Trials@uspto.gov
`571-272-7822
`
`Paper 50
`Date: October 28, 2020
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`APPLE INC.,
`Petitioner,
`v.
`ANDREA ELECTRONICS CORP.,
`Patent Owner.
`
`IPR2017-00626
`Patent 6,363,345 B1
`
`
`
`
`
`
`
`
`
`Before MICHAEL R. ZECHER, JEREMY M. PLENZLER, and
`MIRIAM L. QUINN, Administrative Patent Judges.
`PLENZLER, Administrative Patent Judge.
`
`
`
`JUDGMENT
`Final Written Decision on Remand
`Determining Challenged Claims 6–9 Are Unpatentable
`35 U.S.C. §§ 144, 318(a)
`
`INTRODUCTION
`I.
`A. Background and Summary
`Apple Inc. (“Petitioner”) requested inter partes review of claims 1–25
`and 38–47 of U.S. Patent No. 6,363,345 B1 (Ex. 1001, “the ’345 patent”).
`
`
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`Paper 1 (“Pet.”). We issued a Decision instituting inter partes review.
`Paper 7 (“Inst. Dec.”).
`After institution of trial, Andrea Electronics Corp. (“Patent Owner”)
`filed a Patent Owner Response (Paper 11, “PO Resp.”), to which Petitioner
`filed a Reply (Paper 18, “Pet. Reply”). An oral argument was held on April
`25, 2018. A transcript of the oral argument is included in the record. Paper
`25.
`
`Our Final Written Decision was issued on July 12, 2018. Paper 28
`(“Original Decision”). The Original Decision determined that Petitioner had
`established unpatentability of claims 1–3, 12–25, 38, and 47 of the ’345
`patent, but had not established unpatentability of claims 4–11 and 39–46 of
`the ’345 patent. Original Decision 24.
`On appeal, the U.S. Court of Appeals for the Federal Circuit vacated
`our Original Decision only as to claims 6–9 of the ’345 patent and remanded
`the case for further proceedings. Apple Inc. v. Andrea Elecs. Corp., 949
`F.3d 697 (Fed. Cir. 2020). After conferring with the parties, we permitted
`additional briefing addressing the issues on remand from the Federal Circuit.
`Paper 36. Petitioner and Patent Owner simultaneously filed opening briefs
`(Paper 41 (“Pet. Remand Br.”); Paper 42 (“PO Remand Br.”)), followed by
`simultaneously filed Reply Briefs (Paper 44 (“Pet. Remand Reply”); Paper
`43 (“PO Remand Reply”)). After further conferring with the parties, we
`authorized an additional round of briefing (Paper 45), which was filed to
`address claim 9 of the ’345 patent. Paper 46 (“Pet. Remand Sur-Reply”);
`Paper 49 (“PO Remand Sur-Reply”).
`This is a Final Written Decision on Remand only as to the
`patentability of challenged claims 6–9. For the reasons discussed below, we
`
`2
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`determine that Petitioner has shown by a preponderance of the evidence that
`these challenged claims are unpatentable.
`B. Related Matters
`Petitioner and Patent Owner identify a number of proceedings, both in
`district court and before the Patent Trial and Appeal Board, involving
`patents related to the ’345 patent, including a district court proceeding
`specifically directed to the ’345 patent with Petitioner as a party, and
`IPR2017-00627, which also is directed to the ’345 patent and involves the
`same parties as this proceeding. Pet. viii–x; Paper 4, 1. Our Final Written
`Decision in IPR2017-00627 was affirmed by the Federal Circuit. Apple, 949
`F.3d at 710.
`
`C. The ’345 Patent
`The ’345 patent “relates to noise cancellation and reduction and, more
`specifically, to noise cancellation and reduction using spectral subtraction.”
`Ex. 1001, 1:19–21. The ’345 patent explains that its system receives a noise
`signal and converts that signal to the frequency domain through a Fast
`Fourier Transform (FFT). Id. at 4:50–5:14. Separate thresholds are set for
`each frequency bin to determine the location of noise elements for each
`frequency bin separately. Id. at 6:10–13. The ’345 patent determines the
`thresholds by setting two minimum values, which are described as a future
`minimum and a current minimum. Id. at 6:23–41.
`At predetermined time intervals (e.g., every 5 seconds), the future
`minimum value is initialized as the value of the current magnitude of the
`signal. Id. at 6:24–28. Over that time interval, and before the next
`initialization, the future minimum value of each bin is compared with the
`current magnitude value of the signal. Id. If the current magnitude is
`
`3
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`smaller than the future minimum, the value of the future minimum is
`replaced with that current magnitude. Id. at 6:28–32.
`At the start of each time interval, the current minimum is set as the
`value of the future minimum that was determined over the previous time
`interval. Id. at 6:34–38. The current minimum then follows the minimum
`value of the signal for the next time interval by comparing its value with the
`current magnitude value. Id. The current minimum value is used by the
`spectral subtraction process to remove noise from the signal. Id. at 6:38–41.
`D. Illustrative Claims1
`Claims 1, 4, 5, 6, and 9 of the ’345 patent are illustrative for this
`Decision on Remand and are reproduced below:2
`1. An apparatus for canceling noise, comprising:
`an input for inputting an audio signal which includes a
`noise signal;
`a frequency spectrum generator for generating the
`frequency spectrum of said audio signal thereby generating
`frequency bins of said audio signal; and
`a threshold detector for setting a threshold for each
`frequency bin using a noise estimation process and for detecting
`for each frequency bin whether the magnitude of the frequency
`
`1 The broadest reasonable interpretation was applied in construing claim
`terms of the ’345 patent in the Original Decision. Inst. Dec. 5–6. The ’345
`patent is now expired. The Federal Circuit made clear that, “[w]hen th[e]
`court reviews the claim construction of a patent claim term in an IPR appeal
`after the patent has expired, such as in this case, we apply the standard
`established in Phillips, not the ‘broadest reasonable interpretation.’” Apple,
`949 F.3d at 707 (citations omitted). Neither Patent Owner nor Petitioner
`allege in any way that the claim construction standard applied would affect
`the outcome in this remand decision. Any difference in claim construction
`standard does not affect the outcome of this proceeding.
`2 We include claims 1, 4, and 5 because claims 6 and 9 ultimately depend
`from claim 5, which depends from claim 4, which depends from claim 1.
`
`4
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`bin is less than the corresponding threshold, thereby detecting the
`position of noise elements for each frequency bin.
`Ex. 1001, 9:35–46.
`4.
`The apparatus according to claim 1, wherein said threshold
`detector sets the threshold for each frequency bin in accordance
`with a current minimum value of the magnitude of the
`corresponding frequency bin; said current minimum value being
`derived in accordance with a future minimum value of the
`magnitude of the corresponding frequency bin.
`Id. at 9:54–60.
`5.
`The apparatus according to claim 4, wherein said future
`minimum value is determined as the minimum value of the magnitude
`of the corresponding frequency bin within a predetermined period of
`time.
`Id. at 9:61–64.
`6.
`The apparatus according to claim 5, wherein said current
`minimum value is set to said future minimum value periodically.
`Id. at 9:65–66.
`9.
`The apparatus according to claim 5, wherein said future
`minimum value is set to a current magnitude value periodically; said
`current-magnitude value being the value of the magnitude of the
`corresponding frequency bin.
`Id. at 10:9–12.
`E. Prior Art and Asserted Grounds Relevant to Remand
`Petitioner asserts that claims 6–9 would have been unpatentable on the
`following ground:
`Claim(s) Challenged
`6–9
`
`Reference(s)/Basis
`Hirsch3, Martin4
`
`35 U.S.C. §
`103
`
`
`3 H.G. Hirsch & C. Ehrlicher, “Noise Estimation Techniques for Robust
`Speech Recognition,” IEEE 1995 (Ex. 1005, “Hirsch”).
`4 Ranier Martin, “An Efficient Algorithm to Estimate the Instantaneous SNR
`of Speech Signals,” Eurospeech 1993 (Ex. 1006, “Martin”).
`
`5
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`
`F. CAFC Remand
`On appeal, the Federal Circuit held that our Original Decision erred
`by failing to consider Petitioner’s Reply Brief arguments directed to claims
`6–9. Apple, 949 F.3d at 699. Patent Owner did not appeal our
`determination from IPR2017-00626 that claims 1–3, 12–25, 38, and 47 of
`the ’345 patent are unpatentable or our determination from IPR2017-00627
`that claims 1–5, 10–25, and 38–47 of the ’345 patent are unpatentable.
`Those determinations are now final. Accordingly, claims 6–9 are the only
`claims remaining on remand from the originally challenged claims.5
`
`II. ANALYSIS
`A. Hirsch
`There is no dispute as to any particular finding regarding Hirsch in
`this remand proceeding. Our Original Decision determined that Hirsch
`discloses each limitation in claim 1. Original Decision 6–7, 24. As noted
`above, Patent Owner did not appeal that determination.
`B. Martin
`Petitioner cites Martin as teaching the additional limitations recited in
`claims 4–9, which relate to setting the threshold for the frequency bins. Pet.
`38–46. The Original Decision did not reach whether Martin teaches the
`features recited in claims 4–9 because we determined that one skilled in the
`
`
`5 We address the challenge to claims 4 and 5 in this Decision on Remand
`because each of claims 6–9 ultimately depends from claim 5, which depends
`from claim 4, and the Original Decision did not reach whether Martin
`teaches the features recited in claims 4 and 5. The parties did not raise any
`collateral estoppel arguments with respect to the IPR2017-00627 decision
`holding that Helf (U.S. Patent No. 5,550,924, iss. Aug. 27, 1996) renders
`claims 4 and 5 unpatentable.
`
`6
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`art would not have combined a scenario with a single subwindow in Martin
`with Hirsch’s teachings. Original Decision 14. In view of the Federal
`Circuit’s remand instructions, we address Petitioner’s contentions regarding
`Martin’s teachings below.
`Martin is directed to “estimat[ing] the instantaneous signal-to-noise
`ratio of speech signals” (Ex. 1006, 1093), and is specifically referenced in
`Hirsch as a known way to estimate noise characteristics in a past speech
`segment (Ex. 1005, 153 n.6). Martin explains that its “noise power estimate
`is based on the minimum of signal power within a window of L samples,”
`and “[f]or reasons of computational complexity and delay the data window
`of length L is decomposed into W windows of length M such that M * W =
`L.” Ex. 1006, 1094.
`Martin distinguishes between two scenarios, which involve slowly
`varying noise power and rapidly varying noise power. Ex. 1006, 1094.
`When the minimum power within each window (W) of samples (M) are
`monotonically increasing (i.e., monotonic), a rapid noise power variation is
`determined, and the noise power estimate Pn(i) is set to the minimum noise
`power of the last window (W) from samples (M). Id. When the minimum
`power within each window (W) of samples (M) are not monotonically
`increasing (i.e., non-monotonic), a slow noise power variation is determined,
`and the noise power estimate Pn(i) is set to the minimum noise power of the
`windows W occurring during the non-monotonic case. Id. That is, whether
`monotonic or non-monotonic, at the end of each series of M samples (i.e., at
`the end of a subwindow (W)), Martin sets its noise power estimate Pn(i) to a
`minimum noise power value. See id. Fig. 2.
`Martin provides an example where M = 1250, W = 4, and L = 5000.
`Ex. 1006, 1094. In that example, when the minimum noise power is
`
`7
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`monotonically increasing over the 4 subwindows (the monotonically
`increasing case), the noise power estimate Pn(i) is set to the minimum noise
`power in the fourth window (i.e., the minimum of the 1250 samples in the
`last window). Id. When not monotonically increasing from subwindow to
`subwindow (the non-monotonic case), the noise power estimate Pn(i) is set
`to the minimum noise power for the total data window (i.e., the minimum of
`the 5,000 samples in the total data window in this example). Id.
`1. Claims 4 and 5
`Petitioner contends that Martin’s noise power estimate Pn(i)
`corresponds to the “current minimum value” recited in claim 4 and Martin’s
`minimum observed noise power PMmin corresponds to the “future minimum
`value” recited in claim 4.6 Pet. 41–42. Patent Owner disputes Petitioner’s
`contentions regarding whether Martin teaches the “current minimum value”
`and the “future minimum value” recited in claim 4. PO Resp. 23–28.
`For the monotonically increasing case, Patent Owner contends that
`PMmin cannot be the “future minimum value” because it is not the minimum
`of all samples in the window L. PO Resp. 25. Patent Owner provides a
`
`
`6 As explained in our Original Decision, and not disputed by Patent Owner,
`Petitioner contends that, “[w]hile Martin shows its algorithm operating on
`signal power, it teaches that the algorithm can be adapted to work with
`spectral subtraction techniques that operate on the magnitude of a signal.”
`Pet. 40 (citing Ex. 1006, 1096). Petitioner further contends that “a person of
`ordinary skill in the art would have understood that Martin’s algorithm also
`could be used to track the magnitude of the noise signal instead of its
`power.” Id. at 34 (emphases omitted) (citing Ex. 1003 ¶ 123).
`Dr. Hochwald testifies that “[t]he power of an audio signal is its magnitude
`squared” and “a person of ordinary skill can readily move between the two.”
`Ex. 1003 ¶ 123.
`
`8
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`diagram illustrating the monotonically increasing case, which is reproduced
`below.
`
`
`Id. at 24. The figure reproduced above includes a portion of Martin’s Figure
`2, which illustrates its algorithm along with a diagram of a 20 sample (L =
`20) example with 4 subwindows (W = 4). Id. Consistent with our
`description of Martin’s teachings above, Patent Owner is correct that “for a
`monotonically increasing signal, the last subwindow will always contain the
`largest subwindow minimum value,” and “[t]his most recent (and largest)
`subwindow minimum value, PMmin, (in the example above, 16) . . . is not the
`minimum of the samples in the window L.” Id. at 25. Patent Owner’s
`contention disputing Petitioner’s challenge is based on the claim requiring
`the “future minimum” to be the minimum of the entire window. Id.
`(“[C]laim 4 recites the ‘future minimum’ as a noise floor tracker. See, e.g.,
`Ex. 1001 at 6:28–32 (describing continuously adjusting the future minimum
`
`9
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`value downward based on smaller noise magnitudes)”); see also PO Remand
`Br. 6–8.
`Petitioner responds that Patent Owner’s “argument rests on a
`limitation found nowhere in the claims.”7 Pet. Reply 9. We agree. As
`Petitioner explains, “[n]othing in claim[] 4 . . . specifies the period over
`which the ‘future minimum’ must be calculated,” and claim 5 is “the only
`claim[] that specif[ies] anything about the period over which the ‘future
`minimum’ must be calculated, and [it] specif[ies] simply that the ‘future
`minimum’ is calculated ‘over a predetermined period of time’; [it is] silent as
`to how long that period must be.” Id.
`Claim 4 specifies that “said current minimum value [is] derived in
`accordance with a future minimum value of the magnitude of the
`corresponding frequency bin,” and claim 5 further specifies that “said future
`minimum value is determined as the minimum value of the magnitude of the
`corresponding frequency bin within a predetermined period of time.”
`Ex. 1001, 9:54–64. As Petitioner correctly explains, Martin’s PMmin is a
`noise floor tracker. Pet. Remand Reply 4 (citing Ex. 1006, 1093 (Martin’s
`algorithm “estimate[s] the noise floor… [as] the minimum of a smoothed
`power estimate within a window of finite length.”)). Petitioner additionally
`notes Martin’s predetermined period of time for its windows. See Pet. 33
`(“Martin’s algorithm tracks the noise floor of an audio signal over a
`predetermined period (e.g., 0.625 seconds).”); Reply Br. 4 (“Martin teaches
`a noise estimation algorithm that tracks the noise floor of an audio signal
`
`
`7 Patent Owner’s argument is not based on any dispute regarding claim
`construction.
`
`10
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`over windows of L digital samples, and each window can correspond to a
`period of, e.g., 0.625 seconds.”).
`There is no dispute that Martin’s windows have a duration that is “a
`predetermined period of time” or that Martin’s monotonically increasing
`case sets PMmin as the minimum of the last subwindow. Martin does this
`after each series of M samples (at the end of each subwindow) during the
`monotonically increasing case. That is, Martin’s monotonically increasing
`case sets PMmin as “the minimum value of the magnitude of the
`corresponding frequency bin within a predetermined period of time” (the
`minimum value of the fourth subwindow in the example provided by Patent
`Owner). Accordingly, we are persuaded that Martin’s monotonically
`increasing case teaches the features of claims 4 and 5.
`Patent Owner additionally contends that in the non-monotonic case,
`Martin’s PMmin cannot be the “future minimum value” because “the
`parameter PMmin is used to find the minimum of the first subwindow, and this
`minimum is then stored in the min_vec array” and “PMmin . . . is then reset
`. . . and used to find the minimum of the second subwindow.” PO Resp. 26
`(citing Ex. 1006, 1094). Patent Owner explains that “[t]his process repeats
`itself until all subwindows have been analyzed.” Id.
`There is no dispute that Martin determines a “PMmin” for each
`subwindow or that Martin uses the minimum across windows to set the noise
`power estimate Pn(i) in the non-monotonic case. As Petitioner explains,
`“min_vec stores the PMmin values and that Martin uses min_vec to determine
`which PMmin value in the data window L is the smallest.” Pet. Reply 10
`(citing Ex. 1006, 1094). Patent Owner contends that “PMmin merely
`determines the minimum power of the current sub-window and does not
`purport to represent the minimum magnitude of the frequency bin,” and “the
`
`11
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`minimum power is determined by finding the minimum value of the min_vec
`array, which is stored as Pn(i) – which [Petitioner] insists is the ‘current
`minimum,’ not the ‘future minimum.’” PO Remand Br. 8. Petitioner
`responds that “min_vec is used to store PMmin values, and even [Patent
`Owner’s Declarant] Dr. Douglas admitted that use of min_vec results in
`setting Pn(i) to the smallest PMmin value in a window.” Pet. Remand Reply 4
`(citing Pet. Remand Br. 6–7; Pet. Reply 10–11).
`Again, Petitioner has the better position, as confirmed by Dr. Douglas.
`As noted by Petitioner (Reply Br. 11), Dr. Douglas testifies that the min_vec
`values are the PMmin values for each subwindow and the minimum of those
`PMmin values (i.e., the minimum min_vec) is then used to set the noise power
`estimate Pn(i) (Ex. 1030 177:5–16). This is consistent with Martin’s
`disclosure. See Ex. 1006, Fig. 2. That is, Martin’s non-monotonic case
`teaches the recited “future minimum value” because it determines the
`minimum PMmin after every window (W) of samples (M), and uses the
`minimum value over the duration of the non-monotonic case to set the new
`noise power estimate Pn(i).
`For the reasons explained above, Petitioner has established by a
`preponderance of the evidence that Martin’s monotonic and non-monotonic
`cases each teach the recited “future minimum” and “current minimum.”
`2. Claim 6
`Claim 6 additionally requires that “said current minimum value is set
`to said future minimum value periodically.”8 Ex. 1001, 9:65–67. Petitioner
`contends that “Martin updates the value Pn(i) after every M samples, as Dr.
`
`8 Consistent with our decision in IPR2017-00627, the Federal Circuit
`determined that “periodically” means “at regular intervals of time.” Apple,
`949 F.3d at 709.
`
`12
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`Douglas admitted.” Pet. Reply 12 (citing Ex. 1026, 77:16–80:19).
`Petitioner contends that “whether the signal is monotonically increasing or
`not, Pn(i) is set equal to PMmin at the end of every subwindow of M samples,
`as shown in the blue box of the figure below.” Id. (citing Ex. 1023 ¶ 20).
`The Figure referenced by Petitioner is reproduced below.
`
`
`Id. at 13. The Figure reproduced above is a portion of Martin’s Figure 2,
`which depicts its algorithm, and additionally includes Petitioner’s
`annotations in the form of colored boxes. Id.
`Patent Owner responds that, “[w]hile Martin does set Pn(i) to PMmin in
`the rapidly varying monotonically increasing noise case, it does not do so
`periodically,” and, “[i]nstead, it sets Pn(i) to PMmin randomly.” PO Remand
`Br. 9 (citing (Ex. 1006, 1094; Ex. 2002 ¶¶ 72–75; PO Resp. 28–30)). Patent
`Owner further contends that, “even if the values in the min_vec array were
`considered PMmin values, the non-monotonically increasing case also occurs
`randomly.” Id.
`As explained above, we agree with Petitioner that Martin’s values in
`its min_vec array are properly considered PMmin values. Regardless of
`
`13
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`whether it is a monotonic or non-monotonic case, Martin sets Pn(i) to a PMmin
`value periodically. Petitioner’s declarant Dr. Hochwald provides testimony
`supporting this understanding. Dr, Hochwald explains that “[i]f
`monotonically increasing is determined in the decision diamond, Pn(i) is set
`to the most recent PMmin,” and “[i]f not monotonically increasing, Pn(i) is set
`to the smallest of the W past PMmin values.” Ex. 1023 ¶ 20. Patent Owner’s
`arguments to the contrary are not persuasive. See PO Remand Br. 9
`(asserting that Martin sets Pn(i) to PMmin randomly); PO Remand Reply 5
`(asserting same).
`Although Patent Owner provides testimony from Dr. Douglas to
`support its contention, we do not find this testimony persuasive and credit
`Dr. Hochwald’s testimony over that of Dr. Douglas in this regard, because
`Dr. Hochwald’s testimony is consistent with and supported by the disclosure
`of Martin depicted in the figure above. Dr. Douglas testifies, for example,
`that “[t]he setting of Pn(i) to PMmin does not occur ‘periodically,’ but instead
`it occurs (if ever) based on the characteristics of the given signal,” and “[a]
`skilled artisan would understand that a signal’s characteristics are
`unpredictable.” Ex. 2002 ¶ 72 (citing Ex. 1006, 1094). Regardless of
`whether the signal is unpredictable, as seen above, Martin sets Pn(i) to PMmin
`at the end of each subwindow of M samples. That number of samples (M) is
`fixed, meaning that Martin sets Pn(i) to the PMmin value at regular intervals
`(i.e., periodically). That value is the minimum of the collection of PMmin
`values in the non-monotonic case and the most recent PMmin in the monotonic
`case. Ex. 1006, 1094.
`For the reasons explained above, Petitioner has established by a
`preponderance of the evidence that Martin teaches that Pn(i) is set to PMmin
`
`14
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`periodically (i.e., the recited “said current minimum value is set to said
`future minimum value periodically”).
`3. Claims 7 and 8
`Claim 7 depends from claim 6 and recites that “said future minimum
`value is replaced with the current magnitude value when said future
`minimum value is greater than said current magnitude value.” Ex. 1001,
`10:1–4. Claim 8 depends from claim 6 and recites that “said current
`minimum value is replaced with the current magnitude value when said
`current minimum value is greater than said current magnitude value.” Id. at
`10:5–8
`There is no dispute that Martin teaches the additional features recited
`in claims 7 and 8. See Pet. 44–45; PO Resp. 23–28 (not addressing
`Petitioner’s contentions regarding claims 7 and 8)9. With respect to claim 7,
`citing Figure 2 of Martin, for example, Petitioner explains that Martin’s
`PMmin is set to the smoothed power estimate (i.e., the “current magnitude”)
`whenever the smoothed power estimate is less than PMmin. Pet. 44–45. With
`respect to claim 8, also citing Martin’s Figure 2, Petitioner explains that, if
`Martin’s estimated noise floor Pn(i) (the “current minimum value”) is greater
`than the observed smoothed power estimate (the “current magnitude value”),
`the estimated noise floor value is replaced with the smoothed power value
`(i.e., the “current minimum value is replaced with the current magnitude
`value”). Petitioner’s contentions are supported by Martin’s disclosure (see,
`e.g., Ex. 1006, Fig. 2), as well as the testimony from Dr. Hochwald (see,
`e.g., Ex. 1003 ¶¶ 135–36, 143).
`
`9 Patent Owner’s Remand Brief and Remand Sur-Reply also fail to dispute
`Petitioner’s contentions regarding claims 7 and 8. See generally PO
`Remand Br., PO Remand Sur-reply.
`
`15
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`Accordingly, based on the arguments and evidence before us,
`Petitioner has established by a preponderance of the evidence that Martin
`teaches the additional features recited in claims 7 and 8.
`4. Claim 9
`Claim 9 requires that “said future minimum value is set to a current
`magnitude value periodically; said current-magnitude value being the value
`of the magnitude of the corresponding frequency bin.” Ex. 1001, 10:9–12.
`Petitioner explains that Martin’s PMmin is set to the smoothed power estimate
`(i.e., the “current magnitude”) whenever the smoothed power estimate is less
`than PMmin. Pet. 46 (citing Ex. 1006, 1093–94, Fig. 2; Ex. 1003 ¶ 141).
`Petitioner explains that PMmin is reset by setting it equal to PMax, and then is
`set to the first smoothed power estimate value at the beginning of the next
`period. Pet. Remand Sur-Reply 3 (citing Pet. 41; Ex. 1003 ¶¶ 138–140).
`Patent Owner responds, alleging that Petitioner’s “argument is not
`supported by the evidence and is incorrect.” PO Remand Sur-Reply 3.
`Patent Owner additionally contends that we should disregard Petitioner’s
`contentions as new. See, e.g., id. at 1. As seen above, however, Petitioner’s
`contentions were presented in the Petition and supported by testimony from
`Dr. Hochwald.
`We credit Dr. Hochwald’s testimony, as it is consistent with the
`disclosure of Martin. Martin’s Figure 2 and the cited testimony from Dr.
`Hochwald, relying on Martin’s Figure 2, supports Petitioner’s contention.
`See Ex. 1003 ¶ 140. As seen above in the portion of Martin’s Figure 2
`reproduced in the discussion regarding claim 6, Martin resets PMmin by
`setting it equal to PMax. As Dr. Hochwald explains, when the smoothed
`power estimate (the “current magnitude”) is less than PMmin (the “future
`minimum”), PMmin is set equal to the smoothed power estimate. Id. ¶ 141.
`
`16
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`Dr. Hochwald provides an annotated portion of Martin’s Figure 2,
`reproduced below, supporting this understanding.
`
`
`Id. The Figure reproduced above is a portion of Martin’s Figure 2, which
`illustrates its algorithm, including Dr. Hochwald’s annotations highlighting
`setting PMmin to the smoothed power estimate. Id. Dr. Hochwald’s
`testimony is consistent with Martin’s disclosure, explaining that by resetting
`PMmin equal to PMax, the next cycle of Martin’s algorithm (in the red box
`above) sets PMmin to the smoothed power estimate. Id. ¶ 140.
`Based on the record before us, which includes unrebutted testimony
`from Dr. Hochwald, the preponderance of the evidence supports Petitioner’s
`contentions regarding the additional features recited in claim 9.
`C. Combination of Hirsch and Martin
`Patent Owner contends that “[t]he Board already rejected
`[Petitioner]’s attempt to combine Hirsch with Martin.” PO Remand Br. 1.
`Patent Owner is correct that we rejected a particular combination based on
`Martin and Hirsch, but that was based on our limited review of the Petition’s
`reliance on the scenario where Martin has a single subwindow (W = 1).
`Original Decision 14. In connection with the discussion of Petitioner’s
`
`17
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`challenge to claims 4–11, 39–42, and 46 of the ’345 patent (i.e., the features
`related to the recited algorithm), however, our Original Decision held that
`“one skilled in the art would have considered Martin’s teachings, generally,
`when reviewing the teachings of Hirsch, as Martin is specifically referenced
`in Hirsch itself.” Id. at 13. That is, one skilled in the art would have
`considered Martin’s noise floor estimates distinguishing between monotonic
`and non-monotonic cases in the context of Hirsch because Hirsch includes
`an explicit reference to these teachings in Martin. Patent Owner did not
`appeal that determination. The Federal Circuit also found that “Hirsch refers
`to Martin as a ‘known’ approach ‘to avoid the problem of speech pause
`detection and to estimate the noise characteristics just from a past segment
`of noisy speech.’” Apple, 949 F.3d at 703. Accordingly, whether one
`skilled in the art would have combined Martin’s teachings with those of
`Hirsch has already been resolved by the Federal Circuit.
`To the extent Patent Owner’s contentions regarding Petitioner’s
`rationale to combine the teachings of Martin with those of Hirsch are still at
`issue, we agree with Petitioner that one skilled in the art would have
`considered using the multiple subwindow approach taught by Martin in
`Hirsch’s system. “When a work is available in one field, design incentives
`and other market forces can prompt variations of it, either in the same field
`or in another.” KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 401 (2007).
`Based on the record before us, which includes an express suggestion in
`Hirsch to look to Martin’s teachings, Hirsch does not teach away from the
`proposed combination, and Petitioner has established by a preponderance of
`the evidence that one skilled in the art would have combined the teachings of
`Martin with those of Hirsch.
`
`
`18
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`
`III. CONCLUSION
`
`In summary:
`Claims
`35
`
`U.S.C. §
`
`Reference(s)/Basis
`
`103
`
`Hirsch, Martin
`
`6–9
`
`
`
`
`Claims
`Shown
`Unpatentable
`6–9
`
`Claims
`Not shown
`Unpatentable
`
`
`IV. ORDER
`In consideration of the foregoing, it is hereby:
`ORDERED that Petitioner has established by a preponderance of the
`evidence that claims 6–9 of the ’345 patent are unpatentable; and
`FURTHER ORDERED that, because this is a Final Written Decision
`on Remand, parties to this proceeding seeking judicial review of the
`Decision must comply with the notice and service requirements of 37 C.F.R.
`§ 90.2.
`
`
`
`19
`
`

`

`IPR2017-00626
`Patent 6,363,345 B1
`FOR PETITIONER:
`
`Jeffrey P. Kushan
`Steven S. Baik
`Thomas A. Broughan III
`Sidley Austin LLP
`jkushan@sidley.com
`sbaik@sidley.com
`tbroughan@sidley.com
`
`
`FOR PATENT OWNER:
`
`William D. Belanger
`Frank D. Liu
`Andrew P. Zappia
`Pepper Hamilton LLP
`belangerw@pepperlaw.com
`liuf@pepperlaw.com
`zappiaa@pepperlaw.com
`
`
`20
`
`