`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`RPX CORP., ERICSSON INC.,
`TELEFONAKTIEBOLAGET LM ERICSSON,
`Petitioners,
`
`v.
`
`IRIDESCENT NETWORKS, INC.,
`Patent Owner.
`____________
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`____________
`
`Record of Oral Hearing
`Held: September 24, 2018
`____________
`
`
`
`
`Before THOMAS L. GIANNETTI, MATTHEW R. CLEMENTS and
`SCOTT B. HOWARD, Administrative Patent Judges.
`
`
`
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`ADAM C. FOWLES, ESQUIRE
`J. ANDREW LOWES, ESQUIRE
`Haynes and Boone, LLP
`2505 N. Plano Road
`Suite 4000
`Richardson, TX 75082-4101
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`ROBERT R. BRUNELLI, ESQUIRE
`JASON H. VICK, ESQUIRE
`Sheridan Ross, PC
`1560 Broadway
`Suite 1200
`Denver, CO 80202-5141
`
`
`
`
`
`The above-entitled matter came on for hearing on Monday, September
`24, 2018, commencing at 3:05 p.m., at the U.S. Patent and Trademark
`Office, 600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`2
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`
`P R O C E E D I N G S
`- - - - -
`JUDGE GIANNETTI: Please be seated. So we're back from our
`break and this is the second case we will be hearing this afternoon, case IPR
`2017-01661. The panel is the same panel as we had for the previous case,
`myself, Judge Giannetti, Judge Clements and Judge Howard, and again
`Judge Howard will be presiding. Judge Howard.
`JUDGE HOWARD: Thank you. We'll skip the preliminaries and get
`right down to the argument. Mr. Lowes, will you be arguing again?
`MR. LOWES: No, Your Honor. My colleague Adam Fowles will
`argue.
`JUDGE HOWARD: Okay. And there's 45 minutes for this argument.
`How much time would you like to reserve?
`MR. FOWLES: Your Honor, I'd like to reserve ten minutes of time
`for rebuttal.
`
`JUDGE HOWARD: Okay. And I will let you know when you come
`within five minutes of that rebuttal time.
`
`MR. FOWLES: Thank you.
`
`JUDGE HOWARD: You may begin when you're ready.
`
`MR. FOWLES: Thank you, may it please the Board, and as already
`indicated my name is Adam Fowles. I'm representing Petitioners RPX and
`Ericsson, just for the record.
`Today we'll be focusing on the areas of dispute between the parties
`and as was indicated in the previous proceeding, because claim construction
`is shared between these two I'll be skipping that portion of my presentation
`today unless the Board has any additional questions on that point.
`
`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
`
`3
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`
`Otherwise, turning to slide 2. The other areas of discussion we plan
`on addressing today include first, that a POSITA would have been motivated
`to combine Lee and Fichou with Golden, second, that a Golden combination
`with Fichou and Lee renders obvious every limitation of independent claim
`1. Independent claim 1 is the independent claim for which all the limitations
`were argued in Patent Owner's reply briefing and the limitations are in
`common with those in independent claim 13.
`So turning to slide 3. This reproduces the independent claim 1
`language and specifically starting at line 62 there the language of “providing
`by the controller to the portal routing instructions for traffic corresponding to
`the connection so that the traffic is directed by the portal based only on the
`routing instructions” is one point that we'll be talking about today. The next
`is the "required route supported by the portal and dynamically provisioned
`by the controller.” And finally, about the teaching of control paths for the
`connection are supported only between each of the originating and
`terminating end points and the controller.
`Just to complete this overview, turning to slide 4 there were three
`instituted grounds for this IPR. The first ground is the combination of
`Golden, Fichou and Lee with respect to the independent claim and the other
`grounds rely on Golden, Fichou and Lee in combination as well with the
`addition of Har and Pillai for certain aspects of those additional claims.
`Turning now to slide 8. We'll just dive right into an overview of
`Golden and the other references. Golden describes ways to provide a
`guaranteed quality of service between any end stations providing on demand
`reserve band with connections, what it calls virtual circuit connections, that
`are set up when requested and when done they are released or torn down.
`
`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
`26
`
`4
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`
`JUDGE GIANNETTI: Pardon me, counsel. Which slide are you on
`
`now?
`
`MR. FOWLES: I'm still on slide 8.
`JUDGE GIANNETTI: Okay fine, I have it.
`MR. FOWLES: Okay. We have another copy if you need it as well.
`JUDGE GIANNETTI: I have it.
`MR. FOWLES: So, for example, Golden teaches that one of the types
`of services that it supports for which applications may request service is
`video conferencing which is one of the types of high quality service
`connections that the 119 patent itself indicates as an example. Golden
`teaches managing these requests with what it calls an enterprise control point
`and I've reproduced here figure 9 from Golden which illustrates this
`enterprise control point 50 which is a controller and as well in this figure we
`see on the left hand side what is called an upgraded host 102 that is in
`communication with this ECP50 and we also see the ECP50 in
`communication with switches 56 along a path, and finally we see here on the
`right hand side a host/router 94 and you'll notice that this host/router 94 does
`not have a signaling channel 58 to the ECP50.
`If we could turn to slide 9, however, we see that Golden teaches that
`this host 94 can be similarly upgraded as the host 102 on the left hand side,
`and what would that mean? That would mean that this host 102 on the right
`hand side when similarly upgraded would likewise include a daemon
`process 106 and a signaling interface 104.
`JUDGE GIANNETTI: So counsel, let me just stop you there. Is it
`correct that this figure that you've shown in slide 9 did not appear in the
`patent?
`
`5
`
`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
`26
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`
`MR. FOWLES: That's correct.
`JUDGE GIANNETTI: So this is your rendition of what it would take
`or what it would look like if you replaced that element; is that correct?
`MR. FOWLES: This is Dr. Reddy's rendition of what a POSITA
`would understand that rendition to look like.
`JUDGE GIANNETTI: Thank you.
`MR .FOWLES: Yes. Finally, Dr. Reddy testified that a POSITA
`would understand from this that there would be a signaling channel 58 from
`the right hand side post 102 when similarly upgraded to the ECP50. Golden
`also teaches the desirability of determining whether a connection should be
`admitted in the network but it does not provide details about how to do so.
`So if we turn to slide 10 Fichou provides details about how to admit
`connections within a network. Like Golden, Fichou describes a system that
`uses a centralized controller to provide a required quality of service for a
`requested connection from a source work station. So in Fichou's source
`work station that needs a reservation of a connection to a destination work
`station sends that request to a central controller and Fichou teaches that the
`server first authenticates those requests and then follows it up with what it
`calls a user rights verification where it checks a database that defines what
`types of requests each user is allowed to perform and then verifies based on
`parameters such as bandwidth, request of quality of service, and so forth.
`Now upon verification, Fichou teaches a distribution of what it calls a
`flow ID which can be, according to Fichou, an MPLS label. However
`Fichou and Golden do not provide details about that MPLS operation.
`So turning to slide and looking at Lee with an E -- slightly confusing
`from the prior proceeding, but this is a U.S. printed publication -- looking at
`
`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
`26
`
`6
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`Lee it teaches centralized controllers controlling an MPLS network. In Lee,
`the centralized controller calculates a label switch path and distributes the
`appropriate labels to those switches in that path with what it calls an LFIB, a
`label forwarding information base, and the information in that LFIB as Dr.
`Reddy testified that is the information used by MPLS switches when
`performing look-ups for forwarding packets received that include an MPLS
`label as opposed to using a routing table.
`So let's move to slide 12. Slide 12 just provides an overview of the
`architecture of Golden and how the hardware elements taught in Golden read
`on the recited elements of independent claim 1. So we see here on the left
`hand side an originating end point that is the host 102 that is upgraded. It is
`in communication by signaling channels 58, or control paths, to ECP50
`which is a controller and that controller is in turn also in communication via
`respective signaling channels to these switches in the path and the network,
`and we've identified here on the left hand side a portal, according to the
`claim, that receives packets from the originating end point and decides how
`to forward them based on the MPLS labels according to the combined
`teachings of Lee, and finally, on the right hand side we see the terminating
`end point which again is this similarly upgraded host 102 which used to be
`the host/router 94.
`So let's turn now to slide 13 and really the first area of discussion
`today, and that is that a POSITA would have been motivated to combine
`Fichou and Lee with Golden.
`Turning now to slide 14. The petition demonstrated at first that a
`POSITA would have been motivated to combine Fichou with Golden for
`multiple reasons based on the evidence of record, including Golden itself,
`
`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
`26
`
`7
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`Fichou's teachings and the declaration testimony of Dr. Reddy. For
`example, both Golden and Fichou teach the use of a centralized controller
`architecture to control the quality of service connections within a network.
`Further, Golden and Fichou both teach the desirability of controlling what
`types of connections to admit within the network.
`Golden, again as I noted, doesn't teach exactly how that admission is
`done. That's because Golden was an architectural disclosure. It was focused
`on many other aspects and left open to a POSITA how to fill in admitting
`connections within the network and so a POSITA would have been
`motivated to turn to Fichou because Fichou teaches such details, as Dr.
`Reddy testified and further Fichou teaches that its admission approach is
`advantageous by providing a way for the customer to manage the
`authorization for each user of an end point.
`JUDGE GIANNETTI: Why Fichou though? I mean there are
`probably lots of places you could look for a network architecture in details.
`Why Fichou of all of them? I hear your theory that you’ve got this broad
`architectural disclosure and fill in the detail if you turn to Fichou, but why
`specifically Fichou?
`MR. FOWLES: Well as noted Fichou teaches that it's advantageous
`to do this user rights verification so that each user is controlled and managed
`by the customer and that's relevant, for example, where there's a source work
`station and different people can sit down at that same work station and try
`and log into the network and request a service.
`JUDGE GIANNETTI: But why? Why would they want to do that? I
`mean you're giving me the end result, you're not giving me the rationale for
`why one would do it.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`8
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`
`MR. FOWLES: Why one would turn to Fichou?
`JUDGE GIANNETTI: Well, why would one make this combination,
`this particular combination?
`MR. FOWLES: Because if they want to know how to control
`admission to the network like Golden suggests they want to turn to
`something else that teaches how that's done and I believe you're right that
`there are many different references that could have been turned to by a
`POSITA.
`JUDGE GIANNETTI: Right.
`MR. FOWLES: However, Fichou is exemplary of how that is done
`and Fichou as a U.S. printed publication was readily accessible by a
`POSITA and they would have been able to find it.
`JUDGE GIANNETTI: All of that is interesting but it still doesn't
`answer my question. What is it about Fichou that suggests that you could
`use that particular source of details for this architecture? Why would you
`use it? Is there some suggestion (indiscernible) that this is the
`(indiscernible)? I mean why would one do this?
`MR. FOWLES: I think the suggestion in Golden that it's desirable to
`control admission of connections in the network is that suggestion to point
`one to Fichou because Fichou has details about how to control admission of
`requests in the network.
`JUDGE GIANNETTI: Thank you.
`MR. FOWLES: Looking here on the slide, Patent Owner's arguments
`about why the petition failed to establish focused on this concept that they
`were nothing more than conclusory statements provided in the petition.
`However, the quotes that were provided in the Patent Owner's response do
`
`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
`26
`
`9
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`nothing more than selecting portions from the petition and ignoring the
`surrounding evidence and rationale. For example, as we've already
`addressed, Golden itself teaches that it's desirable to determine whether to
`admit a connection and here from column 10, lines 9 through 12, provides
`the quoted language from Golden that ECP50 could communicate with a
`quality server within the network for further determination of whether to
`admit the connection.
`Looking at the next portion there, Fichou provides those details and
`there are advantages in doing so, and finally Golden provides the that
`statement a POSITA to apply common knowledge, now as something that
`Dr. Reddy testified to and testified that Fichou is representative of that
`common knowledge. All of these together provide evidence from references
`themselves and the understanding of the POSITA as declared to by Dr.
`Reddy.
`JUDGE HOWARD: What about Patent Owner's argument that the
`Petitioner hasn't supplied enough evidence of how the references should be
`combined? What evidence is there of the mechanics of how you would take
`these three different references and put them all together and why?
`MR. FOWLES: Yes, Your Honor. If we could turn to slide 16.
`There are a couple of reasons that this argument should fail and first because
`the petition explained that a POSITA would apply Fichou's verification
`teachings to Golden's enterprise control point which is an example controller
`and specifically Fichou taught that its teachings could be separated or
`incorporated into existing functionality and Dr. Reddy declared that that
`existing functionality, for example, is the ECP50 in Golden, and so that's
`one way in which how the teachings of Fichou would be incorporated into
`
`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
`26
`
`10
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`Golden. It's modifying Golden's ECP50 with this further verification
`teachings of Fichou.
`As another example, the petition explained that a POSITA would
`understand Golden to teach that its ECP50, that its control includes control
`over MPLS network elements generally. It was a known quantity and was
`known to exist and Lee, turning to Lee for a second, Lee provides details
`about how a centralized controller, again modifying Golden's ECP50, would
`determine the label switch paths and provide those labels to the switches
`determined in the path.
`These are all examples about how a POSITA would have modified
`Golden with the different teachings of Fichou and Lee, and beyond that it
`would sound like Patent Owners are also requiring bodily incorporation
`which is not the standard as I understand it. Rather, the petition and Dr.
`Reddy put forth evidence about how the combined teachings, about what the
`combined teachings would suggest to a person having ordinary skill in the
`art.
`
`JUDGE HOWARD: Thank you.
`MR. FOWLES: Yes. Turning back to slide 15, just to provide some
`examples about why a POSITA would have been motivated to combine Lee
`with Golden and Fichou in particular. Both Golden and Lee again teach the
`use of centralized controllers to control different aspects of their operations.
`They both teach operability with MPLS but again Golden doesn't teach
`details about how the labels will be determined and distributed, and Lee
`provides such details. As was already discussed today, MPLS was well
`known and Lee is but one example of how this could be done and
`specifically a POSITA would have been motivated to make this
`
`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
`26
`
`11
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`combination, as Dr. Reddy testified to, because as Lee describes it doing so
`at a centralized controller reduces the complexity and reduces the load for
`the switches and the network and all this evidence from Golden, Fichou and
`Lee, as well as Dr. Reddy, provides the evidence and the rationale about
`why a POSITA would have been motivated to make this combination.
`
`Really again Golden is inviting a POSITA to fill in the details
`and Lee provides those details, as well as provides the assurance just like
`you saw with Surdila that the routing would be done based on the labels
`according to the quality of service connection determined instead of the IP
`lookup. That's just by way of example. Dr. Reddy declared to this as well
`that a POSITA would have known that with MPLS an IP lookup would be
`replaced by an MPLS lookup in the LFIB.
`
`Turning now to slide 17. Unless there are any other questions
`on that point we move on to another area of discussion today and that is
`specifically that Golden in combination with Fichou and Lee renders
`obvious directing traffic based only on the routing instructions provided by
`the controller with no independent routing of the traffic.
`Slide 18 shows the relevant claim language for the based only
`language, and we move to slide 19 Golden teaches that the ECP50, and this
`is looking at what was laid out in the petition, that the ECP50 sends
`reservation requests to switches 56 and specifically the portal identified in
`figure 9 that we annotated and saw on slide 12. But staying here on this
`slide, a POSITA would have appreciated that the ECP in Golden performs
`centralized control functions relating to bandwidth reservation, and so then
`the petition turned to Lee which teaches specifically how to generate MPLS
`labels at the centralized controller as part of the LFIB information that is
`
`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
`26
`
`12
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`sent to the MPLS switches, and finally, as Dr. Reddy declared, the LFIB
`taught in Lee constitutes routing instructions as used in this claim.
`Let's turn to slide 20. So the limitation specifically recites that the
`traffic for the connection is directed by the portal based only on the routing
`instructions. Much has already been discussed in the earlier proceeding
`about the use of MPLS which replaces the regular IP routing lookup with the
`MPLS lookup in the LFIB.
`If we turn to slide 23, we'll address specifically the argument used in
`the Patent Owner's response on this point, and that first focused on Lee and
`its description of the LFIB as information and that because it's information it
`can't be instructions, and this ignores the substance of Lee and the
`understanding of a POSITA what MPLS teaches with the form of a word
`choice, and that's based on the declaration testimony of Dr. Reddy again
`both in the original declaration of Dr. Reddy and as put here Dr. Reddy's
`confirmation of that in his supplemental declaration.
`"The LFIB's inclusion of an input label and output label and output
`interface and so on means the LFIB has all the information the MPLS switch
`needs to now forward using the LFIB instead of a traditional lookup."
`Turning to slide 24. Interestingly, Dr. Sharony cited to in Exhibit
`2003 in his declaration. Exhibit 2003 is from Cisco Press entitled MPLS
`Fundamentals Forwarding Label Packets and there it actually confirms and
`is consistent with Dr. Reddy's declaration testimony and the position put
`forward in the petition that an IP lookup is replaced with a lookup of the
`label in the LFIB when it is MPLS operation and also here looking at this
`middle quote,
`
`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 IPR2017-01661
`Patent 8,036,119 B2
`
`
`"When a router receives an IP packet the lookup done is a IP lookup.
`But when a router receives a label packet the lookup is done in the LFIB of
`the router."
`So here we see again that all the evidence concentrates to the
`conclusion that when there is an MPLS label, the label is used to determine
`how to forward that packet out of the switch instead of a traditional IP
`lookup.
`JUDGE GIANNETTI: Just going back to the previous case where we
`had this discussion about the role of MPLS, I believe that -- and Mr. Vick --
`that there was a table lookup involved, even in the MPLS operation. Is that
`your understanding or is that some other process going on here? I know
`your argument focuses on the labels rather than on the lookup process, but
`Mr. Vick's argument suggested that there was a routing table involved in
`both cases.
`MR. FOWLES: For MPLS Routing I do not understand that to be the
`case, but that is not an issue that was raised in Patent Owner's briefing.
`JUDGE GIANNETTI: Right.
`MR FOWLES: We'd like to have our expert be able to look at that in
`more detail so that we could inform our decision based on that, but talking
`here today when the label is there the lookup done to determine how to
`cause that packet to leave the switch is done in the LFIB instead of a
`traditional routing table lookup.
`JUDGE GIANNETTI: Okay. And tell me, remind me what LFIB
`stands for.
`MR. FOWLES: Label forwarding information base.
`
`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 IPR2017-01661
`Patent 8,036,119 B2
`
`
`JUDGE GIANNETTI: Label forwarding information base. So this
`suggests there's still some sort of table lookup going on there, right?
`MR. FOWLES: Yes, Your Honor. It's in the LFIB.
`JUDGE GIANNETTI: But not a routing table.
`MR. FOWLES: That's right. The switch does not make any kind of
`independent decision on how to forward the packet because the centralized
`controller gave those labels that tell it how to forward that packet.
`JUDGE GIANNETTI: Okay.
`MR. FOWLES: Any additional detail is not something that the claim
`requires and the specification of the 119 patent does not explain how that
`forwarding is accomplished. It simply claims the resulting function of new
`independent routing, and Dr. Reddy testified that MPLS labels, because
`they're used, means that there is no independent routing because it doesn't
`look at what the traditional concept of routing is of that IP lookup.
`JUDGE GIANNETTI: So what is your rationale for using MPLS as
`opposed to IP routing? Is that something that's suggested by your principal
`reference here? What's your argument for why we should be looking in
`MPLS?
`MR. FOWLES: Well first Golden does teach the use of MPLS in
`some of its network elements and specifically teaches that its architecture is
`designed to be interoperable with multiple protocols and as we've discussed
`here today and both sides seem to be in agreement, MPLS was a very well
`known protocol even at the time of 2006.
`JUDGE GIANNETTI: And more efficient, faster? Is that its virtue?
`MR. FOWLES: Well it's definitely --
`
`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 IPR2017-01661
`Patent 8,036,119 B2
`
`
`JUDGE GIANNETTI: What are the virtues of MPLS over IP
`routing?
`MR. FOWLES: One virtue is that it forces packets to adhere to a
`specific flow and here in this evidence, as Lee teaches, that flow is
`determined by the centralized controller and that centralized controller when
`those teachings of Lee are incorporated into Golden's ECP50 means that that
`ECP50 sends those labels to the switch and two things, 1) that flow now
`must follow those labels so that the requested quality service is actually
`achieved and 2) it actually, as Lee teaches, provides the advantage of
`reducing the computational burden at the switch itself, and so that does make
`the switch better, more efficient and able to better achieve its enhanced
`functionality goal of sending things to the centralized controller so that
`something that has a picture of the entire network is able to decide how best
`to meet a request of quality of service.
`JUDGE GIANNETTI: So why hasn't MPLS just replaced IP
`addressing? Why isn't that the standard now?
`MR. FOWLES: MPLS still exists, but beyond that I couldn't answer
`that question. I think that it's because there are different needs for different
`customers and so it's really tailored to what the particular customer wants for
`their network and for their operations.
`JUDGE GIANNETTI: Thank you.
`MR. FOWLES: Yes. I'll note that actually MPLS is still very much
`discussed in the literature even today. Turning now to slide 26. There's
`another argument that Patent Owner raised in their response that Lee does
`not suggest let alone teach that the MPLS OAM, which stands for operation
`administration and maintenance function, could be implemented exclusively
`
`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
`26
`
`16
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`in the centralized control apparatus. This is apparently because, "Lee
`explicitly teaches that the switch can receive an LSP which stands for label
`switch path from the MPLS OAM function instead of from the centralized
`control apparatus."
`But what this argument fails to consider is that that description in Lee
`is of just one embodiment that addresses what happens in failure scenarios
`and in deposition Dr. Sharony acknowledged that in non-failure scenarios
`and normal operation just quote here, "In the teachings of Lee in a normal
`situation where there are no failures the central controller will calculate the
`LSPs."
`That's correct. That agrees with Dr. Reddy's testimony that what
`happens in regular operation is that Lee teaches a centralized controller,
`which is modifying Golden's teachings, determines this label switch path
`and sends the labels that are determined from that to the switches that are
`less effective.
`If there are no further questions on this point I'd like to move on to the
`next area of discussion. Let's move to slide 27 then. The next area of
`discussion touches on the fact that the petition established that Golden in
`combination with Fichou and Lee renders obvious a required route
`supported by the portal and dynamically provisioned by the controller.
`Let's turn to slide 29. This lays out how the petition established how
`the combination renders this limitation obvious and this slide focuses on the
`language about the dedicated bearer path and the required route. What's
`worth noting here is that Golden again teaches that it reserves this path from
`beginning to end as modified by Lee's teachings the MPLS labels. So that
`provides the required route that is at the portal that the portal supports, again,
`
`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
`26
`
`17
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`because it receives these labels from the centralized controller per Lee's
`teachings.
`Turning to the next slide, slide 30, and focusing specifically on the
`language a required route that is dynamically provisioned. Each switch in
`the path, according to Golden's teachings, receives reservation commands
`from the ECP, which is the enterprise control point again, and each switch
`including the portal dynamically provisions its route that is part of the end to
`end path. That is a quote from Dr. Reddy's initial declaration testimony and
`Golden throughout teaches that it really is establishing what it calls an on
`demand reserve bandwidth virtual circuit connections and finally it teaches
`that you can take down these connections once they're done or released, in
`Golden's terminology.
`This is an example of dynamically provisioned and why is that the
`case? Let's turn to slide 32. Let's look first at the comparison that Patent
`Owner has made of what Golden has in its pre-computed path list to the
`claimed required route that is dynamically provisioned. This is again
`comparing apples to oranges because Golden doesn't describe those pre-
`computed lists as being allocated yet. This is merely Golden describing the
`centralized controller becoming aware of what's in the network. The petition
`never compared the pre-computed list to the required route. Rather it looks
`at the virtual circuit connection that is set up in response to a request. This is
`a dynamic provisioning of a required route because it is making a link
`available with requested parameters.
`I say that in that way turning now to slide 33, actually 34, because at
`deposition when Dr. Sharony was asked what provisioning means, for
`example, dynamic provisioning, he answered that it is, "You basically sort of
`
`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
`26
`
`18
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`set up certain parameters or metric, certain parameters in order to support a
`given performance," and this actually agrees with Dr. Reddy's testimony that
`what a POSITA would understand with dynamic provisioning is that it
`would involve insuring enough of a desired parameter, for example,
`bandwidth, would be made available for the requested purpose. Dr. Sharony
`never testified that dynamic provisioning would preclude looking to pre-
`existing paths to then set up a connection. Just because those paths are
`known, does not mean they're already reserved for the end point that's
`requested. In fact, Golden goes through this whole process to demonstrate
`how to actually get one of those paths reserved for its desired quality of
`service connection.
`If there are no further questions, I'll turn to the next area of discussion.
`JUDGE HOWARD: Counsel, just to remind you you have about five
`minutes left before you get into your rebuttal time.
`MR. FOWLES: Thank you, Your Honor. Slide 35 is the last area of
`discussion for today so I'll wrap up with that and that is this question
`whether Golden teaches a control path from the controller to the terminating
`end point. I've already explained how Dr. Reddy testified how a POSITA
`would understand Golden's teaching that figure 9 is host/router 94.
`Let's turn to slide 37, please. Reproduced again here this figure that
`that host/router 94 on the right hand side would be "similarly upgraded" as
`host 102 on the left hand side and let's turn to slide 38. Dr. Sharony --
`JUGE HOWARD: Just going back to 37 for a second.
`MR. FOWLES: Oh, yes please.
`JUDGE HOWARD: Is it fair to say that your argument on this
`limitation is dependent on our agreeing with the modification that Dr. Reddy
`
`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
`26
`
`19
`
`
`
`Case IPR2017-01661
`Patent 8,036,119 B2
`
`made to figure 9? That if we didn't believe that supported making that
`modification, the limitation wouldn't be met?
`MR. FOWLES: The way that we set it up in the petition, yes, You
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
