`____________
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`XILINX, INC.,
`Petitioner
`v.
`QUICKCOMPILE IP, LLC,
`Patent Owner
`____________
`Case IPR2016-00059
`Patent 7,073,158
`____________
`
`PATENT OWNER’S PRELIMINARY RESPONSE TO
`PETITION FOR INTER PARTES REVIEW OF
`U.S. PATENT NO. 7,073,158
`UNDER 35 USC §§ 311-319 AND 37 CFR §42.100 ET SEQ.
`
`
`
`
`
`
`TABLE OF CONTENTS
`
`
`
`INTRODUCTION. .................................................................................. 1
`
`
`I.
`
`II. OVERVIEW OF THE ’158 PATENT. .................................................... 2
`
`
`
`
`
`
`
`
`
`
`III. ARGUMENT ........................................................................................ 4
`
`A. Claim Construction. .............................................................................. 4
`
`B. Petitioner’s Allegations of Unpatentability Over the Combination of
`Banerjee and Benkrid are Insufficient to Meet the Requirements of the
`Claims. ......................................................................................................... 6
`
`C. Petitioner’s Evaluation of the Analyzing Step of Claim 1 is Also
`Deficient. ................................................................................................... 12
`
`D. Petitioner’s Conclusions of Obviousness with Respect to Claim 6
`Should Not Lead to Institution of Trial. .................................................... 15
`
`E. Petitioner’s Conclusions of Obviousness with Respect to Claims 7 and
`8 Should Not Lead to Institution of Trial. ................................................. 18
`
`IV. CONCLUSION. ................................................................................. 19
`
`
`
`
`
`ii
`
`
`
`
`
`TABLE OF AUTHORITIES
`
`CASES
`Abbott Laboratories v. Baxter Pharm. Products, Inc.,
`334 F.3d 1274 (Fed. Cir. 2003) ................................................................. 10
`
`CCS Fitness, Inc. v. Brunswick Corp.,
`288 F.3d 1359 (Fed. Cir. 2002) ................................................................... 6
`
`Corning Incorporated v. DSM IP Assets B.V.,
`Case No. IPR2013-00045 (PTAB May 9, 2014) ....................................... 17
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Google Inc. v. Jongerious Panoramic Techs., LLC,
`Case No. IPR2013-00191 (PTAB Feb. 13, 2014) ....................................... 5
`
`Hartness Int’l. Inc. v. Simplimatic Engineering Co.,
`819 F.2d 1100 (Fed. Cir. 1987) ................................................................. 12
`
`In re Am. Acad. of Sci. Tech. Ctr.,
`367 F.3d 1359 (Fed. Cir. 2004) ................................................................... 5
`
`In re Kotzab,
`217 F.3d 1365 (Fed. Cir. 2000) ................................................................. 17
`
`In re Rouffet,
`149 F.3d 1350 (Fed. Cir. 1998) ................................................................. 17
`
`In re Royka,
`490 F.2d 981 (CCPA 1974) ................................................................. 11, 14
`
`In re Skvorecz,
`580 F.3d 1262 (Fed. Cir. 2009) ................................................................... 4
`
`In re Wilson,
`424 F.2d 1382 (CCPA 1970) ..................................................................... 11
`
`Innova/Pura Water, Inc. v. Safari Water Filtration Sys., Inc.,
`381 F.3d 1111 (Fed. Cir. 2004) ................................................................... 5
`
`iii
`
`
`
`
`
`
`
`
`
`
`
`
`KSR Int’l Co. v. Teleflex Inc.,
`550 U.S. 398 (2007) .................................................................................. 15
`
`Marine Polymer Techs., Inc. v. Hemcon, Inc.,
`672 F.3d 1350 (Fed. Cir. 2012) ................................................................... 5
`
`Phillips v. AWH Corp.,
`415 F.3d 1303 (Fed. Cir. 2005) ................................................................... 5
`
`Renishaw PLC v. Marposs Societa’per Azioni,
`158 F.3d 1243 (Fed. Cir. 1998), .................................................................. 5
`
`Rhine v. Casio, Inc.,
`183 F.3d 1342 (Fed. Cir. 1999) ................................................................... 5
`
`Samsung Elecs. Co., Ltd. v. Va. Innov. Scis., Inc.,
`Case No. IPR2013-00569 (PTAB Oct. 30, 2013) ....................................... 6
`
`Stumbo v. Eastman Outdoors, Inc.,
`508 F.3d 1358 (Fed. Cir. 2007) ................................................................. 11
`
`Wright Medical Technology Inc. v. Orthopheonix, LLC,
`Case No. IPR2014-00912 (PTAB Dec. 16, 2014) ............................... 12, 15
`
`STATUTES
`35 U.S.C. § 314(a) .......................................................................................... 1
`
`
`
`REGULATIONS
`37 C.F.R. § 1.75(d)(1) ..................................................................................... 6
`
`37 C.F.R. § 42.100(b) ..................................................................................... 4
`
`37 C.F.R. § 42.108(c).................................................................................. 1, 7
`
`
`
`
`
`37 C.F.R. § 42.65(a)...................................................................................... 17
`
`
`
`iv
`
`
`
`
`
`
`
`EXHIBIT LIST
`
`Description
`
`Excerpts from Microsoft Computer Dictionary, 5th ed.,
`Microsoft Corp. (2000): (a) analyze (p. 27), (b) compile
`(p. 115), (c) parse (p. 127).
`
`v
`
`
`Exhibit No.
`
`2001
`
`
`
`
`
`
`
`I. INTRODUCTION.
`
`The Patent Trial and Appeal Board (“PTAB” or “Board”) should deny
`
`Petitioner’s request for inter partes review of claims 1-20 of US Pat.
`
`7,073,158 (the “’158 Patent”), because Petitioner has not met its burden to
`
`show a reasonable likelihood that it would prevail with respect to at least one
`
`of the challenged claims.1 For example, although claim 1 recites, “accepting
`
`a user-defined algorithm specified in a source code of a high level language
`
`and designed to process data vectors with one, two, or more dimensions,”
`
`Ex. 1001 at 18:57-59, Petitioner’s analysis, analysis ignores any user-
`
`defined algorithm and focuses instead on the high-level language that is
`
`“designed to process a data vector with at least one dimension . . . .” Pet. at
`
`17. The mere existence of a high level language capable of processing data
`
`vectors does not teach the existence of a user-defined algorithm that
`
`comports with the limitations of the claim. Accordingly, Petitioner’s analysis
`
`
`
`1 35 U.S.C. § 314(a) (An inter partes review may be instituted only if “the
`
`information presented in the petition . . . and any response . . . shows that
`
`there is a reasonable likelihood that the petitioner would prevail with respect
`
`to at least 1 of the claims challenged in the petition.”); 37 C.F.R. §
`
`42.108(c).
`
`
`
`1
`
`
`
`of claim 1 (which affects all of its challenges to all of the claims) is
`
`inadequate and the Board should deny institution of inter partes review.
`
`
`
`II. OVERVIEW OF THE ’158 PATENT.
`
`The ’158 Patent describes and claims methods for programming a
`
`field programmable gate array (“FPGA”). Ex. 1001 at 1:18-26. More
`
`specifically, the patent teaches methods of programming an FPGA “to
`
`implement a desired algorithm for processing data vectors with one, two or
`
`more [ ] dimensions.” Id. at 3:3-4. As illustrated in Fig. 7, reproduced below,
`
`this process involves taking the algorithm (as expressed in a high-level
`
`source code language) and “merging” it with existing “gateware” to produce
`
`one or more low-level files (hardware-gate-programming or HGP files) for
`
`programming the FPGA using conventional software tools. Id. at 3:34-41.
`
`The HGP files contain the code that specifies the required connections to be
`
`programmed into the target FPGA, allowing programming of the FPGA in a
`
`
`
`2
`
`
`
`fully automated manner. Id. at 3:54-67.
`
`As explained in the ’158 Patent, the source code describes the
`
`intended operation of the FPGA. Id. at 9:56-57; 11:16-23. The source code is
`
`written in a high-level programming language, id. at 9:64-67, and an
`
`analyzer module processes the source code to identify vector elements and
`
`provide information concerning the overall vector processing flow and the
`
`order and dependencies of the procedures or operations required to
`
`implement the algorithm. Id. at 10:2-14; 11:30-36. From this, a mapper
`
`module generates a mapping of the necessary gates or logic components
`
`onto the FPGA to accomplish the required functions. Id. at 10:15-21; 11:40-
`
`48. The output of the mapper module specifies configuration, routing, and
`
`location of FPGA logic bocks necessary to implement the original algorithm,
`
`id. at 11:49-55, which information is used to produce a bit stream that will
`
`ultimately program the target FPGA. Id. at 11:55-60.
`
`Claim 1 is the sole independent claim of the ’158 Patent, and it reads
`
`as follows:
`
`1. A method for programming a field programmable gate
`array (FPGA) comprising the following steps:
`accepting a user-defined algorithm specified in
`a source code of a high level language and designed to
`process data vectors with one, two, or more
`
`3
`
`
`
`
`
`dimensions;
`analyzing the user-defined algorithm, including
`identifying the vector processing operations of the
`source code;
`mapping the vector processing operations onto
`logic components of an FPGA; and
`programming the FPGA with the user-defined
`algorithm based on the mapping of the logic
`components.
`
`III. ARGUMENT
`A. Claim Construction.
`
`
`
`In the arguments below, Patent Owner distinguishes the claims over
`
`the art cited by Petitioner and, in doing so, specifies constructions of certain
`
`claim terms. Patent Owner notes that the Board has indicated that it will
`
`interpret the claims of a challenged patent using a “broadest reasonable
`
`construction” approach. Office Patent Trial Practice Guide, 77 Fed. Reg.
`
`48756, 48766 (Aug. 14, 2012); and see 37 C.F.R. § 42.100(b). However,
`
`this standard is an examination expedient, not a rule of claim construction,
`
`In re Skvorecz, 580 F.3d 1262, 1267 (Fed. Cir. 2009). Inasmuch as the
`
`present proceeding is adjudicatory in nature, and is not an examination or
`
`even a reexamination of the challenged claims, see, e.g., Google Inc. v.
`
`
`
`4
`
`
`
`Jongerious Panoramic Techs., LLC, Case No. IPR2013-00191, Paper No.
`
`50, slip op. at 4 (PTAB Feb. 13, 2014), the claims should be construed as in
`
`other adjudicatory proceedings: to give the claims the meaning they would
`
`have to a person of ordinary skill in the art at the time of the invention,
`
`Innova/Pura Water, Inc. v. Safari Water Filtration Sys., Inc., 381 F.3d 1111,
`
`1116 (Fed. Cir. 2004), to cover what was actually invented and what the
`
`inventor intended them to cover, Renishaw PLC v. Marposs Societa’per
`
`Azioni, 158 F.3d 1243 (Fed. Cir. 1998), and, where possible, to preserve
`
`their validity. Marine Polymer Techs., Inc. v. Hemcon, Inc., 672 F.3d 1350,
`
`1368 (Fed. Cir. 2012); Rhine v. Casio, Inc., 183 F.3d 1342, 1345 (Fed. Cir.
`
`1999); and see Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005) (en
`
`banc).
`
`Even if the Board applies the broadest reasonable construction
`
`standard, it is important to recognize that such an interpretation of a claim’s
`
`language is not one that permits any reading thereof. Instead, it is one that
`
`must be made “in light of the specification as it would be interpreted by one
`
`of ordinary skill in the art.” In re Am. Acad. of Sci. Tech. Ctr., 367 F.3d
`
`1359, 1364 (Fed. Cir. 2004). For example, unless the patentee has clearly
`
`demonstrated an intention to stray, there is a “heavy presumption” that a
`
`claim term carries its ordinary and customary meaning. CCS Fitness, Inc. v.
`
`
`
`5
`
`
`
`Brunswick Corp., 288 F.3d 1359, 1366 (Fed. Cir. 2002). Furthermore, patent
`
`claims must “conform to the invention as set forth in the remainder of the
`
`specification and the terms and phrases used in the claims must find clear
`
`support or antecedent basis in the description so that the meaning of the
`
`terms in the claims may be ascertainable by reference to the description.” 37
`
`C.F.R. § 1.75(d)(1).2
`
`
`
`B. Petitioner’s Allegations of Unpatentability Over the Combination
`of Banerjee and Benkrid are Insufficient to Meet the
`Requirements of the Claims.
`
`Claim 1 recites, “accepting a user-defined algorithm specified in a
`
`source code of a high level language and designed to process data vectors
`
`with one, two, or more dimensions.” Ex. 1001 at 18:57-59. The literal
`
`language of the claim thus requires that it be the user-defined algorithm, not
`
`merely a high level language, that is designed to process data vectors with
`
`one, two or more dimensions. Such an understanding is consistent with the
`
`teachings provided in the specification: “the present invention pertains to an
`
`
`
`2 Patent Owner reserves the right to pursue claim constructions in a district
`
`court according to the standards applicable in that venue. See Samsung
`
`Elecs. Co., Ltd. v. Va. Innov. Scis., Inc., Case No. IPR2013-00569, Paper 9,
`
`slip op. at 2 (PTAB Oct. 30, 2013).
`
`
`
`6
`
`
`
`automated system for programming field programmable gate arrays
`
`(FPGAs) to implement a desired algorithm for processing data vectors with
`
`one, two or more of the dimensions.” Ex. 1001 at 3:1-4 (emphasis added).
`
`Accordingly, Petitioner’s focus on a compiler that that can process data
`
`vectors, as opposed to the user-defined algorithm, Pet. at 17, misses the
`
`mark, and cannot serve as a basis for concluding that Petitioner has
`
`demonstrated a reasonable likelihood that at least one of the claims
`
`challenged in the petition is unpatentable. Thus, no inter partes review
`
`should be instituted. 37 C.F.R. § 42.108(c).
`
`In both claim 1 and the specification, the inventors specified that it is
`
`the user-defined algorithm, not the high level language in which it is
`
`expressed, that is “for processing data vectors with one, two or more of the
`
`dimensions.” This distinction is important to understanding the underlying
`
`nature of the invention claimed by the ‘158 Patent, and the failings of the
`
`instant petition. A key benefit of the invention is its ability to “speed up the
`
`process of implementing and testing a fully written high-level user-defined
`
`algorithm to a matter of a few minutes, rather than the days, weeks or even
`
`months presently required using conventional software tools.” Id. at Abstract
`
`(emphasis added). The “conventional software tools” contemplated by the
`
`‘158 Patent (including programming languages like MATLAB), do not, in
`
`
`
`7
`
`
`
`and of themselves, facilitate this advance and herein lies Petitioner’s failure.
`
`Rather than address the invention recited in the claims, Petitioner instead
`
`addresses capabilities of programming languages, MATLAB in particular, to
`
`express certain operations. Pet. at 17-18.
`
`The ability of a programing language to make use of pre-defined
`
`functions or other constructs does not, by itself, meet the claims. Rather, it is
`
`the complexity captured by the user-defined algorithm as disclosed and
`
`claimed that must eventually be reflected in the programming of an FPGA
`
`through the inventive method of the ‘158 Patent. While there may be
`
`programming languages that could be used as the medium in which the user-
`
`defined algorithm is expressed, the available functions of that medium do
`
`not, by themselves, meet the claim. Instead, it is how that medium is
`
`employed. The Petition nowhere addresses this aspect of the claims and
`
`instead sidesteps it by focusing on the programming language.
`
`According to Petitioner,
`
`Banerjee teaches accepting the MATLAB source code.
`
`Further, Banerjee teaches that a user-defined algorithm is
`specified in the MATLAB source code (a high-level
`language) which is designed to process data vectors with
`at least one dimension. A person of ordinary skill in the
`art would have understood that a two-dimensional
`
`8
`
`
`
`
`
`matrix, as disclosed by Banerjee, is a two-dimensional
`vector.
`
`Thus, Banerjee’s disclosure of the MATCH Compiler
`receiving as input a user defined algorithm specified in
`MATLAB source code designed to process a data vector
`with at least one dimension teaches or suggests this
`limitation of claim 1.
`
`Pet. at 16-17 (citations omitted) (emphasis added). Notably, Petitioner does
`
`not cite Benkrid for any teachings with respect to this feature of claim 1.
`
`
`
`Assuming, solely for sake of argument, that Petitioner has correctly
`
`interpreted Banerjee, at most this demonstrates that Banerjee describes
`
`source code of a high level language designed to process data vectors with
`
`one dimension. Even if one credits Petitioner’s allegation that “A person of
`
`ordinary skill in the art would have understood that a two-dimensional
`
`matrix, as disclosed by Banerjee, is a two-dimensional vector,” id. at 17,
`
`then, at best, Petitioner could be said to alleged that Banerjee describes
`
`source code of a high level language designed to process data vectors with
`
`one or two dimensions. Claim 1 requires something else: “accepting a user-
`
`defined algorithm specified in a source code of a high level language and
`
`designed to process data vectors with one, two or more dimensions.” Ex.
`
`1001 at 18:57-60 (emphasis added). Indeed, this requirement is carried all
`
`
`
`9
`
`
`
`the way through claim 1, which concludes with: “programming the FPGA
`
`with the user-defined algorithm based on the mapping of the logic
`
`components,” id. at 18:64-65 (emphasis added), reiterating that it is the user-
`
`defined algorithm, and not the high-level language, that underpins the
`
`invention as claimed. Petitioner’s analysis does not address this requirement.
`
`It is also important to recognize that claim 1 does not recite a
`
`Markush group. “A Markush group is a listing of specified alternatives of a
`
`group in a patent claim, typically expressed in the form: a member selected
`
`from the group consisting of A, B, and C.” Abbott Laboratories v. Baxter
`
`Pharm. Products, Inc., 334 F.3d 1274, 1280 (Fed. Cir. 2003) (citing
`
`M.P.E.P. § 2173.05(h)). “It is well known that ‘members of the Markush
`
`group are . . . alternatively usable for the purposes of the invention” and
`
`thus, “‘members of the Markush group are used singly.’” Id. at 1280-81
`
`(citing In re Driscoll, 562 F.2d 1245, 1249 (CCPA 1977)). That is not the
`
`situation presented in claim 1. For example, in claim 1 the recited data
`
`vectors which the user defined algorithm is designed to process are not
`
`“alternatively useable” species that may be used “singly”—rather, the
`
`recited data vectors represent a solitary classification.
`
`A contrary construction would render claim terms superfluous. For
`
`example, if one construed claim 1 to mean a user-defined algorithm
`
`
`
`10
`
`
`
`specified in a source code of a high level language and designed to process
`
`data vectors with only one dimension, then the words two, or more
`
`dimensions would be rendered superfluous. The same would be true if the
`
`claim were construed to mean a user-defined algorithm specified in a source
`
`code of a high level language and designed to process data vectors with only
`
`two dimensions, and so on. Such a practice is disfavored. Stumbo v. Eastman
`
`Outdoors, Inc., 508 F.3d 1358, 1362 (Fed. Cir. 2007).
`
`All words in a claim must be considered in determining patentability.
`
`In re Wilson, 424 F.2d 1382, 1385 (CCPA 1970). Claim 1 does not include
`
`language which signals that only one algorithm species would satisfy it.
`
`Therefore, claim 1 must be read so that the entirety of the phrase a user-
`
`defined algorithm designed to process data vectors with one, two, or more
`
`dimensions is given meaning. Although the dimensionality of the data
`
`vectors may vary, the claim specifies that the user-defined algorithm must
`
`have the capacity to process data vectors with one, two or more dimensions.
`
`Petitioner’s analysis, even viewed in its best possible light, fails to
`
`address these actual requirements of claim 1. Absent such an explanation,
`
`there can be no prima facie showing of obviousness. In re Royka, 490 F.2d
`
`981, 985 (CCPA 1974) (obviousness requires a suggestion of all limitations
`
`in a claim). This failure to establish a clear rational for the conclusion of
`
`
`
`11
`
`
`
`obviousness warrants denial of Petitioner’s challenge. Wright Medical
`
`Technology Inc. v. Orthopheonix, LLC, Case No. IPR2014-00912, Paper 9,
`
`slip op. at 9-12 (PTAB Dec. 16, 2014).
`
`Claims 2, 4, 9, 16, and 18-20 depend from claim 1 and, for at least the
`
`same reasons as claim 1, Petitioner has failed to demonstrate that any of
`
`these claims are unpatentable in view of the combination of Banerjee and
`
`Benkrid. Hartness Int’l. Inc. v. Simplimatic Engineering Co., 819 F.2d 1100,
`
`1108 (Fed. Cir. 1987) (a claim that depends from a nonobvious independent
`
`claim is nonobvious because it contains all of the limitations of that
`
`independent claim plus a further limitation). Indeed, because all of
`
`Petitioner’s proposed grounds for institution of trial with respect to claims 1-
`
`20 rest on Petitioner’s faulty analysis of claim 1, all of Petitioner’s grounds
`
`for institution of trial should be denied.
`
`
`
`C. Petitioner’s Evaluation of the Analyzing Step of Claim 1 is Also
`Deficient.
`
`Claim 1 recites, “analyzing the user-defined algorithm.” Ex. 1001 at
`
`18:60. According to the specification, analyzing includes parsing, id. at
`
`9:62-64, identifying vector elements, id. at 10:2-3, providing data-flow
`
`information, id. at 10:4-6, and identifying operators and vector operands to
`
`determine the order and dependencies of procedures and operations of the
`
`
`
`12
`
`
`
`user-defined algorithm. Id. at 10:9-14. The data-flow specifies sequence and
`
`timing information from which necessary processing operations and
`
`mathematical operations can be identified and extracted. Id. at 11:30-33.
`
`Petitioner contends that the analyzing step can be met solely by
`
`“Banerjee’s teaching of compiling and parsing the MATLAB source code
`
`[to] identify [ ] vector processing operations of the MATLAB source code.”
`
`Pet. at 19 (emphasis added). Although the ’158 Patent does indicate that
`
`analysis includes parsing, Ex. 1001 at 9:62-64, and identifying vector
`
`elements, id. at 10:2-3, and operands, id. at 10:9-14; the specification also
`
`teaches that analysis includes more than these functions, id. at 9:62-64; 10:2-
`
`14; 11:30-33. Petitioner fails to provide any convincing argument that
`
`compiling and parsing alone are sufficient to teach such “analyzing.”
`
`Reference materials from the time of the invention provide that
`
`“analysis” is “The evaluation of a situation or problem, including review
`
`from various aspects or points of view. In computing, analysis commonly
`
`involves such features as flow control, error control, and evaluation of
`
`efficiency. Often the overall problem is divided into smaller components that
`
`can more easily be dealt with.” Ex. 2001 at 3. Parsing, in contrast, is merely
`
`“break[ing] input into smaller chunks so that a program can act upon the
`
`information,” id. at 4, and compiling is “translat[ing] all the source code of a
`
`
`
`13
`
`
`
`program from a high level language into object code prior to execution of
`
`the program.” Id. at 5. Thus, while parsing (breaking input into smaller
`
`chunks so that a program can act upon the information) and compiling
`
`(translating source code from a high level language into object code) may be
`
`components of analyzing, indeed, the specification of the ’158 Patent
`
`provides as much, analyzing involves more than just these two activities.
`
`Even Petitioner admits that the combination of Banerjee and Benkrid do not
`
`teach or suggest all aspects of analyzing the user-defined algorithm, and
`
`cites Haldar in an attempt to show identifying the orders and dependencies
`
`of vector operations. Pet. at 43-44.
`
`Thus, Petitioner has failed to demonstrate that the proposed combination
`
`of Banerjee and Benkrid teach or suggest the subject matter recited in claim
`
`1, and, therefore, have failed to make out a prima facie case of obviousness.
`
`In re Royka, 490 F.2d at 985. Accordingly, Petitioner has not met its burden
`
`to show a reasonable likelihood that it would prevail with respect to at least
`
`one of the challenged claims, and no inter partes review should be instituted
`
`on the proposed ground.
`
`
`
`
`
`14
`
`
`
`D. Petitioner’s Conclusions of Obviousness with Respect to Claim 6
`Should Not Lead to Institution of Trial.
`
`Petitioner contends that claim 6 is unpatentable in view of the
`
`combination of Banerjee, Benkrid, Haldar, and Hammes. Pet. at 44 et seq.
`
`Rather than support this contention with evidence, however, Petitioner
`
`merely states the conclusion of obviousness, alleging that, “A person of
`
`ordinary skill in the art would have understood that the final DAG [of
`
`Benkrid], with all attributes, indicates the overall flow of vector processing
`
`results between vector operations.” Id. at 44. Conclusory statements such as
`
`this are an insufficient basis upon which to institute trial. See, e.g., Wright
`
`Medical Technology Inc., Case No. IPR2014-00912, Paper 9, slip op. at 9-
`
`12; KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 415-421 (2007)
`
`(recognizing that the rationale to support a conclusion of obviousness should
`
`be made explicit). Moreover, Petitioner cannot rely on Dr. Walker’s equally
`
`conclusory analysis, which does nothing more than repeat the petition’s
`
`argument. Indeed, when one reviews the petition against Dr. Walker’s
`
`declaration, the lack of any explanatory or expository analysis becomes
`
`clear:
`
`Petition at 44-45
`As discussed above in
`connection with claim 5, the
`
`
`
`Walker Declaration ¶ 188
`
` As discussed above in connection with
`claim 5, the combination of Banerjee,
`
`15
`
`
`
`combination of Banerjee,
`Benkrid, and Haldar teaches or
`suggests analyzing the
`MATLAB program and
`constructing an operational data
`flow graph. Benkrid further
`teaches generating “a final DAG
`with all attributes.” Benkrid at 3;
`see also Benkrid at Fig. 3
`(reproduced above, showing the
`final DAG is provided to the
`code generator). A person of
`ordinary skill in the art would
`have understood that the final
`DAG, with all attributes,
`indicates the overall flow of
`vector processing results
`between vector operations.
`
`
`
`Benkrid, and Haldar teaches or
`suggests analyzing the MATLAB
`program, generating a syntax tree, and
`constructing an operational data flow
`graph. Benkrid further teaches
`generating “a final DAG with all
`attributes.” (Ex. 1006) Benkrid at 3;
`see also id. at Fig. 3 (reproduced
`above, showing the final DAG is
`provided to the code generator). A
`person of ordinary skill in the art
`would have understood that the final
`DAG, with all attributes, indicates the
`overall flow of vector processing
`results between vector operations.
`Also, as the ‘158 Patent recognizes, it
`was known in the art that data flow
`graphs indicate all operations, which is
`understood to include the overall flow
`of vector processing results between
`vector operations. In other words, this
`analyzer will construct a sequence and
`timing diagram known in the art as an
`operation data flow graph from which
`all necessary image processing and
`mathematical operations can be readily
`
`16
`
`
`
`identified and extracted. (Ex. 1001)
`‘158 Patent at 11:25-28.
`
`37 C.F.R. § 42.65(a); and see Corning Incorporated v. DSM IP Assets B.V.,
`
`Case No. IPR2013-00045, Paper 92, slip op. at 14-15 (PTAB May 9, 2014)
`
`(Board accords little weight to expert’s “bare opinion” which “repeats,
`
`verbatim, attorney argument set forth in the petition, but identifies no
`
`objective evidence” explaining the basis for the opinion.).
`
`
`
`Even Petitioner’s further argument concerning Hammes offers no
`
`explanation of why if would have been obvious for the person of ordinary
`
`skill in the art to adopt Hammes’ data dependence and control flow (DDCF)
`
`graph to specify the overall flow of vector processing results between vector
`
`processing operations, as recited in claim 6. A conclusion that using such a
`
`graph is “nothing more than applying a known technique of generating a
`
`data flow graph that indicates the overall flow as taught by Hammes, to the
`
`system of Banerjee, to predictably implement complete algorithms on an
`
`FPGA,” Pet. at 46, whether stated in the Petition or in a purported expert’s
`
`declaration, is not a substitute for a reasoned analysis as to why the person of
`
`ordinary skill would adopt such an approach. See In re Kotzab, 217 F.3d
`
`1365, 1371 (Fed. Cir. 2000); In re Rouffet, 149 F.3d 1350, 1359 (Fed. Cir.
`
`1998). Accordingly, Petitioner’s arguments concerning the obviousness of
`
`
`
`17
`
`
`
`claim 6 are insufficient to warrant institution of trial on the proposed
`
`grounds.
`
`
`
`E. Petitioner’s Conclusions of Obviousness with Respect to Claims 7
`and 8 Should Not Lead to Institution of Trial.
`
`Like claim 6, Petitioner addresses claim 7 in only a conclusory
`
`fashion. Rather than explaining how the cited portions of Banerjee and
`
`Benkrid support the conclusions of obviousness, Petitioner instead recites
`
`only selected quotes from those references and presents the conclusion of
`
`obviousness as the premise for finding the same. Pet. at 42-44. For example,
`
`with respect to claim 7, Petitioner quotes Banerjee for “the compiler see[ing]
`
`a MATLAB statement a = b * c,” Pet. at 42 citing Banerjee at 3, but fails to
`
`articulate why the cited passage teaches the elements of the challenged
`
`claim. To the extent that the petition cites Dr. Walker’s declaration, the
`
`declaration merely repeats the allegations of the petition without further
`
`explanation. As previously recognized, this sort of conclusory analysis (and
`
`incorporation by reference) is insufficient to warrant institution of trial.
`
`
`
`Petitioner’s analysis of claim 8 is equally deficient and again merely
`
`repeats statements excerpted from the cited reference Haldar without
`
`explaining how those statements demonstrate the points being advocated.
`
`Dr. Walker’s declaration repeats these statements (albeit with longer
`
`
`
`18
`
`
`
`excerpts from Haldar), but offers no further analysis. As with claims 6 and 7,
`
`these arguments are insufficient to warrant institution of trial.
`
`
`
`IV. CONCLUSION.
`
`For at least the foregoing reasons, Patent Owner respectfully submits that
`
`all of the grounds presented in the Petition should be denied. Further, as this
`
`is Patent Owner’s Preliminary Response, it is not a comprehensive rebuttal
`
`to all arguments raised by the Petition. If a trial is instituted, Patent Owner
`
`reserves the right to contest the Petition on all grounds instituted by the
`
`Board. Moreover, nothing herein should be construed as a concession or
`
`admission by Patent Owner as to any fact or argument proffered in the
`
`Petition.
`
`
`
`
`
`
`
`
`
`
`
`Respectfully submitted,
`
`
`
`
`
`
`
`
`Dated: February 1, 2016
`
`
`
`
`
`
`
`
`
`
`Ascenda Law Group, PC
`333 W. San Carlos St., Suite 200
`San Jose, CA 95110
`
`Tel: 866-877-4883
`Fax: 408-773-6177
`Email: tarek.fahmi@ascendalaw.com
`
`/Tarek N. Fahmi/
`Tarek N. Fahmi
`Reg. No. 41,402
`
`
`
`
`
`
`
`
`
`
`
`19
`
`
`
`CERTIFICATE OF SERVICE
`The undersigned hereby certifies that a copy of the foregoing
`
`PATENT OWNER’S PRELIMINARY RESPONSE
`
`was served on February 1, 2016, by filing this document though the Patent
`
`Review Processing System to the attorneys of record for the Petitioner at the
`
`following address:
`
`David M. O’Dell
`David L. McCombs
`Henry L. Welch
`Gregory P. Huh
`HAYNES AND BOONE, LLP
`2323 Victory Avenue, Suite 700
`Dallas, Texas 15819
`david.odell.ipr@haynesboone.com
`david.mccombs.ipr@haynesboone.com
`henry.welch.ipr@haynesboone.com
`gregory.huh.ipr@haynesboone.com
`
`Jeffrey E. Danley
`SIMPSON THACHER & BARTLETT LLP
`2475 Hanover Street
`Palo Alto, CA 94304
`jdanley@stblaw.com
`
`
`
`
`
`
`
`
`
`
`
`Respectfully submitted,
`
`
`
`
`
`Dated: February 1, 2016
`
`
`
`
`
`
`
`Ascenda Law Group, PC
`333 W. San Carlos St., Suite 200
`San Jose, CA 95110
`Tel: 866-877-4883
`Email: patents@ascendalaw.com
`
`
`
`/Tarek N. Fahmi/
`Tarek N. Fahmi
`Reg. No. 41,402
`
`
`
`
`
`
`
`20