`
`
`
`
`
`
`BEFORE THE OFFICE OF THE UNDER SECRETARY OF COMMERCE
`FOR INTELLECTUAL PROPERTY AND DIRECTOR OF THE UNITED
`STATES PATENT AND TRADEMARK OFFICE
`
`
`
`
`
`
`APPLE INC.,
`Petitioner,
`
`v.
`
`CPC PATENT TECHNOLOGIES PTY, LTD.,
`Patent Owner.
`
`
`
`
`Case IPR2022-00602
`U.S. Patent No. 9,665,705
`
`
`
`
`
`
`
`
`
`
`PATENT OWNER’S REQUEST FOR DIRECTOR REVIEW
`
`
`
`
`
`
`
`
`
`Pursuant to the Revised Interim Rules Governing the Director Review Process
`
`(Sept. 18, 2023), Patent Owner respectfully requests that the Commissioner review
`
`the Final Written Decision (“FWD”) finding all claims of U.S. Patent No. 9,665,705
`
`(“the ‘705 Patent”) invalid. The issues warranting such review are:
`
`1)
`
`After it adopted for institution purposes a construction of the limitation
`
`“accessibility attribute” previously urged by the Petitioner before the district court,
`
`the Panel changed that construction materially to capture the prior art without
`
`providing notice to the Patent Owner sufficient for due process under the
`
`Administrative Procedures Act (“APA”), thereby prejudicing Patent Owner. See,
`
`e.g, Qualcomm Inc. v. Intel Corp., 6 F.4th 1256, 1265 (Fed. Cir. 2021).
`
`2)
`
`The proposed modification of the Mathiassen reference with the non-
`
`biometric teachings of Anderson does not result in a series of received biometric
`
`signal entries that are mapped into an instruction used to populate the database as
`
`part of an enrollment process, as required by the challenged claims.
`
`3)
`
`The Panel misapplied the law regarding the motivation to combine references
`
`in recognizing the combination of the Mathiassen reference, on the one hand, and
`
`the McKeeth and Anderson references, on the other hand, as valid combinations.
`
`
`
`A.
`
`The Panel’s Changed Construction of “Accessibility Attribute”
`
`The challenged claims of the ‘705 Patent require, inter alia, “a transmitter
`
`sub-system controller configured to match the biometric signal against members of
`
`1
`
`
`
`the database of biometric signatures to thereby output an accessibility attribute.”
`
`See, e.g., Ex. 1001, claim 1 (emphasis added). The parties disputed the meaning of
`
`“accessibility attribute” in a co-pending district court proceeding, during which
`
`Petitioner vociferously argued that the term be construed as an “attribute that
`
`establishes whether and under which conditions access to the controlled item
`
`should be granted to a user.” Ex. 2011 at 26 (emphasis added).
`
`Petitioner argued that its proposed construction was “consistent with the
`
`description of the invention throughout the specification and the claims, which goes
`
`beyond mere matching—the binary decision of ‘yes’ or ‘no’—and instead describes
`
`a system that provides for different types of access.” Id. at 26. The Petitioner then
`
`described the “multi-tiered access” system taught in the ‘705 Patent, which “can only
`
`be facilitated by the different types of fingerprint (or other biometric) input - i.e., the
`
`number of presses and duration—recited elsewhere in the claim” See id.
`
`To drive the point home, Petitioner reiterated that “[b]inary matching—
`
`‘match/no match’—is not what the inventor was trying to invent. Instead, he sought
`
`to provide a more sophisticated system with, inter alia, multiple types of access.” Id.
`
`at 28. In contrast, according to Petitioner, CPC’s proposed plain and ordinary
`
`meaning construction “would gut the clear definition given to it by the patentee, and
`
`improperly broaden the scope of the claims to encompass mere matching, a feature
`
`described as prior art.” Id.
`
`2
`
`
`
`
`
`As the Panel acknowledged in the FWD, it adopted, for purposes of institution
`
`Petitioner’s construction “accessibility attribute,” which includes both the “whether”
`
`and “under which conditions” components of that construction. See FWD at 19 &
`
`21. In doing so, the Panel confirmed that adopted the construction excludes “a
`
`‘binary decision’ to grant or not grant access to a locked structure or device.” Id.
`
`
`
`The principal reference relied upon by Petitioner is Mathiassen, which it
`
`represented teaches a “portable control processor . . . configured to match the user’s
`
`biometric signal against the database of biometric signatures,” and, [i]f there is a
`
`match, the processor will proceed to open (or lock) the car doors.” FWD at 44
`
`(emphasis added) (internal citations omitted). This is clearly a binary operation, as
`
`there only two options – unlock the door if there is a match, or not if there is not.
`
`There is no third option. This was effectively confirmed by Petitioner’s expert, who
`
`acknowledged that “Mathiassen is silent as to any incremental access that a car
`
`owner is granted, as opposed to any other user.” PO Sur-Reply at 22, citing Ex. 2015
`
`at 66:1-67:9. In other words, Mathiassen teaches precisely the type of “mere
`
`matching” that would “gut the clear definition” of “accessibility attribute” that
`
`Petitioner warned against. See Ex. 2011 at 28.
`
`
`
`That Petitioner originally agreed with this proposition is evident from its
`
`Petition - “[a]pplying the [District] Court’s construction, Mathiassen’s ‘open door’
`
`command as modified by McKeeth’s teaching of duress and alert conditions
`
`3
`
`
`
`teaches or renders obvious outputting an accessibility attribute, as claimed.” Petition
`
`at 17 (emphasis added). Put another way, Mathiassen needs McKeeth’s teachings to
`
`satisfy the “accessibility attribute” under the construction proposed by it, and
`
`adopted by both the district court and the panel.
`
`
`
`In dealing with Mathiassen in the FWD, the Panel inexplicably found that,
`
`“[b]ased on the language of the claims and specification, the ‘accessibility attribute’
`
`may include only an ‘access attribute,’ which is ‘unconditional,’” which is precisely
`
`the type of binary (yes/no – lock/unlock) decision that Petitioner adamantly opposed
`
`including in the very construction it successfully urged to the Panel for institution
`
`purposes. See FWD at 21. Nonetheless, the Panel stated that its original construction
`
`(again expressly excluding binary decisions) includes “unconditional access, if no
`
`conditions are imposed.” FWD at 45-46. Notwithstanding this supposed inclusion,
`
`the Panel felt compelled to “modify” the construction of “accessibility attribute” to
`
`include the term “if any,” i.e., potentially none, such that the new construction reads
`
`“an attribute that establishes whether and under which conditions, if any, access to
`
`the controlled item should be granted.” FWD at 21 (emphasis added).
`
`
`
`As a result of this obviously changed construction, the Panel went on to find
`
`that Mathiassen, which it found to teach at most a lock/unlock operation, “discloses
`
`or suggests” the “accessibility attribute” limitation. See FWD at 47. As such,
`
`Petitioner was allowed to benefit from one construction of that limitation before the
`
`4
`
`
`
`district court and before the Panel on institution, but when that construction became
`
`inconvenient after institution, the Panel gave Petitioner a different construction,
`
`allowing the unique capture of prior art. More importantly, however, Patent Owner
`
`had no opportunity to address the impropriety of this changed construction, as the
`
`Panel gave no indication that there would be such a change.1
`
`
`
`The Federal Circuit has ruled that a panel cannot, consistent with the APA,
`
`issue a new claim construction after institution without giving the parties notice
`
`thereof. Qualcomm, 6 F.4th at 1265. The Panel recognized as much in the FWD. See
`
`FWD at 26, n.20 (citation omitted) (“the Board ‘must base its decision on arguments
`
`that were advanced by a party, and to which the opposing party was given a chance
`
`to respond’”). Yet, as explained above, the Panel went on to ignore that
`
`administrative guardrail, changing claim construction with no opportunity for Patent
`
`Owner to respond thereto. Without its “if any” change to the construction of
`
`
`1 A stark example of this lack of opportunity is the Panel’s reliance on claim 3 of the
`
`‘705 Patent as purportedly supporting its changed construction. As the Panel
`
`acknowledged, the parties “did not discuss specifically claim differentiation as part
`
`of their claim construction analysis,” the precise purpose for the Panel’s reliance on
`
`claim 3. See FWD at 26, n.20. As an aside, Patent Owner indeed disagrees that claim
`
`3 supports the Panel’s changed construction.
`
`5
`
`
`
`“accessibility attribute,” which demonstrably allows for a binary access/no access
`
`operation, the Panel could not have found Mathiassen teaches that limitation. As
`
`such, the Panel’s belated claim construction modification is clearly prejudicial to
`
`Patent Owner.2 There cannot, then, be any doubt that this changed construction,
`
`which was neither proposed by Petitioner, nor addressable before the Panel by Patent
`
`Owner, runs afoul of the holding in Qualcomm, 6 F.4th at 1265, and its progeny.
`
`B.
`
`The Mathiassen/Anderson Combination does not Yield a Series of
`Biometric Signals as Part of an Enrollment Process
`
`
`“Illustrative” claim 1 requires receiving “a series of entries of the biometric
`
`
`
`signal, said series being characterised according to at least one of the number of said
`
`entries and a duration of each said entry.” FWD at 10 (emphasis added). This
`
`function is part of the “enrolling feature.” FWD at 32. According to the plain
`
`language of the claim, “the number and/or duration of entries is based on entries of
`
`
`2 Of important note is that the Panel, after discussing the combination of the
`
`Mathiassen and McKeeth references, failed to find that it was that combination that
`
`teaches the “accessibility attribute” limitation – only that Mathiassen purportedly
`
`does so. See FWD at 47 (“Based on Petitioner’s arguments and evidence as
`
`summarized above, we determine Petitioner has sufficiently shown that Mathiassen
`
`discloses or suggests” the “accessibility attribute” limitation).
`
`6
`
`
`
`a biometric signal, such as a finger press on a fingerprint sensor.” FWD at 31
`
`(emphasis added).
`
`
`
`As the Panel noted, “Mathiassen does not teach determining a duration of each
`
`entry.” FWD at 50. Therefore, Petitioner relies upon Anderson for the purported
`
`teaching of “inputting an access code including fingerprint presses of varying
`
`duration.” See id. The substitution proposed by the Petitioner, and found invalidating
`
`by the Panel, is described in the FWD as follows:
`
`A POSITA would have found it obvious to substitute or modify such
`
`directional finger movements [from Mathiassen] with a series of presses of
`
`varying duration, as taught by Anderson, for instructing a command at
`
`portable device 20.
`
`FWD at 51.
`
`
`
`As the Panel noted, Patent Owner argued in response to this proposed
`
`combination that “Mathiassen has no teaching that either the ‘predefined sets of
`
`finger movement sequences’ or the ‘command table’ constitute a series of received
`
`biometric signal entries that are mapped into an instruction used to populate the
`
`database as part of the enrollment process.” FWD at 56. Specifically, the Patent
`
`Owner pointed out the following in the passage from the POR:
`
`The ‘series’ requirement, according to Dr. Sears, is purportedly disclosed in
`
`Mathiassen as ‘omni-directional finger movements across the sensor in two
`
`7
`
`
`
`dimensions.’ For the requirement of ‘mapping,’ according to Petitioner’s
`
`expert, these movements are categorized according to ‘predefined sets of
`
`finger movement sequences.’ A ‘command table’ is then used ‘to translate the
`
`categorized finger movements into control signals.’ This has nothing to do
`
`with user enrollment, as Mathiassen makes clear that the control signal are
`
`‘for controlling the device.’ Mathiassen has no teaching that either the
`
`‘predefined sets of finger movement sequences’ or the ‘command table’
`
`constitute a series of received biometric signal entries that are mapped into an
`
`instruction used to populate the database as part of the enrollment process,
`
`as required by representative Claim 1.
`
`POR at 33 (emphasis added) (internal citations omitted). See also Sur-Reply at 22-
`
`23 (“Mathiassen’s finger movements providing command functionality are pre-
`
`defined, i.e., they are not part of the enrollment process”).
`
`
`
`After expressly referencing this portion of the POR, the Panel entirely ignores
`
`this point, focusing solely on whether Mathiassen’s finger motions are “biometric.”
`
`See FWD at 57. Thus, even if everything else Petitioner maintains about the
`
`Mathiassen/Anderson combination were true, that combination would still lack a
`
`biometric signal series received as part of an enrollment process – a requirement that
`
`the Panel expressly called out as being part of the challenged claims.
`
`8
`
`
`
`
`
`The Panel does reference Mathiassen’s master minutiae tables, which are part
`
`of Mathiassen’s enrollment process. See, e.g., FWD at 55-56. However, as Patent
`
`Owner has submitted, the master minutiae tables have nothing to do with
`
`Mathiassen’s pre-stored finger movement sequences – the only feature from
`
`Mathiassen that Petitioner proposes to modify with Anderson. See POR at 33. The
`
`Panel, nonetheless, accepts Petitioner’s mapping of the prior art to this claim
`
`limitation, including Petitioner’s reference to the master minutiae tables, which is
`
`needed to tether the prior art’s teachings to an enrollment process, despite its
`
`irrelevance to Mathiassen’s pre-stored finger movements. See FWD at 56. As the
`
`Panel failed to provide any explanation as to how the Mathiassen/Anderson
`
`combination relates in any relevant way to an enrollment process, the FWD, which
`
`rejects Patent Owner’s argument on this issue must be vacated. See In re Nuvasive,
`
`Inc., 842 F.3d 1376, 1383 (2016) (“it is not adequate to summarize and reject
`
`arguments without explaining why the PTAB accepts the prevailing argument”).
`
`
`
`Separately, even if one could get past the fact that the Mathiassen/Anderson
`
`combination is unrelated to an enrollment process, it remains that Anderson does not
`
`teach a “biometric” signal series as required by the challenged claims. According to
`
`the Panel’s findings, “Anderson’s disclosed system inputs an access code ‘via
`
`temporal variations in the amount of pressure applied to a touch interface’” FWD
`
`at 38 (emphasis added). Further, Anderson’s system “may sense only ‘temporal
`
`9
`
`
`
`applications of pressure,’ relying on timing of the pressure applications for entry of
`
`the access code.” Id. (emphasis in original). Petitioner’s own expert admitted that
`
`pressure and duration patterns of the type taught in Anderson are knowledge-based,
`
`i.e., non-biometric, making that point undisputed. PO Sur-Reply at 19, citing Sears
`
`Dep. Tr. [Ex. 2012] at 18, 58:3-10.
`
`
`
`The Panel, however, goes on to refer mistakenly to the purported teachings in
`
`Anderson of a “series of fingerprint pressure pulses of varying duration” (FWD at
`
`50 (emphasis added)), and “fingerprint access code” (id. at 53 (emphasis added)).
`
`Based on these purported teachings, the Panel found that “there can be no reasonable
`
`dispute that Anderson discloses input biometric signals that vary in number and
`
`duration.” Id. at 51 (emphasis added). The problem with the Panel’s reasoning is that
`
`the terms “fingerprint pressure pulses” and “fingerprint access code” appear
`
`nowhere in Anderson, let alone in the portions of Anderson cited by the Panel as
`
`supposedly teaching these features. See Anderson [Ex. 1006] at 6:45-54 & 7:28-47.
`
`Rather, those terms were created by Petitioner’s expert and adopted by the Panel for
`
`institution purposes. See POR at 26
`
`
`
`
`
`Anderson does teach “an optical scanner or thermal sensor for collecting an
`
`image of the user’s fingerprint” that may optionally be part of the “digitizer pad 120”
`
`used “as a touch interface.” FWD at 53 (emphasis added). This collection of “an”
`
`image of a fingerprint is contradistinct from the claimed “series of entries of the
`
`10
`
`
`
`biometric signal” that are received as part of the enrollment process. FWD at 10
`
`(emphasis added). At no point did Petitioner or the Panel identify any teaching in
`
`Anderson that a series of biometric signals is received at all, let alone as part of an
`
`enrollment process. As such, to the extent the Panel’s decision turns on a purported
`
`teaching in Anderson of a “biometric” signal series, the decision must be reversed.3
`
`C.
`
`The Lack of a Motivation to Combine Mathiassen, on the One
`Hand, and Anderson and McKeeth, on the Other Hand
`
`
`The Petitioner and the Panel turned to the combination of Mathiassen and
`
`
`
`McKeeth to yield the “accessibility attribute” limitation, the latter of which “teaches
`
`both a duress instruction and an alert instruction when there is no match.” FWD at
`
`45. The panel gave short shrift to Patent Owner’s argument against a motivation to
`
`combine, stating that, while there may have been “simpler alternative solutions
`
`available,” “[i]t’s not necessary to show that a combination is the best option, only
`
`
`3 The Panel appears to contradict itself regarding Anderson’s teachings, stating that
`
`Anderson only contributes “a number and duration of pulses as inputs,” while
`
`Mathiassen and McKeeth allegedly provide a teaching of “biometric sensing.” FWD
`
`at 52. Even if that is the true nature of the combination, despite the Panel’s great
`
`pains in finding a biometric signal series in Anderson, as explained above, any
`
`resulting biometric signal series from the Mathiassen/Anderson combination would
`
`not be part of an enrollment process.
`
`11
`
`
`
`that it be a suitable option.” FWD at 46 (emphasis in original). The Patent Owner’s
`
`alternative to the proposed combination was to look to the functionality already
`
`taught in Mathiassen, which was undeniably simpler, as the Petitioner’s expert
`
`acknowledged. POR at 18-19; PO Sur-Reply at 5-6.
`
`
`
`The lead case for the “suitable” proposition cited by the Panel, Intel Corp. v.
`
`PACT XPP Schweiz AG, illustrates the risk for mischief in blindly applying the
`
`“suitable” standard when evaluating the motivation to combine. The relevant portion
`
`of that decision reads as follows:
`
`‘[I]f a technique has been used to improve one device, and a person of
`
`ordinary skill in the art would recognize that it would improve similar devices
`
`in the same way, using the technique is obvious unless its actual application
`
`is beyond his or her skill.’ This is the so-called ‘known-technique’ rationale.
`
`And if there's a known technique to address a known problem using ‘prior art
`
`elements according to their established functions,’ then there is a motivation
`
`to combine.
`
`Intel Corp. v. PACT XPP Schweiz AG, 61 F.4th 1373, 1380 (Fed. Cir. 2023).
`
`
`
`The entirety of Petitioner’s reasoning upon which the Panel relies is that it
`
`would have been obvious to a skilled person “to increase user safety of Mathiassen
`
`by providing accessibility attributes indicating duress access or alert access, as
`
`proposed in McKeeth, to thereby increase user security.” FWD at 45. There is no
`
`12
`
`
`
`discussion in the FWD, for example, of a “known technique” in McKeeth,
`
`warranting the application of the rationale in Intel. Indeed, neither the Petitioner nor
`
`the Panel pointed to anything magical about McKeeth specifically that would have
`
`occasioned a skilled person to modify Mathiassen’s teachings therewith, rather than
`
`using Mathiassen alone.
`
`
`
`Further, this “suitability” test must be viewed in light of precedent cited by
`
`the Panel requiring that that “a person of ordinary skill at the time of the invention
`
`‘would have selected and combined those prior art elements in the normal course
`
`of research and development to yield the claimed invention.’” FWD at 13 (emphasis
`
`added) (citation omitted). As a corollary proposition, one cannot view the suitability
`
`of a prior art combination through the lens of hindsight reconstruction, especially
`
`where the functionality to be added already exists in the reference to be modified by
`
`such combination. See Kinetic Concepts, Inc. v. Smith & Nephew, Inc., 688 F.3d
`
`1342, 1369 (Fed. Cir. 2012).4 Intuitively, availing oneself of Mathiassen’s own
`
`
`4 Patent Owner is cognizant of decisions such as in Elekta Limited v. ZAP Surgical
`
`Systems, Inc., ___ F.4th ___, 2023 WL 6152418 (2023), wherein the Federal Circuit
`
`ruled that the difficulty involved in combining references is not dispositive on the
`
`issue of the motivation to combine. However, that decision, and those like it, do not
`
`13
`
`
`
`teachings is a simpler solution to the generation of a duress signal than looking to an
`
`entirely different reference for such modification. Under the circumstances, absent
`
`hindsight reconstruction, there is no motivation to combine Mathiassen with any
`
`other reference, including McKeeth.
`
`
`
`As to motivation to combine Mathiassen and Anderson, again, Patent Owner
`
`contends that the simpler solution, were one to desire adding Anderson’s
`
`functionality to Mathiassen to achieve the heightened security, would be to utilize
`
`Mathiassen’s existing teachings alone. See, e.g., PO Sur-Reply at 4-8. The Panel,
`
`however, simply reiterated that a combination need only be “suitable,” rejecting any
`
`need to consider simplicity as an incentivizing factor when evaluating a purported
`
`motivation to combine. FWD at 54. Simply labeling a combination as “suitable” in
`
`a vacuum is insufficient to establish a motivation to combine references.
`
`
`
`In the case of both prior art combinations discussed herein, neither Petitioner
`
`nor the Panel disputed that: 1) Mathiassen already taught functionality that made
`
`combination with McKeeth and/or Anderson unnecessary; and 2) relying on
`
`Mathiassen alone was simpler than combining Mathiassen with another reference.
`
`As the test is motivation to combine, as opposed to a potential to combine, it was
`
`
`involve instances, such as here, where a reference sought to be modified already
`
`contained the functionality sought in the combination proposed.
`
`14
`
`
`
`error simply to brush aside the simplicity of relying upon Mathiassen alone in favor
`
`of what was purportedly “suitable.”5 This is especially the case given that the Patent
`
`Owner submitted unrebutted expert testimony that, given Mathiassen’s existing
`
`functionality, there would have been no reason to modify Mathiassen in the manner
`
`Petitioner and its expert proposed. See POR at 19, citing Ex. 2013, ¶ 51. That the
`
`Panel did not credit such testimony was error. See, e.g., Polaris Industries, Inc. v.
`
`Arctic Cat, Inc., 882 F.3d 1056 (2018).
`
`
`
`The three-issues discussed above, which were each mishandled by the Panel,
`
`merit Director review and reversal of the FWD.
`
`Dated: October 4, 2023
`
`
`
`Respectfully submitted by
`K&L GATES LLP,
`
`By: /Darlene F. Ghavimi-Alagha/
`Darlene F. Ghavimi-Alagha
`Reg. No. 72,631
`K&L GATES LLP
`darlene.ghavimi@klgates.com
`(512) 482-6919
`2801 Via Fortuna, Suite 650
`Austin, Texas 78746
`
`
`5 Petitioner did not propose a single-reference obviousness challenge based upon
`
`Mathiassen, so whether Mathiassen by itself renders the challenged claims is
`
`irrelevant. In any event, even availing oneself of Mathiassen’s teachings alone would
`
`not yield the “duration” limitation. See FWD at 50.
`
`15
`
`
`
`CERTIFICATE OF SERVICE
`The undersigned hereby certifies that on October 4, 2023, a true and correct
`
`copy of the foregoing Patent Owner’s Request for Director Review was submitted
`
`in the Patent Trial and Appeal Case Tracking System (P-TACTS), and sent by
`
`electronic mail to the following:
`
`Director
`Email: Director_PTABDecision_Review@uspto.gov
`
`Counsel for Petitioner:
`Jennifer C. Bailey
`Adam P. Seitz
`Email: Jennifer.Bailey@eriseip.com
`Email: Adam.Seitz@eriseip.com
`Email: PTAB@eriseip.com
`
`By:
`
`/Darlene F. Ghavimi-Alagha/
`Darlene F. Ghavimi-Alagha
`Reg. No. 72,631
`K&L GATES LLP
`2801 Via Fortuna, Suite 650
`Austin, Texas 78746
`Darlene.Ghavimi@klgates.com
`T: (512) 482-6919
`F: (512) 482-6859
`
`
`
`
`
`
`