`
` UNITED STATED PATENT AND TRADEMARK OFFICE
` BEFORE THE PATENT TRIAL AND APPEAL BOARD
`NOKIA SOLUTIONS & NETWORKS
`US LLC and NOKIA SOLUTIONS
`AND NETWORKS OY,
` Petitioners,
`vs.
`HUAWEI TECHNOLOGIES, CO, LTD.,
` Patent Owner.
`_________________________________/
`
` DEPOSITION OF DR. DAVID LYON
` San Francisco, California
` Tuesday, October 10, 2017
`
`Reported By:
`LINDA VACCAREZZA, RPR, CLR, CRP, CSR. NO. 10201
`JOB NO. 19828
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`1
`
`HUAWEI 2003
`NOKIA V. HUAWEI
`IPR2017-00660
`
`
`
`Page 2
`
`
`
`
` October 10, 2017
` 9:04 a.m.
`
`
` Deposition of DAVID LYON, held at 560
`Mission Street, Suite 2100, San Francisco,
`California, pursuant to Subpoena before Linda
`Vaccarezza, a Certified Shorthand Reporter of the
`State of California.
`
`
`
`
`
`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
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`2
`
`
`
`Page 3
`
`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
`
`A P P E A R A N C E S:
`FOR THE PATENT OWNER:
` FISH & RICHARDSON
` By: Brian Strand, Esq.
` Andrew Patrick, Esq.
` 500 Arguello Street, Suite 500
` Redwood City, California 94063
` Strand@fr.com
` patrick@fr.com
`
`FOR THE PETITIONERS:
` ALSTON & BIRD
` By: Ross Barton, Esq.
` Samuel Merritt, Esq.
` 101 South Tryon Street, Suite 4000
` Charlotte, North Carolina 28280
` ross.barton@alston.com
` sam.merritt@alston.com
`
`
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`3
`
`
`
`Page 4
`
` I N D E X
`WITNESS: PAGE
` DAVID LYON
`EXAMINATION BY:
` MR. STRAND 5
`
` E X H I B I T S
`Exhibit 1
` Declaration of David Lyon 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
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`4
`
`
`
`Page 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
`
` (Exhibit 1 was marked for identification.)
` DAVID LYON,
` Having been duly sworn, by the Certified
`Shorthand Reporter, was examined and testified as
`follows:
` EXAMINATION
`BY MR. STRAND:
` Q. Good morning, Dr. Lyon.
` A. Good morning.
` Q. Could you please state your full name for the
`record?
` A. David Lyon, L-Y-O-N.
` Q. Very good.
` MR. BARTON: Do you want to do introductions
`just for the record?
` MR. STRAND: Absolutely, yes. I'm Brian
`Strand with Fish & Richardson representing Huawei.
` MR. PATRICK: Andrew Patrick with Fish &
`Richardson representing Huawei.
` MR. BARTON: Ross Barton from Alston & Bird
`representing the petitioner NSN.
` MR. MERRIT: Sam Merrit, also with Alston
`Bird for petitioner.
`BY MR. STRAND:
` Q. All right. Dr. Lyon, you've been deposed
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`5
`
`
`
`Page 6
`
`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
`
`before?
` A. Sure. I have.
` Q. About how many times?
` A. Maybe 15 or 20.
` Q. Very good. So I won't waste a lot of
`everybody's time here on procedure. You know it
`probably better than I do. You understand you're
`under oath, right?
` A. Yes.
` Q. And is there any reason that you can't
`testify fully and truthfully today?
` A. There's no reason.
` Q. If you don't understand any of my questions
`at any time, just let me know. Okay?
` A. Of course.
` Q. All right. You submitted a declaration in
`this proceeding, didn't you?
` A. I did.
` Q. And we had the court reporter go ahead and
`mark Exhibit 1. If you could look over and verify
`that's a copy of your declaration?
` A. Appears to all be here, yes.
` Q. And your declaration concerns Huawei U.S.
`Patent Number 9,060,268, right?
` A. Yes.
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`6
`
`
`
`Page 7
`
`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
`
` Q. And it's okay with you if I just refer to
`that as the '268 patent?
` A. Yes.
` MR. BARTON: Just so the record's clear, this
`deposition is being taken in IPR 2017-00661.
` MR. STRAND: That's a lot.
` THE WITNESS: Excuse me for a second. I'm
`going to grab a bottle of water.
` (Pause in proceedings.)
`BY MR. STRAND:
` Q. All right. So Dr. Lyon, the first question I
`had was in your professional background. If you turn
`to Paragraph 9 of your declaration on Page 3. And you
`talk quite a bit about a communication system that you
`helped design. If you just tell me a little bit about
`that?
` A. Sure. So in this Paragraph 9, I'm referring
`to the period from 1982 to 1987. At that time, I was
`working for a company -- a division of, called
`Linkabit, which was owned by a public company called
`M/A-COM, which is M/A-COM. And the company, when I
`joined 1982, had a majority of its business in the
`defense industry, and wished to use its already
`developed technology and capture business in the
`commercial realm.
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`7
`
`
`
`Page 8
`
`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
`
` And when I joined in 1982, all my background
`was in commercial activity, and I was tasked by my
`direct boss, Andrew Viterbi, to take on a project
`where we created satellite-based communication system
`to replace what then was land line, mainly telephone
`line and T1 line communication, and that was the
`nature of the project.
` Q. Okay. So this was the mid '80s. There were
`other satellite communication networks in use at that
`time?
` A. Not of this type. Yes, there were other
`satellite communication networks in use, but not of
`this type.
` Q. Okay. And how was yours different?
` A. It was different because it was suitable for
`use by very large, highly distributed organizations
`that had most of their computing power in a central
`location. Most of the predecessor satellite
`applications were either for military, which tended to
`be point-to-point sort of technical links or
`strategic, but mainly, point-to-point, where the
`emphasis was on security and reliability, or they were
`broadcast type networks.
` There was -- we are all familiar with TV
`delivered by broadcast, but there were also early
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`8
`
`
`
`Page 9
`
`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
`
`examples of networks where data was being delivered by
`broadcast. But this was a two-way interactive network
`that could be used by very large, highly distributed
`commercial organizations.
` Q. So you would say it had a significantly
`different architecture than previous commercial
`satellite systems?
` MR. BARTON: Objection. Vague.
` A. Well, when we decided to take on the project,
`we had to create for ourselves our own architecture.
`There was no archetype or no example that I was aware
`of in the satellite realm that would have met the
`requirements that were placed on us.
` Q. Okay. So without any examples in the
`industry, you then sat down and developed your own
`from scratch; is that right?
` A. Well, we developed our own without their
`being a model from the satellite industry use. We
`were having to connect to equipment that was already
`communicating via land line. So we knew what the
`interface requirements were, we knew what the overall
`performance requirements were. I was quite familiar
`with various kinds of land line-based network
`architecture. So we could study all that. But we had
`to make our own decisions and our own innovations to
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`9
`
`
`
`Page 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
`
`move it to the satellite realm.
` Q. Was there any particular part of the system
`that you focused on?
` A. I was in charge of the whole development.
`And so my focus, if you will, was to make sure that
`everything from the very top level functional
`requirements all the way down to the blocks where the
`teams of engineers were implementing were consistent,
`and were meeting the overall goals of the project.
`And although I spent a fair amount of time managing
`and supervising, I also spent a fair amount of time
`and share of mind watching over the specifications
`that would allow me to monitor and assure myself that
`the quality of what was being created was suitable for
`what we needed.
` Q. And were you actually involved in developing
`some of the specifications themselves?
` A. Oh, yes.
` Q. Could you give me any examples?
` A. Oh, sure. If a customer comes, like our
`customers did, and said, "We deploy our teams anywhere
`in the United States, and within" -- this happened to
`be the very first customer, a company called
`Schlumberger. And when we deploy a team to a site of
`an exploration well, oil well, we need to be -- we
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`10
`
`
`
`Page 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
`
`need them to be communicating to us at speeds of at
`least one megabit per second within one hour of their
`arriving on site. So that's how the customer
`expressed one of the requirements.
` My job, as the lead, both to manage the
`project, but also to be the primary interface to the
`customer, was to interpret that English language, but
`very clear functional requirement, into first high
`level architecture for the network, and then to break
`down the high level architecture into all the sub
`pieces. If you like, I can start explaining how I do
`that, if you want.
` Q. I think that's fine for this answer. Thank
`you.
` A. Okay.
` Q. If you could turn to Paragraph 21 of your
`declaration.
` A. (Witness complies.)
` Q. You list a number of documents that you
`reviewed and considered in your analysis.
` Do you see that?
` A. Yes, I do.
` Q. And that includes Huawei '268 patent, right?
` A. Of course.
` Q. Did you fully review the '268 patent prior to
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`11
`
`
`
`Page 12
`
`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
`
`signing your declaration?
` A. Yes.
` Q. And when you read the '268 patent, did you
`understand its contents?
` A. Yes.
` Q. You don't recall anything that you didn't
`understand?
` A. I don't recall that I had any problems
`understanding how the words read, and I don't recall
`any particular confusion on my part.
` Q. If you hadn't understood something, would you
`have mentioned that your declaration?
` A. Well.
` MR. BARTON: Objection. Calls for
`hypothetical.
` THE WITNESS: You know, I just don't recall
`there being any point of confusion that rose to that
`level that I would mention it, so no.
`BY MR. STRAND:
` Q. Fair enough. And you didn't see anything in
`the '268 patent that you wouldn't have known how to
`implement?
` MR. BARTON: Object to the form. Vague.
` THE WITNESS: Wouldn't have known how to
`implement? So I didn't consider that question. What
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`12
`
`
`
`Page 13
`
`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
`
`I considered was the asserted claims and what the
`context of the specification was so I could understand
`the asserted claims. I didn't try to read every
`reference to a potential embodiment and try to imagine
`how I would build it.
`BY MR. STRAND:
` Q. If you could turn to Paragraph 42 of your
`declaration, Exhibit 1. And you say, "The '268 patent
`was the result of the inventor's work with 3G."
` Do you see that?
` A. Yes, I see that.
` Q. I don't actually see any support in your
`declaration for that statement. Did you include any?
` MR. BARTON: Object to form. Argumentative.
` THE WITNESS: My support is stated right
`below it. The face -- indeed, the face of the '268
`cites a number of the 3GPP documents. And then I go
`on to reference the amended complaint of Huawei. So
`that's my support.
`BY MR. STRAND:
` Q. Okay. And you go on to say that "Below is a
`summary of the 3GPP organization and the state of the
`relevant technology as of the priority date of the
`'268 patent."
` Did I read that right?
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`13
`
`
`
`Page 14
`
`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
`
` A. You did.
` Q. And when you say below, you're talking about
`the next section from Paragraph 43 in Paragraph 68?
` A. Yes.
` Q. So I would like to take a look at the section
`for a little bit. In Paragraphs 46 to 49, you discuss
`various generations of the 3GPP standard.
` Do you see that?
` A. Yes.
` Q. And you talk about GSM and GPRS systems, and
`you refer to that as a second generation or 2G system,
`right?
` A. Yes.
` Q. And then you talk about UMTS, and you refer
`to that as a 3G system, right?
` A. That's correct.
` Q. And then you talk about LTE, and you refer to
`that as a 4G system, correct?
` A. Indeed, yes.
` Q. And that's generally the terminology that you
`use throughout your declaration, right?
` A. Yes.
` Q. So then looking specifically at Paragraph 49,
`you're talking about 4G. You mention the data rate of
`the difference between 4G and earlier generation
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`14
`
`
`
`Page 15
`
`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
`
`networks, right?
` A. Yes.
` Q. And what are some of the other primary
`differences between LTE and the earlier generation
`networks, other than the data rate?
` A. Okay. Well, to be specific, we'll compare,
`let's say, 4G LTE to 3G UMTS. A convenient way to
`think about any of these networks is in terms of the
`so-called protocol stack. The protocol stack starts
`at the bottom with a so-called physical layer. So the
`physical layer of 4G LTE is completely or distinctly
`different than the physical layer of 3G.
` Not only does it permit higher, what I'll
`call raw data rates and also net data rates, but to a
`specialist in the field, they would recognize that one
`is based on so-called OFDM type modulation schemes and
`the other is based more on a CDMA scheme. So the PHY
`layer big, differences. As we go up, there's many
`differences in the Mac layer and so on.
` I think something that's very distinctive
`about LTE, and when we say LTE, we also include in 4G
`the nature of the core network architecture, which is
`SAE, system architecture evolved or evolution. And
`the big distinction there is that the standards were
`written, and they imply that the core network for 4G
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`15
`
`
`
`Page 16
`
`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
`
`will be run on using IP protocols. So Internet
`protocols, as opposed to what was going on before,
`where there were, if you will, custom -- some custom
`protocols that were used in prior implementations of
`the core network.
` So those, to me, stand out. So at the very
`top level and at the very lowest level, there were
`differences. And because of that, at each layer, you
`can find differences. But you have to sort of go
`through it in detail to point at each of the
`differences.
` Q. So focusing more on the core network side,
`what would you say are the other significant
`differences between 4G and 3G other than the use of IP
`protocols?
` MR. BARTON: Object to form.
` THE WITNESS: One of the other very obvious
`sort of high level or top level differences is
`illustrated by comparison of these diagrams I
`included, one right above Paragraph 52, which is
`labeled "4G architecture," and one just above
`Paragraph 51, which are labeled "3G architecture."
` And the distinction here is that in 4G, the
`3GPP decided to separate the path that the user plane
`traffic, whatever kind of traffic it is, data or --
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`16
`
`
`
`Page 17
`
`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
`
`well, it's IP. So the user plane task had its own
`pathway to the core network as opposed to the control
`and signaling traffic, which is designated by the
`dotted lines that are shown in 4G.
` In prior generations, a single node in the
`core network, the so-called SGSN, I'll just say it
`once and I'll just call it SGSN, serving GPRS node,
`the SGSN, actually performed both control as well as,
`if you will, user data functions. And in 4G, those
`functions were separated out so that the control
`portion of the SGSN became part of what was now the
`MME, the mobility management entity. And then the
`serving gateway, SGW, only dealt with user traffic.
` It also -- it reminds me that another modest
`change was the fact that in the radio access part of
`the network in 4G, one relies on so-called eNodeBs for
`base stations, as opposed to prior generations, where
`the base stations also were accompanied with another
`kind of node called network controller, radio network
`controller in 3G.
` Q. Are there any other significant differences
`between the 4G and 3G core networks that come to mind?
` A. There may be. But I mean, as I sit here now,
`you're asking for a summary on two networks that are
`characterized by tens of thousands of pages of
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`17
`
`
`
`Page 18
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`specifications. That's my first cut. That's my best
`first cut.
` Q. So circling back the use of IP protocols.
`Why was that relevant to the development of 4G?
` MR. BARTON: Object to form.
` THE WITNESS: Well, it's relevant just
`because it allowed 3GPP to, if you will, tag onto all
`the development that was going on in the world of the
`Internet protocols. So that it was not incumbent upon
`the 3GPP to continue inventing new features within
`their own protocol. They could use what was being
`developed in this, if you will, parallel set of
`organizations.
` That's not to say that the 3GPP gave up all
`of its proprietary or its distinctive protocols;
`others continue to survive. But by defining the
`working of the 4G network as IP, it instantly enriched
`a few other capabilities and the possibilities for
`future development.
`BY MR. STRAND:
` Q. So 3GPP was an off-the-shelf transport
`protocol, and they were able to focus more on higher
`level of signaling protocols. Is that fair?
` MR. BARTON: Object to form.
` THE WITNESS: Well, that's -- I wouldn't put
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`18
`
`
`
`Page 19
`
`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
`
`it that way. I would say that they used and invoked
`IP protocols where they thought it was appropriate,
`and they kept their -- call it their distinctive
`non-IP protocols where they thought that was
`appropriate. You would have to go through case-by-
`case to see how they interacted.
`BY MR. STRAND:
` Q. Was there also a benefit of being able to use
`the grounding capabilities IPA as opposed to
`hierarchical set of point-to-point connections?
` MR. BARTON: Objection. Vague.
` THE WITNESS: So you asked about routing
`capabilities and point0to-point. I'm not quite sure
`what you're getting at. Maybe if you could sort of
`help me with a little bit better understanding of what
`your question is implying.
`BY MR. STRAND:
` Q. In earlier generation of core networks, how
`were they -- network nodes connected to each other?
` MR. BARTON: Objection. Vague.
` THE WITNESS: So the -- depending on the kind
`of traffic and depending on the kind of node, the
`connections that could be made could be based on
`protocols had already been established and existing in
`prior generations. So-called GTP protocols, GPRS
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`19
`
`
`
`Page 20
`
`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
`
`tunneling protocols.
` But beyond that, in actually building a
`network to make a network that looks as simple as the
`drawing above Paragraph 52, it's not as though a
`cellular operator has four boxes and he has to connect
`four boxes. So each of these boxes might be
`implemented with dozens of similar functionality boxes
`that are either spread out or close together.
` So there's other kinds of connectivity other
`than functional element to functional element
`connectivity. That connectivity would benefit from or
`could benefit from some of the routing protocols that
`exist in the world of Internet protocol.
`BY MR. STRAND:
` Q. So just looking at a concrete example, you
`talk about the -- one of the differences between the
`functionality of the SGSN and the 3G and the MME and
`4G being the fact that in 3G, the SGSN routes' use of
`plane traffic, correct? Carries use of plane traffic,
`sorry.
` MR. BARTON: Object to the form.
` THE WITNESS: So your question -- or you're
`asking me to confirm that SGSN carries user plane
`traffic in a 3G network?
`BY MR. STRAND:
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`20
`
`
`
`Page 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
`
` Q. Right.
` A. Yes, it does.
` Q. And then in 4G, the MME does not carry user
`plane traffic. User plan traffic goes directly from,
`I think you said the SGW to the eNodeB, right?
` A. In 4G networks, the MME does not handle user
`plane traffic. Does not directly transfer user plane
`traffic.
` Q. So in the 3G network, why doesn't the user
`plane traffic just go from the GGSN to the RNC? Why
`did it have to go through the SGSN?
` MR. BARTON: Objection. Compound. Vague.
`Calls for speculation.
` THE WITNESS: So the GGSN and the SGSN in 3G
`each have distinctive roles to play. Because they
`have distinctive roles to play, it gives the
`operator -- and the manufacturers, but in the end,
`it's the operator's decision -- more flexibility about
`how many devices of each type to purchase. How many
`to deploy, where to deploy them geographically.
` It, I would say from the point of view of
`someone who has been in the field for a while, that
`it's my guess, my belief that the objectives had to do
`with making the implementation of the networks as
`efficient as possible in terms of the requirement for
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`21
`
`
`
`Page 22
`
`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
`
`hardware and keeping the proximity of hardware to the
`various other nodes is to say as simple and as tight
`as possible.
` Q. The SGSN carrying user plane traffic caused
`scalability problems, right?
` MR. BARTON: Object to the form.
` THE WITNESS: I don't know that it caused
`scalability problems. There's quite very large
`networks got built on SGSNs. There's -- to me, it's
`obvious that when you put two different kinds of
`functions inside of one node, there's a certain level
`of, if you will, flexibility that may not be
`available.
` So if you need to scale up the number of SGNs
`because there's a lot of user plane traffic, and
`that's the reason that you have to put more SGSNs out,
`but there was really no need because the control
`traffic was not stretching the limits, well, you could
`argue that was not as an efficient way to go and vice
`versa. So my speculation, and that's all it is, is
`that the objective was to give more degrees of freedom
`in the way that the functions could be deployed.
` Q. And when you say the objective, you're
`talking about the objective of the MME not carrying
`user plane traffic, right?
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`22
`
`
`
`Page 23
`
`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
`
` MR. BARTON: Object to the form.
` THE WITNESS: I meant, the objective in very
`general sense. That when the groups that come
`together in the task groups and working groups that
`represent both carriers and manufacturers, when they
`come together, the notion that control plane and user
`plane would be separated in 4G, it's my guess was
`being driven by this desire for higher degree of
`flexibility in deployment scenarios.
`BY MR. STRAND:
` Q. Fair enough. And so the act of separating
`control and user plane traffic in the LTE and core
`network, there's a fairly significant architectural
`change from 3G, right?
` MR. BARTON: Object to the form.
` THE WITNESS: Not as fundamental as you
`think, because -- or let me say it this way. The
`notion of so-called one tunnel operation, where the
`data in a 3G network could be transferred from the
`radio access part of the network to the GGSN was
`already pretty well understood. I am trying to
`remember the first documents that talked about one
`tunnel or direct tunnel were, I think late '90s.
`Might have been early 2000s. So that was already
`known.
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`23
`
`
`
`Page 24
`
`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
`
` And in fact, that was an option for carriers
`who wanted to effectively allow the SGSN to more
`completely deal with control matters and have very
`little interaction with the user plane data. So that
`was already well known. 4G, if you will, just took it
`farther, and completely defined two separate nodes for
`the two functions.
` Q. So if I'm understanding you right, 3G already
`had the capability to separate control and user plane
`traffic from the SGSN?
` A. They had -- within 3G, within the UMTS
`family, there were standards that read on what I think
`in the standard called one tunnel operation, as
`sometimes referred to as direct tunnel operation. To
`be clear, the user data would still traverse through
`the SGSN, but it would not be operated upon. The SGSN
`was merely shuffling it from an input to an output.
`That was the notion of direct tunnel.
` So it achieved one of the goals, which is it
`meant that SGSNs would be spending eventually all of
`their processing power on control matters, which meant
`that the operator, the carrier only had to reproduce
`SGSNs when the control traffic got so great that they
`needed another node. It helped with the flexibility
`issue that we discussed earlier.
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`24
`
`
`
`Page 25
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` Q. So it reduced the processing burden on the
`SGSN, but the SGSN still had to have enough physical
`bandwidth capacity to handle all the traffic for all
`connected users, right?
` A. Well, just to be clear, it did not require
`any -- the objective of direct tunnel was to make sure
`that there was virtually -- I mean, practically no
`processing power of the SGSN devoted to doing anything
`with the user traffic other than moving it from one
`interface to another. So yes, it passed the data, but
`it was not processing it, the user data.
` Q. And so the act of separating the control and
`the user plane in the LTE would have resulted in a
`different set of process flows, if you will, than in
`3G, right?
` MR. BARTON: Objection. Vague.
` THE WITNESS: So I don't know about process
`flows, but when you have -- when you decide to relabel
`elements and rearrange their physical connection, of
`course, that's going to give way to all kind of
`changes in the precise detail specifications that you
`write at the different layers.
`BY MR. STRAND:
` Q. It affects how the nodes would communicate
`with each other to get something done, right?
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`25
`
`
`
`Page 26
`
`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
`
` MR. BARTON: Objection. Vague.
` THE WITNESS: That would be one aspect, yes.
`BY MR. STRAND:
` Q. And so you couldn't just take a signaling
`flow from the 3G network and transplant it to 4G
`network and expect it to do the same thing because of
`those architectural changes, right?
` MR. BARTON: Objection. Vague. Calls for
`speculation.
` THE WITNESS: Well, on the one hand, taking a
`precise -- let's call it a precise process that's
`defined for 3G set of nodes, and then applying that to
`a 4G network without at least recognizing changes in
`nomenclature, and potentially changes in physical
`interconnect, you couldn't operate these devices
`without making at least those changes.
` The key here, though, and where I believe I
`think this is relevant to answering your question, is
`that there's always been an obligation felt by the
`drafters of the new specifications to maintain as much
`compatibility backward and forward with the prior
`general relations. So it is typical in every
`transition, but in particular, with 4G from 3G, to
`maintain as much common use of techniques as possible.
` So although at the most detailed level, you
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`26
`
`
`
`Page 27
`couldn't simply tell someone to take, for example, the
`same software out of a device that had only been
`designed and tested for 3G and have it run a device
`that was going to be used not just for 3G, but for 4G,
`that's true. But there would be a lot of commonality
`to what was being transferred back and forth.
` Q. Can you give me an example of how LTE is
`backwardly compatible to 3G?
` A. Uh-huh. So, for example, in thinking about
`the use of keys for encryption algorithms and for
`purposes of authentication and authorization, it was a
`key objective as expressed in the working documents
`that there be as much commonality and compatibility
`between 3G and 4G as was possible, and that's
`certainly what resulted.
` Q. And I just want to make sure I'm
`understanding what you mean by backward compatibility.
`Are you meaning I can take a 3G phone and use it on an
`LTE network the same way that I would use on a 3G
`network?
` MR. BARTON: Objection. Vague.
` THE WITNESS: Backward compatibility, as I'm
`using it, would mean that you could take 4G devices
`and use them very simply in 3G networks, so you that
`you don't design a 4G functionality that would be
`
`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
`
`TransPerfect Legal Solutions
`212-400-8845 depo@transperfect.com
`
`27
`
`
`
`Page 28
`
`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
`
`inoperable in a 3G environment. That's in general
`what I'm referring to.
` And in the case here, that would be true,
`that, for example, you could operate 4G devices in the
`3G network without any trouble. But more than that, I
`was saying that 3G devices -- or, sorry, concepts from
`3G were carried forward to 4G to make the forward and
`backward compatibility that much easier and simpler.
`BY MR. STRAND:
` Q. The example you just gave was an example of
`backward compatibility on the handset side, right?
` A. No, no. The keys affect both sides of the
`network.
` Q. So again, I guess I'm trying to understand.
`Are you saying that I could take a 3G phone and use it
`on an LTE network?
` MR. BARTON: Objection. Mischaracterizes the
`witness's testimony. Asked and answered. Vague.
` THE WITNESS: So if I have a 3G phone, and I
`step out on to the street here, and for the sake of
`argument, let's say a 3G phone operates under the
`T-Mobile network. The T-