Trials@uspto.

gov Paper 44
571-272-7822 Entered: June 5, 2018

UNITED STATES PATENT AND TRADEMARK OFFICE

_____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD

____________

APPLE INC. and FITBIT, INC.,

Petitioner,

v.

VALENCELL, INC.,
Patent Owner.
_______________

Case IPR2017-00321
Patent 8,923,941 B2 1
_______________

Before BRIAN J. McNAMARA, JAMES B. ARPIN, and

SHEILA F. McSHANE, Administrative Patent Judges.

ARPIN, Administrative Patent Judge.

FINAL WRITTEN DECISION

35 U.S.C. § 318(a) and 37 C.F.R. § 42.73

1
Case IPR2017-01556 has been joined with this proceeding.
IPR2017-00321
Patent 8,923,941 B2

I. INTRODUCTION

Apple Inc. (“Petitioner”) filed a Petition requesting inter partes

review of claims 14–21 (“the challenged claims”) of U.S. Patent
No. 8,923,941 B2 (Ex. 1001, “the ’941 patent”) under 35 U.S.C. §§ 311–
319. Paper 2 (“Pet.”). Valencell, Inc. (“Patent Owner”) filed a Preliminary
Response. Paper 6 (“Prelim. Resp.”). We instituted the instant inter partes
review as to the challenged claims. 2 Paper 11 (“Inst. Dec.”). Fitbit, Inc.
(also “Petitioner”) filed a corresponding Petition (IPR2017-01556, Paper 2),
accompanied by a Motion for Joinder (IPR2017-01556, Paper 3),
challenging claims 14–21 of the ’941 patent, and we granted the Motion for
Joinder and instituted review of the challenged claims based on the
corresponding Petition (IPR2017-01556, Paper 9).
Subsequent to institution, Patent Owner filed a Patent Owner Response
(Paper 23 (“PO Resp.”)), and Petitioner filed a Reply (Paper 30 (“Reply”)).
In addition, Patent Owner filed a contingent Motion to Amend (Paper 24
(“MTA”)), Petitioner filed an Opposition to Patent Owner’s contingent
Motion to Amend (Paper 31 (“MTA Opp.”)), Patent Owner filed a Reply to
Petitioner’s Opposition to the contingent Motion to Amend (Paper 32
(“MTA Reply”)), and Petitioner filed Sur-reply to Patent Owner’s Reply to
Petitioner’s Opposition to the contingent Motion to Amend (Paper 33

2
We instituted inter partes review with respect to each of the claims
challenged and on all of the grounds asserted in the Petition, and our Final
Decision addresses the patentability of each of the challenged claims on all
grounds. See SAS Inst., Inc. v. Iancu, 138 S. Ct. 1348, 1354 (2018).

2
IPR2017-00321
Patent 8,923,941 B2

(“MTA Sur-Reply”)). A transcript of the oral hearing held on

February 27, 2018, has been entered into the record as Paper 41 (“Tr.”). 3
Although Patent Owner filed objections to evidence submitted with the
Petition (Paper 14) and Petitioner filed objections to evidence submitted
with Patent Owner’s Preliminary Response (Paper 13) and to evidence
submitted with the Patent Owner Response (Paper 25), neither party filed a
Motion to Exclude. Consequently, these objections are deemed waived.
37 C.F.R. § 42.64(c) (“A motion to exclude evidence must be filed to
preserve any objection.”). Petitioner also filed a list of alleged
misrepresentations of fact and inconsistent statements made by Patent
Owner in its Preliminary Response. Paper 10. We considered these listed
items in preparation of our Institution Decision (see Inst. Dec. 10–11), and
Petitioner does not raise the listed, alleged misrepresentations of fact and
inconsistent statements in its post-institution filings. Consequently, we do
not consider them further here.
This Final Written Decision is entered pursuant to 35 U.S.C. § 318(a).
For the reasons that follow, we determine that Petitioner has demonstrated
by a preponderance of the evidence that claims 14–21 of the ’941 patent are
unpatentable. We also deny Patent Owner’s contingent Motion to Amend.

A. Related Proceedings

According to the parties, the ’941 patent is involved in the following

civil actions: Valencell, Inc. v. Apple Inc., Case No. 5-16-cv-00010
(E.D.N.C. 2016); Valencell, Inc. v. Bragi Store, LLC et al., Case No. 5-16-

3
This was a consolidated hearing with the following related case: IPR2017-
00319. See Tr. 3:2–5.

3
IPR2017-00321
Patent 8,923,941 B2

cv-00895 (E.D.N.C. 2016); and Valencell, Inc. v. Fitbit, Inc., Case No. 5-16-
cv-00002 (E.D.N.C. 2016). Pet. 52; Paper 5, 1. Further, the ’941 patent is
involved in a related petition for inter partes review, Case IPR2017-00319,
filed by Petitioner on the same day as the instant Petition.

B. The ’941 Patent

The ’941 patent is entitled “Methods and Apparatus for Generating

Data Output Containing Physiological and Motion-Related Information,”
and was filed February 19, 2014, and issued December 30, 2014. Ex. 1001
at [22], [45], [54]. The ’941 patent is a continuation of U.S. Patent
Application No. 12/691,388, filed January 21, 2010, now issued as
U.S. Patent No. 8,700,111 B2 (id. at [63]), and claims priority to four
provisional patent applications: U.S. Provisional Patent Application
Nos. 61/208,567, filed February 25, 2009; 61/208,574, filed February 25,
2009; 61/212,444, filed April 13, 2009; and 61/274,191, filed August 14,
2009 (id. at [60]).
The ’941 patent relates generally to physiological monitoring
apparatus. Ex. 1001, 1:21–23. Figure 5 of the ’941 patent depicts an
exemplary embodiment and is reproduced below.

4
IPR2017-00321
Patent 8,923,941 B2

Figure 5 depicts a side section view of light-guiding earbud 30 for a headset.

In particular, earbud 30 includes light guide or cover 18 that serves the
function of a housing. Id. at 16:16–19. Light guide 18 includes a plurality
of windows 18w formed in cladding material 21 on outer surface 18a of
cover 18. Id. at 16:19–21. Light 111 emitted from light emitter 24 passes
through windows 18w and into the wearer’s body, and scattered light 110
returning from the wearer’s body passes into light guide 18 through
windows 18w and is directed to light detector 26. Id. at 16:21–24. In other
embodiments, earbud housing and light guide 18 may be separate
components, for example, as shown in Figure 3, which depicts cover 18
surrounding housing 16. Id. at 14:6–10. In addition, light guide 18 of
Figure 5 is surrounded by layer 29 of light transmissive material. Id. at
16:30–31. One or more lenses 29L are formed in layer 29 and are in optical
communication with respective windows 18w in the light guide 18, and
lenses 29L are configured to collect returning, scattered light 110 and to

5
IPR2017-00321
Patent 8,923,941 B2

direct scattered light 110 into light guiding region 19 and to light
detector 26. Id. at 16:31–41. An earbud, such as earbud 30, may integrate a
sensor module containing a plurality of sensor elements for measuring
physiological information and at least one noise source for measuring noise
information and may include a microprocessor that is in electrical
communication with the sensor module or modules. Id. at 3:46–55, 4:21–25.
In the apparatus described in the ’941 patent, photoplethysmography
(“PPG”) signals may be pre-conditioned by the microprocessor to reduce
motion artifacts and signal noise. Id. at 4:11–17, 4:25–32, 30:44–48; see id.
at 32:1–15, 3:47–55. In particular, the physiological information may be
filtered to remove signal noise by using various, known signal processing
techniques. See id. at 3:56–67. Thus, the ’941 patent discloses apparatus for
removing motion-related noise artifacts, such as subject footstep noise. See
id. at 3:65–4:5; 31:18–19.

C. Illustrative Claim

Claim 14 is the sole, challenged independent claim of the ’941 patent.

Each of claims 15–21 depends directly or indirectly from claim 1. Claim 14
is illustrative and is reproduced below with disputed limitations emphasized.
14. A wearable device, comprising:
a housing; and
a chipset enclosed within the housing, the chipset
comprising at least one PPG sensor, at least one motion sensor,
and at least one signal processor configured to process signals
from the at least one motion sensor and signals from the at least
one PPG sensor to reduce motion artifacts from the PPG signals;
wherein the housing comprises at least one window that
optically exposes the at least one PPG sensor to a body of a
subject wearing the device, and wherein the housing comprises

6
IPR2017-00321
Patent 8,923,941 B2

non-air light transmissive material in optical communication

with the at least one PPG sensor and the window.
Id. at 32:1–15 (emphasis added).

D. Applied References and Declaration

Petitioner relies on the following references and declaration in support

of its asserted grounds of unpatentability.

Exhibit References and Declaration

1003 Declaration of Dr. Majid Sarrafzadeh
1016 U.S. Patent Application Publication No. 2009/0105556 A1 to
Fricke et al., filed September 29, 2008, published
April 23, 2009 (“Fricke”)
1025 Hyonyoung Han et al., Development of a wearable health
monitoring device with motion artifact reduced algorithm,
International Conference on Control, Automation and
Systems, IEEE (2007) (“Han”)
1027 U.S. Patent Application Publication No. 2004/0186387 A1 to
Kosuda et al., published September 23, 2004 (“Kosuda”)
1029 Japanese Patent Application Publication No. 2005/270544 A
to Maekawa, published October 6, 2005
1030 Certified English-language translation of Japanese Patent
Application Publication No. 2005/270544 to Maekawa,
published October 6, 2005 (“Maekawa”) 4
1031 U.S. Patent Application Publication No. 2005/059870 A1 to
Aceti, published March 17, 2005 (“Aceti”)
1032 G. Comtois & Y. Mendelson, A Comparative Evaluation of
Adaptive Noise Cancellation Algorithms for Minimizing
Motion Artifacts in a Forehead-Mounted Wearable Pulse
Oximeter, IEEE (2007) (“Comtois”)
Pet. v–vii.
As noted above, the ’941 patent issued claiming benefit from
U.S. provisional patent applications having filing dates as early as

4
Citations to Maekawa are to this English-language translation.

7
IPR2017-00321
Patent 8,923,941 B2

February 25, 2009. Ex. 1001 at [60]. Each of the applied references has an
effective date prior to February 25, 2009. See Pet. 8–9.

E. Asserted Grounds of Unpatentability

Petitioner asserted the following grounds of unpatentability:

References Basis Challenged Claim(s)
Kosuda and Maekawa 35 U.S.C. § 103(a) 14, 15, and 21
Kosuda, Maekawa, and
35 U.S.C. § 103(a) 18–20
Han
Aceti and Fricke 35 U.S.C. § 103(a) 14–19 and 21
Aceti, Fricke, and Comtois 35 U.S.C. § 103(a) 20
Pet. 7. We instituted inter partes review of all of the challenged claims and
on all of these asserted grounds.

II. DISCUSSION
A. Claim Interpretation

In an inter partes review, claim terms in an unexpired patent are given

their broadest reasonable construction in light of the specification of the
patent in which they appear. 37 C.F.R. § 42.100(b). Under the broadest
reasonable interpretation standard, claim terms are given their ordinary and
customary meaning, as they would have been understood by one of ordinary
skill in the art in the context of the entire disclosure. See In re Translogic
Tech., Inc., 504 F.3d 1249, 1257 (Fed. Cir. 2007). Any special definition for
a claim term must be set forth with reasonable clarity, deliberateness, and
precision. See In re Paulsen, 30 F.3d 1475, 1480 (Fed. Cir. 1994).
Petitioner sought construction of five (5) claim terms in its Petition.
Pet. 11–14. Patent Owner challenged some of these constructions in its
Preliminary Response and sought construction of the additional claim term

8
IPR2017-00321
Patent 8,923,941 B2

“PPG sensor.” Prelim. Resp. 14–17. In the Institution Decision, we

proposed constructions for these six (6) terms. Inst. Dec. 7–11. In the
Patent Owner Response, Patent Owner contests the preliminary construction
of the term “PPG sensor,” but does not challenge the constructions of the
other terms addressed in the Institution Decision. PO Resp. 7–9; see
Paper 12, 3 (“The patent owner is cautioned that any arguments for
patentability not raised in the response will be deemed waived.”).
Consequently, for this Decision, we adopt our preliminary constructions of
the terms: “body,” “headset,” “housing,” “chipset,” and “window,” and
address the construction of the term “PPG sensor,” below.

1. “body” (Claims 14–21)

We determine that the broadest reasonable interpretation of the term

“body” is “the body of a subject wearing the device.” Inst. Dec. 7–8.

2. “headset” (Claim 17)

We determine that the broadest reasonable interpretation of the term

“headset” is “any type of device or earpiece that may be attached to or near
the ear of a user, including peripheral devices.” Id. at 8.

3. “housing” (claims 14–21)

We determine that the broadest reasonable interpretation of the term

“housing” is “one or more parts that covers, encloses, supports, or protects;
[a] casing.” Id. at 9.

9
IPR2017-00321
Patent 8,923,941 B2

4. “chipset” (claims 14–21)

We determine that the broadest reasonable interpretation of the term

“chipset” is “a collection of one or more chips or integrated circuits.” Id.

5. “window” (claims 14–21)

We determine that the broadest reasonable interpretation of the term

“window” is “an opening through which light can pass.” Id. at 9–10. We
note that the Specification of the ‘941 patent refers to “windows 18w” and to
“openings 18w,” from which we conclude that either an “opening” or a
“window” with a transparent covering may be the recited “window.” See
Ex. 1001, 16:18–23.

6. “PPG sensor” (claims 14–21)

In the Institution Decision, we adopted Patent Owner’s proposed

construction of the term “PPG sensor” as the broadest reasonable
interpretation of the term. Id. at 10–11. In the Patent Owner Response,
Patent Owner proposes a slight modification to this interpretation (PO
Resp. 9), to which Petitioner does not object (see Reply 1–4). Patent Owner
now contends that the broadest reasonable interpretation of the term “PPG
sensor” is “an optical sensor which obtains a plethysmogram that results
from blood flow modulations caused by the subject’s heartbeat.” PO
Resp. 9 (emphasis added); see Ex. 2005, 1. We agree that the term should
include express reference to “an optical sensor,” rather than merely to “an
optically obtained plethysmogram.” Inst. Dec. 10; see Ex. 1001, 1:64–2:6.
We determine that our previous construction, as modified by Patent Owner,
is the broadest reasonable interpretation of this term.

10
IPR2017-00321
Patent 8,923,941 B2

7. Other Claim Terms

Neither party offers specific interpretations of other terms in the

challenged claims. See Pet. 14 (“All other claim terms should be given their
plain and ordinary meaning under the broadest reasonable construction.”).
Only terms which are in controversy in this proceeding need to be construed,
and then only to the extent necessary to resolve the controversy. See, e.g.,
Nidec Motor Corp. v. Zhongshan Broad Ocean Motor Co., 868 F.3d 1013,
1017 (Fed. Cir. 2017) (“[W]e need only construe terms ‘that are in
controversy, and only to the extent necessary to resolve the controversy.’”
(quoting Vivid Techs., Inc. v. Am. Sci. & Eng’g, Inc., 200 F.3d 795, 803
(Fed. Cir. 1999))). Consequently, except as noted below, no other claim
terms require express interpretation.

B. Obviousness over Kosuda and Maekawa,

Alone or in Combination with Han
1. Overview

Petitioner argues that claims 14, 15, and 21 are unpatentable under
35 U.S.C. § 103(a) as obvious over Kosuda and Maekawa, and claims 18–20
are unpatentable as obvious over Kosuda and Maekawa in combination with
Han. See supra Section I.E. To support its argument, Petitioner provides a
mapping of limitations of claims 14, 15, and 21 to structures taught or
suggested by Kosuda and Maekawa and claims 18–20 to structures taught or
suggested by Kosuda, Maekawa, and Han. Pet. 14–31. Petitioner also cites
Dr. Sarrafzadeh’s Declaration for support. See Ex. 1003 ¶¶ 60–101. Patent
Owner limits its response to these arguments to the challenges to
independent claim 14. PO Resp. 1 (“Grounds 1 and 2 fail because the
proposed combination of [Kosuda] and [Maekawa] suffers from at least two

11
IPR2017-00321
Patent 8,923,941 B2

defects, each of which is fatal to Petitioner’s argument of unpatentability of

claims 14 and all the claims that depend from it.”), 11; see Reply 12.
A patent claim is unpatentable under 35 U.S.C. § 103(a) if the
differences between the claimed subject matter and the prior art are “such
that the subject matter as a whole would have been obvious at the time the
invention was made to a person having ordinary skill in the art to which said
subject matter pertains.” KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 406
(2007). The question of obviousness is resolved on the basis of underlying
factual determinations, including: (1) the scope and content of the prior art;
(2) any differences between the claimed subject matter and the prior art;
(3) the level of skill in the art; 5 and (4), when in evidence, objective
evidence of nonobviousness, i.e., secondary considerations. 6 Graham v.

5
Petitioner proposes an assessment of the level of ordinary skill in the art.
Pet. 11; see Ex. 1003 ¶ 54. Petitioner’s declarant, Dr. Sarrafzadeh, and
Patent Owner’s declarant, Dr. Pollonini, exceeds this assessed level.
Ex. 1004; Ex. 2010 ¶¶ 6–11. Patent Owner does not contest Petitioner’s
assessment or propose an alternative assessment. See MTA 11. To the
extent necessary, we adopt Petitioner’s assessment.
6
In the instituted proceeding, Patent Owner does not raise arguments or
present evidence based on the presence of such objective evidence of
nonobviousness. See In re Applied Materials, 692 F.3d 1289, 1299 (Fed.
Cir. 2012) (“The party seeking the patent bears the burden to overcome the
prima facie case of obviousness with evidence of secondary considerations,
such as commercial success.”); Medtronic, Inc. v. Nuvasive, Inc., Case
IPR2014-00087, 2015 WL 1546574, at *11 (PTAB 2015) (“Although it is
Patent Owner's burden to introduce evidence supporting such objective
indicia, see In re Huang, 100 F.3d 135, 139 (Fed. Cir. 1996), the ultimate
burden of persuasion never shifts to Patent Owner, see 35 U.S.C.
§ 316(e).”), aff’d, In re Nuvasive, Inc., 689 Fed. Appx. 954 (Fed. Cir. 2017);
but see Prelim. Resp. 4–6. Therefore, this factor does not play into our
analysis of Petitioner’s challenges to any claim on any ground.

12
IPR2017-00321
Patent 8,923,941 B2

John Deere Co., 383 U.S. 1, 17–18 (1966). Nevertheless, the Supreme
Court cautions us against “the temptation to read into the prior art the
teachings of the invention in issue.” Graham, 383 U.S. at 36.
We begin our analysis of these grounds of unpatentability with a
review of the applied references.

2. Kosuda (Ex. 1027)

Kosuda’s pulse measurement device includes (1) a pulse wave sensor

to detect a pulse wave from the wrist, (2) a motion sensor to detect a body
motion component, and (3) a signal processing circuit to remove body
motion components contained in the pulse wave signal and to calculate a
pulse rate of the user. Ex. 1027 ¶ 10, Fig. 2. In particular, the motion sensor
may detect body motion along three orthoganal axes. Id. ¶ 138, Fig. 5.
Thus, the pulse rate may be calculated accurately by proportionally
subtracting body motion components detected by a triaxial acceleration
sensor from the output of the pulse wave sensor. Id.
Kosuda’s Figure 3, as annotated by Petitioner, is reproduced below.

13
IPR2017-00321
Patent 8,923,941 B2

Pet. 16. Figure 3 depicts device main body 10A, including back lid 14 and
transparent glass 13C, which is held against a subject’s wrist (not shown) by
wristband 10B. Ex. 1027 ¶ 140. “The transparent glass 13C is fixed by
means of a back lid 14 as a component of the device main body 10A.” Id.
¶ 141. The reverse side of main body 10A includes pulse wave sensor 13
and acceleration (body motion) sensor 12. Id. ¶ 140.
Pulse wave sensor 13 may include light emitting diode (“LED”) 13A
(depicted in red) and photo detector (“PD”) 13B (depicted in orange). Id.
¶ 141. LED 13A emits light, and PD 13B receives detection light via
transparent glass 13C (depicted in yellow), which is fixed to the wrist-side of
main body 10A. Id. LED 13A, PD 13B, and acceleration sensor 12
(depicted in blue) may be connected to mainboard 16 (depicted in green).
Id. ¶¶ 142, 143. Further, a central processing unit (“CPU”) and other
integrated circuit (“IC”) circuits (not shown) may be mounted on

14
IPR2017-00321
Patent 8,923,941 B2

mainboard 16 and may comprise processing circuit 17 (depicted in purple).

Id. ¶ 142.
Kosuda’s data processing circuit 17 may utilize adaptive filter 30 to
process signals from the acceleration sensor and the pulse wave sensor to
reduce motion artifacts in the pulse wave signals. Id. ¶¶ 145, 152–158,
Fig. 5. Adaptive filter 30 has filter coefficient generating section 31 and
synthesizer 32. Id. ¶ 154. Filter coefficient generating section 31 applies
adaptive filter coefficient h based on data previously output. Id. ¶ 155,
Fig. 5. By applying the adaptive filter coefficient h to a simulated low-
frequency signal and to body motion component detection signals, filter
coefficient generating section 31 generates body motion removal data h(x),
h(y), and h(z). Id. Synthesizer 32 subtracts body motion removal data h(x),
h(y), and h(z) from the detected pulse wave data (i.e., pulse wave
components and body motion components) and extracts wave components
e(n). Id.

3. Maekawa (Ex. 1029, Ex. 1030)

Maekawa also teaches a wrist mounted, physiological information

measuring device that determines information, such as a pulse rate.
Ex. 1030 ¶ 20. Maekawa’s Figure 10 is reproduced below.

15
IPR2017-00321
Patent 8,923,941 B2

Figure 10 depicts a physiological information measuring device including

physiological sensor 6, including LED 4 that emits light toward the wrist,
photo diode (“PD”) 5 that receives light backscattered from the wrist, and
data processor (not shown) that determines a pulse rate based on the amount
of light received by sensor 6. Id. ¶ 21; see id. ¶ 23, Figs. 5 (depicting sensor
6 in housing 2) and 7 (depicting data processor 7). Further, Figure 10
depicts that PD 5 and cover glass 23 are separated, and a bundle of optical
fibers 40 extend in the gap between them. Id. ¶ 48. One end 40a of optical
fibers 40 is adjacent to cover glass 23, and other end 40b is adjacent to light
receiving surface 5a of PD 5. Id. Optical fibers 40 are arranged, so that
light passing along the surface B of the wrist is reflected by the outer
circumferential surface of optical fibers 40. Id.; see id. ¶ 14.
Light that only passes along the surface of the skin of the living
body does not contain very much physiological information
whereby blocking this light makes it so that most of the light that
enters the optical fiber, propagates in the optical fiber, and is lead
to the light receiving part is light that has passed deeply through

16
IPR2017-00321
Patent 8,923,941 B2

the living body under the inner skin, in other words, light that
contains a lot of physiological information.
Id. ¶¶ 15 (emphases added), 48. Thus, light reflected back through the wrist
contains more useful physiological information (i.e., light that has penetrated
more deeply into the wrist) and is guided to PD 5 through optical fibers 40.
Id. ¶¶ 14–15, 48. Because light passing along surface B is considered noise,
blocking such light improves the pulse signal’s signal-to-noise ratio. Id.
¶ 47 (“Therefore, when measuring physiological information, light
propagating in the cover glass 23 that becomes light noise can be blocked
enabling improving the [signal-to-noise (“SN”)] ratio for generating a pulse
signal.”), Fig. 9 (depicting reflective body 23A to preventing light
propagating in cover glass 23 from reaching PD 5).

4. Han (Ex. 1025)

Han teaches “a real-time, wearable and motion artifact reduced health

monitoring device.” Ex. 1025, Abstract, Fig. 1. The wearable device
includes a “photoplethysmography (PPG) sensor, 3-axis accelerometer,
microprocessor and wireless module.” Id. Han’s PPG sensor may operate
in infrared wavelengths. Id. at pp. 1582. Motion artifacts, such as those
created by finger movements, may cause the PPG sensor to acquire distorted
heart beat signals. Id., Abstract. Han teaches active noise cancellation,
whereby a motion sensor obtains body movement information, and an active
noise cancellation algorithm in an adaptive filter removes motion noises. Id.
Han’s processor conducts pre-processing on raw PPG signals. Id. at
pp. 1582.
The raw signal demands a low pass filter for reducing high
frequency noise and [a] high pass filter for rejecting a DC
component [of the PPG signal] to enhance the AC component.

17
IPR2017-00321
Patent 8,923,941 B2

. . . The filters are designed as a 0.5–3 Hz band pass filter, and

totally fourth order analog active filter and digital filter are used
in this signal processing.
Id. Han further teaches that Normalized Least Mean Square (“NLMS”)
adaptive filters may be used due to their fast processing speeds and low
order filter coefficients. Id.
Han’s Figure 3 is reproduced below.

Figure 3 depicts
a block diagram of an active noise cancellation algorithm, which
reconstructs a raw pulsation signal (s k ) from the corrupted signal
(d k ), using measurable noise signal (x k ). Here, PPG and body
motion data correspond to d k and x k respectively. This research
predominantly used 3-axis accelerometer signals (x k ) for body
motion data (n k ).
Id. Such active noise cancellation algorithm techniques may remove motion
artifacts due to walking and running. Id. at pp. 1584, Table 2.

5. Analysis
a. Claim 14

Petitioner maps the limitations of claim 14 to the teachings of Kosuda.

Pet. 20–27. In particular, Petitioner argues that Kosuda teaches a wearable
device, such as the wrist-mounted, pulse measurement device 10, depicted in
Kosuda’s Figures 2 and 3. Id. at 20 (citing Ex. 1027, Figs. 2 and 3); see id.
¶ 139; Ex. 1003 ¶ 73. The wearable device recited in claim 14, comprises
(1) “a housing,” and (2) “a chipset enclosed within the housing, the chipset

18
IPR2017-00321
Patent 8,923,941 B2

comprising at least one PPG sensor, at least one motion sensor, and at least
one signal processor.” Ex. 1001, 32:2–5. Referring to Kosuda’s Figure 3,
Petitioner argues that Kosuda teaches housing comprising main body 10A
including back lid 14, encompassing pulse wave sensor 13, triaxial
acceleration sensor 12, and microprocessor unit (”MPU”) 24 and data
processing circuit 17. 7 Pet. 20–24 (citing Ex. 1027 ¶¶ 137, 138, 140–142,
146); see Ex. 1003 ¶¶ 75–82.
Petitioner argues that Kosuda’s pulse wave sensor 13 teaches “at least
one PPG sensor.” Pet. 23; see supra Section II.A.6 (modifying construction
of “PPG sensor” to describe an “optical sensor”). In particular, Petitioner
argues that pulse wave sensor 13 includes the components of a PPG sensor,
namely, LED 13A and PD 13B. Pet. 23.
Further, Petitioner argues that Kosuda’s acceleration sensor 12
teaches “at least one motion sensor.” Pet. 24. In particular, Petitioner
argues that “[a]cceleration sensor 12 detects body motion by directly sensing
motion of the acceleration sensor itself” and that Kosuda teaches other types
of motion sensors, such as angle sensors and blood vessel simulation
sensors.” Id. (citing Ex. 1027 ¶¶ 137, 309, 452, Figs. 66A (angle sensor
122) and 113 (blood vessel simulation sensor 232); see Ex. 1003 ¶¶ 81, 82.
In addition, Petitioner argues that Kosuda’s MPU 24 and/or data
processing circuit 17, teach(es) “at least one signal processor configured to
process signals from the at least one motion sensor and signals from the at

7
As Patent Owner notes, the Institution decision incorrectly referred to
“housing 2” and “processor 7.” PO Resp. 16 (quoting Inst. Dec. 17). Those
typographical errors are corrected here, and Petitioner’s mapping of the
limitations of claim 14 onto Kosuda is clarified.

19
IPR2017-00321
Patent 8,923,941 B2

least one PPG sensor to reduce motion artifacts from the PPG signals”
(Ex. 1001, 32:5–9). Pet 24 (citing Ex. 1027 ¶¶ 135–137, 142, 145, 154–158,
Figs. 3–5); see Ex. 1003 ¶¶ 83–85. In particular, Kosuda teaches that
the body motion components originating in the veins are detected
by a triaxial acceleration sensor, and the pulse rate is accurately
detected based on a signal that is free of the effect of venous
blood by subtracting the detected output from the output of the
pulse wave sensor in a specific proportion.
Ex. 1027 ¶ 138. Thus, Kosuda teaches that signals from the PPG sensor and
the motion sensor are processed to reduce motion artifacts from the PPG
signals. Pet. 24; see Ex. 1003 ¶¶ 65, 84–85.
Referring to Figure 3 (reproduced above), Kosuda teaches that each of
acceleration sensor 12, pulse wave sensor 13, and data processing circuit 17
may be mounted on or physically connected to mainboard 19. Ex. 1027
¶¶ 144–145, Fig. 3. Thus, Petitioner argues that these components of
Kosuda teach “a chipset enclosed within the housing.” Pet. 21–23; see
Ex. 1003 ¶¶ 75– 78; supra Section II.A.4. As noted above, “[t]he
transparent glass 13C is fixed by means of a back lid 14 as a component of
the device main body 10A.” Ex. 1027 ¶ 141. Thus, main body 10A, back
lid 14, and glass 13C together may form the “housing.” Pet. 25 (“Kosuda
discloses that the housing (i.e., main body/watchcase 10A) comprises a
window (i.e., transparent glass 13C in the opening in back lid 14) that
optically exposes the PPG sensor (i.e., pulse wave sensor 13) to a body (i.e.,
user’s arm 11) of a subject wearing the device.” (citations omitted)); see
Reply 1–2.
Finally, Petitioner argues that Kosuda’s transparent glass 13C teaches
the at least one window in the recited housing. Pet. 25 (citing Ex. 1027
¶¶ 139–141, Fig. 3); see Ex. 1003 ¶ 86; supra Section II.A.5. Nevertheless,

20
IPR2017-00321
Patent 8,923,941 B2

Petitioner acknowledges that “Kosuda does not explicitly state that a non-air
light transmissive material exists between sensor 13 and transparent glass
13C.” Pet. 26. Petitioner argues, however, that “Maekawa teaches placing a
non-air light transmissive material (i.e., optical fibers 40) in optical
communication with the PPG sensor (i.e., pulse sensor 6) and the window
(i.e., cover glass 23).” Id. Therefore, Petitioner argues that a person of
ordinary skill in the art would have had reason to combine the teachings of
Kosuda and Maekawa “to place Maekawa’s non-air light transmissive
material (such as an optical fiber) in optical communication with Kosuda’s
PPG sensor (i.e., pulse sensor 13) and window (i.e., transparent glass 13C)
to improve the signal-to-noise ratio of the received pulse signal.” Id. at 26–
27 (emphasis added); see Ex. 1030 ¶¶ 47, 48, Figs. 9 and 10; Ex. 1003 ¶ 89.
As noted above, Maekawa teaches that it is desirable to prevent light
propagating in the cover glass or only passing along the surface of the skin
from entering the photo diode. See supra Section II.B.3. Dr. Sarrafzadeh
testifies that a person of ordinary skill in the art would understand that the
signal-to-noise ratio would be improved by preventing such light from being
received by the Maekawa’s photo diode or Kosuda’s pulse wave sensor. See
Ex. 1003 ¶¶ 38, 69–71, 88–90; Ex. 1070, 150:17–153:9. Because Kosuda
and Maekawa are directed to physiological monitoring devices and to the
extraction of physiological and activity related information from subjects, it
is alleged that a person of ordinary skill in the art would have had reason to
use a technique known to improve a similar device to improve Kosuda’s
monitoring device in a similar way. Pet. 27; Reply 4–9; see KSR, 550 U.S.
at 417 (“For the same reason, if a technique has been used to improve one
device, and a person of ordinary skill in the art would recognize that it would

21
IPR2017-00321
Patent 8,923,941 B2

improve similar devices in the same way, using the technique is obvious
unless its actual application is beyond his or her skill.”); In re Ethicon, 844
F.3d 1344, 1351 (Fed. Cir. 2017) (“The normal desire of artisans to improve
upon what is already generally known can provide the motivation to
optimize variables such as the percentage of a known polymer for use in a
known device.”).

b. Patent Owner’s Contentions

Patent Owner contends that Petitioner fails to demonstrate that the

combined teachings of Kosuda and Maekawa render the device of
challenged claim 14 obvious for two significant reasons. See PO Resp. 15–
24. First, Patent Owner contends that neither Kosuda nor Maekawa teaches
or suggests a chipset within a housing that encloses a PPG sensor. Id. at 15–
18. Second, Patent Owner contends a person of ordinary skill in the art
would not have had reason to combine the teachings of Kosuda and
Maekawa to achieve “the housing compris[ing] non-air light transmissive
material in optical communication with the at least one PPG sensor and the
window.” Id. at 18–24. For the reasons set forth below, we are not
persuaded that Petitioner fails to demonstrate unpatentability in view of
either of Patent Owner’s contentions.
First, Patent Owner contends that neither Kosuda nor Maekawa
teaches or suggests a chipset within a housing that encloses a PPG sensor.
Id. at 15–18. In particular, Patent Owner contends that Petitioner does not
argue that Maekawa teaches or suggests this limitation; thus, Petitioner relies
solely on Kosuda to teach or suggest this limitation. Id. at 15 (citing
Pet. 21–24). More specifically, Patent Owner contends that, although
Petitioner asserts that “Kosuda discloses a PPG sensor . . . enclosed within

22
IPR2017-00321
Patent 8,923,941 B2

the housing (i.e., main body watchcase 10A)” (Pet. 23), Petitioner does not
identify a “housing,” in which a chipset including a PPG sensor is enclosed.
PO Resp. 15. Referring to Petitioner’s annotated version of Kosuda’s
Figure 3, reproduced above, Patent Owner contends that Kosuda’s main
body/watchcase 10 A and back lid 14 do not enclose light emitting diode
13A, photodetector 13B, and transparent glass 13C, “which Petitioner argues
collectively disclose the claimed PPG sensor.” Id. at 16 (citing Ex. 2010
¶ 82). Specifically, Patent Owner’s declarant, Dr. Pollonini, states that
“because Petitioner argues that Transparent glass 13C is part of the pulse
wave sensor, it cannot also be part of the housing which must enclose the
pulse wave sensor. Thus, the entire pulse wave sensor, based on Petitioner’s
argument, cannot be part of a chipset enclosed within a housing.” Ex. 2010
¶ 83. We disagree.
Initially, although neither party offers an interpretation of the word
“enclosed,” we give the word “enclosed” its plain and ordinary meaning
consistent with its usage in the Specification of the ’941 patent. 8 Pet. 14; PO
Resp. 7–8; see Translogic, 504 F.3d at 1257. The verb “to enclose” is
defined as “1. to close in on all sides shut in. 2. to surround, as with a fence:
to enclose land. . . . 4. to contain or hold.” RANDOM HOUSE WEBSTER’S
COLLEGE DICTIONARY 433 (2nd Random House ed. 1999) (Ex. 3001). This
definition of “to enclose” is consistent with the use of that word in the
Specification of the ’941 patent.

8
Although Dr. Sarrafzadeh expressed concern regarding a “scientific
technical definition” of the word “enclosing,” we do not discern that the
record suggests that anything beyond the plain and ordinary meaning is
indicated here. Ex. 2011, 138:25–139:24; see PO Resp. 16.

23
IPR2017-00321
Patent 8,923,941 B2

Figure 6 of the ’941 patent is reproduced below.

Ex. 1001, Fig. 6. Figure 6 depicts a side section view of a light-guiding

earbud for a headset, according to some embodiments of the claimed
invention. Id. at 7:16–18. Referring to Figure 6, the Specification explains
that “[t]he earbud housing 16 encloses the speaker 22, an optical emitter
24 and an optical detector 26 as illustrated. An additional light detector 26 is
located on the base 50 but is not surrounded by the earbud housing 16.” Id.
at 16:61–65 (emphases added). Thus, in the context of claim 14, we
understand “enclosed within,” as recited in claim 14, to encompass
“surrounded by.” See id. at 2:35–39 (“In some embodiments, a housing is
secured to and overlies the base so as to enclose and protect the speaker,
optical emitter and optical detector, as well as other electronic components
secured to the base e.g., sensors, processor, transmitter etc.).”); 12:7–12
(Referring to Figure 1, “[t]he headset housing 14 is secured to the base 12
and is configured to enclose and protect the various electronic components

24
IPR2017-00321
Patent 8,923,941 B2

mounted to the 10 base (e.g., main circuit board 20 and components secured
thereto, etc.) . . .” (emphasis added)).
We are not persuaded that Petitioner argues that light emitting
diode 13A, photodetector 13B, and transparent glass 13C, together, teach
the recited, PPG sensor. Instead, Petitioner asserts that “Kosuda discloses a
PPG sensor (i.e., pulse wave sensor 13) enclosed within the housing (i.e.,
main body/watchcase 10A).” Pet. 23 (emphasis added); cf. PO Resp. 15
(quoted above). Although Kosuda discloses that
the pulse wave sensor 13 has an LED 13A for emitting light lo
detect pulse waves, a PD (Photo Detector) 13B for receiving the
detection light reflected by the body, and transparent glass 13C
for protecting the LED 13A and the PD 13B, transmitting the
light incident on the LED 13A and reflected light obtained via
the body, and directing the light onto the PD 13B;
(Ex. 1027 ¶ 141), Petitioner distinguishes between sensor 13 and transparent
glass 13C in its arguments. In particular, Petitioner argues that “Kosuda
discloses that the housing (i.e., main body/watchcase 10A) comprises a
window (i.e., transparent glass 13C in the opening in back lid 14) that
optically exposes the PPG sensor (i.e., pulse wave sensor 13) to a body (i.e.,
user’s arm 11) of a subject wearing the device.” Pet. 25; see id. at 26
(“Kosuda does not explicitly state that a non-air light transmissive material
exists between sensor 13 and transparent glass 13C.” (emphases added)). In
other portions of the Petition, Petitioner also refers to the recited, PPG
sensor as “pulse wave sensor 13.” See, e.g., Pet. 24 (referring to “signals
from the PPG sensor (i.e., pulse wave sensor 13)”), 25 (referring to “the PPG
sensor (i.e., pulse wave sensor 13)”). Thus, considering the entirety of
Petitioner’s arguments and the cited text and figures of Kosuda, we find that
Petitioner persuasively argues that the recited, PPG sensor is taught by

25
IPR2017-00321
Patent 8,923,941 B2

Kosuda’s LED 13A and PD 13B and that those components are enclosed
within or surrounded by a “housing” comprising Kosuda’s main
body/watchcase 10A and back lid 14, with or without the inclusion of
transparent glass 13C. See Ex. 1027 ¶ 141 (“The transparent glass 13C is
fixed by means of a back lid 14 as a component of the device main body
10A.” (emphasis added)); see also Ex. 2011, 137:12–15 (“I believe a person
of ordinary skill in the art would have understood Kosuda’s main body
watch 10A including, of course, the back lid 14 to satisfy the claimed
housing.”), 138:7–12 (“I’m not claiming that transparent panel 13C is part of
the housing, that’s correct.”). Therefore, we are persuaded that Petitioner
demonstrates that the combined teachings of Kosuda and Maekawa teach or
suggest “a chipset enclosed within the housing, the chipset comprising at
least one PPG sensor.”
Second, Patent Owner contends that Petitioner fails to provide
sufficient reason to combine the teachings of Kosuda and Maekawa. PO
Resp. 18–24. In particular, Patent Owner contends that (1) “Maekawa’s
disclosure (and Dr. Sarrafzadeh’s adoption of it) where light passing only
superficially along the surface of the wrist does not contain ‘useful
physiological information’ is incorrect” (id. at 20–21); (2) “[l]ight entering
[Maekawa’s glass 23] from certain angles would be transmitted entirely
within the glass, and some of that light may exit into the optical cables” (id.
at 21–23); (3) “using Maekawa’s ‘fiber optic cables’ in Kosuda’s device
would not reduce motion artifacts in Kosuda’s pulse wave sensor” (id. at 23–
24); and (4) “a [person of ordinary skill in the art (“POSA”)] would not have
looked to the solution offered in Maekawa because there is a much simpler

26
IPR2017-00321
Patent 8,923,941 B2

way of solving Kosuda’s purported signal-to-noise problem” (id. at 24). We

disagree.
First, although Petitioner argues that “[l]ight that only passes along
the surface of the wrist does not contain useful physiological information”
(Pet. 19 (citing Ex. 1030 ¶¶ 14, 15, 48)), Petitioner explains that:
By blocking this surface light, the light that contains more useful
physiological information (i.e., light that has penetrated more
deeply into the wrist) is led to the PD 5 through optical fibers 40.
Ex. 1030, ¶¶ 0014-0015, 0048; Ex. 1003, ¶¶ 69–71. Because
light propagating along the surface of the skin would be
considered noise, blocking such light noise improves the signal-
to-noise ratio of the pulse signal. Ex. 1030, ¶¶ 0047-0048;
Ex. 1003, ¶ 71.
Pet. 19–20 (emphasis added). Notably, Maekawa discloses that:
Light that only passes along the surface of the skin of the living
body does not contain very much physiological information
whereby blocking this light makes it so that most of the light that
enters the optical fiber, propagates in the optical fiber, and is lead
to the light receiving part is light that has passed deeply through
the living body under the inner skin, in other words, light that
contains a lot of physiological information.
Ex. 1030 ¶ 15 (emphases added). Thus, light that passes along the surface of
the skin may contain some, but certainly contains less, physiological
information than light that passes more deeply through the skin. Reply 5–6.
Second, although Patent Owner asserts that “[l]ight entering
[Maekawa’s glass 23] from certain angles would be transmitted entirely
within the glass, and some of that light may exit into the optical cables,”
Maekawa clearly teaches that this is undesirable. PO Resp. 21. Referring to
its Figure 9, Maekawa teaches that reflective surface 23a within cover
glass 23 deflects light propagating within cover glass 23 and prevents that
propagating light from reaching PD 5. Ex. 1030 ¶ 47; see Reply 6.

27
IPR2017-00321
Patent 8,923,941 B2

Referring to its Figure 10, Maekawa’s bundle of optical fibers 40 limits

propagating light from reaching PD5, even if it does not prevent all
propagating light from reaching PD5. Ex. 1030 ¶ 48; see Reply 6–7.
Petitioner concludes that “even assuming that light traveling within
Kosuda’s transparent glass was a problem, it was a problem recognized in
the prior art to which there were known solutions as taught by Maekawa and
admitted to by [Patent Owner].” Reply 7 (citing PO Resp. 21–23). Thus,
we agree with Petitioner that a person of ordinary skill in the art would have
had reason to combine the teachings of Kosuda and Maekawa even if the
combination did not eliminate all undesirable light from reaching the sensor.
Id.
Third, despite Patent Owner’s contention that “using Maekawa’s
‘fiber optic cables’ in Kosuda’s device would not reduce motion artifacts in
Kosuda’s pulse wave sensor” (PO Resp. 23–24 (emphasis added)), Petitioner
does not rely on Maekawa’s fiber optic cables for this purpose (Reply 7).
See Ex. 2010 ¶ 91. Instead, Petitioner argues that “it would have been
obvious to a POSA to place Maekawa’s non-air light transmissive material
(such as an optical fiber) in optical communication with Kosuda’s PPG
sensor (i.e., pulse sensor 13) and window (i.e., transparent glass 13C) to
improve the signal-to-noise ratio of the received pulse signal.” Pet. 26–27
(emphasis added, citing Ex. 1003 ¶ 89). Petitioner notes that Kosuda’s
motion sensor, not Maekawa’s bundle of fiber optic cables, is intended to
reduce motion artifacts. Reply 7–8.
Finally, Patent Owner’s contention that there exists “a much simpler
way of solving Kosuda’s purported signal-to-noise problem” (PO Resp. 24)
is not determinative of whether a person of ordinary skill in the art would

28
IPR2017-00321
Patent 8,923,941 B2

have had reason to combine the teachings of Kosuda and Maekawa in the
manner proffered by Petitioner. Initially, we note that although Patent
Owner has suggested an alternative solution to the signal-to-noise problem,
neither Patent Owner nor its declarant provides support for their contention
that the alternative solution is superior to that taught by the combination of
the teachings of Kosuda and Maekawa. See PO Resp. 24; Ex. 2010 ¶ 90.
But even if it had, as Petitioner notes, “the law does not require that a
particular combination be the preferred, or the most desirable, combination
described in the prior art in order to provide motivation for the claimed
invention.” Reply 8; see In re Fulton, 391 F.3d 1195, 1200 (Fed. Cir. 2004)
(“The question is whether there is something in the prior art as a whole to
suggest the desirability, and thus the obviousness, of making the
combination, not whether there is something in the prior art as a whole to
suggest that the combination is the most desirable combination available.”
(emphasis added)). Thus, the existence of potentially superior alternatives
does not negate the combination of teachings proposed by Petitioner.
Accordingly, we are persuaded that Petitioner demonstrates
sufficiently that a person of ordinary skill in the art would have had reason
for combining the teachings of Kosuda and Maekawa to achieve the device
recited in claim 14 of the ’941 patent.

c. Claims 15 and 21

Petitioner argues that claims 15 and 21 also are rendered obvious over
the combined teachings of Kosuda and Maekawa and provides a mapping
reading of the additional limitations of these dependent claims on the
combined teachings of Kosuda and Maekawa, as follows:

29
IPR2017-00321
Patent 8,923,941 B2

Claim Limitation Applied Teachings

15 “wherein the non-air light Maekawa teaches or suggests the
transmissive material addition of a non-air light
comprises at least one light transmissive material (such as a
guide in optical bundle of optical fibers 40) in
communication with the at optical communication with
least one PPG sensor and photo diode 5 and cover glass
the window” 23. Ex. 1030 ¶ 48; see Pet. 26.
Kosuda teaches that light is
directed to the back of the
wearer’s wrist and returned to a
detector via the transparent
glass. Pet. 27 (citing Ex. 1027 9
¶¶ 310, 315, 445, 449); see
Ex. 1003 ¶¶ 91, 92.
21 “wherein the at least one Kosuda teaches that the motion
motion sensor comprises at sensor may comprise at least one
least one of the following: of an optical sensor (Ex. 1027
an optical sensor, an ¶ 356), an inertial sensor (id.
inertial sensor, an ¶ 351), a capacitive sensor (id.
electrically conductive ¶ 316), and an accelerometer (id.
sensor, a capacitive sensor, ¶ 140). Pet. 28; see Ex. 1003
an inductive sensor, an ¶¶ 93, 94.
accelerometer, and
a blocked channel sensor”
(emphases added)
Pet. 27–28; see Ex. 1003 ¶¶ 91–94.
Patent Owner does not contest this mapping, but, instead, contends
that Petitioner fails to demonstrate the combined teachings of Kosuda and
Maekawa teach or suggest all of the limitations of the base claim, claim 14,
and/or fails to provide a sufficient reason to combine the teachings of the
Kosuda and Maekawa to achieve the device recited in claim 14. PO Resp. 1;
Reply 12; see supra Section II.B.5. For the reasons set forth above, we are

9
Petitioner incorrectly cites to “Ex. 1030,” instead of “Ex. 1027.” Pet. 27.

30
IPR2017-00321
Patent 8,923,941 B2

persuaded that Petitioner has demonstrated that the combined teachings of

Kosuda and Maekawa teach or suggest all of the limitations of claim 14, as
well as those of claims 15 and 21, and provides sufficient reason to have
combined the teachings of Kosuda and Maekawa. Pet. 20–28. Thus, we are
persuaded that Petitioner has demonstrated by a preponderance of the
evidence claims 15 and 21 are unpatentable under 35 U.S.C. § 103(a) as
rendered obvious by the combined teachings of Kosuda and Maekawa.

d. Claims 18–20

Petitioner argues that claims 18–20 are rendered obvious over the
combined teachings of Kosuda, Maekawa, and Han. See supra Section I.E.
Claim 18 depends from claim 14 and recites that “at least one processor is
configured to reduce motion artifacts by removing frequency bands from the
signals that are outside of a range of interest using at least one band-pass
filter to produce preconditioned signals.” Ex. 1001, 32:24–28. Claim 19
depends from claim 18 and recites that any one of a plurality of types of
filtering may be applied to filter the preconditioned signals in the device of
claim 18. Id. at 32:29–33. Finally, claim 20 depends from claim 14 and
recites that “the subject motion artifacts comprises subject footstep-related
motion artifacts.” Id. at 32:34–35.
Petitioner relies on the combined teachings of Kosuda and Maekawa
to render the base claim, claim 14, of claims 18–20 unpatentable, and
Petitioner specifically identifies the limitations of claims 18–20 that are
missing from the combined teachings of Kosuda and Maekawa. 10 Pet. 28–

10
In its Preliminary Response, Patent Owner contended that Petitioner fails
to demonstrate which limitations of each claims are missing from one or

31
IPR2017-00321
Patent 8,923,941 B2

29 (“Though Kosuda does not expressly mention removing footstep-related

motion or disclose that the frequency bands are removed using a band-pass
filter to produce pre-conditioned signals, such techniques were well known
in the art.”); see Ex. 1003 ¶ 96. Petitioner provides a mapping of the
additional limitations of these dependent claims onto Han, as follows:
Claim Limitation Han’s teaching
18 “wherein the at least one Han teaches that “[t]he raw
processor is configured to signal demands a low
reduce motion artifacts by pass filter for reducing high
removing frequency bands frequency noise and high
from the signals that are pass filter for rejecting a DC
outside of a range of interest component to enhance the
using at least one band-pass AC component. As filters,
filter to produce preconditioned second order active analog
signals” high and low pass filters
(Sallen-Key Filter) are used.
Filtering signals are amplified
to enhance and acquire
discriminable signals by a
thousand times.
The last part is digital
signal processing. High order
filtering has good performance
to extract wanted signal, but
more number of components
are required to increase
filtering order. Therefore,
digital filtering is employed to
satisfy both circuit size and
filtering performance.
The filters are designed as a
0.5 – 3 Hz band pass filter,

more applied reference(s) and supplied by another reference. Prelim.

Resp. 36–39. Patent Owner does not maintain this contention in its Patent
Owner Response. Paper 12, 3.

32
IPR2017-00321
Patent 8,923,941 B2

and totally fourth order analog

active filter and digital filter
are used in this signal
processing.” Pet. 30 (citing
Ex. 1025, pp. 1582 (emphasis
added)); see Ex. 1003 ¶ 100.
19 “wherein the at least one Han teaches “an active noise
processor is configured to filter cancellation algorithm to filter
the preconditioned signals via the preconditioned signals via
at least one of the following: a Normalized Least Mean
FIR (Finite Impulse Response) Square (NLMS) adaptive
filtering, IIR (Infinite Impulse filtering.” Pet. 30–31
Response) filtering, adaptive (emphasis added, citing
filtering. phase filtering, and Ex. 1025, pp. 1582, Fig. 3);
frequency filtering” see Ex. 1003 ¶ 100.
20 “wherein the subject motion In Figure 2, Han teaches that
artifacts comprises subject motion artifacts may include
footstep-related motion running or walking. Ex. 1025,
artifacts” pp. 1584, Table 2, Fig. 3; see
Pet. 30; Ex. 1003 ¶ 100.
Pet. 28–31 (citing Ex. 1025, pp. 1581–82). Moreover, Petitioner identifies a
reason why a person of ordinary skill in the art would have combined the
teachings of Kosuda, Maekawa, and Han to achieve the device recited in
claims 18–20. Pet. 31; see Ex. 1003 ¶ 101.
Patent Owner does not contest this mapping, but, instead, contends
that Petitioner fails to demonstrate the combined teachings of Kosuda and
Maekawa teach or suggest all of the limitations of the base claim, claim 14,
or a reason to combine the teachings of the Kosuda and Maekawa to achieve
the device recited in claim 14. PO Resp. 1; Reply 12; see supra Section
II.B.5. For the reasons set forth above, we are persuaded that Petitioner has
demonstrated that the combined teachings of Kosuda and Maekawa teach or
suggest all of the limitations of claim 14, and provides sufficient reason to

33
IPR2017-00321
Patent 8,923,941 B2

have combined the teachings of Kosuda and Maekawa. Pet. 20–27. Further,
we are persuaded that Petitioner demonstrates where Han teaches the
additional limitations of the dependent claims (Pet. 29–31 (citing Ex. 1003
¶¶ 97–100)), and, despite the similarities between the Petitioner’s arguments
and Dr. Sarrafzadeh’s testimony, we credit Dr. Sarrafzadeh’s testimony
regarding the combined teachings (see Ex. 1003 ¶¶ 1–14, 23). Thus, we are
persuaded that Petitioner has demonstrated by a preponderance of the
evidence claims 18–20 are unpatentable under 35 U.S.C. § 103(a) as
rendered obvious by the combined teachings of Kosuda, Maekawa, and Han.

e. Summary

For the reasons set forth above, we are persuaded that Petitioner
demonstrates by a preponderance of the evidence that claims 14, 15, and 18–
21 of the ’941 patent are unpatentable under 35 U.S.C. § 103(a) as obvious
over Kosuda and Maekawa, alone or in combination with Han.

C. Obviousness over Aceti and Fricke,

Alone or in Combination with Comtois
1. Overview

Petitioner argues that claims 14–21 are unpatentable under 35 U.S.C.

§ 103(a) as obvious over Aceti and Fricke, alone or in combination with
Comtois. See supra Section I.E. To support its argument, Petitioner
provides a mapping of limitations of claims 14–21 to structures taught or
suggested by Aceti and Fricke or by Aceti, Fricke, and Comtois. Pet. 31–50.
Petitioner also cites Dr. Sarrafzadeh’s Declaration for support. See Ex. 1003
¶¶ 102–143. Patent Owner limits its response to these arguments to the
challenges to independent claim 14. PO Resp. 2, 25 (“As with the
combination of Kosuda and Maekawa for Grounds 1 and 2, Petitioner has

34
IPR2017-00321
Patent 8,923,941 B2

failed make out a prima facie case that claim 14 and its dependent claims are
obvious over Aceti and Fricke. Thus, Grounds 3 and 4 also fail.”); see
Reply 12.
We begin our analysis of these grounds of unpatentability with a
review of the applied references.

2. Aceti (Ex. 1031)

Aceti teaches a pulse oximetry sensor (i.e., a PPG sensor) including an

optical emitter and optical detector for sensing physiological information.
Ex. 1031 ¶ 27. Aceti’s Figure 1 is reproduced below.

Figure 1 depicts a partially exploded view of an exemplary monitoring

device in accordance with Aceti’s disclosure. Id. ¶ 7. Monitoring device
100 includes processor portion 102 and conductor portion 104. Id. ¶ 16. In

35
IPR2017-00321
Patent 8,923,941 B2

an exemplary embodiment, conductor portion 104 may be removably

coupled to processor portion 102 and may be disposable. Id. Processor
portion 102 may include housing 106 with cover 108, which may contain
electrical and/or electronic components 110. Id. ¶ 17. Additionally,
electrical and/or electronic components 110 may be found within conductor
portion 104. Id. Conductor portion 104 includes first end 112 configured
for insertion at least partially within the auditory canal of the wearer and
second end 114 coupled to processor portion 102. Id.
Referring to Aceti’s Figure 3, Aceti teaches components 110 may
include pulse oximetry sensor 336, accelerometer 340, and processor 314,
and signal processing circuitry 342, to determine pulse rate. Id. ¶ 27. Aceti
notes that pulse oximetry was a well-known application of
photoplethysmography, and, in Figure 3, Aceti teaches that such a PPG
sensor may emit light in the red (660 nm) and infrared (805 nm)
wavelengths. Id., Fig. 3 (LED 330 (660 nm) and LED 332 (805 nm)); see
Ex. 1034 ¶ 38 (660 nm and 805 nm wavelengths were known to be used in
PPG sensors); Ex. 1003 ¶¶ 32, 33.

36
IPR2017-00321
Patent 8,923,941 B2

Aceti’s Figure 4 is reproduced below.

Figure 4 depicts a section of first end 112 of conductor portion 104 for
insertion into the wearer’s auditory canal. Ex. 1031 ¶¶ 22, 33. Fiber optic
cables 402 terminate in optically transparent elastomer 408 of first end 112
to allow the communication of light between fiber optic cable 402 and the
auditory canal wall. Id. ¶ 33.

3. Fricke (Ex. 1016)

Fricke describes a method for obtaining physiological parameters and

reducing artifacts by removing frequency bands from the signals that are
outside of a range of interest using a band-pass filter. Ex. 1016 ¶ 53.
Fricke’s Figure 3 is reproduced below.

37
IPR2017-00321
Patent 8,923,941 B2

Fricke’s Figure 3 depicts band-pass filtering by band-pass filters 302 and

304. Id. Fricke explains that, once band-pass filtering is performed on
IR plethysmograph parameters signals 300, additional filtering may be
performed on the pre-conditioned (i.e., filtered) signals by, for example,
instantaneous frequency computation component 306 or 318. Id. ¶¶ 53–55,
67–78, 92. Additional filtering may include Hilbert transform (IIR, FIR)
filtering, Least Mean Square (“LMS”) adaptive filtering (IIR), and Kalman
filtering (IIR). Id. ¶¶ 54, 69, 92.

4. Comtois (Ex. 1032)

Comtois recognized that a significant factor limiting the accuracy of

pulse oximetry is poor signal-to-noise ratio because PPG signals, from
which SpO 2 and heart rate (“HR”) measurements are derived, are
compromised by movement artifacts. Ex. 1032, Abstract. Comtois teaches
that “processing motion-corrupted PPG signals by least mean squares (LMS)
and recursive least squares (RLS) algorithms can be effective to reduce

38
IPR2017-00321
Patent 8,923,941 B2

SpO 2 and HR errors during jogging, but the degree of improvement depends
on filter order.” Id.
Comtois’s Figure 2 depicts the analysis of data acquired during
jogging experiments and shows that adaptive noise cancellation (“ANC”)
implemented using LMS and RLS algorithms may improve the accuracy of a
pulse oximeter. Id. at 1530. Thus, Comtois explains that the performance
effectiveness of wearable physiological monitoring devices may be
improved by applying algorithms to reduce limitations imposed by footstep-
related motion (e.g., jogging) artifacts. Id. at 1531.

5. Analysis
a. Claim 14

Petitioner provides a mapping of the limitations of claim 14 on the

teachings of Aceti. Pet. 38–42. In particular, Petitioner argues that Aceti
teaches a wearable device, such as device 100, depicted in Aceti’s Figure 2.
Id. at 38 (citing Ex. 1031 ¶¶ 16–19, Fig. 2; see Ex. 1003 ¶ 111. The
wearable device recited in claim 14 comprises (1) “a housing,” and (2) “a
chipset enclosed within the housing, the chipset comprising at least one PPG
sensor, at least one motion sensor, and at least one signal processor.”
Ex. 1001, 32:2–5. Referring to Aceti’s Figure 2, Petitioner further argues
that Aceti teaches housing 106 of portion 102 or first end 112 of conductor
portion 104 encompassing a set of electric and/or electronic components
110, including pulse oximetry sensor 302, accelerometer 306, pulse
oximetry circuitry 336, signal processing circuitry 342, and processor 314.
Pet. 38–41 (citing Ex. 1031 ¶¶ 17, 20, Fig. 1 and 3); see Ex. 1003 ¶ 113.
Petitioner argues that a person of ordinary skill in the art “would have
understood that Aceti’s housing 106 (with cover 108) and conductor portion

39
IPR2017-00321
Patent 8,923,941 B2

104 collectively function as a casing that covers, encloses, supports, and

protects electrical and/or electronic components 110, and would thus have
considered processor portion 102, including housing 106 and cover 108, and
conductor portion 104 a ‘housing.’” Pet. 38 (citing Ex. 1003 ¶ 112); Reply
10–11; see supra Section II.A.3. Aceti teaches that device 100 includes
processor portion 102 and conductor portion 104 that includes first end 112,
configured for insertion in wearer’s auditory canal and having optically
transparent elastomer 408. Ex. 1031 ¶¶ 16, 33; see Ex. 1003 ¶ 125.
Referring to Aceti’s Figure 3, Petitioner argues that Aceti’s pulse
oximetry sensor 302, which includes 660 nm LED 330, 805 nm LED 332,
and photo diode 334, is a PPG sensor. Ex. 1031 ¶ 27; see Ex. 1003 ¶¶ 26–
33, 104; Ex. 1034 ¶ 38 (describing a typical PPG sensor, in which “two
different wavelengths of light (e.g. 660 and 805 nm) are applied to one side
of a finger and the received intensity is detected on the opposite side after
experiencing some absorption by the intervening vascular tissues”).
Moreover, Aceti explains that its device may include one or more of the
disclosed sensors. See Ex. 1031 ¶ 20. In addition, referring to Figure 4
(reproduced above), light emitted from LEDs 330 and 332 may be
conducted via fiber optic cables 402 to optically transparent elastomer 408
and into the wearer’s tissue. Ex. 1031 ¶¶ 27, 33; see Ex. 1003 ¶ 125.
Similarly, light returning from the wearer’s tissue passes through optically
transparent elastomer 408 and is conducted to photo diode 334 via fiber
optic cables 402. Ex. 1031 ¶¶ 27, 33; see Ex. 1003 ¶ 125. Consequently,
Aceti teaches that “the housing comprises at least one window that optically
exposes the at least one PPG sensor to a body of a subject wearing the
device, and wherein the housing comprises non-air light transmissive

40
IPR2017-00321
Patent 8,923,941 B2

material in optical communication with the at least one PPG sensor and the
window.” Pet. 43–44.
As noted above, Aceti teaches that electric and/or electronic
components 110 include processor 314 in housing 106. See Ex. 1031 ¶¶ 17,
22. Further, Aceti teaches that processor 314 “is configured to process
signals from the sensors, present information (e.g., via the presentation
device 312), and communicate information (e.g., via the data input/output
circuitry 322/324 and the transceiver 320).” Id. ¶ 21. Petitioner
acknowledges that “Aceti does not expressly disclose that the processor is
configured to reduce motion artifacts.” Pet. 41. Nevertheless, Petitioner
argues that “processing signals from a motion sensor to reduce motion
artifacts from a PPG sensor was an old and well-known technique at the time
of the alleged invention” (id. (citing Ex. 1003 ¶ 121)) and that “Fricke’s
processor (i.e., signal processing unit 114) is configured to remove motion
artifacts from signals (i.e., detector signal 112) produced by an optical
detector (i.e., photodetector 110) in response to signals produced by the
motion sensor (i.e., accelerometer 130) to improve the signal-to-noise ratio”
(id. (citing Ex. 1016 ¶¶ 7, 44, 67–69); see Ex. 1003 ¶ 122). Thus, Fricke
teaches the limitations of claim 14 missing from Aceti.
Petitioner further argues that a person of ordinary skill in the art
would have had reason to combine the teachings of Aceti and Fricke “to add
motion reduction signal processing like that of Fricke to Aceti’s device.” Id.
at 41–42; see Ex. 1003 ¶¶ 122–124. Specifically, because the problem of
motion artifacts in PPG signals was known (see Ex. 1003 ¶ 121) and because
Aceti and Fricke are directed to the processing of PPG signals (see Ex. 1031
¶ 27; Ex. 1016, Fig. 3), according to Petitioner, a person of ordinary skill in

41
IPR2017-00321
Patent 8,923,941 B2

the art would have had reason to use a known technique, such as that taught
by Fricke, to improve a similar device to improve Aceti’s monitoring device
in a similar way. Id. at 42–43 (citing Ex. 1003 ¶¶ 123–124); see KSR, 550
U.S. at 417; Ethicon, 844 F.3d at 1351.

b. Patent Owner’s Contentions

Patent Owner contends that Petitioner fails to demonstrate that the

combined teachings of Aceti and Fricke render the device of challenged
claim 14 obvious. In particular, Patent Owner contends that the combined
teachings of Aceti and Fricke fail to teach or suggest all of the limitations of
claim 14. PO Resp. 25–33. For the reasons discussed below, we disagree.
Patent Owner acknowledges that Aceti teaches a PPG sensor and
motion sensor, but that it does not teach a processor that is configured to
reduce motion effects. Id. at 27 (citing Ex. 1031 ¶¶ 27, 33). Patent Owner
further acknowledges that Fricke teaches “the use of an accelerometer to
sense motion and use this information to process the PPG signals to remove
motion-related artifacts.” Id. at 29 (citing Ex. 1016 ¶¶ 44, 67–69). Thus, as
Petitioner argues, the combined teachings of Aceti and Fricke teach or
suggest the “chipset,” recited in claim 14. Pet. 38–42. Nevertheless, Patent
Owner contends that the combined teachings of Aceti and Fricke fail to
teach or suggest “a window that optically exposes a PPG sensor to the body
of a subject and a chipset in the same housing.” PO Resp. 29.
Although Patent Owner does not dispute our construction of the term
“housing” (see supra Section II.A.3.) which includes “one or more parts that
covers, encloses, supports, or protects,” Patent Owner contends that the
recitations of claim 14 require that the “chipset,” the “at least one window,”
and the “non-air light transmissive material” must be within the same

42
IPR2017-00321
Patent 8,923,941 B2

“housing” (id. at 29–30). Petitioner relies on Aceti to teach or suggest this

arrangement (Pet. 38–40, 43–44), but Patent Owner contends Aceti teaches
that these components are contained in processing portion 102 and/or
conductor portion 104 and that these portions are different housings. PO
Resp. 30–31.
In particular, Patent Owner contends that these portions are different
housings because (1) conductor portion 104 is removably coupled to
processing portion 102 (PO Resp. 31 (citing Ex. 1031 ¶ 16)), (2) each part of
the alleged housing does not perform each of the recited functions (id.
(citing Ex. 2010 ¶ 96)), and (3) the parts of the alleged housing may be made
from different materials (id. at 32–33 (citing Ex. 2010 ¶¶ 97, 98)). We do
not find these contentions persuasive.
First, Patent Owner’s contention that conductor portion 104 can only
be removably coupled to processor portion 102 is incorrect. As noted above,
Aceti discloses that “conductor portion 104 is removably coupled to the
processor portion 102 and is considered disposable” only in an exemplary
embodiment. Ex. 1031 ¶ 16. Thus, Aceti is understood to teach
embodiments, in which conductor portion 104 is not removably coupled to
processing portion 102. Consequently, we are not persuaded by Patent
Owner’s argument that Aceti’s device requires that conductor portion 104
and processor portion 102 are removably coupled. Moreover, Patent
Owner’s declarant testified that the Specification of the ’941 patent does not
limit the disclosed device’s housing to removable or fixed components. Ex.
1070, 133:14–134:19. We credit this testimony. Therefore, even if
conductor portion 104 is removable from processor portion 102, that is not

43
IPR2017-00321
Patent 8,923,941 B2

sufficient to distinguish Aceti’s housing teachings from the housing recited

in claim 14.
Second, Patent Owner contends that, “[w]hile the Board’s
construction permits the housing to be comprised of ‘one or more parts,’
those parts must cover, enclose, support, or protect the claimed
components.” PO Resp. 31. This incorrectly suggests that each part of the
housing must perform each of the recited functions of the housing. Patent
Owner’s contention shows a misunderstanding of our construction of the
term “housing.” We construed the term “housing” to mean “one or more
parts that covers, encloses, supports, or protects; [a] casing.” See supra
Section II.A.3. Thus, the one or more parts comprising the housing, alone
or together, covers, encloses, supports, or protects the housed items; and, to
the extent permitted by the claim language, different parts comprising a
housing may perform different functions. As Petitioner argues, “[a] POSA
would have understood that Aceti’s housing 106 (with cover 108) and
conductor portion 104 collectively function as a casing that covers, encloses,
supports, and protects electrical and/or electronic components 110, and
would thus have considered housing 106 and conductor portion 104 a
‘housing.’” Pet. 38 (emphasis added, citing Ex. 1003 ¶ 112); see Reply 10–
11. Patent Owner’s declarant testifies that different parts may form the
housing. Ex. 1070, 131:11–133:13. Again, we credit this testimony.
Therefore, just as different parts of the body of an automobile enclose the
engine, the passenger compartment, and the trunk; the “housing” comprising
Aceti’s processor portion 102 and conducting portion 104 enclose the pulse
oximetry sensor 302, accelerometer 306, pulse oximetry circuitry 336, signal
processing circuitry 342, and processor 314, i.e., the “chipset”; fiber optic

44
IPR2017-00321
Patent 8,923,941 B2

cables 402 and 406, i.e., the “non-air light transmissive material”; and
optically transparent elastomer 408, i.e., the “windows.”
Third, Patent Owner contends that, because “first end 112 is
configured for comfort, biocompatibility, durability, and ease of
manufacture,” a person of ordinary skill in the art “would have considered
elastomer 408 to comprise a different housing or to sit atop the housing
comprised of housing 106 and conductor 104.” PO Resp. 32 (emphasis
added, citing Ex. 2010 ¶¶ 97, 98). Nevertheless, because different parts of
the housing may perform different functions, those parts may be made of
materials tailored to their function. Reply 11–12. Patent Owner’s declarant
testified that the parts of a housing need not be made from the same material.
Ex. 1070, 159:19–160:2. Once again, we credit this testimony. Returning to
our automobile body analogy, an automobile body may comprise parts made
from steel, aluminum, and polymers, depending upon their function, but
together they form the automobile body. Although the parts of the housing
may be made from different materials, we are not persuaded that this
prevents the parts to collectively serve as a housing.
Accordingly, we are persuaded that Petitioner demonstrates by a
preponderance of the evidence that the combined teachings of Aceti and
Fricke teach or suggest all of the limitations of claim 14 and demonstrates
sufficiently that a person of ordinary skill in the art would have had reason
for combining the teachings of Aceti and Fricke to achieve the device recited
in claim 14 of the ’941 patent.

c. Claims 15–19 and 21

Petitioner argues that claims 15–19 and 21 also are rendered obvious
over the combined teachings of Aceti and Fricke and provides a mapping of

45
IPR2017-00321
Patent 8,923,941 B2

the reading of the additional limitations of these dependent claims on the

combined teachings of Aceti and Fricke, as follows:
Claim Limitation Applied Teachings
15 “wherein the non-air light Aceti teaches that connector
transmissive material portion 104 includes fiber optic
comprises at least one light cables 402 in optical
guide in optical communication with pulse
communication with the at oximeter 302, including
least one PPG sensor and photodetector diode 334, and
the window” optically transparent elastomer
408. Ex. 1031 ¶¶ 27, 33, Fig. 4;
see Pet. 44 (citing Ex. 1003
¶ 127).
16 “wherein the wearable Referring to its Figure 2, Aceti
device is an earbud” discloses that exemplary
monitoring device
100 may be inserted into the
auditory canal of a wearer.
Ex. 1031 ¶ 19. Petitioner argues
that a person of ordinary skill in
the art would understand this to
teach an earbud.” Pet. 44 (citing
Ex. 1003 ¶ 128).
17 “wherein the wearable Referring to its Figure 2, Aceti
device is a headset” discloses that exemplary
monitoring device
100 may be positioned relative
to ear 202 on head 204 of a
wearer. Ex. 1031 ¶ 19.
Petitioner argues that a person of
ordinary skill in the art would
understand this to teach a
“headset.” Pet. 44 (citing
Ex. 1003 ¶ 128).
18 “wherein the at least one Referring to its Figure 3, Fricke
processor is configured to discloses using band-pass filters
reduce motion artifacts by 302 and 304 to reduce motion
removing frequency bands artifacts and produce a

46
IPR2017-00321
Patent 8,923,941 B2

from the signals that are preconditioned signal. Ex. 1016

outside of a range of interest ¶ 53. Thus, Petitioner argues
using at least one band-pass that Fricke teaches removing
filter to produce frequency bands from signals
preconditioned signals” that are outside a range of
interest using band-pass filtering.
Pet. 45 (citing Ex. 1003 ¶¶ 108,
130).
19 “wherein the at least one Fricke teaches that, after band-
processor is configured to pass filtering is performed,
filter the preconditioned additional filtering may be
signals via at least one of performed on the band-pass
the following: FIR (Finite filtered signals by, for example,
Impulse Response) filtering, instantaneous frequency
IIR (Infinite Impulse computation component 306 or
Response) filtering, 318. Ex. 1016 ¶¶ 55, 92. Such
adaptive filtering. phase additional filtering may include
filtering, and frequency low-band filtering, and high-pass
filtering” (emphasis added) filtering, Hilbert transform, LMS
adaptive filtering, Kalman
filtering, and matched filtering.
Id. Thus, Fricke teaches at least
adaptive filtering, phase
filtering, and frequency filtering.
Pet. 45–46 (citing Ex. 1003
¶¶ 109, 110, 130–134).
21 “wherein the at least one Aceti discloses that the motion
motion sensor comprises at sensor comprises at least one
least one of the following: accelerometer. Ex. 1031 ¶ 29,
an optical sensor, an inertial Fig. 3; see Pet. 47 (citing
sensor, an electrically Ex. 1003 ¶ 137).
conductive sensor, a
capacitive sensor, an
inductive sensor, an
accelerometer, and
a blocked channel sensor”
(emphases added)
Pet. 44–47 (citing Ex. 1003 ¶¶ 127–137).

47
IPR2017-00321
Patent 8,923,941 B2

Patent Owner does not contest this mapping, but, instead, contends
that Petitioner fails to demonstrate the combined teachings of Aceti and
Fricke teach or suggest all of the limitations of the base claim, claim 14. PO
Resp. 2; Reply 12; see supra Section II.C.5. For the reasons set forth above,
we are persuaded that Petitioner has demonstrated that the combined
teachings of Aceti and Fricke teach or suggest all of the limitations of
claim 14, as well as those of claims 15–19 and 21, and provides sufficient
reason to have combined the teachings of Aceti and Fricke. Pet. 38–44.
Thus, we are persuaded that Petitioner has demonstrated by a preponderance
of the evidence claims 15–19 and 21 are unpatentable under 35 U.S.C.
§ 103(a) as rendered obvious by the combined teachings of Aceti and Fricke.

d. Claim 20

Petitioner argues that claim 20 is rendered obvious over the combined

teachings of Aceti, Fricke, and Comtois and provides a mapping of the
additional limitations of this dependent claim on the combined teachings of
Aceti, Fricke, and Comtois. Pet. 47–50 (citing Ex. 1003 ¶¶ 138–143). In
particular, claim 20 recites that “the subject motion artifacts comprises
subject footstep-related motion artifacts” (Ex. 1001, 32:34–35) and that,
although “Aceti and Fricke do not expressly mention removing footstep-
related motion artifacts” (Pet. 47), Comtois teaches processing PPG signals
to remove such artifacts (Ex. 1032, Abstract, 1530, Fig. 2). See Ex. 1003
¶¶ 141–142. Further, Petitioner argues that, because Fricke and Comtois
teach similar processing techniques to remove artifacts, a person of ordinary
skill in the art would have had reason to combine their teachings to achieve
the device as recited in claim 20. Pet. 50 (citing Ex. 1003 ¶ 143).

48
IPR2017-00321
Patent 8,923,941 B2

Patent Owner does not contest this mapping, but, instead, contends
that Petitioner fails to demonstrate an adequate reason to have combined the
teachings of Aceti and Fricke to achieve the device recited in the base claim,
claim 14, of challenged claim 20. PO Resp. 2; Reply 12; see supra
Section II.C.5. For the reasons set forth above, we are persuaded that
Petitioner has demonstrated by a preponderance of the evidence claim 20 is
unpatentable under 35 U.S.C. § 103(a) as rendered obvious by the combined
teachings of Aceti, Fricke, and Comtois.

e. Summary

For the reasons set forth above, we are persuaded that Petitioner
demonstrates by a preponderance of the evidence that claims 14–21 of the
’941 patent are unpatentable under 35 U.S.C. § 103(a) as obvious over Aceti
and Fricke, alone or in combination with Comtois.
D. Unpatentability of Challenged Claims
In consideration of the above, we conclude that Petitioner has
demonstrated by a preponderance of the evidence that claims 14–21 of the
’941 patent are unpatentable based on the challenges asserted in the Petition.
Specifically Petitioner has demonstrated by a preponderance of the evidence
that:
Challenged Determination
References Basis
Claim(s)
Kosuda and Unpatentable
35 U.S.C. § 103(a) 14, 15, and 21
Maekawa
Kosuda, Maekawa, Unpatentable
35 U.S.C. § 103(a) 18–20
and Han
Aceti and Fricke 35 U.S.C. § 103(a) 14–19 and 21 Unpatentable
Aceti, Fricke, and Unpatentable
35 U.S.C. § 103(a) 20
Comtois

49
IPR2017-00321
Patent 8,923,941 B2

III. MOTION TO AMEND

In an inter partes review, amended claims are not added to a patent as

of right, but rather must be proposed as a part of a motion to amend. 35
U.S.C. § 316(d). We must assess the patentability of the proposed substitute
claims “without placing the burden of persuasion on the patent owner.”
Aqua Prods., Inc. v. Matal, 872 F.3d 1290, 1328 (Fed. Cir. 2017). Thus, we
determine whether the preponderance of the evidence based on the entirety
of the record shows that the substitute claims are unpatentable. Patent
Owner’s proposed substitute claims still must meet the statutory
requirements of 35 U.S.C. § 316(d) and the regulatory requirements of
37 C.F.R. § 42.121. See “Guidance on Motions to Amend in view of Aqua
Products” (Nov. 21, 2017). 11 Accordingly, Patent Owner must demonstrate
(1) the amendment responds to a ground of unpatentability involved in the
review; (2) the amendment does not seek to enlarge the scope of the claims
of the patent or introduce new subject matter; (3) the amendment proposes a
reasonable number of substitute claims; and (4) the original disclosure sets
forth written description support for each proposed claim. See 35 U.S.C.
§ 316(d)(2) and (3); 37 C.F.R. § 42.121.

A. Proposed Substitute Claims

Patent Owner filed a contingent motion to substitute independent

claim 22 for claim 14, if found unpatentable; to substitute dependent claim
26 for claim 18, if found unpatentable; and to substitute claims altering the

11
The guidance memorandum is publicly available at
https://www.uspto.gov/sites/default/files/documents/guidance_on_motions_t
o_amend_11_2017.pdf

50
IPR2017-00321
Patent 8,923,941 B2

dependencies of claims 15–17, 19, and 21 from claim 14 to claim 22 and the
dependency of claim 27 from claim 18 to claim 26, if the claims from which
they depend are found unpatentable. MTA 23–25 (App’x A). In particular,
Patent Owner proposes to limit claim 14 by reciting that the
at least one signal processor [is] configured to process
signals from the at least one motion sensor and signals from the
at least one PPG sensor to reduce motion artifacts from the PPG
signals and to extract physiological and motion parameters;
wherein the at least one signal processor configured to
process data to be output, wherein the output data comprises
physiological information and motion related information, and
wherein the output data is parsed out such that an application-
specific interface (API) can utilize the physiological information
and motion-related information for an application;
and to limit claim 18 by also reciting that the at least one processor is
configured “to generate the parsed output data by executing one or more
processing methods to provide information that is fed into a multiplexed
output serial data string of motion-related and physiological information.”
MTA 23–25 (App’x A) (added limitations underlined).
Because the sole substitute independent claim, claim 22, includes
additional limitations that were not part of the challenged claims and
includes all of the limitations of challenged claim 14, we are persuaded that
the substitute claims do “not enlarge the scope of the claims of the patent.”
35 U.S.C. § 316(d)(3); see 37 C.F.R. § 42.121(a)(2)(ii). Moreover, because
Patent Owner proposes only one substitute claim for each of the challenged
claims, we are persuaded that Patent Owner has proposed a reasonable
number of substitute claims. 37 C.F.R. § 42.121(a)(3) (“The presumption is
that only one substitute claim would be needed to replace each challenged
claim”).

51
IPR2017-00321
Patent 8,923,941 B2

B. Responsiveness to a Ground of Institution

Our Rules provide that “[a] motion to amend may be denied where:
(i) The amendment does not respond to a ground of unpatentability involved
in the trial.” 37 C.F.R. § 42.121(a)(2)(i). Patent Owner makes the general
statement that “[t]he references asserted in this Inter Partes Review, and all
other references known to [Patent Owner], fail to anticipate or render the
substitute claims obvious. See Nike, Inc. v. Adidas AG, 812 F.3d 1326,
1350-51 (Fed. Cir. 2016).” MTA 2. Nevertheless, Patent Owner does not
explain how the substitute claims are distinguishable over the combinations
of references upon which we instituted review or, in fact, even mention
Kosuda, Maekawa, Han, Aceti, Fricke, or Comtois in its Motion to Amend.
See id. at ii, iv.
Petitioner contends that Patent Owner’s Motion to Amend is not
responsive to any ground of unpatentability involved in the review because
the Motion to Amend does not discuss the Kosuda, Maekawa, Aceti, or
Fricke. MTA Opp. 4–5 (citing Aqua Prod., 872 F.3d at 1341 (Reyna J.
concurring)); see supra Section I.E. Although the Motion to Amend
includes a heading entitled “The Claim Amendments Overcome the Asserted
Grounds of [Un]Patentability,” the Motion to Amend contains only the most
conclusory statements in this regard and provides no analysis or evidence in
support of this assertion. MTA 11–12. Instead, Patent Owner identifies and
attempts to distinguish the substitute claims over other references. Id. at 12–
20. Although we do not place the burden of persuasion on the Patent
Owner, Patent Owner’s silence provides no information as to how the
proposed claim amendments distinguish over the applied references in the
challenges on which we instituted inter partes review. Thus, the

52
IPR2017-00321
Patent 8,923,941 B2

significance of the amendments in the substitute claims in the context of the

challenges presented in the Petition is somewhat unclear.
In its Reply to Petitioner’s Opposition to Motion to Amend, Patent
Owner contends that as long as the substitute claims are responsive to a
ground of unpatentability involved in the review, the requirements of our
Rules have been satisfied. MTA Reply 12 (quoting 37 C.F.R.
§ 42.121(a)(2)). Patent Owner further contends that “[t]he Patent Trial and
Appeal Board in Idle Free clarified this requirement stating, ‘[a] proposed
substitute claim is not responsive to an alleged ground of unpatentability of a
challenged claim if it does not either include or narrow each feature of the
challenged claim being replaced.’” Id. at 12 (quoting Idle Free Sys., Inc. v.
Bergstrom, Inc., Case IPR2012-00027, slip op. at 5 (PTAB June 11, 2013)
(Paper 26) (“Idle Free”) 12). The fact that the substitute claim is narrowing is
a necessary, but not a sufficient, condition for distinguishing the substitute
claims over a ground of unpatentability involved in the review. See MTA
Sur-Reply 1 (citing Blackberry Corp. v. MobileMedia Ideas LLC, Case
IPR2013-00016, slip op. at 15–18 (PTAB Feb. 25, 2014) (Paper 32)).
Nevertheless, Patent Owner explains in its Reply to Petitioner’s Opposition
to Motion to Amend how it believes these substitute claims are
distinguishable over the combined teachings of Kosuda and Maekawa, alone
or in combination with Han, and of Aceti and Fricke, alone or in
combination with Comtois. MTA Reply 7–10; see id. at 7 n.2. Thus,
although Patent Owner did not address these grounds in its Motion to

12
Although Idle Free was designated as “informative” at the time of the
briefing in this case, Idle Free was de-designated as “informative” on
June 1, 2018.

53
IPR2017-00321
Patent 8,923,941 B2

Amend, we find that Patent Owner has adequately explained how the
proposed, substitute claims respond to the grounds of unpatentability
involved in the review. See Western Digital Corp. v. SPEX Tech., Inc.,
IPR2018-00082, slip op. at 5–7 (PTAB Apr. 25, 2018) (Paper 13)
(informative) 13 (“Western Digital”) (“Thus, in considering the motion, we
review the entirety of the record to determine whether a patent owner’s
amendments respond to a ground of unpatentability involved in the trial.”).

C. Written Description Support

Our Rules further require that “[a] motion to amend claims must . . .
set forth: (1) The support in the original disclosure of the patent for each
claim that is added or amended; and (2) The support in an earlier-filed
disclosure for each claim for which benefit of the filing date of the earlier
filed disclosure is sought.” 37 C.F.R. § 42.121(b). Patent Owner contends
that the substitute claims are supported by the Specification of the ’941
patent application (Ex. 2116 (the “’941 Application”)); its parent
application, U.S. Patent Application No. 12/691,388, now U.S. Patent No.
8,700,111 B2 (filed on Jan. 21, 2010, the “’111 Application”) (Ex. 2108);
and U.S. Provisional Application No. 61/274,191 (filed on Aug. 14, 2009,
the “191 Prov. Application”) (Ex. 2109), from which the ’941 patent claims
earliest priority. MTA 4. In support of its contention, Patent Owner

13
This order was designated as “informative” on June 1, 2018. “An
informative opinion is not binding authority,” but may identify “Board
norms on recurring issues,” “guidance on issues of first impression,” and/or
“guidance on Board rules and practices.” PTAB SOP 2, rev. 9 (Sec. IV).
Consequently, neither Idle Free nor Western Digital, or the de-designation
of one as informative or the designation of the other, is determinative of the
outcome of this case.

54
IPR2017-00321
Patent 8,923,941 B2

provides a claim chart juxtaposing the recitations of substitute claims 22 and

26 with lists of citations from the ’941 Application, the ’111 Application,
and the 191 Prov. Application, which allegedly provide “Exemplary
Support.” Id. at 4–7. Patent Owner provides no argument or further
explanation in the Motion to Amend of how the cited text provides written
support for the proposed, substitute claims.
Petitioner argues that Patent Owner’s claim charts, including bare
citations to applications in the priority chain of the ’941 patent, without
further argument or explanation, fails to satisfy the requirement that Patent
Owner identify written description support for the substitute claims under
37 C.F.R. § 42.121(b). MTA Opp. 1–2. As the Board has previously
determined, “the motion should account for the claimed subject matter as a
whole, i.e., the entire proposed substitute claim, when showing where there
is sufficient written description support for each claim feature.” MTA Sur-
Reply 2–3 (quoting Facebook, Inc. v. B.E. Tech., LLC., Case IPR2014-
00052, slip op. at 27 (PTAB Mar. 31, 2015) (Paper 45); aff’d, B.E. Tech.,
L.L.C. v. Google, Inc., No. 2015-1827, 2016 WL 6803057, at *7 (Fed. Cir.
Nov. 17, 2016) (citing Respironics, Inc. v. Zoll Med. Corp., Case IPR2013-
00322, slip op. at 24 (PTAB Sept. 17, 2014) (Paper 46) (“Zoll’s string
citations amount to little more than an invitation to us (and to Respironics,
and to the public) to peruse the cited evidence and piece together a coherent
argument for them. This we will not do; it is the province of advocacy.”) 14,

14
See Stampa v. Jackson, 78 USPQ2d 1567, 1571 (BPAI 2005) (quoting
Ernst Haas Studio, Inc. v. Palm Press, Inc., 164 F.3d 110, 111–12 (2d Cir.
1999) (“Appellant’s Brief is at best an invitation to the court to scour the
record, research any legal theory that comes to mind, and serve generally as
an advocate for appellant. We decline the invitation.”)).

55
IPR2017-00321
Patent 8,923,941 B2

vacated and remanded on other grounds, 656 Fed. Appx. 531, 534 (Fed. Cir.
2016)).
In its Reply to Petitioner’s Opposition to the Motion to Amend, Patent
Owner chose not to explain the relevance of the citations set forth in its
claim chart, but, instead, to contend that the bare listings of citations were
adequate to satisfy the requirement in our Rules that Patent Owner establish
written description support for the substitute claims. MTA Reply 10.
Moreover, Patent Owner characterizes its amendments to claims 14 and 22
as not presenting “extensive modifications” and its showing of support as
containing multiple, string citations, rather than a single, string citation. Id.
at 11. Patent Owner also cites decisions in which Board panels have
accepted tables of citations as an adequate showing of written description
support for substitute claims. Id. at 10. Thus, Patent Owner maintains that
its attempt to show written description support is distinguishable from the
unsuccessful attempts described in the decisions cited by Petitioner. Id. at
11; see MTA Sur-Reply 2–3.
Our Rules do not state that a claim chart in the motion to amend
listing bare citations is or is not adequate to show written description support
in a motion to amend. See 37 C.F.R. § 42.121(b). 15 Nevertheless, Petitioner
objects primarily to the form by which Patent Owner attempts to show

15
In Western Digital, the panel states that “[t]he motion to amend itself, not
the claim listing . . . , must set forth the written description support. In
addition, the motion must set forth written description support for each
proposed substitute claim as a whole, and not just the features added by the
amendment.” Western Digital at 8 (citation omitted). Because the claim
listing does not count toward page limits, the panel further states in the
“Claim Listing” section of the order: “All arguments and evidence in
support of the motion to amend shall be in the motion itself.” Id. (Sec. 6).

56
IPR2017-00321
Patent 8,923,941 B2

written description support, rather than to the showing itself. MTA Opp. 1–
2; MTA Sur-Reply 2–3. Here, we find that Patent Owner has made a
sufficient and substantially unrebutted showing of written description
support for the substitute claims.

D. Obviousness Analysis
1. Overview

Patent Owner asserts that each of the substitute claims is patentable

over “the references asserted in this Inter Partes Review, and all other
references known to [Patent Owner].” MTA 2. In particular, Patent Owner
asserts that the substitute claims are patentable over U.S. Patent Application
Publication No. 2008/0200774 A1 to Luo (Ex. 2126); U.S. Patent No.
6,513,532 B2 to Mault et al. (Ex. 2136); U.S. Patent Application Publication
No. 2008/0133699 A1 to Craw et al. (Ex. 2127 (“Craw”)); U.S. Patent
Application Publication No. 2003/0181798 A1 to Al-Ali (Ex. 2137); and
R.G. Lee et al., “A Mobile Care System With Alert Mechanism,” IEEE
Transactions on Information Technology in Biomedicine, Vol. 11, Issue 5,
September 2007 (Ex. 2138), alone or in combination with other references.
MTA 13. Petitioner does not contest these assertions, but, instead, argues
that the claims are rendered obvious over Kosuda, Maekawa, and Gupta,
alone or in combination with Han, or over Aceti, Fricke, and Craw, alone or
in combination with Comtois. MTA Opp. 12–25. Therefore, we address
only Petitioner’s challenges to the substitute claims below. See Aqua
Prods., 872 F.3d at 1327–28 (the patent owner does not bear the burden of
persuasion with respect to patentability); Western Digital at 3–4.

57
IPR2017-00321
Patent 8,923,941 B2

2. Claim Construction

Patent Owner acknowledges that we apply the broadest reasonable

interpretation to the terms of the substitute claims and states that “no
additional claim terms need construction beyond the terms previously
construed by the Board in its Decision to Institute.” MTA 11; see supra
Section II.A. Although Patent Owner relies in part on the addition of the
term “application-specific interface (API)” to distinguish the substitute
claims over the identified references (MTA 8–10, 12–19), Patent Owner
does not construe this term (id. at 10–11; see MTA Opp. 2–3). Although
Patent Owner contends that we construed the term “application-specific
interface (API)” in the related proceeding, IPR2017-00319, we did not. See
MTA Reply 1; Tr. 34:19–25.
In the related proceeding, Petitioner argued that there was a
typographical error in the term “application-specific interface (API)” and
that this term should be read as “application programming interface
(API).” 16 IPR2017-00319, Paper 2, 14–15; see Tr. 18:11–18. Patent Owner
proposed no alternative construction in that proceeding, but objected to
Petitioner’s proposed construction. IPR2017-00319, Paper 6, 13. We
rejected Petitioner’s proposed construction for three reasons. First, because
the term was used consistently in the claims and the Specification of the
’941 patent, we were not persuaded that Petitioner had shown that the use of
a known abbreviation, “API,” demonstrated that the term contained a

16
Petitioner argued that, unlike “application-specific interface (API),”
“application programming interface (API)” has a well-known meaning and
“API” is a common abbreviation for the well-known term. See MICROSOFT
COMPUTER DICTIONARY, 33 (5th ed. 2002) (Ex. 3002).

58
IPR2017-00321
Patent 8,923,941 B2

typographical error. IPR2017-00319, Paper 10, 8–10. Second, we were not

persuaded that we should prefer extrinsic evidence, which allegedly
supported Petitioner’s proposed construction, over intrinsic evidence, which
did not. Id. at 10–11. Third, we determined that Petitioner’s proposed
construction for the term was inconsistent with the explanation of the
meaning of the term in the Specification of the ’941 patent. Id. at 11–12.
Thus, although we rejected Petitioner’s proposed construction, we
determined that it was not necessary for us to construe the term ourselves in
order to deny institution of Petitioner’s challenges to claim 3 of the ’941
patent. Id. at 12; see IPR2017-00319, Paper 13, 2 (“the Board did not
provide an actual definition for ‘application specific interface (API)’ in the
Institution Decision”); IPR2017-00319, Paper 15, 5–6. (“[W]e were not
persuaded by Petitioner’s proposed construction; and, because Petitioner’s
arguments with respect to the unpatentability of claim 3 are based on that
construction, we were not persuaded that Petitioner had shown a reasonable
likelihood of prevailing in showing the unpatentability of claim 3 on the
grounds asserted.”). Thus, because we found Petitioner’s proposed
construction of the term “application-specific interface (API)” to be deficient
in IPR2017-00319, we determined that Petitioner’s challenges to claim 3
based on that deficient construction could not succeed. At the time, it was
not necessary for us to construe this term which appeared only in a claim
upon which we did not institute review, and we did not construe this term.
See SAS Inst., 138 S. Ct. at 1354.
In context of the substitute claims presented in the Motion to Amend,
we now must consider the construction of the term “application-specific
interface (API).” Although Patent Owner has provided us with arguments

59
IPR2017-00321
Patent 8,923,941 B2

and evidence regarding the construction of this term, Patent Owner does not
state what this term means or even what it believes that we construed this
term to mean. MTA 7–11.
The Specification of the ’941 patent states that:
The multiplexed data outputs 604 may be a serial data string of
activity and physiological information 700 (FIG. 18) parsed out
specifically such that an application-specific interface (API) can
utilize the data as required for a particular application. The
applications may use this data to generate high-level
assessments, such as overall fitness or overall health.
Furthermore, the individual data elements of the data string can
be used to facilitate better assessments of other individual data
elements of the data string.
Ex. 1001, 26:15–23 (emphasis added, referring to Fig. 17); see MTA 8–9.
The Specification of the ’941 patent gives examples of these “particular
applications,” e.g., a “Blood Flow” application for determining first and
second derivatives of each blood pulse and an application to determine
“blood volume over each blood pulse.” Ex. 1001, 26:23–31. Referring to
the Abstract of the ’941 patent, Patent Owner contends that “[o]nce the data
string is obtained, it is parsed such that an application specific interface can
use both sets of data to generate statistical relationships between the
physiological parameters and the physical activity parameters.” MTA 8.
Specifically, the Abstract of the ’941 patent states that “[t]he serial data
string is parsed out such that an application-specific interface can utilize the
physiological information and motion-related information for an application
that generates statistical relationships between subject physiological
parameters and subject physical activity parameters in the physiological
information and motion-related information.” Ex. 1001, [57] (emphasis

60
IPR2017-00321
Patent 8,923,941 B2

added). None of Patent Owner’s arguments or evidence, however, amounts

to a construction of this term.
Petitioner argues that Patent Owner does not present any specific
construction of the term “application-specific interface (API).” MTA
Opp. 2–3. We agree. Moreover, Petitioner argues that Patent Owner and its
declarant have taken inconsistent positions regarding the meaning of this
term. Id. Again, we agree.
In IPR2017-00319, Petitioner argued that the term “application-
specific interface (API)” in claim 3 of the ’941 patent should be construed as
“application programming interface (API).” IPR2017-00319, Paper 2, 14–
15. Petitioner explained that “[a]n API specifies how components should
interact with each other. In practice, an API is often a library that includes
specifications for routines, data structures, object classes, and variables.
APIs are thus characterized by their broad applicability to different
applications—and not ‘application specific’ as such.” Id. at 14 (citations
omitted); see MICROSOFT COMPUTER DICTIONARY at 33 (defining
“application programming interface” as “[a] set of routines used by an
application program to direct the performance of procedures by the
computer’s operating system”) (Ex. 3002). As noted above, we rejected this
construction as inconsistent with the Specification of the ’941 patent, which
describes “an application-specific interface (API) can utilize the data as
required for a particular application.” Ex. 1001, 26:17–19; see MTA
Opp. 2–3. Nevertheless, Patent Owner now contends that Figure 17 of the
’941 patent “teaches how to create a serial data string by pulling multiple
metrics from the sensors by outside Application Programming Interfaces
(‘APIs’).” PO Resp. 4 (emphasis added); Tr. 20:15–21; see Ex. 2010 ¶ 40;

61
IPR2017-00321
Patent 8,923,941 B2

Ex. 1070, 125:3–128:24; but see Tr. 34:4–18. We find this Patent Owner
contention, as well as the cited testimony of its declarant, to be inconsistent
with the Specification of the ’941 patent. Moreover, despite its rejection of
Patent Owner’s arguments and evidence regarding the construction of this
term, Petitioner offers essentially the same, previously rejected, construction
of this term. MTA Opp. 8–9; see Ex. 1072 ¶¶ 30–35.
After considering the parties’ arguments and the cited evidence,
especially, the Specification of the ’941 patent and the ordinary meaning of
the word “interface” in this field, 17 in the context of substitute claim 22, we
construe the term “application-specific interface (API)” to mean “an
interface which enables a particular application to utilize data obtained from
hardware, such as the at least one motion sensor and the at least one PPG
sensor.” Because substitute claims 23–29 depend from independent,
substitute claim 22, this construction also applies to those claims. In light of
this construction, we now consider Petitioner’s challenges to the substitute
claims.
3. Obviousness Over Kosuda, Maekawa, and Gupta, Alone or in
Combination with Han
Petitioner argues that claim 22, 23, and 26–29 are unpatentable under
35 U.S.C. § 103(a) as obvious over Kosuda, Maekawa, and Gupta, alone or
in combination with Han. MTA Opp. 12–16. To support its argument,
Petitioner provides a mapping of the limitations of claims 22 and 26 to

17
A relevant definition of “interface” is “[s]oftware that enables a program
to work with the user (the user interface, which can be a command-line
interface, menu-driven interface, or a graphical user interface), with another
program such as the operating system, or with the computer’s hardware.”
MICROSOFT COMPUTER DICTIONARY at 279–80 (Ex. 3002).

62
IPR2017-00321
Patent 8,923,941 B2

structures taught or suggested by Kosuda, Maekawa, and Gupta and by

Kosuda, Maekawa, Gupta, and Han, respectively. Id. Petitioner also cites
Dr. Sarrafzadeh’s Declaration for support. See Ex. 1072 ¶¶ 53–90. Patent
Owner limits is response to these arguments to the challenges to independent
claim 22. MTA Reply 7–8.
We begin our analysis of these grounds of unpatentability with a
review of the Gupta.

a. Gupta (Ex. 1045)

Gupta is entitled “Design of a Low-cost Physiological Parameter

Measuring and Monitoring Device” and teaches a monitoring device used to
measure physiological parameters, such as motion (e.g., fall impact) and
heart rate, of a human subject. Ex. 1045, Abstract. Gupta’s Figure 1 is
reproduced below.

Figure 1 depicts a physiological parameter measurement and monitoring

system. Id. at 2; Ex. 1072, ¶48. The system includes of three sensors: a
temperature sensor, a heart rate sensor, and an impact sensor. Ex. 1045, 2.
Each of the sensor circuitries used in this system generates analog voltages

63
IPR2017-00321
Patent 8,923,941 B2

which are fed as Analog-to-Digital converter (“ADC”) inputs to the

microcontroller. Id. The ADC inputs are time-multiplexed and sampled at
different rates, then encoded by the microcontroller to generate a packet of
heart rate, skin temperature, and impact information. Id. at 2, 4; see
Ex. 1072 ¶ 51.
Gupta’s Figure 5 is reproduced below.

Figure 5 depicts the composition of a data packet. Data packets are sent to
the receiver, which is connected to a personal computer (“PC”) and which is
constantly receiving data packets. Ex. 1045, 4; see Ex. 1072 ¶ 52. “A
program, running on the PC, receives the packetized information from [a]
serial port, decodes the packet and then displays this information on the PC
monitor.” Ex. 1045, 4. When the readings from the patient exceed a
threshold, an alarm is issued. Id.; see Ex. 1072 ¶ 52.

b. Claim 22

We determine above, that the combined teachings of Kosuda and

Maekawa teach or suggest all of the limitations of original claim 14. See
supra Section II.B.5. Claim 22 includes all of the limitations of claim 14
and adds certain additional limitations. In the Petition, Petitioner argues that

64
IPR2017-00321
Patent 8,923,941 B2

Kosuda teaches MPU 24/signal processing circuit 17 is configured to

process signals from triaxial acceleration sensor 12, i.e., at least one motion
sensor, and from pulse wave sensor 13, i.e., at least one PPG sensor, in order
to reduce motion artifacts from the PPG signals. See Pet. 24.
MPU 24/signal processing circuit 17 performs a type of frequency
analysis on the residual data, extracts the harmonic components of the pulse
wave, and calculates the pulse rate from the frequency. Ex. 1027 ¶¶ 151–
152. Thus, MPU 24/signal processing circuit 17 is configured to extract
pulse rate, i.e., a physiological parameter, from its pulse wave signal, i.e., a
PPG signal. Ex. 1072 ¶¶ 64–67.
Nevertheless, Petitioner acknowledges that although MPU 24/signal
processing circuit 17 monitors the motion signal from the motion sensor to
reduce motion artifacts, Kosuda does not teach that the MPU 24/signal
processing circuit 17 is configured to extract a particular motion parameter.
MTA Opp. 13. Moreover, although Kosuda teaches that MPU 24/signal
processing circuit 17 is configured to process data as output, and that this
output data comprises physiological information, Petitioner acknowledges
Kosuda does not teach that the output data includes motion-related
information or that the output data is parsed 18 out, such that an application-
specific interface (API) can utilize the physiological information and
motion-related information for a particular application. Id. (citing Ex. 1072
¶ 68).

18
The verb “to parse” means “[t]o break input into smaller chunks so that a
program can act upon the information.” MICROSOFT COMPUTER
DICTIONARY at 392 (Ex. 3002); see Tr. 37:12–17.

65
IPR2017-00321
Patent 8,923,941 B2

Like Kosuda, Gupta teaches a wrist-worn pulse monitoring device that

includes a heart rate sensor, i.e., a PPG sensor; a temperature sensor; and an
accelerometer, i.e., a motion sensor, to monitor impacts. Ex. 1045, 2; see
Ex. 1072 ¶ 69 (Gupta’s monitoring device also uses photoplethysmography
to measure heart rate). Further, “Gupta’s processor is configured to process
signals from the motion sensor to determine a fall detection parameter (i.e., a
motion parameter) for transmission to a remote computer.” MTA Opp. 13–
14 (citing Ex. 1045, 2; Ex. 1072 ¶¶ 69–71).
Petitioner argues that
Gupta’s system receives the pulse signal and accelerometer data
from the sensors in a time-division multiplexed manner, and
processes the data to provide a data packet (serial data string) of
heart rate and impact (motion) information. Ex. 1045, pp. 2–4,
FIG. 5; Ex. 1072, ¶70. The data is output to a receiving unit.
Ex. 1072, ¶70. The data is arranged such that an application
specific interface can decode the packet and a particular
application can display the information from the packet. Id. The
application program may also raise an alarm when the readings
exceed a threshold. Id.
MTA Opp. 14.
Patent Owner contends that Petitioner relies on Gupta to teach that
“the data includes motion-related information or that the output data is
parsed out such that an application-specific interface (API) can utilize the
physiological information and motion-related information for an
application,” which teachings are missing from Kosuda. MTA Reply 7
(citing MTA Opp. 13). Patent Owner acknowledges that Gupta teaches that
its system receives the pulse signal and accelerometer data from the sensors
in a time-division, multiplexed manner (Ex. 1045, 2–4) and processes the
data to provide a data packet, i.e., a serial data string, of heart rate and

66
IPR2017-00321
Patent 8,923,941 B2

impact information (id., Fig. 5 (reproduced above)). Patent Owner contends,

however, that Petitioner improperly concludes that, because Gupta system
creates a data package, the data packet arranges data “such that an
application specific interface can decode the packet and a particular
application can display the information from the packet.” MTA Opp. 14.
We agree that to the extent that Gupta teaches a packetized data, it is logical
to assume that Gupta further teaches an application that can decode the
packet. See Ex. 1072 ¶ 71.
Petitioner further argues that
a person of ordinary skill in the art would have found it obvious
to configure the device of Kosuda to add the functionality of
impact detection (a motion-related parameter) and transmission
to remote devices, in the same way as Gupta (i.e., by configuring
Kosuda’s processor to process signals from an accelerometer
using Gupta’s impact detection algorithm, to output a serial data
string of physiological information and motion-related
information parsed out such that an specific application interface
is able to utilize the physiological information and motion-
related information for a particular application).
Id. at 15 (citing Ex. 1072 ¶ 72). Specifically, Petitioner asserts that
combining Gupta’s known processing techniques with Kosuda’s known
device, would have involved nothing more than using known techniques to
improve a known device. Id. Thus, Petitioner argues that a person of
ordinary skill in the art would have recognized that Gupta’s processor
outputs physiological information and motion-related information parsed out
in a specific data format, such that a specific application interface is able to
utilize the physiological information and motion-related information for an
application. Ex. 1072 ¶ 71.

67
IPR2017-00321
Patent 8,923,941 B2

In order to demonstrate obviousness, Petitioner must show that a

person of ordinary skill in the art would have had reason to combine the
teachings of Kosuda, Maekawa, and Gupta to achieve the device recited in
the substitute claim. It is not sufficient that the techniques of Kosuda and
those of Gupta were known. This is no more than an assertion that they
could have been combined in the manner proposed by Petitioner.
Petitioner’s declarant, Dr. Anthony, provides no more reason to have
combined the teachings of Kosuda and Gupta than that argued in the
Opposition to the Motion to Amend. Ex. 1072 ¶ 72; see id. ¶ 90 (In
combining the teachings of Kosuda, Gupta, Maekawa, and Han,
Dr. Anthony relies on the reasons discussed above to combine the teachings
of Kosuda and Han.). Despite the fact that Kosuda and Gupta relate to wrist
worn monitoring devices, we find Dr. Anthony’s testimony conclusory and
insufficient to demonstrate a reason to combine the teachings of the Kosuda
and Gupta in the manner proposed. See 37 C.F.R. § 42.65(a).
We are not persuaded that Petitioner has shown a reason to combine
the teachings of these references, other than hindsight improperly gleaned
from substitute claim 22. Consequently, because Petitioner’s challenge to
substitute, independent claim 22 is deficient, we also are not persuaded that
Petitioner prevails in its challenges to claims 23 and 26–29.

4. Obviousness Over Aceti, Fricke, and Craw, Alone or in

Combination with Comtois
Petitioner argues that claim 22–29 are unpatentable under 35 U.S.C.
§ 103(a) as obvious over Aceti, Fricke, and Craw, alone or in combination
with Comtois. MTA Opp. 16–25. To support its argument, Petitioner
provides a mapping of the limitations of claims 22 and 26 to structures

68
IPR2017-00321
Patent 8,923,941 B2

taught or suggested by Aceti, Fricke, and Craw and by Aceti, Fricke, Craw,
and Comtois, respectively. Id. Petitioner also cites Dr. Sarrafzadeh’s
Declaration for support. See Ex. 1072 ¶¶ 91–151. Patent Owner limits its
response to these arguments to the challenges to independent claim 22.
MTA Reply 8–10.
We begin our analysis of these grounds of unpatentability with a
review of Craw.

a. Craw (Ex. 2127)

Craw teaches methods for communicating medical information

between network devices. Ex. 2127 ¶ 2; see id. ¶ 13, Fig. 9A (displaying
physiological information based definitions derived from data dictionaries).
The health care computing environment includes a variety of medical
monitoring and analysis devices that process physiological data, including
heart rate and respiration rate, and communicate that physiological data via a
network. Id. ¶ 4. For example, Craw teaches “a system for interoperability
of medical devices on a network and particularly measurements of non-
invasive blood pressure (‘NIBP’), but it is understood that this example is
merely illustrative and other uses and fields of use are contemplated.” Id.
¶ 51.
Craw teaches serializing data for transmission using a classification
scheme to enable extraction of physiological parameters by a recipient
device, such as for displaying information. See id. ¶¶ 200–216.
“Serialization may be thought of as a process for taking an instance of
software structure or class and turning the attributes/members of the class
into transferable data encoding.” Id. ¶ 235. Craw further teaches the use of
string tables to provide an interface that may be used by software to manage

69
IPR2017-00321
Patent 8,923,941 B2

and access strings of data. Id. ¶ 255. A data dictionary may be used with a
string table as an interface for managing, extracting, and displaying
information from binary information streams. Id. ¶ 256, Fig. 7G. Thus,
Craw’s system may include a protocol for the serializing and deserializing
byte streams of information. Id. ¶ 15. Accordingly, physiological
information may be communicated via known serial communications
channels. Id. ¶¶ 68–70.

b. Claim 22

We determined above, that the combined teachings of Aceti and

Fricke teach or suggest all of the limitations of original claim 14. See supra
Section II.C.5. Claim 22 includes all of the limitations of claim 14 and adds
certain additional limitations.
Substitute claim 22 recites that “at least one signal processor [is]
configured to process signals from the at least one motion sensor and signals
from the at least one PPG sensor to reduce motion artifacts from the PPG
signals and to extract physiological and motion parameters.” MTA 23
(App’x A). Petitioner argues that Aceti discloses pulse oximetry circuitry
336, signal processing circuitry 342, and processor 314. MTA Opp. 19–20
(citing Ex. 1031 ¶ 27, Fig. 3). Aceti further teaches that pulse oximetry
circuitry 336 monitors PPG signals to determine physiological parameters,
i.e., pulse rate and/or blood oxygenation levels. Id. Aceti explains that
signal processing circuitry 342 may process signals from accelerometer 306
to extract a variety of different types of motion-related parameters, including
ovulation monitoring, fall detection, inclination and activity, wakefulness,
patient movement, sleep apnea, and the like. Id. (citing Ex. 1031 ¶¶ 53–55;
Ex. 1072 ¶ 114).

70
IPR2017-00321
Patent 8,923,941 B2

Substitute claim 22 also recites that “the at least one signal processor
configured to process data to be output, wherein the output data comprises
physiological information and motion related information.” MTA 23
(App’x A). Petitioner argues that Aceti teaches this added limitation, as
well. MTA Opp. 20. In particular, Aceti teaches processing elements within
processor portion 102 to determine physiological information, such as heart
rate, and motion related information, such as physical activity. Ex. 1031
¶¶ 18, 27, 29, 46–49, Fig. 9. This determined information may be output via
output circuitry 324. Id. ¶¶ 21–22, 46–47. Thus, Petitioner asserts that
Aceti teaches that signal data is processed by at least one signal processor to
generate output data including both physiological information and motion-
related information. MTA Opp. 20 (citing Ex. 1072 ¶ 119).
Substitute claim 22 further recites that “the output data is parsed out
such that an application-specific interface (API) can utilize the physiological
information and motion-related information for an application.” MTA 23
(App’x A). Petitioner argues that Aceti teaches that “one or more remote
devices 800 (FIG. 8) may be deployed as one or more nodes (e.g., rooms)
within a facility (e.g., home, hospital, care facility)” (Ex. 1031 ¶ 49) and that
“[e]ach node 800 within the facility can receive, from the monitoring device
100, emergency alerts, physiological characteristics and/or physiological
parameters for processing and/or routing to a central processing device 804”
(id. ¶¶ 46, 49, 53). MTA Opp. 20–21. Petitioner further argues that Aceti
teaches that numerous applications can use these physiological parameters,
including physiological information and motion-related information. Id. at
21 (citing Ex. 1072 ¶ 120).

71
IPR2017-00321
Patent 8,923,941 B2

Petitioner acknowledges that, although Aceti teaches that

processor 314 is configured to communicate data, including physiological
information and motion-related information, via the data input/output
circuitry 322/324 and transceiver 320, Aceti does not teach or suggest that
“the output data is parsed out such that an application-specific interface
(API) can utilize the physiological information and motion-related
information for an application.” Id. at 21. Nevertheless, Petitioner
maintains that “Craw’s data structure and classification scheme is directed to
the problem of inoperability and seamless transmission of physiological data
between varied computing environments.” Id. (citing Ex. 2127 ¶ 123). In
particular, Craw teaches use of a serial data format, e.g., “string data,” to
enable extraction of multiple physiological parameters and motion-related
information for display of the extracted information. Ex. 2127 ¶¶ 200–216,
Fig. 7H; see Ex. 1072 ¶ 123.
Petitioner argues that a person of ordinary skill in the art would have
had reason to combine the teachings of Aceti and Craw because the logical
scheme of serializing data output, as taught by Craw, would allow for
transmission of a variety of parameters, such as heart rate, respiration rate,
activity state, activity strength, and activity duration, and subsequent
extraction, i.e., deserialization, of those parameters from the serial data by an
application specific interface of an external receiving device for further
processing and display. MTA Opp. 21 (citing Ex. 1072 ¶¶ 123–125).
Referring to Aceti’s Figure 8, we note that Aceti teaches a system including
monitoring device 100 and a plurality of remote devices 800 and that:
The monitoring device 100 may attach an identification code to
each communication with the remote devices 800 so that a
particular monitoring device 100 is distinguishable from other

72
IPR2017-00321
Patent 8,923,941 B2

monitoring devices. In addition, each remote device 800 may

attach a unique monitoring code to communications
communicated from the monitoring device 100 through the
remote devices 800 to a central processing device 804 in order
to provide an indication of the remote device 800 through which
the monitored information was received.
Ex. 1031 ¶ 40 (emphasis added). Thus, Aceti teaches or suggests
transmission of information in data packets. Consequently, Petitioner argues
that modifying Aceti’s processing techniques with a data formatting
techniques taught be Craw would have amounted to use of a known
processing technique to improve a similar device in the same way. Id. at
22–23 (citing Ex. 1072 ¶¶ 120–126). We agree.
Patent Owner contends that Petitioner has failed to demonstrate that
Craw teaches parsing output data, such that an application-specific interface
(API) can utilize the physiological information and motion-related
information for an application. MTA Reply 9. Specifically, Patent Owner
contends that Craw’s use of serial data packets does not teach parsing out
data for use of an application-specific interface (API). Id. We disagree.
Given our understanding of the word “parsing,” Craw’s use of serial
data packets teaches or suggests the recited parsing. Moreover, Craw relates
to the communication of information to various medical devices (Ex. 2127,
[57], ¶ 2) and Craw explains that:
The design of each medical device, or any other machine
performing health assessment, is dependent upon the particular
subset or subsets of physiological data that the medical device or
other machine processes and communicates. The design of the
software residing on the medical devices is also dependent upon
the subset or subsets of physiological data or clinical outcomes
that the medical device processes and communicates.

73
IPR2017-00321
Patent 8,923,941 B2

Id. ¶ 5 (emphasis added). Thus, we are persuaded that Craw teaches or

suggests that “the output data is parsed out such that an application-specific
interface (API) can utilize the physiological information and motion-related
information for an application,” as recited in claim 22.
We are persuaded that Petitioner has shown by a preponderance of the
evidence that substitute claim 22 is rendered obvious by the combined
teachings of Aceti, Fricke, and Craw. Each of substitute claims 23–25, 28,
and 29 depends directly from independent claim 22 and their recited
limitations correspond to challenged claims 15–17, 20, and 21. Because we
determined above that the limitations of those challenged claims were taught
or suggested by the combined teachings of Aceti and Fricke, alone or in
combination with those of Comtois, we also are persuaded that substitute
claims 23–25, 28, and 29 are rendered obvious over the combined teachings
of Aceti, Fricke, and Craw, alone or in combination with those of Comtois.
See supra Section II.C.5.

c. Claim 26

Substitute claim 26 recites that, in the device of substitute claim 22,

the at least one processor is configured to (i) reduce motion
artifacts by removing frequency bands from the signals that are
outside of a range of interest using at least one band-pass filter to
produce preconditioned signals and (ii) to generate the parsed
output data by executing one or more processing methods to
provide information that is fed into a multiplexed output serial
data string of motion-related and physiological information.
MTA 24 (App’x A). Petitioner argues that the combined teachings of
Aceti’s monitoring device, Fricke’s motion artifact reduction methods, and
Craw’s data formatting scheme teach or suggest the limitation added to this
substitute claim. MTA Opp. 23. Moreover, Petitioner argues that a person

74
IPR2017-00321
Patent 8,923,941 B2

of ordinary skill in the art, in view of Aceti’s teaching regarding the

transmission to multiple remote devices 800 and the extraction of different
parameters from the transmitted data (Ex. 1031 ¶ 140), would have had
reason to combine Craw’s teachings regarding serial data formatting “such
that Aceti’s pulse rate, blood-oxygen content, etc., could be extracted from
the physiological information, and such that subject physical activity
parameters could be extracted from the motion-related information.” See
Ex. 1072 ¶¶ 140–144. Patent Owner does not contest this combination or its
alleged teachings. MTA Reply 10.
We are persuaded that Petitioner has shown by a preponderance of the
evidence that substitute claim 26 is rendered obvious by the combined
teachings of Aceti, Fricke, and Craw. Substitute claim 27 depends directly
from claim 26 and its recited limitations correspond to challenged claim 19.
Because we determined above that the limitations of this challenged claim
were taught or suggested by the combined teachings of Aceti and Fricke, we
also are persuaded that substitute claim 27 is rendered obvious over the
combined teachings of Aceti, Fricke, and Craw.
5. Summary
In consideration of all the evidence discussed above, we deny Patent
Owner’s Motion to Amend.

IV. PROCEDURAL ISSUES

Patent Owner objects to the use of inter partes reviews as

unconstitutional based, at least, upon the reasons presented in the petition for
certiorari that was granted in Oil States Energy Services, LLC v. Greene’s
Energy Group, LLC. PO Resp. 33–34; see Reply 12. On April 24, 2018, the
U.S. Supreme Court upheld the constitutionality of inter partes review; thus,

75
IPR2017-00321
Patent 8,923,941 B2

Patent Owner’s arguments are moot. Oil States Energy Servcs. LLC v.
Greene’s Energy Grp., LLC, 138 S.Ct. 1365, 1370 (2018).

V. CONCLUSION

For the foregoing reasons and on this record, we are persuaded that
Petitioner establishes by a preponderance of the evidence that claims 14, 15,
and 18–21 of the ’941 patent are unpatentable as obvious over Kosuda and
Maekawa, alone or in combination with Han, and that claims 14–21 of the
’941 patent over Aceti and Fricke, alone or in combination with Comtois.
Further, for the foregoing reasons and on this record, we are not persuaded
that Patent Owner is entitled to grant of its Motion to Amend with respect to
substitute claims 22–29.

VI. ORDER

In consideration of the foregoing, it is

ORDERED that claims 14–21 of the ’941 patent are unpatentable;
FURTHER ORDERED that Patent Owner’s Motion to Amend is
denied; and
FURTHER ORDERED, that because this is a final written decision,
parties to the proceeding seeking judicial review of the decision must
comply with the notice and service requirements of 37 C.F.R. § 90.2.

76
IPR2017-00321
Patent 8,923,941 B2

For PETITIONER

Michelle K. Holoubek
Michael D. Specht
Richard M. Bemben
STERNE, KESSLER, GOLDSTEIN & FOX
[email protected]
[email protected]
[email protected]
[email protected]

Harper Batts
Jeremy Taylor
BAKER BOTTS LLP
[email protected]
[email protected]

For PATENT OWNER

Justin B. Kimble
Nicholas C Kliewer
BRAGALONE CONROY PC
[email protected]
[email protected]

77