`
`
`
`
`
`Trials@uspto.gov
`571-272-7822 Entered: March 12, 2018
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`TOMTOM, INC.,
`Petitioner,
`
`v.
`
`BLACKBIRD TECH, LLC d/b/a BLACKBIRD TECHNOLOGIES, LLC
`Patent Owner.
`____________
`
`Case IPR2017-02023
`Patent 6,434,212
`____________
`
`
`
`Before DEBRA K. STEPHENS, THOMAS L. GIANNETTI, and
`CHRISTA P. ZADO, Administrative Patent Judges.
`
`STEPHENS, Administrative Patent Judge.
`
`
`
`DECISION
`Institution of Inter Partes Review
`37 C.F.R. § 42.108
`
`
`
`
`
`
`
`
`
`
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`
`INTRODUCTION
`
`TomTom, Inc. (“Petitioner”) filed a Petition requesting an inter partes
`
`review of claims 1–8 of U.S. Patent No. 6,434,212 B2 (Ex. 1001, “the ’212
`patent”) (Paper 1 (“Pet.”)). Blackbird Tech LLC (“Patent Owner”) filed a
`Preliminary Response (Paper 6 (“Prelim. Resp.”)).
`
`We have authority to determine whether to institute an inter partes
`review under 35 U.S.C. § 314, which provides that an inter partes review
`may not be instituted unless the information presented in the petition “shows
`that there is a reasonable likelihood that the petitioner would prevail with
`respect to at least 1 of the claims challenged in the petition.”
`
`For the reasons set forth below, we institute an inter partes review of
`claims 1–8 of the ’212 patent.
`
` BACKGROUND
`
`Related Matters
`The parties advise us that the ’212 patent is at issue in the following:
`
`Blackbird Tech LLC d/b/a Blackbird Technologies v. Garmin
`International, Inc. and Garmin USA, Inc., Case No. 16-CV-689 (D. Del.),
`
`Blackbird Tech LLC d/b/a Blackbird Technologies v. Fitbit, Inc., Case
`No. 16-CV-683 (D. Del.),
`
`Blackbird Tech LLC d/b/a Blackbird Technologies v. Aliphcom d/b/a
`Jawbone, Case No. 16-CV-684 (D. Del.),
`
`Blackbird Tech LLC d/b/a Blackbird Technologies v. Sony Corp. et
`al., Case No. 16-CV-685 (D. Del.),
`
`2
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`
`Blackbird Tech LLC d/b/a Blackbird Technologies v. Timex Group
`USA, Inc., Case No. 16-CV-686 (D. Del.),
`
`Blackbird Tech LLC d/b/a Blackbird Technologies v. TomTom, Inc.,
`Case No. 16-CV-687 (D. Del.), and
`
`Blackbird Tech LLC d/b/a Blackbird Technologies v. Wahoo Fitness,
`Inc., Case No. 16-CV-688 (D. Del.)
`(Pet. 1–2; Paper 4, 2).
`
`The ’212 patent is additionally at issue in IPR2017-01058 (Garmin
`International, Inc. v. Blackbird Tech LLC d/b/a Blackbird Technologies),
`now terminated.
`
`
`The ’212 Patent
`The ’212 patent, entitled “Pedometer,” relates to a “pedometer having
`
`improved accuracy by calculating actual stride lengths of a user based on
`relative stride rates” (’212 patent, Abstract). More particularly, the patent
`relates to “pedometers having a waist mounted stride-counting device and
`transmitter, and a wrist-mounted receiver and display” (id. at 1:9–11). The
`device calculates a distance walked or run based on converting a base stride
`length and a base stride rate to an actual stride length and using that to
`calculate distance traveled (id. at 1:12–17).
`
`Specifically, a step counter, which is an inertia device, counts the
`number of steps a user takes (id. at 3:7–8). A data processor includes a data
`archive that stores historic data on stride length and pace and closed loop or
`fuzzy logic programming that continually or periodically replaces the base
`
`3
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`stride rate and length with recently calculated stride rates and lengths (id. at
`3:39–47).
`
`The pedometer of the ’212 patent may optionally require the user to
`operate a “‘sampling mode’” (id. at 3:56–57). In this mode, a user walks or
`runs a predetermined distance with the distance then divided by the number
`of strides counted (id. at 3:58–62). The result is the average stride length,
`which is stored in the data archive as the “Base Stride Length” (id. at 3:62–
`64). The data processor further divides the number of strides by the time of
`the run or walk to calculate a “‘Base Stride Rate’” (id. at 3:65–67).
`
`According to the ’212 patent, using a fixed average stride length does
`not account for changes in the user’s pace or improved performance (id. at
`4:19–29). To correct for this, a “‘Use Mode’” is activated that causes the
`data processor to calculate an “Actual Stride Rate” (id. at 4:30–33). The
`“Actual Stride Rate” is calculated periodically, based on data from the stride
`counter and the clock (id. at 4:30–36). An “Actual Stride Length” is
`calculated by determining a percentage change between the Actual Stride
`Rate and the Base Stride Rate (id. at 4:36–38). More specifically, the Actual
`Stride Length is calculated by:
`Actual Stride Length=Base Stride Length + Base Stride Length
`*(((Actual Stride Rate-Base Stride Rate)N)/Base Stride Rate)
`Where: N=1 when Actual Stride Rate is less than or equal to Base
`Stride Rate multiplied by 1.02, and N=3 when Actual Stride Rate
`is greater than Base Stride Rate multiplied by 1.02, although
`other N values in the range of one to three can be used
`
`4
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`(id. at 4:50–58). To further improve accuracy, an N value is derived for the
`user by using a number of samples to establish Stride Length and N (id. at
`5:1–6:9).
`Once the actual stride length is calculated for a given period of
`time, the value can be multiplied by the number of strides in that
`period to obtain a total distance for that period to be stored in a
`data archive file for that particular walk or run and added to other
`actual stride lengths or distances for other periods in which stride
`length was calculated
`(id. at 6:34–38).
`
`
`Challenged Claims
`Petitioner challenges claims 1–8 of the ’212 patent (Pet. 4–5). The
`
`’212 patent has four independent claims, 1, 2, 5, and 6, and four dependent
`claims, 3 and 4 (depending from claim 2), and 7 and 8 (depending from
`claim 6) (Ex. 1001, Claims). Claim 1 is illustrative of the challenged claims
`and is reproduced below:
` 1. An exercise monitoring device comprising:
`
`a strap for releasably securing the exercise monitoring
`device to a user;
`
`a step counter joined to the strap;
`
`a heart rate monitor joined to the strap; and
`a data processor programmed to calculate a distance
`traveled by multiplying a number of steps counted by the
`step counter by a stride length that varies according to a
`rate at which steps are counted.
`
`5
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`
`
`Prior Art Relied Upon
`Petitioner relies upon the following references in asserting the
`
`unpatentability of claims 1–8 of the ’212 patent (Pet. 4–5):
`
`References
`
`Patent Number
`
`Exhibit
`
`1003
`US 5,583,776
`Levi et. al., (hereinafter, “Levi”)
`1002
`US 4,367,752
`Jimenez et al.
`1004
`US 6,145,389
`Ebeling et. al., (hereinafter, “Ebeling”)
`Petitioner also relies on the Declaration of Mr. Thomas Blackadar
`
`(Ex. 1005) (hereinafter “Blackadar Decl.”).
`
`
`The Asserted Grounds of Unpatentability
`Petitioner challenges the patentability of the claims of the ’212 patent
`
`on the following grounds (Pet. 4–5):
`
`Claims
`
`Basis
`
`References
`
`§ 103(a)
`
`§ 103(a)
`
`Jimenez, Levi, and “knowledge of a person having
`ordinary skill in the art”
`Jimenez, Ebeling, and “knowledge of a person
`having ordinary skill in the art”
`
`1–8
`
`1–8
`
`
`
` ANALYSIS
` Claim Construction
`Petitioner proposes specific constructions for the claim terms “step
`counter,” “stride rate,” and “from a range of stride lengths calculated from a
`6
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`range of corresponding stride rates” based on Patent Owner’s proposed
`constructions in the concurrent patent litigation between the parties on this
`patent (Pet. 13–15 (citing Exh. 1011, 2)). Patent Owner contends Petitioner
`has not explained why it is not proposing the constructions Petitioner
`asserted in the District Court litigation (Prelim. Resp. 9–11). Accordingly,
`Patent Owner asserts Petitioner has not complied with 37 C.F.R.
`§ 42.104(b)(3)–(4) because Petitioner has not explained how the challenged
`claims are to be construed and how they read on the prior art (id. at 12).
`We are not persuaded, at this stage of the proceeding, that Petitioner
`has failed to comply with 37 C.F.R. § 42.104(b)(3)–(4).
`
`Additionally, in view of our analysis, we determine that at this stage,
`no claim terms require express construction (see Vivid Techs., Inc. v. Am.
`Sci. & Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999) (Only terms which
`are in controversy need to be construed, and only to the extent necessary to
`resolve the controversy)).
`
`
`Level of Ordinary Skill in the Art
`The level of skill in the art is a factual determination that provides a
`
`primary guarantee of objectivity in an obviousness analysis (Al-Site Corp. v.
`VSI Int’l Inc., 174 F.3d 1308, 1324 (Fed. Cir. 1999) (citing Graham v. John
`Deere Co., 383 U.S. 1, 17–18 (1966); Ryko Mfg. Co. v. Nu-Star, Inc., 950
`F.2d 714, 718 (Fed. Cir. 1991))).
`
`Petitioner asserts that a person of ordinary skill in the art at the time of
`the invention
`
`7
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`
`would have been [1] a person with a bachelor’s degree in
`mechanical engineering, electrical engineering, or a similar field
`with at least two years of experience in motion tracking, motion
`analysis, inertial sensing, or signal analysis, or [2] a person with
`a master’s degree
`in mechanical engineering, electrical
`engineering, or a similar field with a specialization in motion
`tracking, motion analysis, inertial sensing, or signal analysis
`(Pet. 16 (citing Blackadar Decl. ¶ 36)). Patent Owner does not appear to
`dispute the educational level or experiential aspects of Petitioner’s definition
`“[f]or the purpose of this inter partes review” (Prelim. Resp. 13). Patent
`Owner, however, emphasizes “[a]lthough degrees in electrical engineering
`or mechanical engineering are broadly applicable, the specific experience
`(i.e., motion tracking, motion analysis, inertial sensing, or signal analysis) of
`one having ordinary skill in the art is related to the field of exercise
`monitoring” (Prelim. Resp. 14 (citing Blackadar Decl. 2–4)).
`
`At this stage in the proceeding, we determine that Petitioner’s
`description of a skilled artisan as possessing (1) a bachelor’s degree in
`mechanical engineering, electrical engineering or a similar field with at least
`two years of experience in motion tracking, motion analysis, inertial sensing,
`or signal analysis, or (2) a master’s degree in mechanical engineering,
`electrical engineering, or a similar field with a specialization in motion
`tracking, motion analysis, inertial sensing, or signal analysis, is supported by
`the current record. For purposes of this Decision, therefore, we adopt
`Petitioner’s description.
`
`We note also that the applied prior art reflects the appropriate level of
`skill at the time of the claimed invention (see Okajima v. Bourdeau, 261
`F.3d 1350, 1355 (Fed. Cir. 2001)).
`
`8
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`
` Overview of the Asserted Prior Art
`1. Jimenez
`Jimenez is a patent entitled “Apparatus for Testing Physical Condition
`
`of a Subject” (Jimenez, [54]). Jimenez is related to “apparatus for testing the
`physical condition of a subject,” and in particular, “for testing the physical
`condition of a subject in response to signals indicative of heart activity of the
`subject and of the distance traversed by a limb of the subject during a timed
`testing period” (Jimenez, 1:6–12). Jimenez teaches combining monitored
`heart rate functionality and monitored distance traveled functionality into a
`single device that monitors both heart rate and distance traveled (id. at 1:8–
`2:10, 2:38–44).
`
`2. Levi
`Levi is a patent entitled “Dead Reckoning Navigational System Using
`
`Accelerometer to Measure Foot Impacts.” Levi is directed to an electronic,
`portable navigational method and system that use radionavigational data and
`dead reckoning for foot navigation (Levi, 1:8–11). According to Levi,
`[t]he term “dead reckoning” (DR) refers to a position solution
`that is obtained by measuring or deducing displacements from a
`known starting point in accordance with motion of the user. Two
`types of DR
`are known:
`inertial navigation
`and
`compass/speedometer
`(id. at 1:13–17). Levi teaches incorporation of DR functions through use of
`a digital electronic compass with a silicon pedometer and a barometric
`altimeter, with Global Positioning System (GPS) position information and
`digital maps to arrange an integrated navigation system that “provides
`advantages during GPS outages” (id. at 1:60–64, 2:5–11). Specifically,
`9
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`during GPS outages, “DR continuously tracks the user’s position without
`references to external aids or signals” (id. at 2:12–14).
`
`Levi teaches the frequency of a user’s footsteps is used in determining
`the size of footsteps taken by the user (id. at 2:57–60). A silicon
`accelerometer, mounted or attached to the user, senses harmonic motions
`and impact accelerations that result from walking or running, and thus,
`provides acceleration data indicative of footsteps (id. at 3:13–20). Levi
`further teaches three different algorithms: peak detection algorithm,
`frequency measurement algorithm, and dynamic step size algorithm to be
`used (id. at 4:29, 5:20, 6:6).
`
`The peak detection algorithm “allows determination of distance
`directly by a scale factor” (id. at 4:30–31). In the peak detection algorithm,
`accelerometer samples are taken and from each sample, it is determined if a
`peak over a threshold, exists –– thus, indicating a step (id. at 4:36–60, 5:9–
`11, Fig. 4). The frequency measurement algorithm is performed “primarily
`to obtain step size” because “[s]tep size is related to frequency” (id. at 5:21–
`23, Fig. 5).
`
`In the dynamic step size algorithm “[a]s a user walks, faster, both the
`step size and the frequency of steps increases. This can be simply modeled
`as a linear fit to observed data at different walking speeds” (id. at 6:6–9, Fig.
`5). The step size is initialized from stored default values, particular to an
`individual person, generated during a calibration process (id. at 6:22–28,
`Fig. 7, step 701).
`
`10
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`
`3. Ebeling
`Ebeling is a patent entitled “Pedometer Effective for Both Walking
`
`and Running” (Ebeling, [54]). Ebeling is related to a pedometer that
`“accurately calculates the length of the strides taken by a user when walking
`and running,” by “using measurements of the acceleration of the wearer’s
`foot during each stride” (id. at Abstract). In Ebeling, “[t]he total distance
`traveled is computed as the sum of the individual stride lengths” (id. at 4:4–
`5). The length of each individual stride is calculated independently, from a
`profile data vector which includes “values that characterize the running
`stride” and walking strides length coefficients (id. at 4:5–6, 9:51–55, 10:24–
`34).
`
`
`
`Alleged Obviousness of Claims 1–8 over Jimenez, Levi, and
`“knowledge of a person having ordinary skill in the art”
`
`1. Analysis
`Petitioner contends that claims 1–8 are obvious over the combination
`
`of Jimenez, Levi, and “knowledge of a person having ordinary skill in the
`art,” providing an element-by-element analysis identifying where each
`element is found (Pet. 19–34). Petitioner provides supporting testimony
`from its expert, Mr. Thomas Blackadar (Blackadar Decl.).
`Patent Owner first contends Levi is non-analogous art (Prelim. Resp. 44–
`50). Patent Owner contends Petitioner has not established that the
`combination of Jimenez and Levi teaches “the stride length is determined
`with reference to a plurality of calibrations that each calculate a stride length
`as a function of a known stride rate,” as recited in claim 2 and “a data
`processor . . . further programmed to derive an actual stride length from a
`11
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`range of stride lengths calculated from a range of corresponding stride
`rates,” as recited in claims 5 and 6 (Prelim. Resp. 19–24). Patent Owner
`additionally asserts Jimenez fails to teach a “transmitter” and a “receiver,” as
`recited in claim 6 (Prelim. Resp. 28–30); and Jimenez fails to teach “a heart
`rate monitor joined to the strap,” as recited in claims 1, 2, and 5. Patent
`Owner additionally asserts an ordinarily skilled artisan would not have been
`motivated to combine the teachings and suggestions of Jimenez and Levi to
`achieve the recited claims (Prelim. Resp. 34–43).
`a. Alleged Non-Analogous Art
`Petitioner asserts Levi is analogous prior art to the ’212 patent
`
`because both the ’212 patent and Levi “relate to pedometers that count the
`number of steps taken to calculate distance traveled” and thus, “Levi is in
`the same field of endeavor as and is reasonably pertinent to the claimed
`invention [of] the ’212 Patent” (Pet. 22 (citing Blackadar Decl. ¶ 73)).
`Patent Owner contends Petitioner fails to show “Levi is either from the same
`field as the ‘212 Patent, or that Levi reasonably pertains to the problems
`addressed by the ‘212 Patent” (Prelim. Resp. 45). According to Patent
`Owner, “[n]either Petitioner, [nor] its expert, ever actually claim that Levi is
`from the same field as the ’212 Patent, instead only asserting ‘a [person
`having ordinary skill in the art] would have understood that Jimenez and
`Levi are concerned with the same field of invention’” (id. at 46 (citing Pet.
`25)). Patent Owner asserts although the ’212 patent “‘relates generally to
`pedometers’ in the context of exercise monitoring devices” (Prelim. Resp.
`46 (citing ’212 patent)), Levi is directed to the field of navigational systems
`and “in particular, to electronic, portable navigation systems that use
`12
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`radionavigational data and dead reckoning for foot travel” (id. (citing Levi,
`1:7–11)). Patent Owner further asserts Petitioner has not explained why
`navigational systems are in the same field as that of the ’212 patent ––
`exercise monitoring devices (id. at 47 (citing Pet. 25; Blackadar Decl. ¶ 78)).
`
`According to Patent Owner, navigational systems and exercise
`monitoring devices are not from the same field of endeavor because
`“[n]avigational systems determine the geographic location of a user to help
`the user find his or her way” and “[e]xercise monitoring devices . . . monitor
`users’ physical activity” which may include distance determination without
`regard to the user’s geographical location (id. at 47).
`
`Patent Owner further argues “Petitioner also fails to establish that
`Levi reasonably pertains to the problems addressed by the ’212 Patent”
`(Prelim. Resp. 47). In particular, Patent Owner contends the ’212 patent
`addresses problems all related to exercise monitoring and specifically
`“recognizes a need to improve pedometer distance calculations, among other
`exercise monitoring facilities such as heart rate monitoring, in order to
`monitor the user’s physical activity levels more accurately” (id. at 48 (citing
`’212 patent, 2:8–12). In contrast, Patent Owner contends, Levi addresses
`problems related to navigation and specifically, “relates to providing
`additional navigational capability –– i.e., more information about the user’s
`position –– in the absence of radionavigational data from satellites” and not
`distance determinations (id. at 48–49 (citing Levi, 1:59–63)).
`
`As a result, Patent Owner argues, “Levi would not have commended
`itself to the inventor because Levi relates to improving navigation systems
`and specifically, to improving dead reckoning determinations in the context
`13
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`of such systems”; Petitioner has not shown an ordinarily skilled artisan “in
`the art of exercise monitoring equipment would know what dead reckoning
`is” (id. at 49–50).
`Two separate tests define the scope of analogous prior art: (1)
`whether the art is from the same field of endeavor, regardless of
`the problem addressed and, (2) if the reference is not within the
`field of the inventor’s endeavor, whether the reference still is
`reasonably pertinent to the particular problem with which the
`inventor is involved
`(In re Bigio, 381 F.3d 1320, 1325 (Fed. Cir. 2004)).
`
`Here, based on the record before us, Petitioner has persuaded us that
`Levi is from the same field of endeavor as the ’212 patent ––pedometers that
`count the number of steps taken to calculate distance traveled (see Pet. 20).
`Levi is directed to a pedometer that, in pertinent part, adjusts step size
`according to step frequency in order to estimate distance traveled by a user
`(Levi, 6:6–43). Levi states “[t]he present invention for a ground
`speed/distance sensor is an improvement over a common hiker’s pedometer”
`and “[t]he present invention analyzes the frequency of a user’s footsteps to
`aid in the detection of future footsteps, and further to aid in determining the
`size of footsteps taken by the user” (Levi, Abstract, 2:57–60, 2:65–67).
`
`Similarly, the ’212 patent is directed to a pedometer and specifically,
`a pedometer that calculates actual stride lengths of a user based on relative
`stride rates (’212 patent, Abstract). Based on the record before us, we are
`persuaded that both the ’212 patent and Levi are from the same field of
`endeavor –– pedometers or devices that include or rely on pedometers.
`Furthermore, based on the record before us, we are persuaded Levi is
`pertinent to the particular problem with which the inventor is involved ––
`14
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`accurate distance measurement. Levi’s specific use of “radionavigational
`data” and “dead reckoning” for foot navigation in its pedometers does not
`persuade us of a different result. Thus, we are persuaded by Petitioner’s
`arguments with respect to this issue.
`
`
`b. “stride length is determined with reference to a plurality of
`calibrations that each calculate a stride length as a function of a
`known stride rate” – claims 2, 5, and 6
`
`
`Petitioner asserts that this limitation is met by Levi. In particular,
`
`Petitioner asserts Levi teaches “the stride length is determined with
`reference to a plurality of calibrations that each calculate a stride length as a
`function of a known stride rate,” as recited in claim 2 (Pet. 35); “a stride
`length that varies according to the rate at which steps are counted,” as
`recited in claims 5 and 6 (Pet. 44–45). According to Petitioner, Levi teaches
`“an algorithm wherein the stride length is determined with reference to a
`plurality of calibrations that each calculate a stride length as a function of a
`known stride rate” (id. at 35 (citing Levi, 6:22–28)). Petitioner identifies
`step 701 of Levi as teaching the step size is initialized from default values
`stored in a configuration file and “[t]he configuration is particular to the
`individual person using the system and is generated during the calibration
`process” (id.). According to Petitioner, “Levi explains that the calibration
`curve is taken by measuring at least three step sizes at different speeds”
`identifying Figure 5 as support (id. at 36 (citing 6:7–15, 7:37–52)).
`
`Figure 5 of Levi illustrates the linear relationship between frequency
`of a user’s steps and step size (id. (reproducing Ex. 1005, Fig. 5)).
`
`15
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`Petitioner relies on its expert, Mr. Blackadar, to assert that an ordinarily
`skilled artisan would have understood Levi’s Figure 5 illustrates “a
`calibration process involving three different step rates that correlate to three
`different step sizes” (id. (Blackadar Decl. ¶¶ 101–105)). Petitioner contends,
`“Levi’s dynamic step size algorithm adjusts the step sizes (or stride lengths)
`based on the plurality of calibrations” and identifies the discussion of Step
`705 in which a new step size is calculated by S=S0+m * (f-f0), where “‘m is
`the slope of the calibration curve’ based on the three speeds/step size” (Pet.
`36–37 (citing Levi, 6:25–28, 40–42)). Petitioner then asserts “[a person
`having ordinary skill it the art] would have understood that Levi discloses
`and teaches that each step size (or stride length) is determined with reference
`to at least three (or a plurality) of calibrations” (id. at 37). Petitioner argues
`that at least three calibrations “each calculate each step size (or stride length)
`as a function of a known step frequency (or stride rate)” and therefore, an
`ordinarily skilled artisan would have understood Levi as disclosing “two or
`more calibrations that each generate a correspondence between stride rate
`and stride length” (id. (citing Blackadar Decl. ¶ 105)).
`
`Patent Owner responds that “Petitioner’s arguments that Levi teaches
`the ‘plurality of calibrations’” in the recited limitation, are “incorrect”
`(Prelim. Resp. 19–20). More specifically, Patent Owner first contends
`Levi’s calibration process teaches “a single calibration procedure that results
`in a single new step size parameter” (id. at 20). Patent Owner asserts Levi
`teaches a dead reckoning system calibration that is performed by a user
`marking his/her current position on a map, walking to a destination point,
`and marking that ending point on the digital map (id. (citing Levi, 9:59–
`16
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`10:30)). Thus, according to Patent Owner, although Levi’s navigational
`system calculates a new step size and a new compass offset, Levi teaches a
`single calibration (“not even relied upon by Petitioner”) to generate a single
`set of default variables for use in dead reckoning (id. at 21 (citing Levi
`10:14–19, 28)).
`
`Specifically, Patent Owner contends Levi’s discussion of a curve
`generated from calibration samples is shown in Figure 1 (not relied upon by
`Petitioner); however, Patent Owner contends Figure 1 illustrates two
`different “calibration curves” for two different people with two different
`strides (id. at 21–22 (citing Levi 2:25–26, 3:66–67, 4:4–9)).
`Patent Owner asserts, “[t]he ‘calibration’ in the claims of the ’212
`Patent refers to determining the correlation between stride rate and stride
`length for a particular user” (id.). Patent Owner asserts Levi describes
`“FIG. 5 illustrates the relationship between frequency of a user’s step and
`step size” discovered through “lab experiments” (Prelim. Resp. 22 (citing
`Levi, 2:34–35, 37–38)). Specifically, Patent Owner contends this
`relationship is shown as a curve having slope “m,” the points of which
`shows that as number of steps increase, the step size increases, not three
`calibration samples from a single test subject (id. at 23). “Nowhere does
`Levy teach or suggest that those two [] points [(1.7 steps per second/0.72
`meters and 2.1 steps per second/2.1 meters)] are the result of calibration
`samples for a particular user” (id.).
`Initially, we note none of the independent claims recite “for a particular
`user.” We further note the ’212 patent does not define explicitly the term
`
`17
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`“calibration,” and neither Petitioner nor Patent Owner provides a proposed
`construction.
`Figure 5 of Levi, illustrates the linear relationship between frequency
`of a user’s steps and step size, is shown below:
`
`
`
`(Pet. 36 (reproducing Ex. 1005, Fig. 5)).
`Levi teaches that Figure 5 “illustrates the relationship between
`frequency of a user’s step and step size” (Levi, 2:34–35). Levi describes
`[a]s a user walks faster, both the step size and the frequency of
`steps increases. This can be simply modeled as a linear fit to
`observed data at different walking speeds. Looking at the
`calibration data shown in FIG. 5, as the number of steps increases
`
`18
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`
`from 1.7 to 2.1 steps per second, for example, the step size
`increases from 0.72 meters (2.36 feet) to 0.90 meters (2.95 feet).
`(id. at 6:7–15). Figure 5 describes m as the slope of the line created in
`plotting step size vs steps per second (id. at Fig. 5). Levi further teaches
`adjusting step size according to the step frequency through use of a dynamic
`scaling algorithm (id. at 6:15–20; Pet. 36).
`
`Petitioner contends an ordinarily skilled artisan “would have
`understood [Figure 5] shows a calibration process involving three different
`step rates that correlate to three different step sizes” (Pet. 36 (citing
`Blackadar Decl. ¶¶ 101–105)). Petitioner then asserts “Levi’s dynamic step
`size algorithm adjusts the step sizes (or stride lengths) based on the plurality
`of calibrations in Step 705” (id. at 36–37).
`
`In Levi, the default step size (default stride length) is originally set to
`an initialized value S0 and a new step size is calculated by:
`S=S0+m * (f-f0)
`where S0 is the initial step size, m is the slope of a step-size versus step-
`frequency calibration curve, f is the step frequency, and f0 is the frequency at
`which there is no correction to S0 (Levi, 6:25–43, 12:17–41, Figs. 6, 9).
`Because step size “directly affects the estimated [dead reckoning] distance,”
`“[a] dynamic scaling algorithm . . . adjust[s] the scale factor as a function of
`step frequency” (id. at 6:17–20). Levi also teaches collecting various
`samples, “model[ing] a[] linear fit to observed data at different walking
`speeds” (id. at 6:8–9). “Looking at the calibration data shown in FIG. 5, as
`the number of steps increases from 1.7 to 2.1 steps per second, for example,
`the step size increases from 0.72 meters (2.36 feet) to 0.90 meters (2.95
`
`19
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`feet)” (id. at 6:10–13). The modeled linear fit is then used to determine the
`slope of the step-size versus step frequency calibration (id. at 6:8–9; Fig. 5).
`Levi further teaches using the determined slope with a known stride rate
`(S0) to calculate a new stride rate (id.).
`Figure 7 of Levi, as shown below, “illustrates the dynamic step size
`algorithm of the presently preferred embodiment” (id. at 2:39–40).
`
`
`(Levi, Fig. 7). As noted by Petitioner, Levi describes “[t]he step size may be
`initialized from default values stored in a configuration file . . . particular to
`the individual person using the system and [] generated during the
`
`20
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`calibration process” (Pet. 35–36 (citing Levi, 6:22–28). Levi further
`describes the constants S0, m, and f0 are saved (id.).
`
`As Levi explains, the step size is determined with reference to an
`initial step size, step frequency, initial step frequency, and the slope of a
`step-size versus step frequency calibration curve, (Levi, 6:42–43, Fig. 5).
`Petitioner asserts an ordinarily skilled artisan “would have understood, FIG.
`5 shows a calibration process involving three different step rates that
`correlate to three different step sizes” (Pet. 36). According to Petitioner,
`“Levi’s dynamic step size algorithm adjusts the step sizes (or stride lengths)
`based on the plurality of calibrations in Step 705, which states that ‘[t]he
`new step size is calculated by S=S0+m*(f-f0)’ . . . where ‘m is the slope of
`the calibration curve’ based on the three speeds/step sizes” (Pet. 36–37
`(citing Levi 6:40–42, 6:25–28)).
`
`Thus, based on the record before us, we are persuaded by Petitioner’s
`assertion “wherein the stride length is determined with reference to a
`plurality of calibrations that each calculate a stride length as a function of a
`known stride rate,” as recited in claim 2. In particular, Petitioner has
`persuaded us Levi teaches the number of steps (counted by the pedometer) is
`multiplied by a stride length S that varies in accordance with a stride rate (f-
`f0), wherein the stride length S is determined with reference to a plurality of
`calibrations that each calculate a stride length (S) as a function of a known
`stride rate.
`Patent Owner contends these points are not calibration samples as
`claimed, because the claims require determining the correlation “between
`stride rate and stride length for a particular user” whereas Levi describes
`21
`
`
`
`IPR2017-02023
`Patent 6,434,212
`
`“endpoints of a graph” (Prelim. Resp. 23). The claims, however, do not
`recite that the calibration samples are for a particular user. Moreover, at this
`stage, and on this record, we determine the claims do not require the
`calibration samples are “for a particular user.”
`
`Claim 5 recites “deriv[ing] the stride length from a range of stride
`lengths calculated