`Date: April 27, 2015
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`LINEAR TECHNOLOGY CORPORATION
`Petitioner
`
`v.
`
`IN-DEPTH TEST LLC
`Patent Owner
`
`
`
`Case IPR2015–00421
`Patent 6,792,373 B2
`
`
`
`IN-DEPTH TEST LLC’S PRELIMINARY RESPONSE
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`I.
`II.
`III.
`
`TABLE OF CONTENTS
`INTRODUCTION .......................................................................................................... 1
`GROUNDS ..................................................................................................................... 3
`CLAIM CONSTRUCTION .......................................................................................... 4
`A.
`Petitioner Relies upon Expert Who Uses the Wrong Standard ....... 4
`B.
`Proposed Terms .............................................................................................. 5
`1.
`“outlier” (claims 1, 6–8, 11, 13–16 and 18) ...................................... 5
`a. Claims Define “Outlier” as a Test Result, Not the
`Tested Component ......................................................................... 5
`b. The ’373 Patent Explicitly Defines “Outlier” as a
`Measured Value that Statistically Strays from a Data
`Set ........................................................................................................ 6
`c. Petitioner’s Own References Support Patent Owner’s
`Definition ......................................................................................... 10
`d. Petitioner’s Support Consists of a Misinterpretation
`of Figure 9 and Misleading Quotation ................................... 10
`e. Petitioner Uses Patent Owner’s Definition in
`Argument ........................................................................................ 14
`f. Prior Claim Construction and Inventor Agree with
`Patent Owner ................................................................................. 14
`g. Petitioner’s Litigation Codefendant Agrees with
`Patent Owner ................................................................................. 15
`h. The Correct Definition of “Outlier” .......................................... 16
`“test data” (claims 1, 3–5, 7–8, 10–12, 14–15 and 17–20) ........ 16
`
`2.
`
`i
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`3.
`
`4.
`5.
`6.
`7.
`
`8.
`
`“output report [including the identified outlier]” (Claims
`1 and 8) ................................................................................................. 18
`“[identify the outlier at] run time” (Claims 6, 13 and 15) ....... 21
`“component” (Claims 1, 3, 10, 15 and 17) ................................... 22
`“recipe file” (Claims 2, 9 and 16) .................................................. 23
`“section group [of components on a wafer]” (Claims 3,
`10 and 17) ............................................................................................ 24
`“correlate/correlating the test data” (Claims 5, 12 and
`20) .......................................................................................................... 25
`EKSTEDT IN LIGHT OF GHAEMI FAILS TO RENDER OBVIOUS
`INDEPENDENT CLAIMS ......................................................................................... 26
`A.
`Claim 1 (“A computer . . . configured to . . . identify an outlier
`in the test data”); Claim 8 (“a supplementary data analysis
`element configured to identify outliers in the semiconductor
`test data”); Claim 15 (“automatically identifying an outlier in
`the test data”) ................................................................................................ 26
`Claim 1 (“a computer . . . configured to . . . generate an
`output report including the identified outlier”); Claim 8
`(“output element configured to generate an output report
`including the identified outliers”) ............................................................ 28
`Ghaemi Fails to Correct Ekstedt’s Missing Limitations ...................... 31
`C.
`O’NEILL FAILS TO ANTICIPATE ANY OF THE ’373 PATENT CLAIMS ....... 35
`A.
`Claim 1 (“A computer connected to the tester and configured
`to . . . identify an outlier in the test data”); Claim 8 (“a
`supplementary data analysis element configured to identify
`outliers in the semiconductor test data”); Claim 15
`(“automatically identifying an outlier in the test data”) ................... 35
`
`IV.
`
`V.
`
`B.
`
`ii
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`B.
`
`C.
`
`D.
`
`E.
`
`Claim 1 (“a computer . . . configured to . . . generate an
`output report including the identified outlier”); Claim 8
`(“output element configured to generate an output report
`including the identified outliers”) ............................................................ 40
`Claim 8 (“supplementary data analysis element configured to
`identify outliers in the semiconductor test data”; “output
`element configured to generate an output report including
`the identified outliers.”) .............................................................................. 42
`Claim 15 (“identifying an outlier in the test data at run time”);
`Claim 6 (“the computer is configured to identify the outlier at
`run time”); Claim 13 (“supplementary data analysis element is
`configured to identify the outliers at run time”) ................................ 44
`DEPENDENT CLAIMS ARE NONOBVIOUS OVER SIX
`REFERENCES .................................................................................................. 45
`1.
`Claim 2 (“the computer is configured to operate in
`conjunction with a set of configuration data in a recipe
`file”); Claim 9 (“Supplementary data analysis element is
`configured to operate in conjunction with a set of
`configuration data in a recipe file”); Claim 16
`(“identifying the outlier according to the configuration
`data in the recipe file”) ................................................................... 46
`Claim 3 (“test data corresponds to a section group of
`components on a wafer”); Claim 10 (same); Claim 17
`(same) ................................................................................................... 48
`Claim 4 (“computer is configured to automatically
`calibrate a sensitivity of the computer to the test data”);
`Claim 11 (“supplementary data analysis element is
`configured to automatically calibrate a sensitivity of the
`outlier identification element to the test data”); Claim
`
`2.
`
`3.
`
`iii
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`4.
`
`18 (“calibrating a sensitivity for identifying the outlier in
`the test data using the computer system.”) .............................. 51
`Claim 5 (“computer further comprises a data
`correlation element configured to correlate the test
`data”); Claim 12 (“supplementary data analysis element
`includes a data correlation element configured to
`correlate the test data”); Claim 20 (“correlating the test
`data to identify similarities in the test data using the
`computer system”) ........................................................................... 52
`a. O’Neill .............................................................................................. 52
`b. Ekstedt ............................................................................................. 54
`c. La ....................................................................................................... 54
`d. Ghaemi ............................................................................................ 56
`Claim 7 (“a data smoothing element configured to
`receive the test data and smooth the test data, and
`wherein the computer is configured to receive the
`smoothed test data and identify the outlier in the
`smoothed test data”); Claim 14 (“a data smoothing
`element configured to receive the test data and
`smooth the test data, and wherein the supplementary
`data analysis element is configured to identify the
`outliers in the smoothed test data”); Claim 19
`(“smoothing the test data using the computer system.”) ... 57
`a. O’Neill .............................................................................................. 57
`b. Ekstedt ............................................................................................. 58
`c. Daasch ............................................................................................. 58
`CONCLUSION ........................................................................................................... 59
`
`5.
`
`VI.
`
`iv
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`TABLE OF AUTHORITIES
`CASES
`01 Communique Lab., Inc. v. LogMeIn, Inc.,
`687 F.3d 1292 (Fed. Cir. 2012) ...................................................................................... 42
`Am. Piledriving Equip., Inc. v. Geoquip, Inc.,
`637 F.3d 1324 (Fed. Cir. 2011) ...................................................................................... 32
`Arthur A. Collins, Inc. v. N. Telecom Ltd.,
`216 F.3d 1042 (Fed. Cir. 2000) ....................................................................................... 10
`Baldwin Graphic Sys., Inc. v. Siebert, Inc.,
`512 F.3d 1338 (Fed. Cir. 2008) ...................................................................................... 53
`Bettcher Indus., Inc. v. Bunzl USA, Inc.,
`661 F.3d 629 (Fed. Cir. 2011) .......................................................................................... 31
`Broadcom Corp. v. Emulex Corp.,
`732 F.3d 1325 (Fed. Cir. 2013) ...................................................................................... 47
`Continental Can Co. v. Monsanto Co.,
`948 F.2d 1264 (Fed. Cir. 1991) ...................................................................................... 53
`Kinetic Concepts, Inc. v. Smith & Nephew, Inc.,
`688 F.3d 1342 (Fed. Cir. 2012) ...................................................................................... 48
`Merrill v. Yeomans,
`94 U.S. 568 (1876) .............................................................................................................. 6
`Phillips v. AWH Corp.,
`415 F.3d 1303 (Fed. Cir. 2005) (en banc) .................................................................... 6
`Vitronics Corp. v. Conceptronic,
`90 F.3d 1576 (Fed. Cir. 1996) ...........................................................................................7
`PATENT AND TRADEMARK APPEALS BOARD
`Ex parte Grafenauer,
`No. 2010–001906, pp. 8–9 (BPAI Mar. 2, 2012) ...................................................... 43
`Fairchild Semiconductor Corp. v. In-Depth Test LLC,
`No. CBM2015–00060, Paper 1 (PTAB Jan. 13, 2015) ............................................... 15
`
`v
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`Marvel Semi v. Intellectual Ventures,
`No. IPR2014-00547, Paper 19 (PTAB Jan. 14, 2015)............................................... 47
`Space Exploration Tech v. Blue Origin LLC,
`No. IPR2014-01378, Paper 6 (PTAB Mar. 3, 2015) ................................................... 5
`STATUTES
`35 U.S.C. § 112 ¶ 6 ................................................................................................................ 25
`REGULATIONS
`37 C.F.R. § 42.100(b) ...................................................................................................... 3, 4, 5
`37 C.F.R. § 42.104(b)(4) ......................................................................................................... 4
`37 C.F.R. § 42.22(a)(2) ............................................................................................................ 4
`37 C.F.R. § 42.6(a)(3) .............................................................................................................. 3
`OTHER AUTHORITIES
`77 Fed. Reg. 48,419, 48,627 cmt. 17 (Aug. 14, 2012) .................................................. 18
`
`vi
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`CURRENT EXHIBIT LIST
`
`Brief Description
`No.
`1001 U.S. Patent 6,792,373 to Tabor (“’373 patent”)
`1002 U.S. Patent 6,792,373 Non-Final Office Action
`1003 U.S. Patent 6,792,373 Amendment and Response
`1004 U.S. Patent 6,792,373 Notice of Allowance
`1005 Reexamination 90/008,313 Request for Reexamination
`1006 Reexamination 90/008,313 Order Granting Reexamination
`1007 Reexamination 90/008,313 Patent Owner’s Statement
`1008 Reexamination 90/008,313 Petitioner’s Reply to Patent Owner’s
`Statement
`1009 Reexamination 90/008,313 Non-Final Office Action
`1010 Reexamination 90/008,313 Patent Owner’s Response to Office Action
`1011 Reexamination 90/008,313 Final Office Action
`1012 Reexamination 90/008,313 Patent Owner’s Appeal Brief
`1013 Reexamination 90/008,313 Notice of Intent to Issue a Reexamination
`Certificate
`1014 Andrew Grochowski, Integrated Circuit Testing for Quality Assurance
`in Manufacturing: History, Current Status, and Future Trends, IEEE
`TRANS. ON CIR. & SYS., Vol. 36, No. 8, 610-33 (Aug. 1997)
`1015 Wesley H. Higaki, Remote Monitoring and Control of Semiconductor
`Processing, Hewlett-Packard Journal, Vol. 36, No. 7, 30-34 (Jul. 1985)
`1016 Computer Focus – International, Hewlett Packard (Sept. 1985)
`1017 Semiconductor Process Analysis Software Incorporates Graphics
`Package, IEEE CG&A, 8-10 (Dec. 1985)
`1018 Zarrin Ghaemi, Application of Data Screening to the Characterization
`of Integrated Circuits, CAN. J. PHYS., Vol. 67, No. 4, 221-224 (Apr.
`1989)
`1019 U.S. Patent 6,366,108 to O’Neill et al.
`1020 W. Robert Daasch et al., Variance Reduction Using Wafer Patterns in
`IddQ Data, Proceedings of International Test Conference, 189-198 (Oct.
`3, 2000)
`1021 U.S. Patent 5,206,582 to Ekstedt et al.
`
`vii
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`1022 Peter Maxwel et al., Current Ratios: A Self-Scaling Technique for
`Production IDDQ Testing, International Test Conference, 1148-1156
`(2000)
`1023 U.S. Patent 5,789,933 to Brown
`1024 U.S. Patent 5,442,282 to Rostoker
`1025 U.S. Patent 5,761,064 to La et al.
`1026 Thomas Gneiting et al., Measurement Related and Model Parameter
`Related Statistics (Oct. 11, 2000)
`1027 U.S. Patent 6,574,760 to Mydill
`1028 Declaration of Adit Singh
`1029 Proof of Service of Summons showing the date of service of
`September 29, 2014
`1030 U.S. Patent 6,792,373 C1 to Tabor (Ex Parte Reexamination Certificate)
`2001 Claim Construction Order, Test Advantage, Inc. v. Tabor, No. 2005-cv-
`03392 (D. Az. filed Feb. 16, 2007)
`2002 Bullock, Statistical Outliers Impossible in Small Samples, The Secret
`Garden of Maths (Oct. 13, 2013), available at
`http://www.mickybullock.com/blog/2013/10/statistical-outliers-
`impossible-in-small-samples/
`2003 Ex parte Grafenauer, No. 2010-001906 (BPAI filed Mar. 2, 2012)
`2004 File, Computer-Dictionary-Online (2015), available at
`http://www.computer-dictionary-online.org/index.asp?q=file
`
`viii
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`I.
`
`INTRODUCTION
`The heart of petitioner’s argument is that the ’373 patent claim scope
`
`encompasses semiconductor pass-fail threshold testing, but, as patent owner
`
`demonstrates in this preliminary response, the ’373 patent claims are drawn to
`
`a different invention: identification of outlier data from parts that have already
`
`passed threshold testing, not detection of inoperable components themselves.
`
`On one hand, the primary references petitioner puts forth as purportedly
`
`rendering the ’373 patent claims obvious, Ekstedt and O’Neill, are concerned
`
`with conventional testing of semiconductor parts by subjecting these
`
`components to an input, and classifying the part as defective and discarding it
`
`if a known good output is not sensed. For example, O’Neill or Ekstedt might
`
`engage in threshold testing by simply measuring the current drawn by a device
`
`to see if the amperes exceed or fall below the part tolerances, and, if so, the
`
`part is marked “bad.” The PTO extensively considered this type of testing
`
`during the initial prosecution and reexamination of the ’373 patent, and found
`
`the ’373 patent to be non-obvious in light of such art.
`
`On the other hand, the ’373 patent invention analyzes many data
`
`measurements from chips that have already passed threshold testing (i.e., had
`
`measurements that did not exceed the part control limits), and performs
`
`statistical analysis within this test data to find any measurements that stray from
`
`
`
`Page 1
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`the main group of data (outlier data points). As the ’373 patent explains, this
`
`identification of outliers may be used to “bin” functional parts already marked
`
`as “good” into further subcategories as “marginal” or “critical” based upon the
`
`number of outliers and the degree to which the outliers stray from the main
`
`data set. [Ex. 1001 15:24–67.] Additionally, the outlier detection can also be
`
`used to scrutinize the tests themselves, rather than the parts, by identifying
`
`incorrect, questionable, or unusual results, repetitive tests, and/or tests with a
`
`relatively high probability of failure. [Id. 5:11–13].
`
`In reality, petitioner does not simply argue the ’373 patent’s claims
`
`encompass defective component threshold testing, as much as petitioner
`
`attempts to rewrite the claim language, through creative and improper
`
`constructions, to make the claims cover the prior art, contrary to the ’373
`
`patent’s teaching. In particular, despite the ’373 patent’s explicit claim
`
`language that the machine or method is analyzing test results, i.e., measured
`
`data points, petitioner attempts to define terms to include the identification of
`
`a failed part. In spite of a specification that makes it clear that outliers for
`
`identification do not include values that are outside of device control limits,
`
`petitioner ignores this differentiation in its proposed claim construction.
`
`Notwithstanding the fact that it is impossible to have an outlier in a data set
`
`without enough points for there to be a main group for one datum to be an
`
`
`
`Page 2
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`outlier to, petitioner attempts to define terms such that one data point can be
`
`an outlier simply based upon the single point’s value. And, in the face of claim
`
`language requiring an output report showing the identified outlier, petitioner
`
`attempts to manipulate the limitations so that the indication of a failed part
`
`falls within the claim scope.
`
`However, if the claims are construed with the appropriate § 42.100(b)
`
`standard, grounded in the claim and specification language, and petitioner’s
`
`conspicuously litigation-driven proposals are not adopted, petitioner’s cited
`
`prior art clearly does not render any of the claims anticipated or obvious. And
`
`even with petitioner’s calculated claim definitions, the prior art still falls outside
`
`the ’373 patent claim scope as described in full below. Because petitioner relies
`
`on prior art that is cumulative over the prosecution and reexamination, drawn
`
`to a different invention altogether, and asserts the same arguments repeatedly
`
`considered and rejected by the PTO, petitioner fails to demonstrate a
`
`reasonable likelihood that any of the ’373 patent claims are unpatentable.
`II. GROUNDS
`Petitioner embeds claim charts, reference to exhibits, and other
`
`arguments to support its unpatentability challenges in its expert declaration
`
`that do not appear in the petition. [See, e.g., Ex. 1028 ¶ 61.] Because such
`
`incorporation by reference is against PTAB rules, 37 C.F.R. §§ 42.6(a)(3),
`
`
`
`Page 3
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`42.22(a)(2), 42.104(b)(4), patent owner will limit its preliminary response to the
`
`arguments actually put forth in the petition.
`III. CLAIM CONSTRUCTION
`A. Petitioner Relies upon Expert Who Uses the Wrong Standard
`As a preliminary matter, petitioner almost exclusively relies upon its
`
`expert, to support its proposed definitions; however, Dr. Singh is using the
`
`wrong standard for claim construction, and, as such, all of petitioner’s proposed
`
`definitions are suspect. As Dr. Singh repeatedly states, “My understanding is
`
`that the claims are to be given their broadest reasonable construction.” [Ex.
`
`1028 ¶¶ 67, 110, 176; see also ¶ 191 (“Given the broadest reasonable construction
`
`. . .”); ¶ 20 (“[B]roadest reasonable ordinary meaning”); ¶ 19 (“[B]roadest
`
`reasonable interpretation that is consistent with the specification of the
`
`patent.”).] But the correct standard is that a claim “shall be given its broadest
`
`reasonable construction in light of the specification.” § 42.100(b).
`
`Apparently, as a consequence of using the wrong standard, petitioner
`
`and its expert first find the broadest possible construction that can be thought
`
`of, and then only reject the definition if it is not “consistent with the
`
`specification.” [Pet. 15–17 (citing expert for every term and arguing “[t]his is
`
`consistent with the ‘373 Patent . . .”).]. This is wrong. The words “in light of”
`
`make clear that proper claim construction takes into account “the specification
`
`
`
`Page 4
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`of the patent in which it appears” before determining the broadest reasonable
`
`construction, § 42.100(b), not using the specification as some afterthought
`
`sanity-check. As such, petitioner’s definitions riddled with language such as
`
`“any,” “associated,” “on some measure,” “any form,” any means,” “something
`
`else,” and “any kind” are simply unreasonably broad. See Space Exploration
`
`Tech v. Blue Origin LLC, No. IPR2014-01378, Paper 6, at 6 (PTAB Mar. 3, 2015)
`
`(rejecting petitioner’s claim definition of “any structure” as violating the
`
`‘broadest reasonable construction’ standard) (emphasis in original).1
`B. Proposed Terms
`Patent owner’s proposed definitions using the correct standard, and
`
`specific issues that render petitioner’s proposed definitions incorrect, are
`
`
`
`addressed below.
`1. “outlier” (claims 1, 6–8, 11, 13–16 and 18)
`Petitioner
`Patent Owner
`any semiconductor device or
`a test result whose value strays from
`part or data point associated
`a set of test results having statistically
`with the device or part that on
`similar values, but does not exceed
`some measure falls above or
`control limits or otherwise fail to be
`below a threshold value
`detected [Ex. 1001 6:32–46]
`a. Claims Define “Outlier” as a Test Result, Not the Tested Component
`Petitioner’s proposed definition of “outlier,” beginning with “any
`
`
`1 In this response, all emphasis in quotations is added unless otherwise noted.
`
`
`
`Page 5
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`semiconductor device or part . . .,” is a transparent, but impermissible, attempt
`
`to unreasonably broaden the term in an effort to bring prior art into the ’373
`
`patent claim scope. Specifically, petitioner is trying to include the component
`
`as an outlier, not just the measured data for the component.
`
`Inspection of the claims, however, shows that the term “outlier” is
`
`consistently used in the larger phrase “outlier in the test data” (claims 1 and 15)
`
`or “outliers in the semiconductor test data” (claim 8). The claims themselves,
`
`therefore, make it clear that an outlier is one of the component’s
`
`measurements, not the component itself. Phillips v. AWH Corp., 415 F.3d 1303,
`
`1312 (Fed. Cir. 2005) (en banc) ("[T]he Supreme Court made clear that the claims
`
`are ‘of primary importance, in the effort to ascertain precisely what it is that is
`
`patented.’”) (citing Merrill v. Yeomans, 94 U.S. 568, 570 (1876)). Therefore, an
`
`outlier is “a test result,” not a “semiconductor device or part.”
`b. The ’373 Patent Explicitly Defines “Outlier” as a Measured Value that
`Statistically Strays from a Data Set
`The ’373 patent specifically defines “outliers” as so: “For the present
`
`purposes, those test results that stray from the first set [of test results having
`
`statistically similar values] but do not exceed the control limits or otherwise fail
`
`
`
`Page 6
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`to be detected are referred to as ‘outliers.’”2 [Ex. 1001 6:41–46.] When a
`
`patentee specifically defines a term, as here, this definition controls. Vitronics
`
`Corp. v. Conceptronic, 90 F.3d 1576, 1582 (Fed. Cir. 1996) (“[A] patentee may
`
`choose to be his own lexicographer and use terms in a manner other than their
`
`ordinary meaning, as long as the special definition of the term is clearly stated
`
`in the patent specification.”). And the ’373 patent’s definition of outlier is clear:
`
`an outlier is a measured test data that “statistically strays” from a main set of
`
`data points. It is not, as petitioner attempts to define the term, simply a “data
`
`point associated with the device or part that on some measure falls above or
`
`below a threshold value.”
`
`The difference between the ’373 patent’s definition of outlier and
`
`petitioner’s not-so-subtle attempt to bring traditional threshold testing within
`
`the purview of the patent claims, is that “outlier,” as used in the ’373 patent, is
`
`found based upon the relationship of one datum to the data set, not simply by
`
`comparing the datum to a static threshold value. Indeed, the ’373 patent
`
`acknowledges that the tester 102 may conduct traditional pass-fail part
`
`classification (threshold testing) using a device’s “upper test limit and . . . lower
`
`
`2 Petitioner acknowledges the ’373 patent’s definition, but maintains, without
`explanation, that “the ’373 patent is not limited to this definition.” [Pet. 25.]
`
`
`
`Page 7
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`test limit” control value to see if a “component may be classified as a ‘bad part.’”
`
`[Id. 6:38–40.] See also annotated Figure 1 below (left):
`
`But such part classification is explicitly excluded from the ’373 patent’s
`
`definition of “outlier” that says outliers do not include measurements that
`
`“exceed the control limits.” Therefore, it is incorrect to include this part
`
`threshold testing using the device’s predefined maximum and minimum ratings
`
`in the definition of “outlier,” the type of testing done by the prior art.3 [See,
`
`e.g., O’Neill, Ex. 1019 9:2–4 (“It is well known that most circuits 14 have an
`
`absolute maximum value for IDDQ that is established by the circuit’s design and
`
`3 Additionally, the ’373 patent definition of “outlier” also specifically excludes
`measurements that “fail to be detected.” This makes sense, because when a
`tester cannot measure a part, that part would be classified as broken, but the
`invention cannot operate to identify outlier data from this part, because there
`is no data to operate upon.
`
`
`
`Page 8
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`specifications of the process in which it is manufactured.”); see also Ekstedt, Ex.
`
`1021 8:51–52 (“Another set of input variables relate to design characteristics of
`
`the device.”).]
`
`More importantly, the ’373 patent explains that the traditional threshold
`
`testing done by the tester 102, is different than the claimed outlier identification
`
`subsequently done by the analysis element 206 operating on the computer
`
`108. See Ex. 1001 Fig. 2, shown in part & annotated (above on right). The data
`
`from the tester, including the “bad part” classification [id. 6:58–59], is actually
`
`stored in “a tester data file [that] is then provided to the computer 108” [id.
`
`6:61–62]. This test-gathered data is subsequently processed by the computer
`
`108 that has “the supplementary data analysis element 206 [which] analyzes the
`
`data to provide enhanced output results” [id. 6:64–66], including the claimed
`
`“outlier identification” [id. 7:8–15].
`
`The ’373 patent further clarifies the distinction between part classification
`
`and outlier identification noting that the components may be further classified,
`
`once the outliers have been identified, based upon the number of outliers and
`
`in what range outliers fall. [Id. 15:58–
`
`67; see also id. 14:59–62 (explaining
`
`that it is “the outlier classification
`
`element
`
`[212
`
`that]
`
`is
`
`suitably
`
`
`
`Page 9
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`configured to classify the components 106 into critical/marginal/good part
`
`categories, for example in conjunction with user-defined criteria.”).]
`
`But part classification, done by comparing a data measurement to the
`
`component characteristic (i.e., threshold testing) or other user defined criteria,
`
`is not the claimed identification of an outlier that is done by comparing a data
`
`measurement to the rest of the measured data in a set. And petitioner’s
`
`attempt to shoehorn part classification into the definition, against the ’373
`
`patent teaching, is improper.
`c. Petitioner’s Own References Support Patent Owner’s Definition
`Even artisans in the field, including the very references that petitioner
`
`relies upon, such as O’Neill (“[a]s known in the art, an outlier is defined as points
`
`outside of the distribution of a population” [Ex. 1019 6:52–54]) and Ghaemi
`
`(“outliers (one or several points lying outside the main group . . .)” [Ex. 1018, p.
`
`221]), define outlier as being in relation to other measured component data
`
`rather than threshold comparison. See Arthur A. Collins, Inc. v. N. Telecom Ltd.,
`
`216 F.3d 1042, 1044–45 (Fed. Cir. 2000) (“Even when prior art is not cited in the
`
`written description or the prosecution history, it may assist in ascertaining the
`
`meaning of a term to a person skilled in the art.”).
`d. Petitioner’s Support Consists of a Misinterpretation of Figure 9 and
`Misleading Quotation
`Petitioner and its expert’s primary support for the contention that
`
`
`
`Page 10
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`“outlier” should include any device “that on some measure falls above or below
`
`a threshold value” is a mischaracterization of Figure 9 that directly contradicts
`
`the ’373 patent’s explanation of this figure. [See Pet. 13; Ex. 1028 ¶¶ 42–43.] In
`
`particular, petitioner incorrectly says that Figure 9 shows that “[a]n outlier is
`
`shown as any part that produces a test result beyond the set threshold.” [Id.]
`
`To the contrary, the ’373 patent explains that Figure 9 shows “test results
`
`for a single test of multiple components” [Ex. 1001 6:33–34] and notes there can
`
`be more than one result per component [id. 6:32–33 (“Each test generates at
`
`least one result for at least one of the components.”); see also 1:48–49 (“For
`
`example, each tester may perform 200 tests on a single component.”).]
`
`Additionally, the ’373 patent describes Figure 9 as having two sets of test
`
`results: “a first set of test results having statistically similar values and a second
`
`set of test results characterized by values that stray from the first set.” [Ex. 1001
`
`6:34–37.] As clarified, the second set
`
`of test results are outliers because
`
`they statistically “stray from the first
`
`set,” not because they are outside a
`
`“set threshold” like the lower test limit
`
`(“LTL” shown as 0.001 on the bottom
`
`of the y-axis) or upper test limit (“UTL”
`
`
`
`Page 11
`
`
`
`In-Depth Test’s Pre-Trial Preliminary Response
`
`IPR2015-00421 (6,792,373 B2)
`
`shown as 0.005 on the top of the y-axis). [Id. 6:34–37.]
`
`Further, the patent explains that the second set is additionally broken
`
`down in to two types of test results: “Some of the test results in the second set
`
`that stray from the first set may exceed the control limits [circles of Figure 9],
`
`while others do not [triangles of Figure 9].” [Id. 6:42–44.] The patent explains
`
`in no uncertain terms that it is the subset in the second set that do not exceed
`
`the control limits that are “outliers” to be identified by the supplemental data
`
`analysis element based upon their relationship to the first data set. [Id. 6:44–
`
`46 (“[T]est results that stray from the first set but do not exceed the control
`
`limits . . . are re