`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`SONY CORPORATION,
`Petitioner,
`
`v.
`
`COLLABO INNOVATIONS, INC.,
`Patent Owner.
`____________
`
`Case IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`____________
`
`Record of Oral Hearing
`Held: May 9, 2018
`____________
`
`
`
`
`
`
`
`
`
`Before DAVID C. McKONE, GREGG I. ANDERSON, and JENNIFER
`MEYER CHAGNON, Administrative Patent Judges.
`
`
`
`
`
`
`
`Case IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`MATTHEW A. SMITH, ESQUIRE
`Smith Baluch, LLP
`100 M Street, S.E.
`Suite 600
`Washington, D.C. 20003
`
`
`
`ON BEHALF OF PATENT OWNER:
`
`
`TERRY A. SAAD, ESQUIRE
`Bragalone Conroy, P.C.
`2200 Ross Avenue
`Suite 4500W
`Dallas, Texas 75201-7924
`
`
`
`
`The above-entitled matter came on for hearing on Wednesday, May 9,
`
`2018, commencing at 1:00 p.m., at the U.S. Patent and Trademark Office,
`600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`2
`
`
`
`Case IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`
`P R O C E E D I N G S
`- - - - -
`JUDGE ANDERSON: If you are not already seated, hopefully let
`me just make it clear you can be seated. And welcome. This is the hearing
`for IPR2017-00958 and 960. Petitioner is Sony Corporation. Patent owner
`is Collabo Innovations, Inc., the owner of U.S. patent number 7,023,034,
`which is the challenged patent in both of the IPR proceedings.
`I am obviously joining you remotely. Judge McKone is on the
`other screen there in the hearing room joining you remotely as well. Judge
`Chagnon is there with you in person. Because Judge McKone and I are
`remote, any demonstratives that you decide to use need to be described by
`slide number so that both of us can follow along.
`Per our order, each party is going to have 60 minutes to present its
`arguments for both proceedings. You are not required to take all of the time.
`Let me repeat that. You are not required to take all of the time. Because
`petitioner has the burden to show unpatentability of the original claims,
`petitioner will proceed first followed by patent owner, who will, during that
`time, address its motion to exclude in both cases. Petitioner may reserve
`time to rebut patent owner's opposition and the motion to exclude. And if
`the patent owner desires to, it may reserve some time to rebut petitioner's
`opposition to its motion to exclude.
`For the record, we have a recent Supreme Court case, SAS versus
`Iancu. These particular cases, we instituted on all grounds on all claims, so
`we have no need to do anything additional to comply with that Supreme
`Court decision which happened about a little over a week ago.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`3
`
`
`
`Case IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`
`Also, any objection to argument presented by the opposing party
`shall be made during the time set aside for the objecting party's argument.
`In other words, we don't want interruptions to the arguments. People should
`understand that we have experienced judges here. We can sort through this
`at such time as you make the point, which can be during the presentation of
`the party that has the objection.
`At this time, let me have the counsel make their entry of
`appearances. First for petitioner, please.
`MR. SMITH: Thank you, Your Honor. This is Matthew Smith,
`Smith Baluch, LLP, for the petitioner, Sony Corporation.
`JUDGE ANDERSON: Can you say that again. I did not get that.
`MR. SMITH: Yes. Matthew Smith.
`JUDGE ANDERSON: Okay, Mr. Smith. Thank you. And for
`patent owner?
`MR. SAAD: Yes, Your Honor. Terry Saad on behalf of the patent
`
`owner.
`
`JUDGE ANDERSON: Thank you, counsel. We did receive some
`demonstratives from the patent owner. We didn't receive any objections.
`We assume that the petitioner has no objection to those. Is that true,
`Mr. Smith?
`MR. SMITH: Yes, Your Honor, we have no objection to the
`demonstratives.
`JUDGE ANDERSON: And on the same subject of
`demonstratives, we did not receive any demonstratives from petitioner. So
`we assume you are going to make your presentation without any, right?
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`4
`
`
`
`Case IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`
`MR. SMITH: That is correct. There are no petitioner
`demonstratives, Your Honor. We may refer to portions of the record that
`have already been filed.
`JUDGE ANDERSON: I'm sorry, what was that? You need to be
`closer to a microphone. I don't know whether Judge McKone is having a
`problem, but I'm having a hard time hearing you.
`MR. SMITH: Can you hear me from the podium microphone?
`JUDGE ANDERSON: Pull it closer up, Mr. Smith.
`MR. SMITH: Can you hear me now?
`JUDGE McKONE: I don't think the microphone is on.
`MR. SMITH: So to be clear, petitioner has no demonstratives of
`their own. We may show portions of the record that have already been filed
`or patent owner's demonstratives, but none of our own.
`JUDGE ANDERSON: Okay. Well, remember, Judge McKone
`and I are going to have a hard time seeing any of that. So you are going to
`need to be particularly scintillating with your discussion here today.
`Okay. So do you want to reserve any rebuttal time, Mr. Smith?
`MR. SMITH: Yes, Your Honor, could I reserve a half hour,
`
`please.
`
`JUDGE ANDERSON: Okay. You can do that. And I should say
`since how Judge Chagnon is there, she's going to keep track of the time for
`us. And with that, Mr. Smith, you may begin your presentation right now.
`Thank you.
`MR. SMITH: Thank you, Judge Anderson, and members of the
`Board. I will be very clear as to what we are talking about, which exhibits,
`when. We are pulling up right now the patent owner's demonstrative slide,
`
`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
`
`5
`
`
`
`Case IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`so not our demonstratives, Collabo's demonstratives. And in particular, slide
`20, if you could look at that.
`I'll talk about today, I think, three categories of issues that we have.
`The issues are actually relatively limited, so I will take into account that we
`don't need to use the entire time necessarily, but I think there are three things
`the parties are disputing. One is the issue in the 958 IPR of whether the
`Tomoda reference has the required reflecting walls. Second, in the 960 IPR
`is whether or not the Takahashi and Kimura references can be combined to
`give the claimed reflecting walls. And then third, in the 960 IPR are the
`arguments related to the Abe reference. And I'll do them in that order.
`Beginning with the 958 IPR proceeding, I have a few comments on
`where the patent owner seems to be going with the reflecting walls argument
`in that proceeding. If you look at this slide, the patent owner has said the
`reflecting walls construction which the petitioner put forward and which was
`adopted in the institution decision is too broad because everything visible
`reflects light. And I'm reading from the bottom of the patent owner's slide
`here. It would render any vertical surface a reflecting wall, and a broad
`interpretation enables petitioner to reclassify Tomoda's light-shielding layer
`as a reflecting wall.
`A couple of observations on this. First, if you could look at the
`patent owner response in the 958 case on page 22, in the first full paragraph
`of that page, the patent owner says, For the purposes of this proceeding only,
`the patent owner applies the Board's construction for its analysis but reserves
`the right to seek alternative constructions in other proceedings.
`We asked the patent owner's expert, Professor Afromowitz, about
`this in his deposition. He also did not have any dispute with the construction
`
`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 IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`the Board adopted in the institution decision, meaning that on slide 20 of the
`demonstratives, it seems like the patent owner is saying the construction to
`which we have already agreed and acquiesced would cover the structures
`within Tomoda's system.
`And there's no dispute that there is a vertical surface in the place
`that is required by the claim in the Tomoda system. The only question is
`whether that would be a reflecting wall or not. And the patent owner seems
`to be saying here that, yes, it would be a reflecting wall because the claim
`construction would cover that. They are now appearing to assert that the
`claim construction is too broad. But that's not something they took. That's
`not something they did in the patent owner response.
`And if the claim construction is thought to be too broad by the
`patent owner, what is the construction that they think would be appropriate?
`They have never told the Board what they think the appropriate construction
`would be. There's no patent owner proposal of what a reflecting wall should
`mean. And I think that's because --
`JUDGE ANDERSON: Mr. Smith, let me ask you something here.
`Do you agree that in order to actually anticipate claim 1 or 10, that the
`light-shielding layers of Tomoda do have to reflect light, right?
`MR. SMITH: Yes, Your Honor, I do agree.
`JUDGE ANDERSON: So you also understand that Tomoda
`describes in the structure that you point to as showing the reflecting wall a
`light-shielding layer. And we all agree with that?
`MR. SMITH: Yes, Your Honor.
`JUDGE ANDERSON: And you do agree or not that a
`light-shielding layer has a different function than to reflect light, and that's
`
`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 IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`specifically to shade certain areas of the semiconductor device; isn't that
`right?
`
`MR. SMITH: I think I would answer that strictly speaking, no,
`although it's probably a nuanced yes/no, and that the overall design intent of
`a light-shielding layer certainly is to shield parts of the circuit. But it often
`does that by reflecting. For example, you can see that clearly in the Kimura
`reference, which is part of both IPRs, but this suggests a background
`statement that that is -- what I'm saying is actually true. The Kimura
`reference has both what we would call the wall and the light-shielding layer
`made out of the same material, both called light-shielding, in other words,
`light-shielding layer, light-shielding wall. So one of the ways in which one
`can accomplish a light-shielding design objective is to use reflective
`materials. So yes and no to that question, Judge Anderson.
`JUDGE ANDERSON: I understand from the papers that you are
`relying in part on the fact that the light-shielding layer is made of a metallic,
`I think it's tungsten, as disclosed in Tomoda; is that right?
`MR. SMITH: That is correct, Your Honor. Tungsten or
`aluminum, I believe, is the correct metal.
`JUDGE ANDERSON: Okay. Proceed.
`MR. SMITH: Thank you, Your Honor. The emphasis of the '034
`patent was not, however, on those kinds of details, Your Honor. I think if
`you look at the '034 patent, I'm referring specifically to the bottom of
`column 5. The last four lines of that column says about what there is to be
`said in the '034 patent. It says, The reflection wall 62 reflects a portion of
`the light, specifically oblique light, incident from above onto the aperture 65.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`8
`
`
`
`Case IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`
`And then it goes on to the top of column 6 by saying with
`reference to Figures 1 and 2, The placement of reflecting wall 62, which is
`one of the features of the present invention, will be described in detail.
`That's about what you know about the reflecting walls in the '034 patent.
`There is certainly a description of how the reflecting walls are positioned,
`and that seems to be what the '034 patent was intending to claim as the
`invention. But the specific makeup of the reflecting walls, the size, the
`shape of the reflecting walls is all left a little bit unclear.
`There is one embodiment, which we pointed to in Figure 5 which
`shows a reflecting wall made of titanium and tungsten, but that particular
`embodiment didn't make it into the claims. All the claims say about
`reflecting walls is you have reflecting walls and they are in a certain position
`relative to other elements. It's not even until you get to dependent claim 5
`where the reflecting walls are even required to be made of a metal. There is
`no claim directed to, say, the combination of titanium of tungsten.
`So all the claims are talking about is having this vertical surface
`that reflects light as, the patent owner seems to acknowledge in slide 20 of
`their demonstratives. And that, I think, makes a certain amount of sense.
`If you could look at Figure 9 of the '034 patent, Figure 9 is a figure
`that the original inventors labeled prior art. And you see in Figure 9 the
`structures 512 which are reflecting walls. This is admitted prior art. The
`concept of the reflecting wall and its function was well known at that time.
`What the inventors were instead focusing on as the invention was shifting
`the reflecting walls and shifting other elements in there relative to the center
`of the aperture.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`9
`
`
`
`Case IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`
`Now if we could pull up the Tomoda reference, and in particular
`Figure 1 of the Tomoda reference, you will see, and I'm sure you have
`appreciated this before, the Tomoda reference has the structures in Figure 9
`with the exception of the center lens. It's certainly got the first microlens
`element and vertical planes made out of a metal silicon dioxide interface that
`are shifted in the way required by the patent.
`So the question is, are those vertical planes and light-shielding
`layers actually reflecting walls? What evidence do you have before you to
`decide that? And certainly you have, first of all, the testimony of
`Mr. Guidash in his declaration in paragraphs 99 to 110 where Mr. Guidash
`explains that having this particular kind of metal and silicon dioxide will
`generate a reflective surface that will be incident with light that is coming in,
`oblique light, and will reflect in the normal way we expect a metal with an
`optically transparent material like silicon dioxide to do.
`Beyond Mr. Guidash's testimony, however, I think you also have
`admissions of the patent owner and the patent owner's expert. Certainly
`Professor Afromowitz admitted in his deposition that the metals used in
`Tomoda, aluminum and tungsten, would have coefficients of reflectivity
`somewhere between 50 to 80 percent. And that means that they would
`reflect 50 to 80 percent of the light that is incident on them.
`You also have, if we could look at the patent owner response at
`page 8 in the 958 proceeding, the patent owner's discussion of how
`light-shielding layers work in reference to the figure seen on that page, if
`you look at the figure, there are three rays illustrated. Of course it's a
`spectrum of incoming electromagnetic radiation but shown as four rays.
`One of the rays hits the top of the light-shielding layer and bounces off. The
`
`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 IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`patent owner's discussion of this is a ray is shown reflecting off a metal line,
`perhaps a light-shielding structure placed over some sensitive electronics,
`and another ray misses the center photodiode completely and hits one on the
`right. So here you see the patent owner's own sort of implicit
`acknowledgment that metal light-shielding layers work by reflection.
`I think one also needs to take into account Professor Afromowitz's
`declaration in total where if you read it carefully, you can see that he never
`actually takes the position that the structures in Tomoda would not reflect
`light. His arguments go more to the amount of light that would be reflected.
`So his arguments along the lines of whether the layer is thick enough, for
`example, whether it's 8 percent of the layer, the surrounding layer that it's in,
`or 80 percent of the surrounding layer that it's in, whether or not the surface
`was roughened or something like that. Those all go to the amount of light
`that's going to be reflected. Nowhere in Professor Afromowitz's direct
`testimony does Professor Afromowitz say, I can say that no light will be
`reflected from these metal lines in Tomoda.
`And that's significant, I think, because you cannot get from the
`'034 patent a limitation of the claims that is directed to the amount of light
`that would be reflected. It's simply not discussed anywhere in the '034
`patent. All the '034 patent says about the amount of light is that a portion of
`the light will be reflected. So it would be very difficult to limit the claims
`to, say, a 5 percent reflection or a 10 percent reflection or something like
`that. There's just no foothold in the specification for that.
`JUDGE McKONE: So it's your position that any de minimus
`amount of reflection on the side or on the wall would be good enough,
`would be sufficient?
`
`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 IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`
`MR. SMITH: Yes, Judge McKone, that is our position. I think it
`makes sense in the context of the '034 patent because they simply weren't
`interested in the reflecting was characteristics themselves as opposed to
`where the reflecting wall is. But if there was some sort of de minimus
`limitation, as Mr. Guidash testified, I think the prior art would meet that. In
`other words, we have a metal reflector inside silicon dioxide in the right
`position. It's going to reflect a substantial amount of light. Do we know the
`exact numbers? No, of course we don't, but we don't know the exact
`numbers in the '034 patent either. The '034 patent is simply about having the
`reflecting walls in the right position. I don't think they were interested in
`distinguishing infringing products that might have 5 percent reflection or 10
`percent reflection.
`JUDGE ANDERSON: Wouldn't you have to modify Tomoda to
`some extent to actually get the kind of performance that would, in fact, be a
`reflecting wall when you already have something that really is primarily a
`light-shielding layer?
`MR. SMITH: I don't think so, Your Honor. I think the
`light-shielding layer being made out of metal means that it's operating by
`reflection to begin with. If it's in right position, it's going to reflect light and
`it's going to reflect a significant amount of light, as Mr. Guidash testified.
`Of course the exact dimensions, at least breadth, for example, is not given in
`Tomoda. Although, the height of that layer is given in Tomoda. So you
`need to do some things when you implement Tomoda, but a person of
`ordinary skill in the art know how to do that. But the walls themselves still
`need to 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
`
`12
`
`
`
`Case IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`
`JUDGE McKONE: What evidence do we have that a significant
`amount of light is reflected by the walls?
`MR. SMITH: You have Mr. Guidash's testimony in paragraphs
`107 to 110 of his declaration, I believe. This is not a quantitative analysis.
`And I'll say that from the outset. But I think there are a couple of things we
`have to emphasize here. One is that the '034 patent itself is not about a
`quantitative analysis, as I have said before. It's about having a reflecting
`wall structure in the right position.
`Second, you are already talking about containing error. As
`Collabo itself argues, the light that comes through the microlens is intended
`to go down to the photodiode without interacting with light-shielding layers.
`It's not physically possible to do that in a real system. There's always going
`to be some oblique light that's escaping, but it's a smaller percentage of the
`overall incoming light to begin with. So you are already operating at a small
`percentage of the incoming light, so the light levels are going to be relatively
`small to begin with. And that's why it is -- in part why it is appropriate to
`have no de minimus limitation in the claims. If there is a reflecting wall in
`that position, it's going to interact with light coming in. Both experts have
`testified along those lines, as we cited in our reply. And it's going to reflect
`a significant amount of light.
`JUDGE ANDERSON: One of the things Dr. Guidash says in
`paragraph 107 of his declaration is he goes through these preferred ranges of
`the thickness of the insulation layer versus the light-shielding layer, and then
`he says using those preferred -- I believe he's using the preferred range,
`which for the insulation film is 600 to 1,000 nanometers, and for the -- I'm
`
`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 IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`sorry, for the light-shielding layer is 400 to 800 nanometers and for the
`insulation layer is 600 to 1,000 nanometers.
`So then in paragraph 108 he flips it and somehow he makes a
`mathematical -- it's not really mathematical. It's an arithmetic error. He
`says, at least as I see it, I think the patent owner pointed this out in the
`response, the light-shielding layers are, in fact, not 66 to 80 percent of the
`thickness of the layer. They are, in fact, the inverse of that. They are the 34
`to 20 percent, because obviously from the preferred ranges, the thickness of
`the insulation layer is 600 to 1,000, which is more than the 400 to 800 of the
`other layer. So all I want to point out is it looks to me like a mistake was
`made in terms of this testimony.
`MR. SMITH: Yes, Judge Anderson, I don't actually agree with
`that, but I'll explain it just briefly. There are two ranges given. One is for
`the layer in which the light-shielding layer is embedded, and one is for the
`light-shielding layer itself. As you pointed out, the layer in which it's
`embedded, I think, is 800 to 1,000, whereas, the light-shielding layer is 400
`to 600. And all he did was match up the lower boundary with the lower
`boundary and the upper boundary with the upper boundary. So 800 divided
`by 1,000 is 80 percent, and 400 divided by 600 is two-thirds, essentially.
`So his 66 to 80 percent is the range that he came up with. Of
`course, there are other potential ranges. One could pick different numbers
`within those ranges, but those are apparent to a person of ordinary skill.
`This is basic arithmetic. Something that Professor Afromowitz, Collabo's
`expert, acknowledged at his deposition, a person of ordinary skill in the art
`would be aware of all of those possible combinations.
`
`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 IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`
`JUDGE ANDERSON: So just so I understand, he took the 800 of
`the light-shielding layer, the highest of the preferred range, and compared
`that to the insulation film of 1,000? That's the thickness of the insulation
`layer. It seems to me that's larger than the light-shielding layer, and so to
`say it's 80 percent of the whole vertical space is wrong.
`MR. SMITH: So we have in paragraph 27 of Tomoda the
`interlayer insulation film is preferably 600 to 1,000. And the thickness of
`each light-shielding layer is preferably 400 to 800. Do we agree on those
`numbers?
`JUDGE ANDERSON: Sure.
`MR. SMITH: So the light-shielding layer is less thick than the
`interlayer insulation film. I think that's accurate, because the light-shielding
`layer is contained within the interlayer insulation film.
`JUDGE ANDERSON: Right.
`MR. SMITH: And if the maximum preferred range of the
`interlayer insulation film is 1,000 nanometers and the maximum preferred
`range of the light-shielding layer is 800 nanometers, that 800 divided by
`1,000 would be 80 percent of the layer. And correspondingly, if the
`minimum preferred range of the interlayer insulation film is 600 and the
`minimum preferred thickness of the light-shielding layer is 400, then 400
`divided by 600 would be 66, 67 percent. So what he's doing is taking the
`preferred range and matching the end points, the highest to the highest and
`the lowest to the lowest.
`JUDGE ANDERSON: Okay.
`MR. SMITH: Does that make sense?
`
`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 IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`
`JUDGE ANDERSON: I'm still thinking about it, but at least you
`have given me some basis to think that maybe that's correct.
`MR. SMITH: These are expressed as ranges. So the experts have
`to look at this and do something reasonable with it, I think. And I think
`Collabo's point on this was you could do other reasonable things with it.
`And our reply to that is that doesn't really matter because whatever you do
`with this that's reasonable, even if you take what Collabo says is the ultimate
`preferred range of 40 percent, that's still pretty thick and it's still going to
`give you significant reflection.
`Bottom line is significant reflection is not part of the claim
`construction that Collabo agreed to. And even if we could come up with
`some sort of de minimus level of reflection, the prior art would be
`significantly above it. These are not layers. These are pure metal layers.
`They are not layers designed to be absorptive, for example, like the blackout
`layer you might see sometimes in the prior art. These are layers that are
`designed to be metal layers and designed to be light-shielding layers, I think,
`by reflectivity.
`Most of Collabo's arguments on Tomoda, I think, do go to the
`proportion of light that is reflected. And unfortunately, that's just not a part
`of the claim language that Collabo has expressly agreed to and that
`Dr. Afromowitz acknowledged in his deposition. He had no dispute with it.
`Let me move on to the 960 IPR unless there are further questions
`on the 958. In the 960 IPR, we have two categories of issues. One is ground
`1, the combination of Takahashi and Kimura and then the combination with
`the Abe reference with the trapezoidal claims, claims 3 and 12. And some
`of the issues overlap with the issues we discussed in the 958 proceeding. In
`
`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 IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`other words, whether the light-shielding layers would actually reflect light in
`the combination of Takahashi and Kimura, a lot of the same arguments will
`apply there except in the 960 proceeding, Kimura is a reference that
`expressly teaches the use of light-shielding layers and teaches their benefit.
`So a person of ordinary skill would be motivated, I think, to make sure that
`there is reflectivity in those light-shielding layers to the extent there's any
`quibble that a metal layer in silicon dioxide would actually reflect light. So
`there is an added motivation to reach the light-shielding reflective layers in
`this particular combination.
`Mike, could you pull up the petition in the 960 case. And in
`particular, let's look at the diagram on page 32 of the original petition which
`discusses the combination of Takahashi and Kimura. The combination of
`Takahashi and Kimura, I just want to point this out, is page 32 of the
`petition, please. So we are looking at the 960 petition, page 32. I want to
`point out that this combination is, physically speaking, relatively simple.
`We are simply taking from Kimura the well-known idea of reflecting walls,
`as acknowledged in the '034 patent in relation to Figure 9, and using the
`reflective properties in Takahashi.
`One of Collabo's arguments -- and there are basically three
`arguments about why this combination would not have been obvious to a
`person of ordinary skill. One of Collabo's arguments has to do with an
`alleged difference in the design philosophies of Takahashi and Kimura.
`Takahashi, Collabo says, is seeking to minimize the vertical thickness of this
`device, while Kimura says, yeah, you want to minimize the vertical
`thickness, but sometimes it's hard to do. So in order to get around that, we
`can provide a different structure.
`
`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 IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`
`That, however, we don't think, is in conflict at all, particularly with
`this kind of combination. We are not adopting the structure of Kimura per
`se, but simply it's teaching that reflecting walls are beneficial to use. They
`are going to help collect light and that you can put them in approximately
`this location over the aperture to get the reflective effect.
`So there's no bodily incorporation of Kimura into Takahashi at all.
`Nor if there were a bodily incorporation are the two design philosophies
`really in conflict. Kimura says, yes, we want to lower the height of the
`overall device but sometimes that's hard to do. Takahashi says you can do
`that, so why wouldn't you just do it in Takahashi. There doesn't seem to be a
`conflict.
`The second issue that Collabo brings up is whether or not
`Takahashi says that all light entering the photocell would be projected on to
`the photodiode and not interact with the light-shielding layers at all.
`And I think it is correct to say that that is generally the intent of
`these devices but not correct to say that in physical reality that will always
`happen. There's always going to be some light which is escaping which is
`causing a little bit of an error. Professor Afromowitz acknowledged this in
`his deposition. And this is testimony that's cited in our reply: The lens
`systems simply aren't perfect that these structures are using. Therefore, you
`are always going to have some oblique light that can be reflected by these
`reflecting walls.
`Mr. Guidash testified in exactly the same way. There's going to be
`a spectrum of light coming in at different angles and they are going to
`interact with the light-shielding layers. Indeed, that is one of the purposes of
`a light-shielding layer, as we discussed earlier, that a light-shielding layer 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
`
`18
`
`
`
`Case IPR2017-00958 (Patent 7,023,034 B2)
`Case IPR2017-00960 (Patent 7,023,034 B2)
`
`there to block some of this light. And in this case, in the case of the
`combination, it will do so by reflection and will do so off the side into the
`photodiode.
`The third thing that Collabo argues is the light could bounce off the
`top of the light-shielding layer in the combination and then it could bounce
`back up, say, off the color filter or off the microlens, bounce back down and
`bounce off some other structure and hit an adjacent photocell. So I think
`you have got three or four reflections there. At each reflection there's going
`to be some attenuation of the light which Collabo acknowledges in its patent
`owner response. And the signal that gets there is going to be relatively
`weak.
`
`So this, we don't think, is a problem for the combination at all and
`it's, I think, e