`
`
`
`
`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 No. IPR2018-00424
`Patent No. 7,881,902
`
`_____________________
`
`
`PETITIONER’S REPLY
`
`
`
`
`
`
`
`

`

`
`
`
`
`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`
`
`
`
`
`I. 
`II. 
`
`TABLE OF CONTENTS
`Introduction. ..................................................................................................... 1 
`Patent Owner’s Response entirely consists of attorney argument.
` ......................................................................................................................... 1 
`III.  Claim Construction .......................................................................................... 2 
`A. 
`The term “dominant axis” is properly construed to include
`the axis most influenced by gravity. ..................................................... 2 
`1. 
`Petitioner’s construction is not required to define
`the full scope of the term. ........................................................... 3 
`Petitioner’s construction does not
`improperly
`exclude embodiments from the specification. ............................ 4 
`The term “cadence window” is properly construed to
`include a window of time since a last step was counted. ...................... 6 
`The term “periodically sampling acceleration data at a
`predetermined sampling rate, wherein each sample
`includes acceleration data measured by the inertial sensor
`over a predetermined period of time.” .................................................. 7 
`IV.  The combination of Pasolini and Fabio was never considered
`during original prosecution. ............................................................................. 8 
`Fabio’s validation interval teaches the “cadence window.” ............................ 8 
`A. 
`Patent Owner’s
`arguments
`are
`based
`on
`a
`mischaracterization of Fabio’s validation interval. ............................... 9 
`Patent Owner’s arguments based on
`its
`incorrect
`understanding of Fabio fail. ................................................................ 12 
`VI.  Fabio discloses the dynamic step cadence window limitations. ................... 14 
`A. 
`Fabio discloses “determining a dynamic step cadence
`window.” ............................................................................................. 14 
`
`2. 
`
`B. 
`
`C. 
`
`V. 
`
`B. 
`
`i
`
`

`

`
`
`
`
`
`
`
`B. 
`
`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`Fabio discloses “using the dynamic step cadence window
`to identify the time frame within which to monitor for the
`next step.” ............................................................................................ 15 
`VII.  Both Fabio and Pasolini teach “assigning a dominant axis based
`on the orientation.” ........................................................................................ 15 
`VIII.  Mitchnick teaches a mobile device. ............................................................... 17 
`A. 
`The combination of Mitchnick’s internal and external
`embodiments render obvious
`the claimed “mobile
`device.” ................................................................................................ 18 
`A POSITA would have combined Mitchnick’s internal
`and external embodiments to teach the claimed mobile
`device. .................................................................................................. 22 
`C.  Mitchnick renders obvious “detecting motion by an
`inertial sensor included in a mobile device.” ...................................... 25 
`D.  Mitchnick discloses “determining, by the mobile device
`whether the motion has a motion signature indicative of a
`user activity that the mobile device is configured to
`monitory.” ............................................................................................ 25 
`IX.  Fabio teaches the same sample of acceleration data to teach the
`claimed “current acceleration measurement” recited in claim 9. .................. 26 
`The dependent claims are obvious ................................................................. 27 
`X. 
`XI.  Conclusion ..................................................................................................... 28 
`XII.  Certificate of Word Count ............................................................................. 29 
`
`
`B. 
`
`ii
`
`

`

`
`
`
`
`Ex.1001
`Ex.1002
`
`Ex.1003
`
`Ex.1004
`
`Ex.1005
`
`Ex.1006
`
`Ex.1007
`
`Ex.1008
`
`Ex.1009
`
`Ex.1010
`
`Ex.1011
`
`Ex.1012
`
`Ex.1013
`
`
`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`PETITIONER’S EXHIBIT LIST
`
`
`
`
`
`Updated: January 17, 2019
`
`U.S. Patent No. 7,881,902
`
`Prosecution History of U.S. Patent No. 7,881,902
`
`Declaration of Joe Paradiso, Ph.D., under 37 C.F.R. § 1.68
`
`Curriculum Vitae of Joe Paradiso
`
`U.S. Patent No. 7,463,997 to Fabio Pasolini et al. (“Pasolini”)
`
`U.S. Patent No. 7,698,097 to Fabio Pasolini et al. (“Fabio”)
`
`U.S. Publication No. 2006/0084848 to Mitchnick (“Mitchnick”)
`
`U.S. Patent No. 6,469,639 to Tanenhaus et al. (“Tanenhaus”)
`
`U. S. Patent No. 5,957,957 to Sheldon (“Sheldon”)
`
`Reserved
`
`Reserved
`
`Prosecution History of U.S. Patent No. 8,712,723
`
`U. S. Patent No. 5,559,497 to Hong (“Hong”)
`
`iii
`
`

`

`
`
`
`Introduction.
`
`I.
`
`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`
`
`
`
`
`The Petition and the record as a whole provide detailed reasons why the
`
`cited art renders the challenged claims of the’902 patent obvious. None of the
`
`arguments in the Response adequately refute the evidence of record. Moreover, the
`
`evidence weighs in Petitioner’s favor because the Response relies on Patent
`
`Owner’s expert, Mr. Easttom, regarding two issues, neither of which rebut the
`
`application of the prior art to the challenged claims.
`
`Patent Owner’s remaining arguments also fail because they rely on
`
`irrelevant prosecution history of a related, later filed continuation (see Response,
`
`pp.3-5), incorrectly argue that the proposed claim constructions exclude other
`
`embodiments (see Response, pp.6-11), mischaracterize Fabio’s teachings regarding
`
`its validation interval (see Response, pp.13-20) and Pasolini’s teachings regarding
`
`using a 3-axis accelerometer and determining the axis most influenced by gravity
`
`(see Response, p.20), and disregard evidence that Mitchnick’s embodiments teach
`
`a mobile device. As discussed below, Patent Owner’s arguments are incorrect and
`
`unsupported, and should thus be rejected.
`
`II.
`
`Patent Owner’s Response entirely consists of attorney argument.
`
`Patent Owner’s Response is nothing more than unsupported attorney
`
`argument. Tellingly, the Response only cites to two portions of Mr. Easttom’s
`
`declaration filed pre-institution, neither of which refutes the application of the prior
`
`1
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`art to the challenged claims as set forth in the Petition. See Response, pp.3,15. The
`
`
`
`
`
`
`
`first citation is directed to Patent Owner’s definition of the level of ordinary skill in
`
`the art, which Patent Owner admits “is inconsequential to the dispute here.” See
`
`Response, p.3. The second citation is intended but fails to support a statement that
`
`“Fabio defines its [validation interval] TV as necessarily starting before the last
`
`step is counted.” See Response, p.15.
`
`“[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. See Apple, Inc. v. Evolved Wireless LLC, IPR2016-00758,
`
`Paper 48, p.20 (P.T.A.B. Mar. 12, 2018) (“we keep in mind that attorney argument
`
`is not evidence and Patent Owner cannot rebut evidence with unsworn attorney
`
`argument.”). Accordingly, the Response amounts to nothing more than attorney
`
`argument that does not sufficiently rebut the evidence of record.
`
`III. Claim Construction
`A. The term “dominant axis” is properly construed to include the
`axis most influenced by gravity.
`
`Claim 10 of the ’902 patent broadly recites “determining an orientation of
`
`the mobile device with respect to gravity” and “assigning a dominant axis based on
`
`the orientation.” Ex.1001, 16:40-42. The Petition establishes that the term
`
`“dominant axis” “includes ‘the axis most influenced by gravity.’” See Petition, p.9.
`
`2
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`This is supported by both the claim language and the ‘902 patent’s specification.
`
`
`
`
`
`
`
`See Ex.1001, 14:37-41, 6:20-26; Ex.1003, p.13.
`
`More specifically, the specification explains that “the dominant axis is
`
`assigned after identifying a gravitational influence. The gravitational influence
`
`may be identified by calculating total acceleration based upon the acceleration on
`
`each axis.” Ex.1001, 14:36-38. The specification also explains that “once the
`
`orientation is determined, a dominant axis is assigned based upon the orientation.
`
`Determining an orientation of the electronic device 100 may include identifying a
`
`gravitational influence.” Ex.1001, 6:13-16. Based on these disclosures, the
`
`specification contemplates that the dominant axis is, at a minimum, “the axis most
`
`influenced by gravity, which may change over time (e.g., as the electronic device is
`
`rotated).” Ex.1001, 6:16-18; Petition, p.6. Accordingly, Petitioner’s construction is
`
`fully supported by the claim language and the specification.
`
`1.
`
`Petitioner’s construction is not required to define the full
`scope of the term.
`
`Rather than provide its own construction of “dominant axis,” Patent Owner
`
`argues that Petitioner’s construction is too narrow. See Response, pp.6-8.
`
`According to Patent Owner, Petitioner’s construction “violates the well-established
`
`rule against reading limitations from the specification into the claim language” and
`
`that “the teachings of the specification cited by Petitioner clearly state that they are
`
`3
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`only example embodiments and are not meant to be limiting.” Response, p.6
`
`
`
`
`
`
`
`(emphasis in the original).
`
`Patent Owner’s argument is irrelevant. Petitioner’s construction is not
`
`required to cover the full metes and bounds of the claim term but need only
`
`construe the term 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.”). Emerson Electric Co., v. Ipco, LLC, IPR2017-00007
`
`(P.T.A.B., March 29, 2018), Paper 42, p.8 (holding that further construction of
`
`terms defined by what they include is not necessary to resolve the dispute
`
`regarding the asserted grounds of unpatentability). Here, Petitioner’s construction
`
`of “dominant axis” includes, at a minimum, “the axis most influenced by gravity,”
`
`which is sufficient to resolve the dispute at issue here.
`
`2.
`
`Petitioner’s construction does not improperly exclude
`embodiments from the specification.
`
`Patent Owner further argues that the construction of “dominant axis” in the
`
`Petition “would impermissibly exclude preferred embodiments.” Response, p.8.
`
`Patent Owner then cites to various embodiments in the ’902 patent that would
`
`allegedly be excluded. See Response, p.11. A closer inspection of these
`
`embodiments shows the opposite.
`
`4
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`For example, Patent Owner refers to the specification’s statement that:
`
`
`
`
`
`
`
`“[t]herefore, a new dominant axis may be assigned when the orientation of the
`
`electronic device 100 and/or the inertial sensor(s) attached to or embedded in the
`
`electronic device 100 changes.” Response, p.7 (citing Ex.1001, 6:19-22). Patent
`
`Owner offers no explanation detailing how Petitioner’s construction excludes this
`
`embodiment, other than the conclusory statement that this is “yet another example
`
`of a different way to determine the ‘dominant axis.’” Response, p.7. Further
`
`description of this embodiment, though, indicates that “an orientation of the
`
`electronic device 100 may include identifying a gravitational influence.” Ex.1001,
`
`6:15-16. Thus, even in this embodiment, the dominant axis is assigned based on a
`
`change in the orientation of the device, determined by the influence of gravity.
`
`Patent Owner also refers to the specification’s statement that “[i]n one
`
`embodiment, the dominant axis corresponds to a virtual axis that is a component of
`
`a virtual coordinate system.” Ex.1001, 6:28-30, Response, p.7. Again, Patent
`
`Owner does not proffer any explanation of how this embodiment excludes the
`
`influence of gravity. Instead, upon further inspection, a virtual axis may also be the
`
`axis most influenced by gravity. See Ex.1001, 6:30-37. Thus, Petitioner’s
`
`construction of “dominant axis” does not exclude embodiments from the
`
`specification.
`
`Patent Owner further alleges that “the ’902 patent reveals that using a
`
`5
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`rolling-average process to determine orientation does not necessarily result in
`
`
`
`
`
`
`
`assigning a dominant axis that is most influenced by gravity.” Response, pp.8-9.
`
`Here, Patent Owner seemingly argues that the dominant axis should be broader
`
`than “the axis most influenced by gravity.” This fails for two reasons. First, the
`
`embodiments disclosed in the specification, even those referenced by Patent
`
`Owner, determine the dominant axis based on the influence of gravity. Second, a
`
`broader construction would still include Petitioner’s construction, which is
`
`disclosed by the prior art as outlined in the Petition. See, e.g., Petition, pp.69-70.
`
`Thus, the Response does not sufficiently refute the Petition’s claim construction
`
`nor the Petition’s showing that the prior art satisfies a broader construction.
`
`B.
`
`The term “cadence window” is properly construed to include a
`window of time since a last step was counted.
`
`Petitioner proposes that the term “cadence window” includes a “window of
`
`time since a last step was counted that is looked at to detect a new step.” See
`
`Petition pp.9-10; Ex.1001, 3:67-4:1. The Board, agreeing with Patent Owner,
`
`believes that no construction is necessary. See Institution Decision, p.15. Whether
`
`under Petitioner’s construction or the plain and ordinary meaning, the analysis set
`
`forth in the Petition and Dr. Paradiso’s expert declaration establishes that the
`
`combination of Pasolini and Fabio renders the claimed “cadence window” obvious.
`
`See Petition, pp.50-55; Ex.1003, pp.76-80.
`
`6
`
`

`

`
`
`
`
`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`C. The term “periodically sampling acceleration data at a
`predetermined sampling rate, wherein each sample includes
`acceleration data measured by the inertial sensor over a
`predetermined period of time.”
`
`
`
`The Board, sua sponte, construed this term (see claim 3) to mean
`
`“periodically obtaining acceleration data by sampling the acceleration data over a
`
`predetermined period of time.” Institution Decision, p.16. Petitioner believes that
`
`no construction is required. Patent Owner holds the same belief, but if construed,
`
`argues for a construction of “periodically obtaining acceleration data at a
`
`predetermined sampling rate by sampling the acceleration data over a
`
`predetermined period of time.” Response, p.13 (emphasis original).
`
`To the extent that there is any difference between the Board’s construction
`
`and Patent Owner’s proposed construction, Mitchnick teaches sampling the
`
`acceleration data at a predetermined sampling rate of, e.g., 10, 20, or 50 Hz.
`
`Ex.1007, ¶73 (“Typical accelerometer sampling rates are approximately at least 10
`
`Hz, or preferably approximately 20 Hz, to approximately 50 Hz or higher.”); see
`
`also Petition, p.24; Ex.1003, p.38. Moreover, regardless of any construction of this
`
`term, Patent Owner does not separately argue the patentability of claim 3 under
`
`any construction, and therefore does not rebut either the Board’s analysis or the
`
`analysis offered in the Petition.
`
`7
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`IV. The combination of Pasolini and Fabio was never considered during
`original prosecution.
`
`
`
`
`
`
`
`Patent Owner notes that Pasolini was considered during prosecution of U.S.
`
`Patent No. 8,712,723, which issued from a continuation of the ’902 patent. See
`
`Response, pp.3-5. Patent Owner, however, has not identified any statute, case law,
`
`or other authority showing the relevance of Pasolini being considered during
`
`prosecution of the later filed continuation. Moreover, the Petition does not rely on
`
`Pasolini alone but the combination of Pasolini with Fabio, which was not
`
`considered as Fabio does not appear in the file history of the ’723 patent. See
`
`Ex.1012. Thus, the Petition presents a new, previously unconsidered argument.
`
`V.
`
`Fabio’s validation interval teaches the “cadence window.”
`
`The Petition identifies Fabio’s “validation interval” as corresponding to the
`
`claimed “cadence window.” Petition, p.52. Patent Owner challenges this
`
`comparison by stating that “a ‘validation interval’ (TV) is used for a prior step—
`
`not the next step.” Response, p.14. Patent Owner then makes several incorrect
`
`statements regarding Fabio’s validation interval as supposed reasons why it is
`
`different than the claimed “cadence window.” Moreover, Patent Owner’s
`
`characterization of Fabio’s validation interval still teaches the claimed “cadence
`
`window.”
`
`8
`
`

`

`
`
`
`A.
`
`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`Patent Owner’s arguments are based on a mischaracterization of
`Fabio’s validation interval.
`
`
`
`
`
`The Petition cites to Fabio’s validation of detected steps that fall within a
`
`validation interval as rendering obvious “using a default step cadence window to
`
`identify a time frame within which to monitor for a next step.” Petition, pp.50-55.
`
`Patent Owner argues that Fabio’s validation interval does not meet the “cadence
`
`window” limitation. Response, pp.11-17. These arguments, though, are based on
`
`the incorrect premise that “Fabio’s [validation interval] is retrospective at least in
`
`that it is used to validate only the immediately preceding step (shown in Fig.6 as
`
`K-1), as opposed to the current step detected (shown in Fig.6 as K).” Response,
`
`pp.13-15. This flawed interpretation is based on a single sentence in Fabio, taken
`
`out of context, stating that “[m]ore precisely, the last step recognized is validated if
`
`the instant of recognition of the current step TR(K) falls within a validation interval
`
`TV[.]” See Response, p.15 (quoting Ex.1006, 4:35-39). Patent Owner apparently
`
`believes that the term “last step recognized” means the step recognized in some
`
`previous step cycle1, not the current step cycle. Fabio, though, does not detect a
`
`
`1 For purposes of this paper, the term “step cycle” refers to Fabio’s iterative
`
`process of recognizing (e.g., 225), validating (e.g., 230) and buffering/counting
`
`(235) a step upon acquisition of an acceleration sample. See Ex.1006, Figs.4,7.
`
`9
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`step in one step cycle, and then proceed to a next cycle to validate and buffer/count
`
`
`
`
`
`
`
`that step. Rather, Fabio recognizes, validates, and either buffers or counts a step
`
`within a single step cycle. See, e.g., Ex.1006, Figs.4,7.
`
`Further, Fabio teaches first recognizing an acceleration signal as a step by
`
`verifying “whether the time plot of the acceleration signal AZ (i.e., the sequence of
`
`the samples acquired) has pre-determined characteristics.” Id., 4:12-15. Fabio then
`
`teaches that “[i]f … the step-recognition test is passed,” the system “executes a
`
`first validation test, corresponding to the regularity of the individual step.” Id.,
`
`4:22-27. Depending on the state of Fabio’s system, the recognized and validated
`
`step will either be buffered (in the first counting procedure) or counted (in the
`
`second counting procedure). Id., 5:10-13, 6:40-43.
`
`This is also shown in Fabio’s Fig.4:
`
`10
`
`

`

`
`
`
`
`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`
`
`
`
`
`Increment a step count
`in a step count buffer
`variable NVC
`
`
`
`
`
`Increment valid step count NVT if regular steps are detected
`
`
`
`Ex.1006, Fig.4 (annotated); Petition, p.49; Ex.1003, p.70. As shown in Fig.4, Fabio
`
`recognizes a step at 225, validates the step immediately upon recognition at 230,
`
`and buffers the step upon successful validation at 255. See Ex.1006, 4:12-5:13;
`
`Petition, pp.50-51. Then, at 260, if the buffer count (NVC) is less than a threshold
`
`(NT2), the process returns to 200 and the cycle repeats. Ex.1006, 5:13-29; Ex.1003,
`
`pp.70-71.
`
`Because the flowchart in Fig.4 and the corresponding text clearly show that
`
`11
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`steps are recognized, validated, and buffered in a single cycle,2 Patent Owner’s
`
`
`
`
`
`
`
`interpretation—that step recognition occurs in one cycle and validation and occur
`
`in another cycle—is incorrect. Moreover, the Response cites to no testimonial
`
`evidence supporting its failed interpretation. Thus, the Response fails to distinguish
`
`the challenged claims over Fabio.
`
`B.
`
`Patent Owner’s arguments based on its incorrect understanding
`of Fabio fail.
`
`Relying on its flawed interpretation of Fabio, Patent Owner makes several
`
`failing arguments. First, Patent Owner argues that: “Fabio’s validation interval TV
`
`is not ‘a window of time since a last step was counted’ (as required by Petitioner’s
`
`construction) at least because Fabio defines its TV as necessarily starting before
`
`the last step is counted.” Response, p.15. However, despite this unsupported
`
`assertion, the validation interval for a currently recognized step is the window of
`
`time since (i.e., after) the last step was counted because the last step (K-1) was
`
`recognized, validated, and counted in the previous step cycle (TR(K-1)). This is
`
`represented in Fabio’s Fig.6, reproduced below:
`
`
`2 Fabio’s Fig.7 shows a similar process in the second counting procedure where
`
`steps are recognized, validated, and counted in a single step cycle.
`
`12
`
`

`

`
`
`
`
`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`
`
`
`
`
`Instant of recognition of current step
`
`Instant of
`recognition of
`previous step
`
`Validation Interval
`
`
`
`Ex.1006, Fig.6 (annotated); see also Petition, p.51; Ex.1003, p.73.
`
`
`
`Second, Patent Owner alleges that time TS1 is not a default cadence window
`
`because “TS1 precedes the step validation procedure, which is backwards-looking
`
`and functions to validate the previous step.” Response, pp.16-17. However, TS1 is
`
`not backward-looking. As the Board pointed out, TS1 is a default cadence window
`
`because it is the maximum time, such as 10 seconds, within which samples of the
`
`acceleration data AZ (read in step 200) can be tested to recognize (step 225) and
`
`validate (step 230) steps. See Ex.1006, 3:60-4:11, Fig.4; Institution Decision, p.42.
`
`Further, as the Petition explains, TS1 is a default cadence window because the first
`
`counting procedure “terminates when a time interval T, that has elapsed from the
`
`last step recognized is longer than a first time threshold TS1, for example 10 s.”
`
`Ex.1006, 3:30-32. In other words, “a sequence of steps corresponding to a regular
`
`gait of a user is recognized” when the last step occurs within TS1. Petition, p.53;
`
`13
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`Ex.1006, 3:27-28; Ex.1003, p.75. The Response thus fails to present any argument
`
`
`
`
`
`
`
`that rebuts the evidence showing Fabio teaching the “default cadence window.”
`
`VI. Fabio discloses the dynamic step cadence window limitations.
`
`The Petition shows that Fabio’s step cadence widow is dynamic because the
`
`validation interval TV changes from step to step, based in part on the previous
`
`step. Petition, p.56. The Response challenges this comparison by alleging that
`
`Fabio’s validation interval does not disclose a step cadence window. Response,
`
`pp.17-19. Many of these arguments are addressed above in section V. The
`
`remaining arguments are addressed below.
`
`A.
`
`Fabio discloses “determining a dynamic step cadence window.”
`
`Patent Owner alleges that Fabio fails to disclose the “a dynamic step
`
`cadence window” because “the last recognized step in Fabio comes at the same
`
`frequency as steps made previously.” Response, p.19. This again mischaracterizes
`
`Fabio’s teaching. Fabio does not teach “the same frequency.” Rather Fabio
`
`explicitly teaches that its validation interval is “defined with respect to the instant
`
`of recognition of the immediately preceding step TR(K-1).” Ex.1006, 4:37-39;
`
`Petition, p.56. Fabio further teaches that validation “occurs when the duration ΔTK
`
`of a current step K is substantially homogeneous with respect to the duration ΔTK-1
`
`of an immediately preceding step K-1” and defines the validation interval TV in
`
`terms of ΔTK-1 , namely as TV=[TR(K−1)+ΔTK−1−TA, TR(K−1)+ΔTK−1 +TB] and
`
`14
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`where TA = ΔTK-1/2 and TB = ΔTK-1. Petition, p.56; Ex.1003, p.79; Ex.1006, 4:36-
`
`
`
`
`
`
`
`49. According to Dr. Paradiso, defining the validation interval TV in this way
`
`compensates for changes in each step. See Ex.1003, p.79; Petition, p.56.
`
`Accordingly, when the duration ΔTK−1 changes from step to step, the values for TA
`
`and TB likewise change, thus changing the validation interval TV from step to
`
`step. See Ex.1003, p.79; Petition, p.56. This teaches a dynamic step cadence
`
`window. See Ex.1003, pp.79-80; Petition, pp.56-57.
`
`Finally, Patent Owner alleges that the Petition’s reliance on Dr. Paradiso’s
`
`declaration is speculative and conclusory. See Response, p.9. However, the Petition
`
`does not rely on the declaration alone, but also cites to evidence in Fabio to show
`
`Fabio disclosing “a dynamic step cadence window.” Ex.1006, 4:28-49.
`
`B.
`
`Fabio discloses “using the dynamic step cadence window to
`identify the time frame within which to monitor for the next step.”
`
`Again, Patent Owner alleges that Fabio fails to disclose this limitation
`
`because Fabio fails to teach the claimed “cadence window.” Response, p.20.
`
`Because Fabio discloses the cadence window for the reasons discussed in Section
`
`V, the Patent Owner fails to present any separate argument that effectively
`
`overcomes the analysis in the Petition and the evidence of record.
`
`VII. Both Fabio and Pasolini teach “assigning a dominant axis based on the
`orientation.”
`
`Patent Owner argues that Pasolini fails to teach “assigning a dominant axis”
`
`15
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`recited in claim 10 because “the dominant axis may be other than vertical
`
`
`
`
`
`
`
`(gravity).” Response, pp.20-21. This argument is irrelevant.
`
`First, there is no evidence in the record that necessarily requires the
`
`dominant axis to not be the axis most influenced by gravity. Instead, there is
`
`substantial evidence that the ’902 patent contemplates the dominant axis including
`
`the axis most influenced by gravity. See, e.g., Ex.1001, 14:34-41 (“[T]he dominant
`
`axis is assigned after identifying a gravitational influence.”).
`
`Further, the Petition provides evidence from both Pasolini and Dr. Paradiso
`
`showing that Pasolini teaches using a 3-axis accelerometer and “identifying the
`
`main vertical axis to be used for step detection as the axis of detection that has the
`
`highest mean acceleration value Accm (on account of gravity).” Ex.1005, 8:11-24;
`
`Petition, p.69; Ex.1003, pp.97-98. The Petition further relies on Dr. Paradiso’s
`
`testimony stating that a “POSITA would understand the main vertical axis to be a
`
`dominant axis because the main vertical axis is the axis most aligned with gravity
`
`(i.e., has the highest mean acceleration value Accm on account of gravity).”
`
`Petition, pp.68-70; Ex.1003, pp.97-98. Similarly, the’902 patent states that “[i]n
`
`one embodiment, the dominant axis is assigned after identifying a gravitational
`
`influence.” Ex.1001, 2:9-10; see also Petition, p.6 (emphasis added); Ex.1003,
`
`p.15.
`
`Second, Patent Owner incorrectly argues that “[n]either Fabio nor Pasolini
`
`16
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`are interested” in a dominant axis, because “both Fabio and Pasolini are interested
`
`
`
`
`
`
`
`in a vertical axis with respect to gravity.” Response, p.21. This argument is
`
`directed to one of Pasolini’s embodiments that uses a single-axis accelerometer,
`
`(see Ex.1005, 2:60-63), but ignores Pasolini’s other embodiment that uses a 3-axis
`
`accelerometer to determine device orientation at each sample iteration (see
`
`Ex.1005, 8:11-24). The use of this 3-axis accelerometer embodiment is what the
`
`Petition identifies as teaching the dominant axis limitations. See Petition, pp.67-68;
`
`Ex.1003, pp.97-98.
`
`Specifically, Pasolini teaches that “accelerometer 2 could be equipped with a
`
`number of axes of measurement, for example three mutually orthogonal axes of
`
`measurement.” Ex.1005, 8:11-24; see also Ex.1003, p.97. When using this
`
`embodiment, Pasolini teaches “identifying the main vertical axis” on account of
`
`gravity “at each acquisition of a new acceleration sample.” Ex.1005, 8:15-22. This
`
`identifying of the main vertical axis with respect to gravity functionally assigns a
`
`dominant axis because it accounts for changes in gravity as the orientation of the
`
`device, and the accelerometer inside the device, changes. See Ex.1003, p.97. Thus,
`
`the combination of Fabio and Pasolini teaches this limitation.
`
`VIII. Mitchnick teaches a mobile device.
`
`The Petition identifies Mitchnick’s internal embodiment implemented as an
`
`external device to teach the “mobile device” in claim 1. Petition, pp.10-11;
`
`17
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`Ex.1003, pp.18-19. Patent Owner challenges this arguing that Mitchnick’s device
`
`
`
`
`
`
`
`resides “in a cavity of, a participant” with a purpose of “sensing and recording data
`
`relevant to [a] clinical trial or study.” Response, p.22. Patent Owner then makes
`
`several incorrect arguments regarding why Mitchnick’s internal device cannot
`
`function externally and be a mobile device, while ignoring explicit teaching that
`
`the device can reside “in or on the body.” Petition, p.9; Ex.1007, ¶43.
`
`A. The combination of Mitchnick’s internal and external
`embodiments render obvious the claimed “mobile device.”
`
`First, Patent Owner alleges that Mitchnick does not disclose a “mobile
`
`device” because Mitchnick’s primary embodiment is “a vaginally implanted
`
`medical device.” Response, p.22. The claims of the ’902 patent, however, provide
`
`no language that specifically excludes an implantable device, nor do the claims
`
`explicitly define the requirements of device mobility in any specific way. The only
`
`supposed restriction in the ’902 patent states that the device “may be carried in a
`
`backpack, pocket, purse, hand, or elsewhere.” Ex.1001, 2:35-36. Not even this
`
`disclosure excludes an implantable device.
`
`Moreover, the combination of Mitchnick’s internal and external
`
`embodiments teaches the claimed “mobile device.” The Petition explains that
`
`Mitchnick’s internal device can be modified to be an external device, because
`
`Mitchnick’s device is small (as shown in Ex.1007, Fig.5, shown below), battery
`
`18
`
`

`

`Petitioner’s Reply
`IPR2018-00424 (Patent No. 7,881,902)
`
`
`powered, can be carried by a human, and can communicate with other devices
`
`
`
`
`
`
`
`wirelessly. See Petition, p.13; Ex.1003, pp.22-24.
`
`Mobile Device
`
`
`
`Ex.1007, Fig.5B (annotated); see also Ex.1003, p.23.
`
`To the extent that Patent Owner’s argument against an implantable
`
`embodiment is even relevant, it only addresses one aspect of Mitchnick and
`
`otherwise ignores Mitchnick’s explicit teachings of its external embodiment.
`
`Patent Owner also fails to rebut expert testimony that Mitchnick’s device can be
`
`externally attached because it can reside “in or on t