`
`
`
`Paper No.
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`_____________________
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`_____________________
`
`
`APPLE INC.,
`Petitioner,
`
`v.
`
`UNILOC LUXEMBOURG, S.A.,
`Patent Owner
`
`_____________________
`
`
`Case IPR2018-00289
`Patent No. 8,872,646
`
`_____________________
`
`
`PETITIONER’S REPLY
`
`
`
`

`

`
`
`
`
`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`
`
`
`
`
`I.
`II.
`
`B.
`
`TABLE OF CONTENTS
`Introduction ...................................................................................................... 1
`The Evidence of Record Weighs in Favor of Petitioner ................................. 1
`A.
`Patent Owner’s response contains only unsupported
`attorney argument and is entitled to little to no weight ......................... 2
`The Declaration provided by Patent Owner with the patent
`owner preliminary response is entitled to little to no weight
` ............................................................................................................... 3
`III. The Petition and the Declaration Define the Level of Ordinary
`Skill in the Art ................................................................................................. 5
`IV. The Term “Glitch” is Properly Construed to Include a Datum
`Outside of an Acceptable Range. .................................................................... 6
`A.
`Petitioner’s proposed construction is supported by the
`claim language and specification .......................................................... 6
`Detecting a glitch
`is separate and distinct from
`determining whether a jostle or bump warrants awakening
`a device .................................................................................................. 8
`A construction defining what a term includes is sufficient
`for this proceeding ...............................................................................10
`V. McMahan Teaches Verifying and Removing Glitches .................................10
`A. McMahan’s error is no different than the claimed glitch ....................11
`B.
`Patent Owner attempts
`to narrow “motion data”
`improperly ...........................................................................................12
`C. McMahan’s modifying sensor values teaches removing
`glitches .................................................................................................13
`VI. A POSITA Would Have Combined McMahan with Pasolini,
`Goldman, and Mizell .....................................................................................15
`
`B.
`
`C.
`
`i
`
`

`

`
`
`B.
`
`C.
`
`
`A.
`
`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`The record shows that a POSITA would have found it
`obvious to combine the teachings of McMahan with
`Pasolini, Goldman, and Mizell ............................................................16
`The teachings of McMahan’s sensors are applicable to the
`accelerometers of Pasolini and Goldman. ...........................................18
`The proper test for obviousness is what the references
`suggest to a POSITA ...........................................................................20
`VII. The Dependent Claims are Obvious ..............................................................23
`VIII. Conclusion .....................................................................................................23
`IX. Certificate of Word Count .............................................................................24
`
`
`
`
`
`
`ii
`
`

`

`
`
`
`
`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`PETITIONER’S UPDATED EXHIBIT LIST
`
`
`
`
`
`October 12, 2018
`
`APPL-1001 U.S. Patent No. 8,872,646.
`
`APPL-1002 Prosecution History of U.S. Patent No. 8,872,646.
`
`APPL-1003 U.S. Patent No. 7,409,291 to Pasolini et al.
`
`APPL-1004 Using the LIS3L02AQ Accelerometer, Ron Goldman, Sun
`Microsystems Inc. Dated February 23, 2007.
`
`APPL-1005 U.S. Patent No. 7,204,123 to McMahan et al.
`
`APPL-1006 U.S. Patent Publication No. 2006/0161377 to Rakkola et al.
`
`APPL-1007 Using Gravity to Estimate Accelerometer Orientation,” David
`Mizell, Proceedings of the Seventh IEEE International Symposium
`on Wearable Computers (ISWC ’03) 2003.
`
`APPL-1008 Declaration of Chris Butler, Under 37 C.F.R. § 1.68.
`
`APPL-1009 Dictionary of Scientific and Technical Terms,” McGraw-Hill.
`
`APPL-1010 Declaration of Joe Paradiso, Ph.D, Under 37 C.F.R. § 1.68.
`
`APPL-1011 Curriculum Vitae of Joe Paradiso.
`
`APPL-1012 Declaration of Ingrid Hsieh-Yee, Under 37 C.F.R. § 1.68.
`
`APPL-1013 Reserved.
`
`APPL-1014 U.S. Patent No. 7,028,220 to Park et al.
`
`APPL-1015 Email Correspondence between Petitioner’s Counsel and the Board.
`
`APPL-1016 Email Correspondence between Petitioner’s Counsel and Patent
`
`iii
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`
`
`
`
`
`
`
`
`
`
`
`Owner’s Counsel.
`
`iv
`
`

`

`
`
`
`Introduction
`
`I.
`
`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`
`
`
`
`
`The Petition and the record as a whole provide detailed reasons why the
`
`claimed subject matter of the ’646 patent would have been obvious to a person of
`
`ordinary skill in the art (“POSITA”) in view of Pasolini, Goldman, Mizell, and
`
`McMahan. None of Patent Owner’s arguments adequately refute the Petition.
`
`Moreover, the evidence weighs in favor of the Petitioner because Patent Owner has
`
`not provided any evidence to refute the statements made by Petitioner’s expert.
`
`Patent Owner’s arguments fail for multiple reasons, as described in more
`
`detail below. For instance, Patent Owner challenges Petitioner’s proposed
`
`construction of a glitch by seeking an unduly narrow and improper characterization
`
`of the term “glitch.” See Response at 5-8. Patent Owner then relies on its improper
`
`construction to argue that the claims of the ’646 patent do not cover McMahan’s
`
`error in the accelerometer data. See Response at 9-15.
`
`Patent Owner further alleges that the Petition does not show sufficient
`
`motivation to combine the prior art references. See Response at 15-18. Petitioner
`
`shows below that each of Patent Owner’s arguments are incorrect and unsupported.
`
`II. The Evidence of Record Weighs in Favor of Petitioner
`
`The Petition shows the limitations of claims 1, 3, 5-11, 13-18, and 20 are
`
`obvious and unpatentable in light of the cited references. For example, the Petition,
`
`citing to Dr. Paradiso’s Declaration, illustrates that a POSITA would have
`
`1
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`combined the teachings of the cited references to arrive at the claimed invention of
`
`
`
`
`
`
`
`the ’646 patent.
`
`A.
`
`Patent Owner’s response contains only unsupported attorney
`argument and is entitled to little to no weight
`
`In the Response, Patent Owner’s arguments against the proposed
`
`combination are unsupported, and there is no controverting evidence presented,
`
`through expert testimony or otherwise. Notably, the Response does not cite to any
`
`expert declaration. See generally, Response at 1-18. “[U]nsworn attorney argument
`
`. . . is not evidence and cannot rebut . . . evidence.” Gemtron Corp. v. Saint-Gobain
`
`Corp., 572 F. 3d 1371, 1380 (Fed. Cir. 2009). Moreover, the Board gives little or
`
`no weight to attorney argument not supported by evidence. For example, in
`
`IPR2016-00758, the Patent Owner did not provide any expert declarations or other
`
`evidence. See Apple, Inc. v. Evolved Wireless LLC, IPR2016-00758, (PTAB Mar.
`
`12, 2018), Paper 48 at 20. The Petitioners in IPR2016-00758, however, presented
`
`an expert declaration, which was given appropriate evidentiary weight in light of
`
`the cross-examination testimony submitted by Patent Owner. See id. In view of
`
`such facts, the Board noted “we keep in mind that attorney argument is not
`
`evidence and Patent Owner cannot rebut evidence with unsworn attorney
`
`argument.” See id; see also Itron Networked Solutions, Inc., IPR2017-01024, Paper
`
`49 at 26 (Aug. 21, 2018) (“Patent Owner provides considerable attorney argument,
`
`2
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`but does not support its assertions with evidence that rebuts … expert testimony”)
`
`
`
`
`
`
`
`(“[i]n the absence of controverting evidence that we might weigh against Dr.
`
`Soliman’s opinion, we are not in a position to discount Dr. Soliman’s expert
`
`view.”); Nike Inc. v. Jezign Licensing, LLC, IPR2017-00246, Paper 33 at 27 (May
`
`29, 2018) (“We are not swayed by Patent Owner’s conclusory attorney arguments
`
`… [r]ather … we agree with Petitioner’s position as supported by the declaration
`
`testimony.”); Celanese International Corporation v. Daicel Corporation,
`
`IPR2017-00166, Paper 61 at 30 (May 4, 2018) (“Patent Owner relies on attorney
`
`argument to criticize the declaration of Dr. Liu, which we give little to no
`
`weight.”).
`
`Accordingly, without any supporting evidence, Patent Owner’s statements
`
`against the combination described by Dr. Paradiso and relied upon in the Petition
`
`amount to nothing more than attorney argument and cannot rebut the evidence of
`
`record.
`
`B.
`
`The Declaration provided by Patent Owner with the patent owner
`preliminary response is entitled to little to no weight
`
`While Patent Owner provided a declaration of William Easttom with its
`
`Preliminary Response to the Petition (“POPR”), Mr. Easttom was not made
`
`available for cross-examination. See APPL-1016. The Board has made clear that
`
`“[a]s a guiding principle of routine discovery, as defined by our Rules, if a party
`
`3
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`proffers a witness’s testimony, that party must make that witness available for
`
`
`
`
`
`
`
`cross-examination by the other party.” HTC Corp. v. NFC Technology, LLC,
`
`IPR2014-01198, Paper 41, at 3 (PTAB Nov. 6, 2015) (emphasis original). And, as
`
`the Board has correctly recognized, “little to no weight is given to testimony of
`
`declarants who are not subject to cross-examination, even if the testimony is not
`
`excluded as hearsay.” The Mangrove Partners Master Fund, LTD, et al., v. Virnetx
`
`Inc., IPR2015-01047, Paper 52 at 6-7 (PTAB Apr. 15, 2016); see also Kolmes v.
`
`World Fibers Corp. 107 F.3d 1534, 1542 (Fed. Cir. 1997) (holding that evidence
`
`was properly disregarded since no opportunity to cross-examine declarant).
`
`Accordingly, the declaration of Mr. Easttom, if not stricken from the record, is
`
`entitled to no weight.1
`
`In sum, the attorney argument in the Response is not evidence and cannot
`
`rebut the evidence presented by Petitioner’s expert, Dr. Paradiso. Further, the
`
`testimony Mr. Easttom provided with the Preliminary Response should be entitled
`
`no weight, because Mr. Easttom was not made available for cross-examination.
`
`Thus, Patent Owner has no evidence to refute Dr. Paradiso’s Declaration, and the
`
`evidence of record weighs strongly in favor of Petitioner.
`
`
`1 Petitioner filed a Motion to Strike the Declaration of William Easttom on
`
`September 19, 2018. The motion has not yet been decided.
`
`4
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`III. The Petition and the Declaration Define the Level of Ordinary Skill in
`the Art
`
`
`
`
`
`
`
`Patent Owner argues “[t]he Petition does not set forth the requisite analysis
`
`necessary to prove obviousness at least because (among other deficiencies) it fails
`
`to provide or expressly rely upon any definition for the level of ordinary skill in the
`
`pertinent art.” Response at 4. However, the Petition consistently cites to the
`
`Declaration’s analysis, which refers to a POSITA as defined in the Declaration, as
`
`explained below.
`
`The level of ordinary skill in the art is clearly spelled out in Dr. Paradiso’s
`
`Declaration. APPL-1005, ¶¶16-17. Each time the Declaration and the Petition refer
`
`to a POSITA, they are referring to a POSITA as defined in paragraphs 16 and 17 of
`
`Dr. Paradiso’s Declaration. The Paradiso Declaration defines a POSITA, explains
`
`prior art teachings limitation-by-limitation from the perspective of a POSITA, and
`
`explains why a POSITA would have combined the teachings. See, e.g., APPL-
`
`1010 at 46-47; Petition at 31-32. The Petition consistently cites to the Declaration’s
`
`analysis. Therefore, the Petition and the Declaration sufficiently define the level of
`
`ordinary skill in the art and consistently apply that definition to the obviousness
`
`analysis. Patent Owner’s allegation has no merit.
`
`5
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`IV. The Term “Glitch” is Properly Construed to Include a Datum Outside
`of an Acceptable Range.
`
`
`
`
`
`
`
`A.
`
`Petitioner’s proposed construction is supported by the claim
`language and specification
`
`In the Petition, Petitioner proposed that, for purposes of this proceeding, “the
`
`broadest reasonable interpretation of a ‘glitch’ includes a datum that is outside of
`
`an acceptable range.” Petition at 7; See APPL-1010, ¶42. This proposed
`
`construction is supported by the claims and encompasses the embodiments of
`
`glitch described in the specification.
`
`First, claim 1 is written broadly with respect to the “glitch”: “receiving
`
`motion data from a motion sensor in a device, the motion sensor sensing motion
`
`along three axes; [and] verifying whether the motion data includes one or more
`
`glitches and removing the one or more glitches from the motion data.” APPL-
`
`1001, 8:34-39. According to the words of claim 1, motion data is the data received
`
`from the motion sensor, and glitch is used without further context.
`
`Turning to the Specification, “glitch” is described broadly:
`
`a glitch is a datum that indicates a motion outside an
`acceptable range. For example, it is extremely unlikely that a
`device would go from idle (e.g., no motion) to moving at an
`acceleration of 64 feet per second squared (equivalent to 2g).
`APPL-1001, 6:35-40 (emphasis added).
`
`Again, the proposed construction of glitch comports with the specification.
`
`6
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`However, Patent Owner argues, without explicitly presenting an alternate
`
`
`
`
`
`
`
`construction, that “[t]he patent uses the word ‘glitch’ to refer to actual motion data
`
`deemed to not fit the signature of human motion indicative of someone preparing
`
`to interface with a device.” Response at 2. Patent Owner further mischaracterizes
`
`the term “glitch” by arguing that “the phrase ‘outside of an acceptable range’ (in
`
`the above-cited description) refers to a range of motion that, although accurately
`
`determined, does not warrant waking up the device from an idle state to an active
`
`state.” Response at 6 (emphasis added).
`
`Patent Owner seeks, through attorney argument, to limit a “glitch” to apply
`
`only to motion data that is “accurately determined.” This is wholly inconsistent
`
`with the specification. The specification provides several examples of “motion
`
`data” that is not accurate or accurately determined, such as “abnormal data,” data
`
`indicative of a “problem,” and data indicative that the accelerometer is
`
`“malfunctioning”:
`
`• “[A] glitch correcting logic which removes abnormal data from the motion
`
`data before the motion data is passed to the long average logic.” APPL-
`
`1001, 5:19-23 (emphasis added).
`
`• “[A]n excessive number of glitches may indicate a problem with the
`
`accelerometer … If the process determines that there have been an excessive
`
`number of glitches, the process, at block 645, generates an alert regarding
`
`7
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`the problem.” APPL-1001, 6:56-65 (emphasis added).
`
`
`
`
`
`
`
`• “In one embodiment, glitch notifier logic 237 may also notify the
`
`manufacturer. The glitches generally are indicative that the accelerometer or
`
`sensor is malfunctioning.” APPL-1001, 3:33-37 (emphasis added).
`
`In other words, the Specification describes numerous ways that a datum can
`
`fall outside an acceptable range, including abnormal data, data that indicative of a
`
`problem or malfunction. Patent Owner’s arguments with respect to “glitch”
`
`exclude these embodiments disclosed in the specification. “[A] claim interpretation
`
`that excludes a preferred embodiment from the scope of the claim is rarely, if ever,
`
`correct.” On-Line Techs., Inc. v. Bodenseewerk Perkin-Elmer GmbH, 386 F.3d
`
`1133, 1138 (Fed. Cir. 2004).
`
`For at least these reasons, Patent Owner’s arguments are unpersuasive. The
`
`term “glitch,” properly construed, includes a datum outside an acceptable range.
`
`Petition at 7; See APPL-1010, ¶42.
`
`B. Detecting a glitch is separate and distinct from determining
`whether a jostle or bump warrants awakening a device
`
`Patent Owner seeks to support is position by citing to passages in the
`
`Specification that describes ignoring jostles and bumps. Response at 5-8. For
`
`instance, “[a]t block 330, the process determines if the movement is a ‘real’ motion
`
`and not a mere jostle or bump. The device may move, for example, as a result of a
`
`8
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`little jostle of a desk or table on which the device is laying, a heavy step nearby, or
`
`
`
`
`
`
`
`something else that creates a very small motion, but which does not warrant
`
`waking up the device.” APPL-1001, 4:61-66.
`
`What Patent Owner fails to address, however, is that block 330 is not the
`
`entirety of the disclosure of the specification. First, block 330 does not describe the
`
`glitch removal process. In fact, nowhere is the term “glitch” found in the
`
`paragraphs describing block 330. Nor is there any discussion of removing the
`
`glitch in such paragraphs. See APPL-1001, 4:53-5:6. Claim 1 requires “removing
`
`the one or more glitches from the motion data.” APPL-1001, 8:34-39. Block 330,
`
`merely determines whether a detected motion “warrants awakening the device”
`
`and provides no discussion of removing one or more glitches. APPL-1001, 5:1-2.
`
`In contrast, Fig. 6 and its accompany text describes glitch removal. See also APPL-
`
`1001, 6:56-65.
`
`Nevertheless, Patent Owner attempts to characterize glitch as “a mere jostle
`
`or bump” such as from a purse being jostled, a desk on which the device rests
`
`being shaken, or a heavy step nearby. Response at 6-7 (citing APPL-1001, 4:62,
`
`1:63, 2:46−47, 2:47, 4:63−64). However, Patent Owner’s characterizations are
`
`inaccurate. In none of the cited passages is the motion (jostle, shake, bump)
`
`referred to as a “glitch.” Patent Owner conflates using motion data to determine
`
`whether to awaken the device with removing a “glitch” (i.e., some value outside an
`
`9
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`acceptable range, such as abnormal data, data that indicative of a problem or
`
`
`
`
`
`
`
`malfunction). Therefore, Petitioner maintains that the originally proposed
`
`construction is sufficient.
`
`C. A construction defining what a term includes is sufficient for this
`proceeding
`
`Patent Owner argues that “Petition offers no argument, evidence, or
`
`explanation for why a ‘glitch’ should be defined in terms of what it purportedly
`
`includes (an unreasonably broadening construction), as opposed to what it is.”
`
`Response at 7.
`
`However, the Petition need not define the entire metes and bounds of the
`
`claim term. Instead, Petitioner need only construe the claim to the extent necessary
`
`to resolve the issue. Vivid Techs., Inc. v. Am. Sci. & Eng’g, Inc., 200 F.3d 795, 803
`
`(Fed. Cir. 1999) (“[O]nly those terms need be construed that are in controversy,
`
`and only to the extent necessary to resolve the controversy.”). Here, the claim
`
`scope of “glitch” includes a datum that is outside of an acceptable range, consistent
`
`with the Specification. APPL-1001, 3:14-16; 5:19-23; 6:35-40.
`
`V. McMahan Teaches Verifying and Removing Glitches
`
`McMahan describes a “sensor 102 compris[ing] a vibrating beam
`
`accelerometer.” APPL-1005, 3:34. “When the output of sensor 102 is not within
`
`the expected range of its normal operation, it is presumed that the output is an
`
`error. This means that the output of the sensor is not an accurate reflection of the
`
`10
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`stimulus that the sensor is designed to monitor.” APPL-1005, 4:26-30. After
`
`
`
`
`
`
`
`determining (verifying) that such an error (glitch) is present, McMahan teaches
`
`that “enhancement circuit 104 ... provid[es] a value to electronic circuit 106 that is
`
`within the normal range of the output of sensor 102.” APPL-1005, 4:35-38. The
`
`sensor output error of McMahan is data outside an acceptable range and, thus,
`
`teaches a glitch. Petition at 31; APPL-1010, p. 46. Therefore, McMahan
`
`determines whether the motion data includes a glitch at step 304 and then removes
`
`the glitch from the motion data at step 308. APPL-1010, p. 46.
`
`A. McMahan’s error is no different than the claimed glitch
`
`Patent Owner argues: “The ‘error’ in McMahan bears no resemblance to the
`
`‘glitch’ disclosed and claimed in the ’646 patent” and that “[u]nlike the ‘error’ in
`
`McMahan, each example ‘glitch’ described in the ’646 patent indicates actual
`
`motion (generally characterized as ’a mere jostle or bump’) that the device is
`
`designed to monitor, though such motion does not warrant waking up the device.”
`
`Response at 10.
`
`Patent Owner’s arguments are entirely predicated on its improper
`
`construction of the term “glitch.” Specifically, Patent Owner conflates assessing
`
`whether jostles and bumps should awaken a device with removing glitches that are
`
`outside an acceptable range. For the reasons described above, the glitches in the
`
`’646 Specification include data that is outside an acceptable range, which includes
`
`11
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`abnormal and problematic values. See supra IV.A. McMahan’s errors (“output of
`
`
`
`
`
`
`
`the sensor is not an accurate reflection of the stimulus that the sensor is designed to
`
`monitor,” APPL-1005, 4:29-30) are similarly abnormal and problematic values.
`
`McMahan’s detection and removal of a value “not within the expected range of its
`
`normal operation” is thus no different than the detection and removal of glitches
`
`described in the ’646 Specification.
`
`B.
`
`Patent Owner attempts to narrow “motion data” improperly
`
`Patent Owner further alleges that McMahan does not teach removing the
`
`glitches from motion data as claimed, stating: “The ‘modify’ block 308 [of
`
`McMahan] refers to processing an erroneous output which, due to its impossible
`
`value, is never included as part of anything that can be considered motion data (and
`
`thus it cannot be removed from such data).” Response at 14.
`
`It appears that Patent Owner asserts that McMahan’s errors are removed
`
`before they become motion data. This is incorrect, as McMahan clearly shows the
`
`errors removed from the “sensor output” of block 302. There is no basis in the
`
`Specification, claim language, or file history of the ’646 Patent to narrow “motion
`
`data” to exclude sensor output such as in McMahan. Furthermore, Patent Owner
`
`undermines its own argument by pointing to McMahan at 4:31-34, which
`
`emphasizes the importance of removing glitches from the sensor output. See also
`
`APPL-1005 at 4:24-26.
`
`12
`
`

`

`
`
`
`
`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`
`
`
`
`
`APPL-1005, Fig. 3
`
`
`
`And if Patent Owner is arguing that “motion data” does not include errors or
`
`impossible values, that is incorrect as well for being inconsistent with Patent
`
`Owner’s own Specification. For instance, the ’646 Specification often refers to
`
`“motion data” as having abnormalities, unlikely, or impossible values that are to be
`
`removed. APPL-1001, 5:19-22 (“the long average logic receives the data from a
`
`glitch correcting logic which removes abnormal data from the motion data.”); also
`
`APPL-1001, 6:25-26 (“FIG. 6 is a flowchart of an embodiment of a process 600 to
`
`detect and correct glitches in motion data.”).
`
`C. McMahan’s modifying sensor values teaches removing glitches
`
`As noted in the Petition: “McMahan determines whether the motion data
`
`includes a glitch at step 304 and then removes the glitch from the motion data at
`
`13
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`step 308.” Petition at 31 (citing to step 308 of McMahan’s Figure 3, as teaching
`
`
`
`
`
`
`
`“remov[ing] the one or more glitches from the motion data.”). The Institution
`
`Decision is consistent in its reading of McMahan, noting, “McMahan teaches not
`
`only ‘modifying’ out of range output to be in range, but also that out-of-range
`
`output ‘is replaced’ with in-range values … When an output data set is modified or
`
`replaced, it no longer includes the original output. The original output has,
`
`therefore, been removed.” Institution Decision at 12.
`
`Patent Owner accuses the Board of impropriety, stating, “the Board argues
`
`on behalf of Petitioner that McMahan teaches ‘[w]hen an output data set is
`
`modified or replaced, it no longer includes the original input.’ … (reiterating that
`
`the PTAB may not adopt arguments on behalf of a petitioner that are not raised in
`
`the petition itself).” Response at 14, citing Rovalma, S.A. v. Bohler-Edelstahl
`
`GmbH & Co. KG, 856 F.3d 1019, 1027 (Fed. Cir. 2017). In Rovalma, the Federal
`
`Circuit stated:
`
`We rejected ‘the PTO's position that the Board is free to adopt arguments on
`behalf of petitioners that could have been, but were not, raised by the
`petitioner during an IPR’ and explained that ‘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.’
`
`
`
`Rovalma, 856 F.3d. at 1027 (citing SAS Inst., Inc. v. ComplementSoft, LLC, 825
`
`F.3d 1341, 1351 (Fed. Cir. 2016)).
`
`14
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`In the present case, the Board did not improperly adopt arguments on behalf
`
`
`
`
`
`
`
`of Petitioner because the Board’s remark concerns an argument explicitly made
`
`and advanced in the Petition. For example, the Petition states: “A POSITA would
`
`have understood that by modifying the sensor output so that the output is within
`
`the acceptable range, the error (glitch) is removed.” Petition at 25. Accordingly,
`
`Patent Owner’s assertion that this argument “was not raised in the petition itself” is
`
`baseless. Additionally, Patent Owner was provided a chance to respond to both the
`
`Petition and the Board’s corresponding statements in its Response. Thus, Patent
`
`Owner’s reliance on Rovalma does not negate the positions presented in the
`
`Petition that McMahan teaches the claimed removal of glitches.
`
`Thus, for the reasons stated in the Petition, McMahan’s modification of
`
`accelerometer data that is determined to be outside an acceptable range renders
`
`obvious removing glitches from the motion data. See Petition at 30-32; APPL-
`
`1010, pp. 46-47.
`
`VI. A POSITA Would Have Combined McMahan with Pasolini, Goldman,
`and Mizell
`
`The record clearly shows that a POSITA would have been motivated to
`
`apply McMahan’s error removal techniques to the teachings of Pasolini. For
`
`instance, a POSITA would have recognized the desirability of removing
`
`accelerometer signal errors to have more reliable accelerometer data, as taught by
`
`15
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`McMahan. Petition at 26, (citing APPL-1010, ¶82). Dr. Paradiso has testified, both
`
`
`
`
`
`
`
`in the originally filed Declaration and during his deposition, that it was well-known
`
`to POSITAs at the time of filing the ’646 patent that accelerometers had glitches
`
`and that it was desirable to remove them. See, e.g., APPL-1010, ¶82; Paradiso Dep.
`
`at 18, 20.
`
`Patent Owner challenges the motivation to combine McMahan with Pasolini
`
`and Goldman through a mix of unsupported attorney argument and a fixation on
`
`bodily incorporation of the references. As shown below, Patent Owner’s arguments
`
`should be dismissed.
`
`A. The record shows that a POSITA would have found it obvious to
`combine the teachings of McMahan with Pasolini, Goldman, and
`Mizell
`
`Petitioner’s expert explains: “[A] POSITA would have also recognized the
`
`desirability of removing accelerometer signal errors in order to have more reliable
`
`accelerometer data, as evidenced by McMahan. ... Here, applying McMahan’s
`
`modification of sensor data determined to be outside an acceptable range would
`
`have improved the accuracy of the accelerometer.” APPL-1010, ¶82 (cited in
`
`Petition at 26). Patent Owner challenges the Petition by arguing that: “McMahan
`
`does not even purport to address ‘more reliable accelerometer data.’ Indeed, no
`
`form of the word ‘reliable’ appears anywhere in the McMahan disclosure.”
`
`Response at 15. However, Patent Owner provides no authority that there must be
`
`16
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`word-for-word correspondence between the prior art and the POSITA’s
`
`
`
`
`
`
`
`understanding of such art. Moreover, Patent Owner misses the point—a POSITA,
`
`looking at the teachings of McMahan and Pasolini, would have recognized the
`
`desirability to improve the reliability of Pasolini’s accelerometer with the teachings
`
`of McMahan. See Petition at 26.
`
`In addition to the clear showing made in the Petition and its accompanying
`
`Declaration, the deposition testimony of Dr. Paradiso further supports the
`
`conclusion of obviousness. Specifically, Dr. Paradiso further testifies that
`
`glitches/errors in accelerometers were well-known to occur and that it was
`
`desirable to remove them.
`
`And a POSITA would be aware of glitches because we saw them, you
`know, we encountered them. The early accelerometers were even
`worse. And we would always put in a glitch detector because these
`glitches would affect the accuracy adversely.
`So a POSITA would be aware of the glitches just by looking at the
`data and would be aware that there's a reason to get rid of them. And
`I think McMahan gives a great example of doing this. So a POSITA
`would want to definitely combine the essence of what McMahan
`says in glitch removal to the devices of the sort that Pasolini and
`Goldman discuss.
`
`Paradiso Dep. at 20 (emphasis added).
`
`This is an example, an example of something out of the range. And
`
`17
`
`

`

`
`
`
`
`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`we can identify the magnitude of the spike, the unphysical spike,
`which doesn't correspond to normal activity, versus the typical
`motion. And we set that threshold high enough so we don't exclude
`normal motion. We will exclude the spike. And this is common
`practice -- was common practice well before the date of that patent
`for people working with things like accelerometers.
`
`Paradiso Dep. at 18 (emphasis added).
`
`
`
`
`
`Patent Owner has not provided any evidence to refute Dr. Paradiso’s
`
`testimony given at deposition. Thus, Patent Owner’s arguments that are in
`
`controversy with Dr. Paradiso’s statements about a POSITA’s desire to remove
`
`glitches amount to nothing more than attorney argument. See Gemtron Corp. v.
`
`Saint-Gobain Corp., 572 F. 3d 1371, 1380 (Fed. Cir. 2009) (“[U]nsworn attorney
`
`argument . . . is not evidence and cannot rebut . . . evidence.”). Accordingly, for the
`
`reasons stated in the Petition, a POSITA would have found it obvious to combine
`
`the teachings of McMahan with those of Pasolini, Goldman, and Mizell.
`
`B.
`
`The teachings of McMahan’s sensors are applicable to the
`accelerometers of Pasolini and Goldman.
`
`Patent Owner’s assertions about the compatibility of McMahan’s
`
`accelerometer with those of Pasolini and Goldman are premised on a
`
`mischaracterization of McMahan. Patent Owner argues that: “These facts further
`
`undermine Petitioner’s unsupported argument that the sensor-specific enhancement
`
`18
`
`

`

`Petitioner’s Reply
`IPR2018-00289 (Patent No. 8,872,646)
`
`
`of McMahan cited in the Petition could be applied to the entirely different
`
`
`
`
`
`
`
`‘accelerometer of the Pasolini and Goldman combination.’” Response at 18
`
`(emphasis added). However, Patent Owner is inappropriately relying on a single
`
`identified embodiment of McMahan.
`
`Despite Patent Owner’s assertions, McMahan’s disclosure is broadly
`
`applicable to accelerometers and sensors in general. For example, McMahan states:
`
`“In one embodiment, sensor 102 comprises an accelerometer or other appropriate
`
`sensor for monitoring a selected stimulus.” APPL-1005, 3:29-31 (emphasis
`
`added). Further, “In