JOSEPH A. PARADISO, PH.D.
`
`1
`
`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
`
` UNITED STATES PATENT AND TRADEMARK OFFICE
` BEFORE THE PATENT TRIAL AND APPEAL BOARD
`C.A. No. IPR2018-00289
`U.S. Patent 8,872,646
`* * * * * * * * * * * * * *
`APPLE INC.,
`*
`Petitioner,
`*
`v. *
`*
`UNILOC LUXEMBOURG S.A., *
`Patent Owner. *
`* * * * * * * * * * * * * *
` VIDEO-STREAMED TELECONFERENCED
`DEPOSITION OF JOSEPH A. PARADISO, PH.D., a
`witness called on behalf of the Patent
`Owner, pursuant to the United States Patent
`and Trademark Office Patent Trial and Appeal
`Board Rules of Procedure, before Jessica L.
`Williamson, Registered Merit Reporter,
`Certified Realtime Reporter, and Notary
`Public in and for the Commonwealth of
`Massachusetts, at the Offices of Hogan
`Lovells US LLP, 100 High Street, Boston,
`Massachusetts, on Thursday, July 19, 2018,
`commencing at 9:11 a.m.
`
`Complete Legal
`
`214-746-5400
`Apple v. Uniloc, IPR2017-00289
`Uniloc's Exhibit 2002
`
`
`
`1
`
`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
`
` UNITED STATES PATENT AND TRADEMARK OFFICE
` BEFORE THE PATENT TRIAL AND APPEAL BOARD
`C.A. No. IPR2018-00289
`U.S. Patent 8,872,646
`* * * * * * * * * * * * * *
`APPLE INC.,
`*
`Petitioner,
`*
`v. *
`*
`UNILOC LUXEMBOURG S.A., *
`Patent Owner. *
`* * * * * * * * * * * * * *
` VIDEO-STREAMED TELECONFERENCED
`DEPOSITION OF JOSEPH A. PARADISO, PH.D., a
`witness called on behalf of the Patent
`Owner, pursuant to the United States Patent
`and Trademark Office Patent Trial and Appeal
`Board Rules of Procedure, before Jessica L.
`Williamson, Registered Merit Reporter,
`Certified Realtime Reporter, and Notary
`Public in and for the Commonwealth of
`Massachusetts, at the Offices of Hogan
`Lovells US LLP, 100 High Street, Boston,
`Massachusetts, on Thursday, July 19, 2018,
`commencing at 9:11 a.m.
`
`Complete Legal
`
`214-746-5400
`Apple v. Uniloc, IPR2017-00289
`Uniloc's Exhibit 2002
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`2
`
`A P P E A R A N C E S
`
`BY: JAMIE H. MCDOLE, ESQ.
` Haynes and Boone, LLP
` 2323 Victory Avenue
` Suite 700
` Dallas, Texas 75219
` (214) 651-5121
` jamie.mcdole@haynesboone.com
` -and-
`BY: THOMAS KELTON, ESQ.
` Haynes and Boone, LLP
` 2505 N. Plano Road
` Suite 4000
` Richardson, Texas 75082-4101
` (972) 739-6923
` thomas.kelton@hanesboone.com
` Counsel for the Petitioner and the
` Deponent
`
`1
`
`23
`
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`3
`
`A P P E A R A N C E S, Continued
`
`BY: JEFFREY HUANG, ESQ.
` Etheridge Law Group
` 2600 East Southlake Boulevard
` Suite 120-324
` Southlake, Texas 76092
` (408) 797-9059
` jeff@etheridgelaw.com
` Counsel for the Patent Owner
` (Present via teleconference.)
`
`ALSO PRESENT:
`
` Anthony Piccirilli, Videographer
`
`1
`
`23
`
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`4
`
` I N D E X
`DEPONENT PAGE
`JOSEPH A. PARADISO, PH.D.
`Examination by Mr. Huang 6
`
` E X H I B I T S
`
` (None.)
`
`1
`2
`3
`4
`
`56
`
`78
`
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` P R O C E E D I N G S
` THE VIDEOGRAPHER: Good morning.
` We're now on the record. Please note that
` the microphones are sensitive and may pick
` up whispering and private conversations.
` Please turn off all cell phones or place
` them away from the microphones as they can
` interfere with the deposition audio.
` Recording will continue until all parties
` agree to go off the record.
` My name is Anthony Piccirilli
` representing Complete Legal. The date today
` is July 19th, 2018, and the time is
` approximately 9:11 a.m. This deposition is
` being held at Hogan & Lovells located at 100
` High Street in Boston, Massachusetts. The
` caption of this case is Apple Incorporated
` versus Uniloc Luxembourg S.A. The name of
` the witness is Joseph A. Paradiso. At this
` time the attorneys present in the room and
` everyone attending remotely will identify
` themselves and the parties they represent,
` after which our court reporter, Jessica
` Williamson, representing Complete Legal will
` swear in the witness, and we can proceed.
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` MR. MCDOLE: Jeffrey, do you want
` to go first?
` MR. HUANG: Sure. This is Jeffrey
` Huang with Ethridge Law Group on behalf of
` the patent owner.
` MR. MCDOLE: Jamie McDole from
` Hanes and Boone on behalf of Apple and the
` witness.
` MR. KELTON: Thomas Kelton, Hanes
` and Boone, on behalf of Apple and the
` witness.
`
` JOSEPH A. PARADISO, PH.D.,
` a witness called on behalf of the Patent
` Owner, having satisfactorily been identified
` by the production of a driver's license, and
` being first duly sworn, was deposed and
` testified as follows:
`
` DIRECT EXAMINATION
`
` BY MR. HUANG:
`Q. Okay. Hi. Good morning. Let's start with
` this: Is there any reason you can't give us
` your full and truthful testimony this
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` morning?
`A. No.
`Q. And if at any time point during this
` deposition you don't understand a question,
` would you please let me know?
`A. Definitely.
`Q. And can we agree that if you do provide an
` answer, that you have understood the
` question?
`A. Yes.
`Q. Okay. Great. Did you submit a declaration
` in this IPR, which is IPR2018-00289?
`A. Let me check.
` (Witness reviews documents.)
`A. Can you repeat the number?
`Q. Sure. It's IPR2018-00289.
`A. That number does not appear in my
` declaration, I don't believe, but I did
` submit a declaration, yes.
`Q. Okay. So you have your declaration in front
` of you --
`A. Yes, sir, I do.
`Q. Does that appear to be your declaration you
` submitted in the IPR we're going to be
` talking about today?
`
`Complete Legal
`
`214-746-5400
`
`
`
`8
`
`JOSEPH A. PARADISO, PH.D.
`
`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. Yes, it does.
`Q. And can you please turn to the last page of
` that declaration, page 86?
`A. Yes.
`Q. And on the last page you see a signature?
` Is that your signature?
`A. I'm getting there in a second.
`Q. Okay. Sorry.
`A. This is my signature, yes.
`Q. Great. Did you review your declaration
` before signing it?
`A. Yes, I did.
`Q. And since you signed your declaration, were
` there any errors or omissions that you have
` identified in this declaration?
`A. There are no errors that I found on
` reviewing it. This appeared to be fine.
`Q. Great. And does your declaration contain
` all of your opinions for this IPR?
` MR. MCDOLE: Objection, form.
`A. This declaration reflects the opinions that
` I make with a prior art that I had found and
` put together on patent -- the '646 patent.
`Q. Okay. And so you do not have any other --
` or any other opinions not specifically in
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` your declaration regarding this IPR; is that
` correct?
`A. I may develop opinions in the future. These
` are the opinions I had at the time I wrote
` the declaration.
`Q. And as you sit here today, do you have any
` further opinions in this IPR other than
` that -- what's in your deposition (sic)?
`A. The declaration reflects my opinions.
` Again, the more I read, the more opinions I
` get.
`Q. Okay. Let's start with page 40 of your
` declaration, paragraph 81.
`A. Sure.
`Q. Please let me know when you're there and you
` have read the paragraph.
`A. I am there.
`Q. Okay. Do you see the paragraph, part of the
` first sentence, it says that "McMahan is
` directed to 'a method of enhancing the
` accuracy of a sensor,'" and "As taught by
` McMahan, this is done by 'determining
` whether the measure [of the sensor] falls
` outside of an acceptable range for the
` output of the sensor, and, when the measure
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` falls outside the acceptable range,
` modifying the measure of the output such
` that the measure falls within the acceptable
` range for the sensor.'"
` Do you see that?
`A. Yes, I see that.
`Q. What value specifically falls within the,
` quote-unquote, acceptable range?
`A. Can you repeat the question?
`Q. Yes. What value specifically falls within
` the, quote-unquote, acceptable range.
`A. Let me review what I wrote.
`Q. Sure.
` (Witness reviews document.)
`A. The accessible range is a signal within the
` normal range of the output of the sensor,
` quoting from McMahan, and I believe that.
`Q. And what would the specific values be for
` the normal range of that sensor?
`A. I cannot give you specific values because it
` depends very much on the sensor. You see
` how the signals are produced from the way
` the sensor's used. You establish a normal
` range or you look at the specifications of
` the sensor, and/or, and then you establish a
`
`Complete Legal
`
`214-746-5400
`
`
`
`11
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` normal range. There are many ways --
`Q. How --
`A. -- to do this.
`Q. Sorry. Go ahead.
`A. There are many ways to do that.
`Q. Okay. What -- for example, what are some of
` the many ways to establish the acceptable
` range for a sensor?
`A. You run tests. You see what their output
` is, and you set your range based on what you
` determine to be a normal output.
`Q. Is there any other way?
`A. That's the main way.
`Q. And when you say you run tests, what sort of
` tests would you run?
`A. You would use it in the context of how the
` device is used.
`Q. And in that case, the values -- the values
` derived in the context that the device is
` meant to be used, those would be all values
` within the acceptable range?
`A. You would determine by the normal operation
` of the device and the response of the sensor
` what the acceptable range is. Anything
` outside -- you would define your bounds of
`
`Complete Legal
`
`214-746-5400
`
`
`
`12
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` what is not acceptable based on that, on the
` response of the sensor, what the sensor
` would normally do in the context also of how
` the sensor would be used.
`Q. And so in this case, I believe in your
` declaration you're applying this -- the
` teachings of McMahan to a proposed
` Pasolini/Goldman combination; is that
` correct?
`A. Yes.
`Q. And so in the proposed Pasolini/Goldman
` combination, what specifically would be the
` acceptable range for that proposed device?
`A. Oh, I cannot tell you because it depends on
` what the sensor is, what the values are,
` what the units are, how it actually is being
` used. There many factors there. To give a
` specific number, I don't think that's
` relevant here.
`Q. Why don't you think it's relevant?
`A. Because it's ambiguous.
`Q. What is ambiguous?
`A. Asking me to give you a specific range.
`Q. I'm sorry. Do you think the question is
` ambiguous, or is something else ambiguous?
`
`Complete Legal
`
`214-746-5400
`
`
`
`13
`
`JOSEPH A. PARADISO, PH.D.
`
`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. What you're asking me does not have a clear
` answer.
`Q. Would it ever have a clear answer?
`A. It would have a clear answer if I had a
` device on the table, I was looking at the
` output, and I knew the units that this
` criteria would be applied to and I saw how
` it was used and I've reviewed the specs of
` the particular sensor. That's how we do --
` we do this all the time, by the way. It's a
` very common operation with the sensors and
` accelerometers. We were doing that in the
` early 2000s.
`Q. Okay. So for -- first of all, for the
` proposed Pasolini/Goldman combination, what
` type of sensor would be present in that
` combination?
`A. Oh, for the proposed combination it could be
` any accelerometer because Goldman can
` accommodate in principle. At least the
` ideas of Goldman can accommodate any
` accelerometer.
`Q. And so generally, then, what would be a
` typical acceptable range for an
` accelerometer such as the one in the
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` proposed Pasolini/Goldman combination?
`A. Oh, this depends very much on what you do
` with it.
`Q. Okay. What are the typical things you would
` do with such a device?
`A. Well, if you look at the patent
` specification, I guess it's if you're
` carrying a device or moving a device or
` jostling a device or you have it on the
` table or it's not moving. There's a range
` of things.
`Q. Okay. So I think let's start with you said
` it was -- you started with moving a device,
` jostling a device. Let's just start with
` moving a device. What would be a typical
` range of -- what would be generally an
` acceptable range when you're moving a
` device --
`A. If you're moving -- if you look at it in
` terms of G forces, if you're moving a
` device, you're in the realm of a G, maybe a
` couple of Gs. It depends. It depends on
` how you move it. If you have it on your
` shoe, you get impact forces that are quite a
` bit higher at heel strike. If you have it
`
`Complete Legal
`
`214-746-5400
`
`
`
`15
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` in your pocket, it will be a little bit
` less, but you will see as you're walking,
` you know, some variation of acceleration on
` the order of some Gs.
`Q. Okay. And so does that range generally from
` the lowest to the highest --
`A. Oh, it depends on --
` MR. MCDOLE: Objection, form.
` (Reporter interruption.)
`A. Can you repeat your question? That would
` probably be the best way.
`Q. Yeah. So the acceptable range is a range of
` values, I would imagine. So for the
` activity we're talking about, which is
` moving, for that activity, what would be
` generally the high and low values in the
` acceptable range?
`A. It depends on whether you look at absolute
` value or whether you look at a bipolar
` value. If -- you would basically see one G
` if it's at rest depending on the
` orientation. If you look at the dynamic
` acceleration, you can see changes that go up
` to a G or maybe a few Gs, on that order. If
` it's mounted on the shoe, you can see peaks
`
`Complete Legal
`
`214-746-5400
`
`
`
`16
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` that are much higher with the heel strike.
` It depends on where this device is located
` on the body and again how it's being used.
`Q. Okay. Great. So when you say given a
` value, which is a G or a few Gs or
` something, is that the upper value in the
` range?
`A. I would want to look at the specific data
` for this device in the context of use. If
` it was on something like a shoe, it would be
` higher. If it was on -- in the pocket,
` maybe it would be lower, but I would want to
` see this thing actually used and look at the
` data stream before I set this threshold.
`Q. I understand. But in your examples, when
` you did give an example of a G or several
` Gs, was that value what you intend to be the
` upper range of the acceptable range?
`A. It varies. It could be that. It could be.
`Q. Okay. And then what would be the lower
` range of the acceptable range?
`A. If it's a bipolar signal, it would be
` negative that, because it can go both ways.
` If it's an absolute value, it's only one
` value. If it's at rest, and if you subtract
`
`Complete Legal
`
`214-746-5400
`
`
`
`17
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` out gravity, it's zero. There are many
` answers to that question.
`Q. Okay. Great. And so going back to what you
` said before about testing to determine the
` acceptable range, I think you had mentioned
` you look at the actual output of the sensor
` data and the activity you are doing to
` determine what the acceptable range was; is
` that right? Or can you correct me if I'm
` wrong about that?
`A. You would do that. You would have to do it,
` or you would have an algorithm do it as
` well, but you would have to have a clear
` idea of what the acceptable range is.
`Q. And how do you get the clear idea of what
` the acceptable range is?
`A. By seeing. As I've said already -- I've
` answered the question -- by seeing the
` response of the device under normal use.
`Q. And so, I'm sorry, just for, again, my
` clarity, so the response of the device in
` normal use, that is -- those are values
` within the acceptable range; is that
` correct?
`A. Yes. Yes, those are values.
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
`Q. Okay.
`A. If you see the device in normal use, those
` are within the acceptable range, except in
` the case -- even within normal use you can
` have a circuit malfunctioning. We've seen
` it. We've built these for years. And we
` can have these glitches that appear. So
` what you do is that you look at the signal.
` You know, you can have an algorithm look at
` it, typically we look at it, and we see
` where there may be spikes appearing in the
` signal. This is an example, an example of
` something out of the range. And we can
` identify the magnitude of the spike, the
` unphysical spike, which doesn't correspond
` to normal activity, versus the typical
` motion. And we set that threshold high
` enough so we don't exclude normal motion.
` We will exclude the spike. And this is
` common practice -- was common practice well
` before the date of that patent for people
` working with things like accelerometers.
`Q. Okay. Thank you. Now, can you please turn
` to page 41 and paragraph 82 of your
` declaration.
`
`Complete Legal
`
`214-746-5400
`
`
`
`19
`
`JOSEPH A. PARADISO, PH.D.
`
`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. Yes.
`Q. If you want to read the whole paragraph,
` please do so, and let me know when you're
` ready.
`A. Okay.
` (Witness reviews document.)
`A. Yes, I've finished.
`Q. Great. In the first sentence you said that
` a person of ordinary skill in the art "would
` have been motivated to" --
` (Reporter interruption.)
`Q. Okay. I'll start over. In the first
` sentence of the paragraph 82 it states a
` person of ordinary skill in the art "would
` have been motivated to combine the teachings
` of McMahan with the Pasolini and Goldman
` combination in order to enhance the accuracy
` of the accelerometer of the Pasolini and
` Goldman combination."
` Do you see that?
`A. I see it.
`Q. Can you tell me where in Pasolini or Goldman
` it states that the accuracy of the
` accelerometer is insufficient?
`A. If you look, in general, using
`
`Complete Legal
`
`214-746-5400
`
`
`
`20
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` accelerometers, any kind of distortion of
` the data is going to affect the accuracy.
` This is just a fact independent of Pasolini
` and Goldman. When you use a device, if
` there's distortion in the device, you have
` difficulty with the accuracy. And a POSITA
` would be aware of glitches because we saw
` them, you know, we encountered them. The
` early accelerometers were even worse. And
` we would always put in a glitch detector
` because these glitches would affect the
` accuracy adversely.
` So a POSITA would be aware of the
` glitches just by looking at the data and
` would be aware that there's a reason to get
` rid of them. And I think McMahan gives a
` great example of doing this. So a POSITA
` would want to definitely combine the essence
` of what McMahan says in glitch removal to
` the devices of the sort that Pasolini and
` Goldman discuss.
`Q. So is there any discussion of this in either
` Pasolini or Goldman?
`A. I would have to review Pasolini and Goldman
` to be sure. I do not believe that Pasolini
`
`Complete Legal
`
`214-746-5400
`
`
`
`21
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` or Goldman talk specifically about glitches
` as we call them, which is why we include
` McMahan. But anybody who worked with these,
` a POSITA working with systems like this at
` the time would be very well aware of
` glitches because we were plagued by them.
`Q. And so you're saying specifically -- and by
` the way, I belive you have Pasolini and
` Goldman there --
`A. Yes, I do.
`Q. -- possibly in front of you? So if you want
` to go ahead and review Pasolini and Goldman,
` please do so and take your time. But my
` question would be, so specifically for
` either Pasolini or Goldman are you saying
` that those devices were also plagued by
` glitches such that its accuracy would be
` insufficient?
`A. Let me review these --
`Q. Sure.
`A. -- quickly and get back to you.
`Q. Take your time.
`A. Sure.
` (Witness reviews documents.)
`A. Okay. I have reviewed these pieces of art.
`
`Complete Legal
`
`214-746-5400
`
`
`
`22
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` Can you repeat the question?
`Q. Yes. So specifically in Pasolini or Goldman
` can you show me where it speaks about these
` -- I'm sorry, let's start over.
` Specifically in either Pasolini or
` Goldman can you show me where it speaks
` about being plagued by glitches, so to
` speak, as you referred to it?
`A. Okay. If you look at Pasolini, he doesn't,
` at least on this quick reread that I made
` here, explicitly talk about glitches, but he
` does talk about effects on accuracy of the
` accelerometer. So he's already concerned
` with these devices not being accurate. So
` that's raised in Pasolini. He talks about
` offsets and how his baseline restore
` basically is removing them dynamically as
` well as compensating for gravity, but he
` talks about offsets and offset drift which
` are issues with accelerometers.
` If you look at Goldman, he doesn't
` talk so much about that. He does talking
` about calibrating it. He tells you how to
` remove the offsets and calibrate by flipping
` it. So you calibrate the range and the
`
`Complete Legal
`
`214-746-5400
`
`
`
`23
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` offset that way. But he also does talk
` about issues in reading these out. So if
` you look at the very end of it, when he
` talks about sampling rates, he actually says
` that -- at the very end "If radio packets
` are sent by reading the accelerometer, it
` might" --
` (Reporter interruption.)
` THE WITNESS: Sorry.
`A. "If radio packets are sent while reading the
` accelerometer, it might cause a sample
` period to be missed as it takes about 4
` milliseconds to send a packet. Sample
` readings may also be lost due to garbage
` collection, though since a generational
` collection scheme is used by the Squawk Java
` VM, the expected delays... are only about
` 20-30 milliseconds."
` So he's talking about sample readings
` being affected, the actually data coming
` from the accelerometer being essentially
` distorted, which can create a glitch.
` That's one way glitches can be produced.
` There are several ways you can get them, but
` that's one way. So both of them discuss or
`
`Complete Legal
`
`214-746-5400
`
`
`
`24
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` point to accuracy issues with
` accelerometers.
` And in Goldman as a POSITA reading
` that, I would be worrying about glitches.
` And as POSITA at the time, I saw them
` constantly. You hook up an accelerometer on
` a scope, you move it around, depending on
` how you build your circuit, what
` accelerometer you use, you see these
` glitches. You know, many sensors can suffer
` from these. Depends oftentimes on how
` they're applied, but it's very common. So
` we knew about them, and I can see the
` pointers in the literature.
`Q. Okay. So let's stay with Goldman because
` we're there. In the passage you read it
` talks about it might cause a sample period
` to be missed, and also sample readings can
` be lost due to garbage collection --
`A. Yes.
`Q. -- is that correct?
`A. Yes.
`Q. And so -- I'm sorry. And so, in your
` opinion, you're saying that a sample period
` being missed or lost is equivalent to a
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` glitch?
`A. Depends on how it's -- how it's actually
` playing out. If you get part of a sample
` coming in or, you know, there's something
` wrong with your conversion, if there's some
` process that interferes with storing the
` data, you can glitch. It's one of the ways
` it happens.
`Q. What do you mean by "part of a sample coming
` in"?
`A. Let's say bits get mixed up. You do an
` A-to-D conversion, there's some other task
` that happens, depending on how you write
` your code you could be making a -- not doing
` it the optimal way, it could be a property
` of the device, many ways this can happen,
` you can, you know, get a bit thrown,
` shouldn't be there. So you suddenly see a
` large spike is a good example.
`Q. So a bit that is thrown or something that is
` there that shouldn't be there --
`A. It's one example.
`Q. -- that is one example --
`A. One example.
`Q. -- that is an example of sample period being
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` missed or lost?
`A. A sample period being missed or lost you
` would see a discontinuity, but there could
` be related things happening in the software
` that could perhaps change the actual data
` byte. Add a bit, mess up the conversion,
` there's a problem with timing so you read a
` register too soon, this is not unusual
` sometimes, and you can get a glitch. There
` are many ways --
`Q. (Inaudible.)
`A. -- glitches happen. This is one way. And
` I've seen it. I've seen it coding up
` accelerometers. It's one way. Another way
` is a problem with the circuit. Another way
` is a problem with the accelerometer.
`Q. Okay. What --
`A. Could be a bad data link. You'd see it
` there as well.
`Q. Okay. So all those various other ways, are
` they talked about in Goldman?
`A. No. But you can see Goldman is concerned
` with the accuracy of the accelerometer
` values, and glitches are in that family. So
` as a POSITA I want to get the best, you
`
`Complete Legal
`
`214-746-5400
`
`
`
`27
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` know, sensor stream that I can. I'm going
` to be concerned that all of these factors --
` these guys are pointing to issues with
` accuracy. Goldman is pointing one way, the
` beginning of one way you can start to get
` glitches. So I don't -- I don't want them.
` Let's say I drop the accelerometer.
` That's another way I can get a glitch. I
` get a huge value. It's not relevant to
` this. It could skew all my data. I get a
` spike.
`Q. But that isn't talked about in Goldman now,
` correct?
`A. No.
`Q. And so the issue that you identified that we
` talked about in Goldman, which is the sample
` period being missed or lost, does Goldman
` state that because of that issue its
` accuracy would therefore be insufficient?
`A. I believe he implies that. That's something
` you want to avoid, yes.
`Q. Okay. Let's go back to Pasolini. Could you
` leading with the column and line numbers
` refer me to the sections in Pasolini that
` you were referring to?
`
`Complete Legal
`
`214-746-5400
`
`
`
`28
`
`JOSEPH A. PARADISO, PH.D.
`
`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. Let me see. If you look at column 5,
` paragraph -- lines 25 to 30, he talks about
` some of the contributions to accura --
` inaccuracy in accelerometers.
`Q. Okay. And so in that -- well, actually,
` let's start with that. Is there any other
` section that you've identified --
`A. There may be. I'd have to read it again,
` but this is what I saw when I read here.
`Q. Okay. And so in the section that you
` referred us to, it's referring to
` "fabrication offsets and deviations" --
`A. Uh-huh.
`Q. -- "that can be caused by the fatiguing of
` the materials" --
`A. Yes.
`Q. -- correct?
`A. Yes.
`Q. And what does that mean?
`A. That means that accelerometers of these
` sorts usually work with a cantilever beam
` done in a MEMS architecture, and that has a
` certain spring constant, certain mechanical
` properties, and he is saying that those can
` change over time, which is causing drift and
`
`Complete Legal
`
`214-746-5400
`
`
`
`29
`
`JOSEPH A. PARADISO, PH.D.
`
`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
`
` the scale factor or, if not, offset of the
` accelerometer, both.
`Q. Okay. In the very next sentence after that
` it states "Subtraction of the static-
` acceleration signals" -- I'm going to skip
` the symbols -- "from the acceleration
` signals," again, some more symbols I'll skip
` -- "advantageously enables compensation of
` said offsets."
` Do you see that?
`A. Yes, I see that.
`Q. So doesn't Pasolini provide the solution to
` its fabrication offsets and mediation issues
` right there in the same paragraph?
`A. Yes. He suggests the solution there, but
` what I called attention to that paragraph
` for isn't to describe glitches per se but to
` talk about the general concern that POSITAs
` had at that time with the accuracy of these
` sensors.
`Q. Okay.
`A. So Pasolini is already flagging accuracy
` issues. He's not saying they're perfect.
`Q. And given the issue and proposed solution by
` Pasolini, does Pasolini suggest that its
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`30
`
`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
`
` accuracy is still nonetheless insufficient?
`A. Insufficient for what?
`Q. For the purpose of the Pasolini device.
`A. That's not what he's suggesting there. He's
` basically saying that there are artifacts
` that come from the special MEMS
` accelerometers. This is one of them. And
` his subtraction, his technique that he is
` putting forth in this patent can compensate
` for those. It's not a common filter or any
` kind of advanced system for this, but it is
` a simple way to compensate for bias drift.
` But he is raising a flag of accuracy as
` being a potential issue in these devices.
` He's suggesting a solution for it.
`Q. Okay. Can you turn to page 42, paragraphs
` 84 and 85 --
`A. Sure.
`Q. -- of your declaration?
`A. Yes.
`Q. So here where you're talking about applying
` Mizell to, again, Paso -- the proposed
` Pasolini and Goldman combination; is that
` correct?
`A. That's correct.
`
`Complete Legal
`
`214-746-5400
`
`
`
`JOSEPH A. PARADISO, PH.D.
`
`31
`
`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 here your talking about applying Mizell
` to the proposed Pasolini and Goldman
` combination "in order to smooth and reduce
` noise in order to obtain a more accurate
` representation of the gravity signal."
` Do you see that?
`A. I see it.
`Q. Do you see that?
`A. Yeah, I do see it.
`Q. So if -- actually, yeah, let me start from
` the top. So where is it in -- let's start
` with Pasolini where it talks about the need
` to sooth and reduce noise for an a
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
