of the citations that the Board has. The first citation, I've put
`
`a portion of it just below, at the very bottom of the slide, the citation
`
`number 1 to column 6 really just describes the carrier sense
`
`mechanism function that was known in IEEE 802.11 standard. The
`
`remaining references, as to the second reference, starts out with
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`column 6, that describes the coordinated scheduling function and the
`
`11
`
`CW, the contention window parameters.
`
`12
`
`The third reference discloses -- talks about the carrier sense
`
`13
`
`mechanism again, that's the column 7, lines 6 through 10. The fourth
`
`14
`
`reference, again, a column 7 reference, describes allowing the station
`
`15
`
`to transmit if the carrier sense mechanism locates an idle slot and the
`
`16
`
`backoff period for a queue has expired.
`
`17
`
`The fifth citation discusses the unique contention window
`
`18
`
`parameters for each queue, based on the priorities of the queue. The
`
`19
`
`sixth citation talks about a per queue backoff calculation that creates
`
`20
`
`the contention window for a particular priority queue. So, that's the
`
`21
`
`six-part algorithm that the Board identified.
`
`22
`
`If you notice, there's some overlap between the algorithm,
`
`23
`
`and in essence, what I derive from that algorithm is that first, you've
`
`24
`
`got to be using a carrier sense mechanism. You've got to have the
`
`25
`
`coordinated scheduling function, because that's what helps apply the
`
`
`
`
`
` 14
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`rules, and you have to have these contention window parameters
`
`involved in applying the rules.
`
`All of those materials are plainly disclosed in the Ayyagari
`
`references. Turning to slide 24, the Ayyagari reference, the '508
`
`patent, clearly discloses the carrier sense mechanism. It expressly
`
`talks about listening to the medium, being free for a predetermined
`
`interval. There's no dispute that it also discloses a CPU and a
`
`transceiver, and if you look at Dr. Roy's testimony, as well as the
`
`claim charts we submitted, there's also absolutely no dispute that the
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`Ayyagari references disclose a coordinated scheduling function, and
`
`11
`
`rules in the form of these contention window parameters. I showed
`
`12
`
`you those right at the beginning going through the background, and
`
`13
`
`that's true for the White Paper as well.
`
`14
`
`15
`
`JUDGE KIM: Counsel, what about figure 3?
`
`MR. STOCKWELL: Okay. So, figure 3, let's jump to
`
`16
`
`figure 3. This is on slide 27. So, this is the Patent Owner's
`
`17
`
`contention, that the boxes in figure 3, we have a CF schedule, and I'm
`
`18
`
`sorry, I can't read that, but there are multiple boxes labeled CF
`
`19
`
`schedule in figure 3, but they're all part of this media access control
`
`20
`
`part.
`
`21
`
`This is what the Patent Owner contends should be part of the
`
`22
`
`structure, and they say, well, there have to be multiple separate
`
`23
`
`schedulers, one for each queue, okay? First off, we don't believe that
`
`24
`
`was part of their contention early on, when they made their
`
`25
`
`preliminary response. It's not clear what the Patent Owner means by
`
`
`
`
`
` 15
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`this is part of the structure. In their preliminary response, at page 15,
`
`the structure they identified in terms of physical components was a
`
`computer and a transceiver.
`
`The schedulers are not the algorithm, at least these boxes
`
`don't seem to be the algorithm. To the extent the Patent Owner is
`
`contending that schedulers have to be hardware, we don't think that's
`
`supported. And, in fact, if you look at the quote to the right there, it
`
`says, "figure 3 is a block diagram depicting the software
`
`implementation of the media access control in accordance with an
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`illustrative embodiment."
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`JUDGE KIM: So, what are they, then?
`
`MR. STOCKWELL: Sorry?
`
`JUDGE KIM: What are the CF boxes?
`
`MR. STOCKWELL: What are the CF boxes?
`
`JUDGE KIM: Why are there multiple CF boxes?
`
`MR. STOCKWELL: Well, because CF boxes represent
`
`17
`
`logically that each queue will have associated with it a particular rule
`
`18
`
`in the form of a particular contention window parameter. It's a logical
`
`19
`
`diagram that lays out that for queue zero, let's suppose that's the
`
`20
`
`high-priority queue, it's going to have associated with it a rule that
`
`21
`
`says you're a high-priority queue, the rule being implemented via
`
`22
`
`these contention window parameters. But it's not a physical device
`
`23
`
`that schedules for that queue. It's simply a software logic diagram that
`
`24
`
`depicts each queue will have these rules associated with it.
`
`
`
`
`
` 16
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`JUDGE KALAN: Where is the association on the diagram
`
`between each queue and its respective CF schedule?
`
`MR. STOCKWELL: I don't think there is an association on
`
`the diagram in that. I think you have to read the description to
`
`understand the -- that the queues -- that the rules for the contention
`
`windows relate back to each of these boxes. There's not a whole lot of
`
`disclosure in the specification about this.
`
`In fact, most of the specification talks about if you look at
`
`that second quote, on the slide, most of the specification stresses that
`
`10
`
`what's really happening is there's a coordination function implemented
`
`11
`
`in a scheduler 52, okay? And it talks about there being these multiple
`
`12
`
`boxes, and it talks about being multiple queues, and it talks about
`
`13
`
`having particular contention window parameters based on priority and
`
`14
`
`access level.
`
`15
`
`Frankly, I can't recall a portion of the specification where it
`
`16
`
`links all of those three pieces together, so I infer that when you look at
`
`17
`
`the diagram and you read the description and you've got a scheduler,
`
`18
`
`you know, the overall scheduler, and you've got these logic boxes, the
`
`19
`
`logic boxes represent queue 1 with a particular rule and a particular
`
`20
`
`contention window, but I don't recall any place in the specification
`
`21
`
`where that's really clearly laid out. As I said, largely because it
`
`22
`
`focuses on the notion that there's this coordination function in a
`
`23
`
`scheduler.
`
`24
`
`JUDGE KIM: So, what you're saying is this diagram is sort
`
`25
`
`of a non sequitur and it needs to be determined by the function solely?
`
`
`
`
`
` 17
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`MR. STOCKWELL: Well, what I'm saying is the
`
`following: First off, let's go back to basic principles. We know we
`
`have a means for sensing and we've got all that language in the
`
`functional part of the clause, okay? And we know that the job of
`
`claim construction is to determine the portions of the specification that
`
`are linked to that means, and that are sufficient. You don't have to add
`
`everything in that the specification discloses in order to perform the
`
`function, but you just have to add enough structure sufficient to
`
`perform the recited function.
`
`10
`
`There is nothing in the specification that the Patent Owner
`
`11
`
`has identified that says in order to sufficiently perform the functional
`
`12
`
`recitation in the means for language, you've got to have multiple
`
`13
`
`schedulers. Instead, the Patent Owner simply says, oh, figure 3 shows
`
`14
`
`what they contend to be multiple schedulers and, therefore, the means
`
`15
`
`for clause requires this. There's no link back to why, you know, as the
`
`16
`
`specification says, this is an illustrative embodiment, there's not a lot
`
`17
`
`of description as to these boxes being actually -- there's no description
`
`18
`
`that they're physical hardware.
`
`19
`
`So, there's nowhere the patentee has identified anything that
`
`20
`
`says, well, you have to have this notion of having multiple boxes, one
`
`21
`
`for each queue, schedulers, as they would say, and they have to be
`
`22
`
`separate and independent in order to perform the function. We have
`
`23
`
`not seen them identify that.
`
`24
`
`JUDGE KIM: I have a problem with the sufficient
`
`25
`
`language, because, you know, that's difficult to define. I mean, I think
`
`
`
`
`
` 18
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`it would be easier if you're doing 112, 6, saying, okay, what is the
`
`thing that does this in the spec, and to me, it would merely jump out
`
`that. So, whatever it is, you get that, and you get the equivalent of
`
`that. I think when you put in the sufficiency language, you sort of --
`
`you're parsing hairs.
`
`MR. STOCKWELL: Well, I understand that may be the
`
`case, but there are Federal Circuit decisions, and I think the one that
`
`comes to mind for me is the Abbott case, where the Federal Circuit
`
`has said, you can't go too far in putting too much stuff from the
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`specification in in order to construe a "means for" language. And
`
`11
`
`that's my argument, and the way you draw the line is you look at
`
`12
`
`linking language in the specification and try to figure out what is
`
`13
`
`necessary to perform the function.
`
`14
`
`I would also say, even apart from the -- this claim
`
`15
`
`construction debate, what Dr. Roy explained, and what our position is,
`
`16
`
`is look, what actually does -- going back to your fundamental
`
`17
`
`question, what does this, you know, what is it that does this? The
`
`18
`
`boxes are not the important part. What's important is a coordination
`
`19
`
`function implemented in a scheduler, exactly as said in the
`
`20
`
`specification, and that scheduler manages rules for each of the queues.
`
`21
`
`So, you've got to have a coordinated scheduler, and you've got to have
`
`22
`
`rules applicable to each of the queues. And the question is, does the
`
`23
`
`Ayyagari reference have that, and the answer is absolutely. And
`
`24
`
`Dr. Roy tied up where that was in the Ayyagari reference.
`
`
`
`
`
` 19
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`JUDGE KIM: Just so you know, counsel, you have about
`
`five minutes.
`
`MR. STOCKWELL: Yes, thank you. I want to just finish
`
`up this argument with another reference to the Patent Owner's
`
`preliminary response. Ms. Gray, if you could pull that up. It's the
`
`page -- it bridges pages 18 and 19 of the response. It's PDF pages 25
`
`and 26.
`
`So, the top part of this is on one of the Patent Owner's slides,
`
`and this is the Patent Owner's argument as to what algorithmic
`
`10
`
`structure performs the second part of the claim function in the means
`
`11
`
`for, and I'm going to read the part that bridges pages 18 and 19. After
`
`12
`
`talking about the coordination scheduling function modules when
`
`13
`
`executed by the CPU, Patent Owner says, "Each output queue
`
`14
`
`competes for transmission opportunities (Tx-Opp;) using a
`
`15
`
`coordinated scheduling function. Each coordinated scheduling
`
`16
`
`function details a set of rules that differs for each traffic classification
`
`17
`
`and, hence, each queue. The set of rules in the scheduling function
`
`18
`
`further specifies a contention window CWmin from which a random
`
`19
`
`backoff is computed for each queue."
`
`20
`
`In other words, we believe the Patent Owner initially agreed
`
`21
`
`with us that what's critical for performing the algorithm is a
`
`22
`
`coordinated scheduling function, and for each queue, you have a set of
`
`23
`
`rules that differ per traffic classification that are implemented via this
`
`24
`
`CWmin parameter, all of which are disclosed in Ayyagari.
`
`
`
`
`
` 20
`
`
`
`Cases IPR2014-00501
`Patent 7,136,392
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`I want to talk briefly about the prior art issue, if