`
`In re patent of Kikinis
`
`U.S. Patent No. 5,779,334
`
`Issued: July 14, 1998
`
`Title: ENHANCED VIDEO
`
`PROIECTION SYSTEM
`
`0O'Z>OO'><DJOO'>O€7>¢£'.>OO'JOO'>OO'3
`
`Petition for Inter Partes Review
`
`Attorney Docket No.: 42299.43
`Customer No.:
`27683
`
`Real Party in Interest: Xilinx, Inc.
`
`Declaration of A. Bruce Buckman, Ph.D. under 37 C.F.R. § 1.68
`
`Directed to Petitioner’s Reply
`
`I.
`
`INTRODUCTION
`
`1.
`
`I, Dr. A. Bruce Buckman, am making this declaration at the request of
`
`Xilinx in IPR2013-00112 ofU.S. Patent No. 5,779,334 (“the ’334 Paten ”) to
`
`Kikinis.
`
`2.
`
`I previously submitted a declaration explaining why the ’334 patent is
`
`invalid. That declaration is marked as XLNX-1005, and sets forth my experience,
`
`qualifications, publications, materials considered and compensation. I also provided
`
`a supplemental declaration that provides additional information regarding my
`
`qualifications. (See XLNX-1008.)
`
`3.
`
`As described in my prior declaration, I have over forty years of
`
`experience in the field of optics, including thirty—five years of experience as a
`
`-1-
`
`IVI LLC EXHIBIT 2018
`XILINX V. IVI LLC
`
`IPR Case 2013-00029
`
`
`
`professor in the electrical engineering department of the University of Texas at
`
`Austin. During this time, my teaching and research have focused on a wide range of
`
`topics in field of optics.
`
`4.
`
`The list of materials I have considered is set forth in my opening
`
`declaration. In addition, I have reviewed the Board decisions, IV’s Oppositions, the
`
`Declarations of Mr. Smith-Gillespie, and all exhibits cited thereto in both the ’545
`
`and ’334 IPRs. I have also reviewed the deposition of Mr. Smith-Gillespie
`
`(XLNX-1013 and XLNX-1014) as well as XLNX-1015, which contains excerpts
`
`from Spatial Light Modulator Technology (Uzi Efron ed., Marcel Dekker 1995). I
`
`have also reviewed the other exhibits cited in this report.
`
`II.
`
`SUMMARY OF OPINIONS
`
`5.
`
`This declaration addresses a variety of issues that have arisen since I
`
`submitted my original declaration. This includes issues raised by (1) the Board’s
`
`Decision to institute review; (2) IV’s Opposition; and (3) the testimony of Mr.
`
`Smith-Gillespie. For the reasons set forth below and in my other declarations, it is
`
`my opinion that the ’334 patent is invalid.
`
`III. CLAIM CONSTRUCTION
`
`6.
`
`The Board’s Decision addresses several claim construction issues.
`
`Although the Board addressed these terms at IV’s request, it did not adopt IV’s
`
`
`
`proposed constructions. The following is my response to the constructions proposed
`
`by IV and the preliminary constructions adopted by the Board.
`
`A.
`
`“light-shutter matrix system”
`
`
`
`
`A set of matrices, such as monochrome
`A two-dimensional array of
`LCD arrays or cells of a monochrome LCD elements that selectively admit and
`
`array, where each matrix comprises a
`block light.
`
`rectangular arrangement of elements
`
`
`
`ht.
`caable of limitin the assae of li
`
`
`
`
`
`
`7.
`
`I generally agree with the Board’s preliminary construction of
`
`“light-shutter matrix system.” In particular, I agree that a light shutter, in the context
`
`of the ’334 patent, is an element that is “capable of limiting the passage of light.”
`
`8.
`
`IV does not propose an alternative construction, but it does provide a
`
`long discussion of how it interprets the ’334 patent on this point. I disagree with
`
`several of IV’s points, as described below.
`
`9.
`
`IV argues that I have provided an opinion on “LCD cell” that is
`
`inconsistent with the specification. Not so. In my deposition testimony, I explained
`
`that the term LCD cell in the context of the ’334 patent refers to the entire LCD
`
`device. Ex. 2010 at 49:22-51 :1. The fact that the words “cell” and “pixel” are
`
`interchangeable in some contexts and not in others is not relevant to my opinions.
`
`10. With respect to the “matrix system” claim element, IV argues that a
`
`person having ordinary skill in the art “would appreciate that the claimed system of
`
`the ’334 patent is an electrically addressed system.” (Response, Paper No. 26 at 12.)
`
`-3-
`
`
`
`I disagree. As I discuss below, all LCD display devices create images using a
`
`continuous thin layer of liquid crystal material. LCD display devices use
`
`electromagnetic fields to organize the liquid crystal into a matrix of pixels arranges
`
`in rows and columns. Each individual pixel is a “light shutter” element that is
`
`capable of limiting (or blocking) the passage of light.
`
`11.
`
`The Board’s preliminary construction limits the phrase “matrix
`
`system” to a “rectangular arrangement.” I do not object to this proposed
`
`construction, although I note that other reasonable constructions may be broader. A
`
`broader interpretation of “matrix system” would not impact my analysis.
`
`12.
`
`I have reviewed the Tannas reference that IV cites on pages 12-15 of its
`
`Response. I generally agree that the Tannas reference describes one or more ways to
`
`make a “matrix” system display. I do not agree, however, that the cited portions of
`
`Tannas describe all possible ways to implement a light-shutter matrix system.
`
`Below, I describe other ways to make a pixelated light-shutter matrix system.
`
`B.
`
`“video controller adapted for controlling the light-shutter
`matrices”
`
` ponent tt c lit
`
` A component tha orol hght-shu
`
`matrices to facilitate the display of video
`in accordance with a video si nal.
`
`
`matrices to facilitate the display of
`video
`
`
`
`13.
`
`I agree with the Board’s proposed construction of “video controller
`
`adapted for controlling the light-shutter matrices.” Specifically, I agree that the
`
`
`
`“video controller” of the ’334 patent is a component that controls the light shutter
`
`matrices to facilitate the display of Video.
`
`14.
`
`IV’s proposed construction requires the video controller to act “in
`
`accordance with a Video signal.” I disagree with this construction because it adds a
`
`limitation to the claims. As a practical matter, all LCD Video projection systems in
`
`the mid-1990s used a video controller to control the operation of the LCD and to
`
`display video. But this “video controller” component was not responsible for
`
`processing the incoming video signal. Instead, the video signal went to a Video
`
`decoder that converted the Video signal into a different format. Nothing in the ’334
`
`patent requires these two functions to be performed in the same component. Mr.
`
`Smith-Gillespie and I agree that in 1996, the Video-signal—decoding and the
`
`light-shutter—matrix controlling functions were not generally performed on the same
`
`chip. (XLNX-1014 at 210123-212:6.) Thus, I disagree that the claims require the
`
`video controller to act “in accordance with a video signal.”
`
`C.
`
`“equivalent switching matrices”
`
`I Switching matrices that are
`corresponding or virtually identical in
`
`function or effect
`
`Sting matrices that are Virtly
`identical in function and effect
`
`15.
`
`I agree with the Board’s proposed construction of “equivalent
`
`switching matrices” as being “switching matrices that are corresponding or virtually
`
`identical in function or effect.”
`
`
`
`16.
`
`I note that both the Board’s construction and IV’s proposed
`
`construction depend on the Merriam Webster definition of the word “equivalen ” as
`
`“corresponding or virtually identical esp. in effect or function.” (See IV-2001.) The
`
`Board’s proposed construction uses the entire definition, whereas IV’s construction
`
`omits the word “corresponding.” I agree with the Board. In my opinion, switching
`
`matrices may be equivalent to one another by having corresponding functions or
`
`effects (even if they are not strictly or virtually identical).
`
`17.
`
`IV argues that “the ’334 patent distinguishes between equivalent
`
`monochrome LCDS versus conventional color LCD systems in which a unique LCD
`
`is used for each color.” I disagree. The ’334 patent actually distinguishes between
`
`using a single AM-LCD to display all colors ofthe video image (the prior art) versus
`
`shining beams of pre-colored light through separate LCDs, one for each light color
`
`beam (what is described in the specification).
`
`IV. OPINIONS REGARDING CHALLENGE NO. 2: OBVIOUSNESS BY
`TAKANASHI AND LEE
`
`18. My opening report explained why the ’334 patent is invalid in View of
`
`the Takanashi and Lee references. IV and Mr. Smith-Gillespie raise several
`
`arguments for why Takanashi and Lee do not invalidate the claims of the ’334
`
`patent. I disagree with those arguments for the reasons set forth below. Based on the
`
`analysis set forth herein and in my original report, it remains my opinion that the
`
`claims of the ’334 patent are invalid.
`
`
`
`A.
`
`Takanashi discloses a light—shutter matrix system
`
`19.
`
`IV and Mr. Smith—Gillespie assert that the ’334 patent is non-obvious
`
`because Takanashi does not satisfy the “light—shutter matrix system” limitation. I
`
`disagree.
`
`20.
`
`Takanashi discloses several different ways to create an image using a
`
`liquid crystal device called a “spatial light modulator” (“SLM”). Takanashi does not
`
`describe the inner workings of its SLM in detail because it was a commonly-known
`
`device in in the early 1990s. In general, the Takanashi SLM operates using two
`
`different types of light—a “read light” and a “write light.” These lights typically
`
`have different wavelengths. In operation, both the read light and the write light shine
`
`on the spatial light modulator. The write light encodes an image on the SLM, which
`
`then encodes that same image on the read light. (In a video projector system, the read
`
`light is visible to humans, whereas the write light may not be in the visible
`
`spectrum). The encoding process works because the SLM is photosensitive to the
`
`write light, causing the liquid crystal elements change their state in response to the
`
`write light. This allows the system to display an image through SLM by controlling
`
`the write light. When the read light passes through or reflects offofthe SLM, it takes
`
`on the image encoded by the write light. The SLM is typically not photosensitive to
`
`the read light wavelength (this is essential because otherwise the read light would
`
`
`
`inject noise into the system or ruin the image). The read light is then directed to
`
`project the displayed image to the viewer.
`
`21. A person having ordinary skill in the art in 1995 would consider the
`
`SLM in Takanashi to be an “optically addressed” SLM or “OASLM.” OASLMS are
`
`described in Spatial Light Modulator Technology (XLNX-1015 at 310-31). The
`
`“optical” addressing refers to the fact that the write light controls the image encoded
`
`on the SLM (as opposed to electrically addressed SLM (“EASLMS”), such as those
`
`described in the specification of the ’334 patent, which use electrical circuits to
`
`encode the SLM). Both OASLMS and EASLMS have several similarities. Most
`
`notably, both types of SLMs create an image by controlling a continuous liquid
`
`crystal layer. An OASLM controls the liquid crystal by shining the write light at a
`
`photosensitive material (i.e., a material that takes on an electric charge in response to
`
`light). By shining the write light at some areas and not others, an “image” of electric
`
`charge is created on the photosensitive material. The continuous liquid crystal layer
`
`changes states in response to this charge “image.” Similarly, in an EASLM, electric
`
`circuitry such as transistors and capacitors are used to generate an electric charge
`
`“image.” The continuous liquid crystal layer changes states in response to this
`
`charge “image,” just as in the OASLM. In either instance, visible light passing
`
`through the continuous liquid crystal layer will receive the image encoded by the
`
`electric charge.
`
`
`
`22.
`
`In practical applications, EASLMS and OASLMS are also similar in
`
`that they create images out of rows and columns. In either type of proj ection system,
`
`images are created by organizing the continuous liquid crystal layer into a pixelated
`
`matrix of rows and colunms. Each pixel in the liquid crystal matrix permits or limits
`
`the passage of light according to the electric field near that location. The electric
`
`field is also organized into rows and columns. In a typical EASLM, the liquid crystal
`
`rows and columns correspond to the locations where electric fields are created by the
`
`electrical elements fabricated on or in the glass. In OASLM systems, the liquid
`
`crystal rows and columns correspond to the locations where electric fields are
`
`created by the write light hitting the photosensitive elements.
`
`23.
`
`One common example of a write light for an OASLM is a cathode ray
`
`tube (“CRT”). (XLNX-1015 at 331.) A CRT in an OASLM system works similarly
`
`to a CRT in a standard television. The CRT shines a moving light beam at the
`
`OASLM. This light beam moves in a scanning pattern across the SLM (e.g.,
`
`sweeping from lefi to right across one row, then sweeping across the next row, and
`
`so forth). This method is called “raster scanning.” During the scanning process, the
`
`CRT paints an image made up ofrows and columns. This process is described in the
`
`Tannas reference cited by Mr. Smith-Gillespie in the ’334 IPR. (IV—20l2, Lawrence
`
`E. Tannas, Flat-Panel Displays and CRTs (1985).) Tannas explains that CRT
`
`televisions use a method called “scan addressing” to display a video image. (Id. at
`
`-9-
`
`
`
`24.) To display an image according to a typical NTSC signal, the CRT must scan
`
`over 480 rows and 320 columns, “for a total of 153600 usable addressable pixels.”
`
`(Id.)
`
`24. A prior art video projection system using CRTs to drive OASLMS is
`
`illustrated below:
`
`~
`'
`r» mxmy
`Hmusmmnwaa
`K‘
`V “W “fig ‘
`"$l5|3lW3l<iidIY¢I!}u§I31!tI..(I-‘lfaoztawtluhlvhjltfi
`
`(XLNX-1015 at 555.) This system was built for the U.S. Navy. It was capable of
`
`displaying an image of 1800 rows by 1024 colurrms at 30 frames per second
`
`(interlaced). (Id. at 556.) This system is similar to the Takanashi system in that it
`
`uses read light (from the arc lamp) split, using light filters, into red, green and blue
`
`-10-
`
`
`
`colors before it reaches three light valves (OASLM’s) and write light (from the 3
`
`CRTs, each one dedicated to one of the colors) to create an image using OASLMS.
`
`At any instant oftime, the write light emanating from each CRT is coming primarily
`
`from a single spot on that CRT surface, where the electron beam in that CRT is
`
`striking that surface. One of ordinary skill in the art could then trace the write light
`
`rays leaving that point on the CRT surface through the optics directing that light onto
`
`the OASLM, determining the location ofthe write light spot on the OASLM. One of
`
`ordinary skill in the art would also recognize that this spot of write light on the
`
`OASLM cannot be made infinitely small, but rather has a lower limit on its size
`
`dictated by the optics in the system. This lower limit on spot size dictates how close
`
`together different light spots can be placed, and hence dictates an upper limit on the
`
`number of rows and columns one can obtain using raster scanning techniques. The
`
`results obtained with this system and quoted above show that it was adequate for
`
`video projection.
`
`25.
`
`IV and Mr. Smith-Gillespie make several arguments for why Takanashi
`
`does not disclose a “light-shutter matrix system.” First, they argue that “none of the
`
`elements of Takanashi can be reasonably construed as a ‘matrix system’ of any kind,
`
`much less a ‘light-shutter matrix system?” (Response, Paper No. 26 at 21.) I
`
`disagree. As explained above, in Takanashi, the write light (which is usually from a
`
`CRT) organizes the liquid crystal elements in the Takanashi SLM into an XY matrix
`
`-11-
`
`
`
`of pixels organized into rows and columns. This is a light shutter matrix—the “light
`
`shutter” is the liquid crystal, and the matrix is the XY organization of the liquid
`
`provided by the write light. IV argues that the Takanashi structure is not a matrix
`
`because it uses structures “formed of continuous layers of material, rather than any
`
`‘rectangular arrangement of elements into rows and columns.” (Id. at 23.) I
`
`disagree. As explained above, the write light organizes the liquid crystal layer into a
`
`two-dimensional array of rows and columns. Each pixel in this array is individually
`
`controlled by the write light to limit (or permit) the passage of light. Thus, these
`
`pixels are a “rectangular arrangement of elements capable of limiting the passage of
`
`light” as required by the Board’s construction.
`
`26. Mr. Smith—Gillespie also argues that creating the light shutter matrix
`
`using an OASLM does not result in a “physical” matrix. (XLNX-1013 at
`
`81 :20-83 :2, 84:19-85:19; XLNX—10l4 at 180213-181 :1 l.) I find this argument to be
`
`based on a misunderstanding of what “physical” means. The CRT creates a spot of
`
`light using a physical process and physical components, which alters the physical
`
`properties of the photoconductor and the liquid crystal at a particular spot (i.e., row
`
`and column intersection) on the OASLM, which finally alters the physical process of
`
`read light propagation through the OASLM. The matrix thus created in the OASLM
`
`liquid crystal from all the available rows and columns is every bit as physical as the
`
`liquid crystal matrix created by the arrangement of conducting lines, transistors, and
`
`-12-
`
`
`
`capacitors in an EASLM. Matrices created both ways are physical; the methods for
`
`creating them require different structures to make an SLM, but both structures are
`
`physical.
`
`B.
`
`Lee discloses a video controller
`
`27.
`
`As explained in my opening report, Takanashi does not disclose a video
`
`controller. I agree with Mr. Smith-Gillespie, however, that any practical video
`
`projection system in the mid-1990s would have a video controller adapted for
`
`controlling the light shutter matrices. (XLNX—10l3 at 59:17-60:8; XLNX-1014 at
`
`206:9-212:7.) Circuits 20 and 21 in Lee are one example of a video controller,
`
`because it they include a “liquid crystal display LCD driver 20” and an “image
`
`controlling circuit 21”. (XLNX—1003 at 3:46-52.)
`
`C.
`
`Takanashi discloses equivalent switching matrices
`
`28.
`
`IV and Mr. Smith-Gillespie also assert that Takanashi does not teach
`
`“equivalent switching matrices.” I disagree.
`
`29.
`
`As explained in my opening report, Takanashi teaches to use a separate
`
`SLM for each of the primary colors in the system—red, green, and blue. These
`
`separate SLMs are equivalent switching matrices because they are corresponding or
`
`identical in function or result. They each encode a light image from a write light onto
`
`a read light beam (an identical function and result), where each read-light beam
`
`corresponds to one color.
`
`-13-
`
`
`
`30.
`
`IV and Mr. Smith-Gillespie argue that the red, green, and blue SLMs
`
`are not “equivalen ” because they operate on different color wavelengths. I disagree.
`
`The ’334 patent teaches that each ofthe light-shutter matrices operates on a different
`
`color. (XLNX-1001 at 3:15-18.) Thus, the fact that light-shutter matrices operate on
`
`different colors does not make them non—equivalent.
`
`31.
`
`IV argues that “the specification of the ’334 patent identifies several
`
`advantages that are realized in a system which uses equivalent switching matrices .
`
`.
`
`. over systems such as Takanashi which utilize a ‘color AM-LCD’ configuration.” I
`
`disagree, because Takanashi does not use a “color AM-LCD” configuration; instead,
`
`it uses a triple monochrome LCD structure. I note that Mr. Smith-Gillespie agrees
`
`that this argument is “probably not accurate.” (XLNX-1014 at 221 :2-222:15.)
`
`32.
`
`IV also argues that the three SLMs in Takanashi are not “equiva1en ”
`
`switching matrices because they are not “corresponding in effect or function” as
`
`required by the Board’s construction. (Response, Paper No. 26 at 33.) I disagree.
`
`The SLMS in Takanashi correspond to one another. Each of the Takanashi SLMs
`
`takes in a read light and a write light and uses those lights to create an image. This is
`
`a “corresponding effect or function.” IV also argues that the “equivalent switching
`
`matrices” must be “essentially interchangeable.” (Id. at 34.) I disagree—the ’334
`
`patent does not contain this limitation. I agree with the Board that the Takanashi
`
`SLMS “correspond to each other and, apart from allowing different colors of light
`
`-14-
`
`
`
`(red, green, or blue) to pass through, appear to function in the same manner.”
`
`(Decision, Paper No. 14 at 21.)
`
`33.
`
`IV attempts to distinguish Takanashi on the grounds that it uses
`
`“wavelength selection filters” that remove all wavelengths of light which are not
`
`desired, and that one could not use those wavelength-specific devices in the claims
`
`of the ’334 patent. (Response, Paper No. 26 at 34-36.) This is not a valid distinction,
`
`however, because the claims require wavelength-specific elements, such as “parallel
`
`beams of light of different colors” (claim 1), which can be created by shining white
`
`light through “three color filters (claim 2).” Thus, Takanashi’s use of color—specific
`
`elements is not a distinction; in fact, it satisfies elements of the claims.
`
`34.
`
`Finally, IV argues that Takanashi does not have “equivalent switching
`
`matrices” because the red, green, and blue components in Takanashi are not
`
`completely interchangeable. (Response, Paper No. 26 at 36.) I disagree for two
`
`reasons. First, Takanashi does not teach that the actual liquid crystal elements (i.e.,
`
`the elements that actually block the light) are different in the red, green, or blue
`
`elements. Instead, only the polarizers and/or color filters are different. Thus, in
`
`Takanashi, the actual “switching matrices” (the liquid crystal elements) are the same
`
`across the red, green, and blue components. Second, as discussed above, the
`
`Takanashi “wavelength selection filter” is a combination of color filters, polarizers,
`
`and SLMs. It is this combination that is color specific. The same is true in the ’334
`
`-15-
`
`
`
`patent—the combination of the color filter and the monochrome LCD array is also
`
`color specific. Thus, full interchangeability is not required by either the claims or the
`
`specification. The fact that the Takanashi components perform the same function
`
`and have the same effect on different colors of light is sufficient to make them fall
`
`within the scope of “equivalent switching matrices” in the ’334 patent.
`
`V.
`
`CONCLUSION
`
`35.
`
`For the reasons stated herein, in my deposition testimony, and in my
`
`opening report, it is my opinion that the ’334 patent is invalid as obvious in View of
`
`Takanashi and Lee.
`
`36.
`
`I declare that all statements made herein on my own knowledge are true
`
`and that all statements made on information and belief are believed to be true, and
`
`further, that these statements were made with the knowledge that willful false
`
`statements and the like so made are punishable by fine or imprisonment, or both,
`
`under Section 1001 of Title 18 of the United States Code.
`
`Executed: October 21, 2013
`
`Dr. A. Bruce Buckman
`
`-15-