UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`CISCO SYSTEMS, INC. AND OCLARO, INC.
`Petitioners
`
`v.
`
`OYSTER OPTICS, LLC
`Patent Owner
`____________
`
`IPR2017-02189
`Patent 6,476,952
`____________
`
`PETITIONERS’ REPLY
`
`
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`CISCO SYSTEMS, INC. AND OCLARO, INC.
`Petitioners
`
`v.
`
`OYSTER OPTICS, LLC
`Patent Owner
`____________
`
`IPR2017-02189
`Patent 6,476,952
`____________
`
`PETITIONERS’ REPLY
`
`
`
`IPR2017-02189
`
`I.
`II.
`
`TABLE OF CONTENTS
`Introduction ...................................................................................................... 1
`Ground 1: Claims 1-3 and 5 Are Rendered Obvious over Kaneda in view of
`Schneider ......................................................................................................... 3
`A.
`A POSITA would have been motivated to combine Kaneda and
`Schneider ............................................................................................... 3
`The combination teaches altering the phase of the phase modulator
`(Claim 1[g]) .........................................................................................20
`The combination teaches rotating the phase imparted by the phase
`modulator by a predetermined amount (Claim 5) ...............................22
`III. Ground 2: Claim 4 is Rendered Obvious over Kaenda in view of Schneider
`and Heflinger .................................................................................................24
`A.
`Kaneda discloses an interferometer ....................................................24
`B.
`The combination discloses and teaches “an additional phase
`modulator” in an arm of the interferometer ........................................24
`The Board should consider all challenged claims .........................................25
`
`B.
`
`C.
`
`IV.
`
`i
`
`
`
`IPR2017-02189
`
`LIST OF EXHIBITS
`
`U.S. Patent No. 6,476,952 to Snawerdt (“the ’952 Patent”)
`
`CV of Daniel Blumenthal
`
`Expert Declaration of Daniel Blumenthal
`
`U.S. Patent No. 6,826,371 to Bauch et al. (“Bauch”)
`
`Japanese Unexamined Patent Application Publication No. S61-127236
`by Tetsuya Kaneda et al. (“Kaneda”)
`
`Declaration Regarding English Translation of Kaneda
`
`English Translation of Kaneda
`
`Oyster Optics, LLC v. Cisco Systems, Inc. et al, Case No. 2:16-cv-
`01301-JRG, Complaint (E.D. Tex. Nov. 24, 2017) (Dkt. 1)
`
`Phase-Modulated Optical Communication Systems by Keang-Po Ho
`
`Digital Processing, Optical Transmission and Coherent Receiving
`Techniques by Le Nguyen Binh
`
`Coherent Optical System Design by Pieter W. Hooijmans
`
`Coherent Optical Communications Systems by Silvello Betti et al.
`
`N. M. Blachman, “The effect of phase error of DPSK error
`probability,” IEEE Trans. Commun., vol. COM-29, no. 3, pp. 364-
`365, 1981.
`
`G. Nicholson, “Probability of error for optical heterodyne DPSK
`system with quantum phase noise,’’ Electron. Lett., vol. 20, no. 24,
`1005-06 (1984)
`
`R. Wyatt, T. G. Hodgkinson et al., “DPSK heterodyne experiment
`featuring an external cavity diode laser local oscillator,” in Electron.
`Lett., vol. 19, no. 14, 550-52 (July 1983)
`
`1001
`
`1002
`
`1003
`
`1004
`
`1005
`
`1006
`
`1007
`
`1008
`
`1009
`
`1010
`
`1011
`
`1012
`
`1013
`
`1014
`
`1015
`
`ii
`
`
`
`IPR2017-02189
`
`1016
`
`1017
`
`1018
`
`1019
`
`1020
`
`1021
`
`1022
`
`1023
`
`1024
`
`1025
`
`1026
`
`1027
`
`1028
`
`S. Yamazaki et al., “1.2 Gbit/s optical DPSK heterodyne detection
`transmission system using monolithic external-cavity DFB LDs,” in
`Electron. Lett., vol. 23, no. 16, 860-62 (1987)
`
`S. Watanabe, T. Naito, T. Chikama, T. Kiyonaga, Y. Onoda, and H.
`Kuwahara, “Polarization-insensitive 1.2 Gb/s optical DPSK
`heterodyne transmission experiment using polarization diversity,”
`presented at ECOC’88 (Brighton, U.K.), vol. 1, 90-93 (1988)
`
`J. M. P. Delavaux et al., “1.4 Gbit/s optical DPSK heterodyne
`transmission system experiment,” 1988 Fourteenth European
`Conference on Optical Communication, ECOC 88 (Conf. Publ. No.
`292), Brighton, UK, 475-78 vol. 1 (1988)
`
`John R. Barry et al., Performance of Coherent Optical Receivers,
`Proceedings of the IEEE, Vol. 78, No. 8 (Aug. 1990)
`
`Eric. A. Swanson et al., “High Sensitivity Optically Preamplified
`Direct Detection DPSK Receiver with Active Delay-Line
`Stabilization,” IEEE Photonics Technology Letters, Vol. 6, No. 2
`(1994)
`
`U.S. Patent No. 6,559,996 to Miyamoto et al. (“Miyamoto”)
`
`U.S. Patent No. 5,543,952 to Yonenaga et al. (“Yonenaga”)
`
`Affidavit of Christopher Butler, Internet Archive
`
`Introduction to Logic Design, Second Edition, Sajjan Shiva (1998)
`
`U.S. Patent No. 6,396,605 to Heflinger et al. (“Heflinger”)
`
`U.S. Patent No. 6,700,907 to Schneider et al. (“Schneider”)
`
`U.S. Provisional Application No. 60/249,438 (“Schneider’s
`provisional application”)
`
`J. Salz, “Coherent Lightwave Communications”, AT&T Technical
`Journal, Vol. 64, No. 10 (Dec. 1985)
`
`iii
`
`
`
`IPR2017-02189
`
`Terumi Chikama et al., “Modulation and Demodulation Techniques in
`Optical Heterodyne PSK Transmission Systems,” Journal of
`Lightwave Technology, Vol. 8 No. 3 (1990)
`
`Maryanne Heinbaugh, The Mach-Zehnder Coupler (August 27, 1997)
`(published thesis, Naval Postgraduate School) (on file with Calhoun
`Naval Postgraduate School Institutional Archive).
`
`K. P. Zetie et al., “How does a Mach-Zehnder Interferometer Work?,”
`Physics Education, Vol. 35, No. 1 (1999).
`
`Full-Text Comparison between the ’952 Patent and U.S. Patent No.
`6,469,816
`
`Full-Text Comparison between the ’952 Patent and U.S. Patent No.
`6,594,055
`
`U.S. Patent Publication No. 2003/0007216 to Chraplyvy et al.
`(“Chraplyvy”)
`
`Excerpts from the Prosecution History of the ’952 Patent
`
`Optical Fiber Telecommunications, Fourth Edition, Vol. B, Ivan
`Kaminow & Tingye Li (2002) (“Kaminow”)
`
`Edward I. Ackerman, “Broad-Band Linearization of a Mach-Zehnder
`Electrooptic Modulator,” IEEE Transaction of Microwave Theory and
`Techniques, Vol. 47, No. 12 (Dec. 1999)
`
`Technical Note, Using the Lithium Niobate Modulator: Electro-
`Optical and Mechanical Connections, Lucent Technologies (Apr.
`1998)
`
`Fundamentals of Photonics, First Edition, Bahaa E.A. Saleh & Malvin
`C. Teich (1991) (“Saleh”)
`
`S.J. Spammer & P.L. Swart, “Differentiating Optical-Fiber Mach-
`Zehnder Interferometer,” Applied Optics, Vol. 34, No. 13 (May 1995)
`
`1029
`
`1030
`
`1031
`
`1032
`
`1033
`
`1034
`
`1035
`
`1036
`
`1037
`
`1038
`
`1039
`
`1040
`
`iv
`
`
`
`IPR2017-02189
`
`1041
`
`Vijaya Poudyal & Mohcene Mezhoudi, Wavelength Sensitivity of
`Ti:LiNbO3 Mach-Zahnder Interferometer, Integrated Optics and
`Microstructures II, SPIE Vol. 2291 (Oct. 1994)
`
`1042
`
`Annotated Declaration of Keith W. Goossen
`
`1043
`
`Deposition Transcript of Keith W. Goossen
`
`1044
`
`Expert Declaration of Daniel Blumenthal, in support of Petitioners’
`Reply
`
`1045
`
`U.S. Patent No. 6,046,838 to Kou et al. (“Kou”)
`
`1046
`
`U.S. Patent No. 5,170,274 to Kuwata et al. (“Kuwata”)
`
`1047
`
`Edward Ackerman et al., “Bias Controllers for Phase Modulators in
`Fiber-Optic Systems,” Lightwave, Vol. 18, No. 5 (2001)
`
`1048
`
`Lightwave Homepage from archive.org (May 13, 2001)
`
`1049
`
`PSI Modulator Bias Controller, Photonic Systems, Inc. (2001)
`
`v
`
`
`
`IPR2017-02189
`
`I. INTRODUCTION
`The issues raised in Patent Owner’s Response (Paper 14 (“Resp.”)) do not
`
`withstand scrutiny. PO inexplicably devotes nearly thirty pages (Resp., 15-46) to
`
`confirming Petitioners’ expert testimony that Schneider “discloses MZ modulator
`
`10 operating to modulate intensity.” Pet., 41; Ex. 1003, ¶¶ 154-162. Those pages
`
`are essentially irrelevant to any material issue. Instead, the Board may focus on PO’s
`
`five substantive challenges: (1) it would not have been obvious to combine elements
`
`taught in Schneider and Kaneda, as Dr. Blumenthal proposed (Ground 1) (Resp., 47-
`
`50); (2) using Schneider’s Mach-Zehnder Modulator (MZM) bias “control” would
`
`not “alter” the phase (Resp., 51-53); (3) claim construction for “rotating a phase
`
`imparted by the phase modulator by a predetermined amount” (claim 5) (Resp., 53-
`
`58); (4) Kaneda does not disclose an “interferometer” (claim 4) (Resp., 58-62); and
`
`(5) Heflinger does not teach an “additional phase modulator” (claim 4) (Resp., 62-
`
`64).
`
`PO’s challenge to Ground 1 principally contends that Schneider teaches away
`
`from using an MZM to modulate intensity to determine a bias and gain setting and
`
`then computing a different set of operational points for “open loop” control of the
`
`MZM to modulate phase, and that the use of “open loop” set points would not
`
`provide a reasonable expectation of success. Resp., 50. PO’s challenge fails, inter
`
`alia, because (a) as Dr. Blumenthal explained, the proposed combination would be
`
`1
`
`
`
`IPR2017-02189
`
`a simple use of set point values obtained via Schneider’s closed loop algorithm; and
`
`(b) it is based on a premise that Schneider shows is false—that a POSITA would
`
`ignore Schneider’s teaching that the control loop may be turned off “for extended
`
`periods.” Ex. 1026, 6:34-35. PO’s second argument simply re-packages the first
`
`one and assumes that a POSITA’s use of Schneider is limited to modulating
`
`intensity. Again, PO’s challenge fails because it was known at the time that MZMs
`
`could be used to modulate intensity or phase. See, e.g., Ex. 1029, 309-10, 314.
`
`PO’s third argument fails because its expert, Dr. Goossen, concedes that
`
`normal use of the phrase “rotating a phase” in the art includes rotating the phase of
`
`an arm by changing the DC bias voltage in prescribed adjustments, as explained by
`
`Dr. Blumenthal (Ex. 1003, ¶ 190-92), and because there is no basis to import
`
`limitations from the specification. PO’s fourth argument fails because Kaneda
`
`discloses an interferometer to a POSITA. And PO’s fifth argument that the
`
`combination with Heflinger does not teach an additional phase modulator is
`
`misplaced because it is undisputed that Heflinger teaches a PZT that alters the optical
`
`path, and because all evidence of how a POSITA would have understood that
`
`teaching shows that Heflinger discloses the claimed “additional phase modulator.”
`
`Petitioners’ obviousness challenge is supported by the asserted art, the testimony of
`
`Petitioners’ expert, and numerous supporting references indicative of a POSITA’s
`
`knowledge at the time. Accordingly, Petitioners’ challenge should succeed.
`
`2
`
`
`
`IPR2017-02189
`
`II. GROUND 1: CLAIMS 1-3 AND 5 ARE RENDERED OBVIOUS OVER
`KANEDA IN VIEW OF SCHNEIDER
`A. A POSITA would have been motivated to combine Kaneda and
`
`Schneider
`
`PO and Petitioners agree that Schneider discloses a Mach-Zehnder Modulator
`
`(MZM) operating to modulate intensity (see Pet., 41; Ex. 1003, ¶¶ 154-162;
`
`Ex. 2030, 113:4-20), that a POSITA could apply Schneider’s algorithm to settle at a
`
`quadrature point (e.g., Vπ/2) along the transfer function (see Ex. 2030, 71:16-72:9,
`
`85:4-6; 89:1-21; Resp., 26-27; Ex. 1043, 29:1-5), and that the transfer function of
`
`Schneider’s MZM is the same irrespective of whether the MZM modulates intensity
`
`or phase (see Pet., 41; Ex. 1003, ¶¶ 69-70, 162; Ex. 2030, 92:13-18; Resp., 17-20;
`
`Ex. 2031, ¶¶ 29-32; Ex. 1043, 7:22-8:11, 145:2-10).
`
`Petitioners and its expert explained that Schneider’s algorithm is executed to
`
`derive information about the MZM’s transfer function, and then use that information
`
`to operate the MZM as either an intensity modulator or a phase modulator, and
`
`specifically as a phase modulator when implemented in Kaneda’s system. Pet., 41-
`
`44; Ex. 1003, ¶¶ 154-162. PO’s principle contention is that a POSITA would not
`
`have used the information from Schneider’s control algorithm to operate the MZM
`
`as a phase modulator. Resp., 48. As explained below in Sections II.A.1-5, a
`
`POSITA would have been motivated to combine Kaneda and Schneider and would
`
`have had a reasonable expectation of success in doing so.
`3
`
`
`
`IPR2017-02189
`
`1. The combination of Kaneda and Schneider
`
`PO asserts that the Petition does not explain the combination of Kaneda and
`
`Schneider. Resp., 48. However, the Petition and Dr. Blumenthal explained that a
`
`POSITA would have been motivated to combine Kaneda and Schneider to use
`
`Schneider’s MZM, with its gain and bias control algorithm, as Kaneda’s phase
`
`modulator so that the modulator would be stable and otherwise able to operate in
`
`“real world” conditions (e.g., noise, temperature, and aging), as disclosed in
`
`Schneider. Pet., 43-44; Ex. 1003, ¶¶ 150-163.
`
`The Petition and Dr. Blumenthal explained the knowledge a POSITA had
`
`about the operation of MZMs, e.g., that the same MZM component could be used to
`
`both modulate intensity and phase. Ex. 1003, ¶ 69; Ex. 1029, 309-10, 314; Ex. 1009,
`
`50. MZMs have a sinusoidal power transfer function (shown by the dashed line in
`
`the figure below):
`
`4
`
`
`
`IPR2017-02189
`
`Ex. 1003, ¶¶ 68-69 (annotating Ex. 1036, Fig. 16.3).
`
`a) An MZM is characterized by the Vπ value
`The MZM’s transfer function is a periodic sinusoid with a minimum,
`
`maximum, and period. Ex. 1044, ¶ 6. A quantity known as Vπ (labeled “minimum
`
`amplitude” above) corresponds to the point along the X-axis (1 on the graph above)
`
`where there is no intensity at the output (0 on the Y-axis). A POSITA would have
`
`understood that Vπ characterizes the transfer function. Ex. 1044, ¶ 6. Another
`
`quantity known as quadrature (e.g., Vπ/2 or 3Vπ/2, labeled “quadrature operating
`
`point” above) corresponds to mid-points along the transfer function between full and
`
`5
`
`
`
`IPR2017-02189
`
`no intensity.1
`
`Id.; Ex. 1003, ¶¶ 69-71; Ex. 1010, 32; Ex. 1040, 1; Ex. 1049, 2;
`
`Ex. 1047, 2; Ex. 1045, 5:51-63, Fig. 2.
`
`b) An MZM would be biased at some function of the Vπ value
`based on modulation mode
`MZMs had DC bias control, which could change the DC operation point of
`
`the modulator to operate at different points along the transfer function. Ex. 1003,
`
`¶¶ 64-71; see Ex. 1038, 7, 9. A POSITA understood that the peaks (i.e., minimums
`
`and maximums) of the transfer function define points at which the output of the
`
`MZM is at a minimum or maximum intensity. Ex. 1003, ¶ 69.
`
`Against this backdrop, the Petition and Dr. Blumenthal explained the
`
`proposed combination: Kaneda’s phase modulator is implemented using Schneider’s
`
`MZM and control circuit. Pet., 42-44. In the combination, a POSITA would have
`
`executed Schneider’s control algorithm and after execution is complete, a POSITA
`
`would have operated the same MZM to modulate phase in Kaneda’s system. Pet.,
`
`1 PO suggests a narrow meaning of “quadrature.” Compare Resp., 26; Ex. 2031,
`
`¶ 40 with Ex. 2030, 36:17-37:13; Ex. 1003, ¶ 39. A POSITA would have understood
`
`that there are multiple quadrature points along a periodic transfer function. Ex. 1044,
`
`¶ 6; Ex. 1049, 2; Ex. 1047, 2; Ex. 1045, 5:51-63, Fig. 2. Dr. Goossen rejected PO’s
`
`narrow meaning and agreed that “quadrature” is used more broadly in the art.
`
`Ex. 1043, 107:2-5, 158:10-18.
`
`6
`
`
`
`IPR2017-02189
`
`41-44, 67-69; Ex. 1003, ¶¶ 160-62, 190-91.
`
`Specifically, Schneider’s algorithm can be implemented to settle on a bias
`
`point at the quadrature corresponding to Vπ/2 on the x-axis of the graph above.
`
`Ex. 2030, 72:3-5 (“quadrature point associated with an intensity output power
`
`function”); Ex. 1043, 29:1-5 (confirming the same settling point).
`
`Schneider discloses that the “objective of the bias control loop is to derive the
`
`peak of the sinusoidal Mach-Zehnder function, where the derivative is zero” and that
`
`the gain control loop provides “the peak-to-peak swing of the electrical data signal
`
`output.” Ex. 1026, 5:18-23, 4:50-54. And Dr. Blumenthal explained that Schneider
`
`teaches to a POSITA to extract information about the transfer function from the
`
`settled point. Ex. 1003, ¶¶ 64-71, 87, 154-162; Ex. 2030, 48:3-49:9, 72:6-8; cf.
`
`Ex. 48:9-52:12, 95:18-23. Thus, a POSITA would have understood that Schneider
`
`teaches identifying or deriving the peak and peak-to-null swing associated with the
`
`transfer function based on the settled values of the control algorithm. Pet., 41-42,
`
`67-69 (citing Ex. 1026, 5:18-23, 4:50-54); Ex. 1003, ¶ 190; Ex. 2030, 48:3-49:9,
`
`72:6-8; see also Ex. 1003, ¶¶ 64-71, 87, 154-162. For example, a POSITA could
`
`derive the peak of the transfer function (e.g., Vπ, which characterizes the transfer
`
`function) by multiplying the control voltages associated with the settled quadrature
`
`point (i.e., Vπ/2) by two and could derive the peak-to-null swing by using the same
`
`control voltages associated with the settled gain. Ex. 1044, ¶¶ 6-7.
`
`7
`
`
`
`IPR2017-02189
`
`Using the derived peak and peak-to-null swing, Dr. Blumenthal explained
`
`specifically why it does not matter whether Schneider’s MZM operates to modulate
`
`data via intensity versus phase modulation. Pet., 66-69; Ex. 1003, ¶ 190; see, e.g.,
`
`Ex. 1029, 310. He described how the knowledge of the transfer function determined
`
`by the control algorithm can be used to operate the MZM to modulate intensity or
`
`phase. Ex. 1003, ¶¶ 156, 162. As shown below, a POSITA would have understood
`
`that in a DPSK system, the MZM would modulate phase by setting the DC bias to
`
`the minimum point (e.g., Vπ) and the gain such that the phase shift corresponds to
`
`180 degrees relative to the input, or twice the peak-to-null swing (e.g., 2Vπ):
`
`Ex. 1003, ¶¶ 70, 162. A POSITA would have set these points by first deriving the
`8
`
`
`
`IPR2017-02189
`
`peak (e.g., Vπ) and peak-to-null swing (e.g., Vπ), and then simply multiplying the
`
`swing by two to arrive at the peak-to-peak swing (i.e., 2Vπ). Id. Alternatively, Dr.
`
`Blumenthal explains that because Schneider’s algorithm settles at a DC bias
`
`corresponding to Vπ/2 and a gain corresponding to Vπ, a POSITA could have
`
`multiplied each value by two (i.e., for the DC bias, Vπ/2 multiplied by two equals
`
`Vπ, and for the gain, Vπ multiplied by two equals 2Vπ) to arrive at the desired points
`
`for operating the MZM as a phase modulator.2 Ex. 1044, ¶ 7. Following Schneider’s
`
`teaching of turning off the modulator control routine for extended periods, the
`
`Petition and Dr. Blumenthal explained that Schneider’s control algorithm may be
`
`periodically re-run to compensate for aging and temperature, and then powered
`
`down. Pet., 41, 67; Ex. 1003, ¶ 160-61, 190; Ex. 1026, 6:20-25.
`
`Thus, the modification proposed in the Petition and supported by Dr.
`
`Blumenthal’s explanation is simple and clear. As summarized above, both parties
`
`agree that Schneider discloses a MZM operating to modulate intensity, that
`
`Schneider’s algorithm settles at a quadrature point along the transfer function, and
`
`that the transfer function of Schneider’s MZM is the same irrespective of whether it
`
`modulates intensity or phase. In the proposed combination, Schneider’s control
`
`2 Dr. Goossen admitted that a POSITA would understand this approach to work for
`
`an ideal MZM. Ex. 1043, 10:12-17, 18:4-10, 50:9-13.
`
`9
`
`
`
`IPR2017-02189
`
`algorithm executes until it converges to derive information about the transfer
`
`function (i.e., peak and peak-to-null swing) for use in operating the MZM as a phase
`
`modulator in Kaneda’s system. Dr. Goossen agrees that the combination would
`
`work for an ideal MZM, and only contends that it wouldn’t for a non-ideal one. And
`
`PO principally argues that Schneider must execute with closed-loop operation
`
`without disruption. But, as explained below, the combination teaches the closed-
`
`loop operation of Schneider’s algorithm that does not disrupt system operation and
`
`would also work for a non-ideal MZM.
`
`2. Schneider’s algorithm executes using closed-loop operation and
`
`a POSITA would have expected success when implementing it in
`
`Kaneda
`
`PO and its expert argue that Schneider is limited to what they deem “closed-
`
`loop control” and thus, Schneider teaches away from biasing its MZM with
`
`parameters derived from the control routine. Resp., 28-30, 49-51. They contend
`
`that unlike open loop control, which “operates without regard to its output” (Resp.,
`
`29; Ex. 2031, ¶ 50), Schneider exclusively teaches closed-loop control, in which the
`
`control “is dependent on a measurement of its output.” Resp., 28; Ex. 2031, ¶ 45.
`
`They further contend that Schneider is limited to operation dependent on direct,
`
`closed loop feedback control. Ex. 2031, ¶ 95. This is all beside the point. In the
`
`combination, the control loop and algorithm are still the exact same as disclosed in
`
`10
`
`
`
`IPR2017-02189
`
`Schneider.
`
`PO and its expert fail to consider that in the proposed combination, a POSITA
`
`would have understood that when Schneider’s control algorithm executes, the
`
`control “is dependent on a measurement of its output” (i.e., closed-loop control).
`
`Specifically, the control algorithm settles at a converged point based on output
`
`measurements and input data with approximately an equal number of 0s and 1s (e.g.,
`
`“0101…”), and then turns off. Ex. 1026, 3:25-31; Ex. 1044, ¶ 8. Schneider explains
`
`that this works because its algorithm relies on the shape of the transfer function and
`
`not specific values. Id., 2:38-39, 5:19-21, 5:50-53; Ex. 1027, 4-5; Ex. 1044, ¶ 8. As
`
`Dr. Blumenthal explained, a POSITA would have understood that Schneider runs its
`
`control algorithm in closed-loop operation, derives information about the transfer
`
`function (i.e., the peak and peak-to-null swing), and then uses the information
`
`derived from the settled point of the control algorithm to run the MZM as a phase
`
`modulator in Kaneda’s system, in manner very similar to how Schneider uses the
`
`settled point to run the MZM as an intensity modulator. Ex. 1003, ¶ 154; Ex. 1044,
`
`¶ 8. Thus, PO’s open-loop control argument is an irrelevant red herring.
`
`Indeed, PO’s binary view of control systems runs contrary to Schneider’s
`
`explicit disclosure of dis-continuous operation. Schneider discloses running its
`
`algorithm, and then powering it down once settled “for extended periods of time to
`
`save power.” Ex. 1026, 6:20-25, 6:36-38; see also Pet., 41; Ex. 1027, 4-6. Although
`
`11
`
`
`
`IPR2017-02189
`
`PO did not address Schneider’s dis-continuous operation in its Response, Dr.
`
`Goossen admitted that Schneider teaches dis-continuation operation and explained
`
`that when Schneider’s control loop is powered down and running with fixed settings,
`
`there is no “control system anymore” (i.e., no closed or open control). Ex. 1043,
`
`56:17-57:1, 149:19-150:15. In his view, running with fixed settings is just “setting
`
`it.” Id., 147:20-148:14; see also 148:15-149:18. Thus, even under Dr. Goossen’s
`
`view, Schneider is not limited to closed-loop control, as it is not even a “control
`
`system” when using fixed settings.
`
`It appears that PO’s and its expert’s actual argument is that Schneider
`
`allegedly requires the values for controlling the MZM be taken directly from the
`
`“settled” point. Ex. 1043, 123:2-9. This is incorrect for several reasons. Schneider
`
`does not describe, let alone require direct values be taken from the “settled” point in
`
`the algorithm. As explained above, Schneider teaches deriving information about
`
`the transfer function (i.e., peak and peak-to-null swing) from the settled values.
`
`Supra, Section II.A.1; Ex. 1026, 5:18-23, 4:50-54. And Dr. Goossen conflates
`
`taking the value directly from the settled point with PO’s requirements that the
`
`algorithm run during data flow and that no adjustment is permitted after the control
`
`algorithm settles. Ex. 1043, 122:9-123:1 (expressing narrow view that “all of the
`
`advantages that [Schneider] states” are specific to “directly using … the operational
`
`output values.”); Ex. 1043, 124:19-25; see also Ex. 1043, 125:1-9, 126:21-127:3,
`
`12
`
`
`
`IPR2017-02189
`
`127:11-18. As explained in Sections II.A.3 and 11.A.5.a below, the combination
`
`does not disrupt system operation and Schneider does not limit use of its derived
`
`values to the settling points of the control algorithm.
`
`Therefore, it is consistent with Schneider’s teachings to run Schneider’s
`
`control algorithm periodically while not transmitting data, to configure the MZM
`
`based on the derived information of the control algorithm (e.g., as a phase
`
`modulator), then to transmit phase-modulated data using the MZM.
`
`3. The combination of Kaneda and Schneider does not disrupt
`
`system operation
`
`PO and its expert argue that Schneider’s control algorithm must be run while
`
`data is transmitted. Resp., 28, 49-51; Ex. 2031, ¶¶ 49, 97 (citing to Ex. 1027, 1). PO
`
`and its expert repeatedly assert a narrow view of Schneider predicated on data flow
`
`while the control algorithm executes. Resp., 29, 49-51; Ex. 2031, ¶¶ 46-49, 97. But
`
`Schneider is not limited to operating its control algorithm during data transmission.
`
`Rather, Schneider teaches a goal of avoiding disruption of the system.
`
`As Dr. Blumenthal explained, “[s]ome systems stop data to do things and there
`
`is not a problem.” Ex. 2030, 61:23-25. A POSITA would have understood that
`
`Schneider’s control routine would be used periodically in a system (e.g., like
`
`Kaneda) that can stop data transmission (e.g., for maintenance). In other words, a
`
`POSITA would have understood that Schneider’s description of “non-disruptive[]”
`
`13
`
`
`
`IPR2017-02189
`
`operation provides for acceptable levels of operation that were known in the art at
`
`the time, and not absolute perfection. Ex. 1044, ¶¶ 8-10; Ex. 2030, 58:3-62:2;
`
`Ex. 1026, 3:15-30, 4:44-54; Ex. 1027, 1.
`
`In combination with Kaneda, Schneider achieves this goal by reducing the
`
`impact of the gain and bias control routine on system operation, just as it does in
`
`Schneider alone. Id.; Ex. 2030, 71:6-9. Schneider discloses its control algorithm
`
`avoids “time consuming manual alignment techniques.” Ex. 1026, Abstract, Fig. 2;
`
`Ex. 1027, 6. Indeed, Dr. Goossen found that Schneider’s bias control routine can
`
`settle in as few as four iterations. Ex. 2031, ¶ 77. A POSITA would have recognized
`
`that Schneider’s control algorithm could execute more quickly than the prior art bias
`
`control routines, limiting disruption. Ex. 1044, ¶ 9.
`
`Moreover, Schneider discloses a control algorithm that settles, powers down,
`
`and re-runs periodically to compensate for temperature and aging. Ex. 1026, 3:25-
`
`30. Dr. Goossen explained that temperature can drift daily and aging could drift
`
`over years. Ex. 1043, 52:19-53:9. In combination with Kaneda, a POSITA would
`
`have understood that Schneider’s control algorithm would execute much more
`
`quickly than prior art systems, power down, and then re-run as needed for the desired
`
`level of Kaneda. Ex. 1044, ¶ 9. Accordingly, a POSITA would have recognized
`
`that the combination achieves Schneider’s goal of avoiding disruption of the system
`
`and its data flow by reducing the impact of the gain and bias control routine on
`
`14
`
`
`
`IPR2017-02189
`
`system operation. Id.
`
`4. PO’s remaining arguments similarly fail to show no motivation
`
`to combine
`
`PO asserts several additional arguments that similarly fail to show no
`
`motivation combine.
`
`a) Schneider does not limit its derived values to the settling
`points of the control algorithm
`PO and its expert assert that setting the DC bias and gain for an MZM to
`
`operate as a phase modulator amounts to “tuning or adjustment,” which Schneider
`
`purportedly seeks to avoid3. Resp., 28, 49-51; see Ex. 1026, 2:14-15; Ex. 1027, 1;
`
`Ex. 2031, ¶ 97.
`
`However, Schneider does not limit its derived values to the settling points of
`
`the control algorithm. Nothing in Schneider restricts its application to intensity
`
`modulation, nor bars arithmetic operation on the values at which Schneider “settles.”
`
`Schneider’s algorithm itself provides for adjustment by setting the DC bias and gain
`
`3 Dr. Goossen admitted that this description in Schneider’s description of offsets
`
`varying in “selection, time, or environment” (Resp., 49) corresponds to components
`
`other than the modulator (i.e., MZM). Ex. 1043, 154:10-155:15. A POSITA would
`
`have understood that Schneider’s varying offsets are not unique to the Kaneda
`
`combination because the same external control circuitry is used. Ex. 1044, ¶ 14.
`
`15
`
`
`
`IPR2017-02189
`
`based on the identified peak and peak-to-null swing along the transfer function.
`
`Ex. 1044, ¶¶ 7, 10; Ex. 1026, 5:19-21; Ex. 1027, 3. And it remains undisputed that
`
`it was well known at the time that MZM could operate to modulate intensity or
`
`phase. Ex. 1003, ¶ 162; see, e.g., Ex. 1029, 310, Fig. 2(b). Dr. Blumenthal
`
`explained how to operate the MZM to modulate phase using the knowledge of the
`
`transfer function determined by the control algorithm. Ex. 1003, ¶¶ 156, 162. In
`
`particular, a POSITA would have understood to set the DC bias to the minimum
`
`point (e.g., Vπ or twice the settled quadrature point) and the gain to 2Vπ (i.e., twice
`
`the derived peak-to-null swing). Id.; see Ex. 1044, ¶ 6, 11.
`
`Moreover, even if modifying the output of Schneider’s algorithm to operate
`
`the MZM as a phase modulator was “additional adjustment and tuning,” this is not
`
`the type of additional adjustment and tuning Schneider seeks to avoid. Schneider’s
`
`provisional application explains that “[t]he invention allows for a simple … control
`
`system which does not require tuning or adjustment” (Ex. 1027, 4), and that previous
`
`methods required adjustment of the control loops and tuning of tone-based control
`
`circuits that were high-speed and data-rate dependent. Ex. 1026, 1:60-2:15;
`
`Ex. 1027, 5. Schneider discloses that the “shortcomings of conventional tone-based
`
`laser modulator control schemes … are effectively obviated by a microcontroller-
`
`based laser modulator control mechanism … .” Ex. 1026, 2:10-15. This sort of
`
`additional adjustment and tuning is not required in the proposed combination.
`
`16
`
`
`
`IPR2017-02189
`
`Schneider’s MZM and control circuitry remain the same—no data-rate dependent
`
`components or tone-based control is added. Ex. 1044, ¶ 10; Ex. 1026, 1:60-2:15.
`
`b) The combination does not require inventive work
`PO asserts that Dr. Blumenthal had not evaluated the combination in which
`
`Schneider’s control algorithm executes while the MZM transmits phase-modulated
`
`data in Kaneda’s system. Resp., 47; see Ex. 2030, 116:17-117:19. However,
`
`Petitioners do not argue such a combination. As Dr. Blumenthal clarified on
`
`redirect, no inventive work is required for the proposed combination. Ex. 2030,
`
`128:4-15. As explained above, the exact same control algorithm is used.
`
`Schneider’s algorithm is executed to derive information about the MZM’s transfer
`
`function based on the settled points, and then that information is used to operate the
`
`MZM as a phase modulator in Kaneda’s system.
`
`c) A POSITA would have understood that Schneider’s control
`algorithm works for an ideal and non-ideal MZM
`PO and its expert assert a POSITA would not have configured Schneider’s
`
`MZM as a phase modulator. Resp., 48-49; Ex. 2031, ¶ 97. Although not argued in
`
`PO’s Response, Dr. Goossen alleged during his deposition that a POSITA would
`
`understand configuring Schneider’s MZM as a phase modulator to work for an ideal
`
`MZM (Ex. 1043, 10:12-17, 18:4-10, 50:9-13), but not for a non-ideal MZM because
`
`it “could give rise to tremendous amounts of noise.” Ex. 1043, 34:14-18, 40:18-25,
`
`45:5-12. Yet Dr. Goossen assumed that the transfer function is still largely the same
`
`17
`
`
`
`IPR2017-02189
`
`as an ideal MZM. Id., 32:1-11, 175:22-176:9 (assuming an S-shaped curve), 176:18-
`
`25 (assuming symmetry around the bias point). Although he eventually provided a
`
`few sources that might cause non-ideal response (id., 85:6-86:1), he failed to explain
`
`even one non-ideal feature for the transfer function to support his allegation that a
`
`non-ideal MZM could give rise to tremendous amounts of noise.
`
`Dr. Goossen’s concerns are unfounded. He readily admitted that he had not
`
`designed or built a working system that included an MZM at the time (Ex. 1043,
`
`170:19-171:6), had “[v]ery limited” experience working with MZMs, and had no
`
`experience whatsoever with operating an MZM as a phase modulator (id., 168:13-
`
`16, 168:23–169:4). See also, id., 174:21–175:4; id., 168:10-11.
`
`By contrast, Dr. Blumenthal has experience with MZMs, controlling the bias
`
`and gain of MZMs, and operating MZMs as a phase modulator. Ex. 1003, ¶¶ 4-10,
`
`Ex. 1002; Ex. 1044, ¶¶ 4-5. As Dr. Blumenthal explained, a POSITA would have
`
`known how to configure the MZM as a phase modulator by using the derived peak
`
`and peak-to-null swing associated with the transfer function. Ex. 1003, ¶¶ 159-163.
`
`Specific
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
Still Working On It
This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.
Give it another minute or two to complete, and then try the refresh button.
A few More Minutes ... Still Working
It can take up to 5 minutes for us to download a document if the court servers are running slowly.
Thank you for your continued patience.
This document could not be displayed.
We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.
Your account does not support viewing this document.
You need a Paid Account to view this document. Click here to change your account type.
Your account does not support viewing this document.
Set your membership
status to view this document.
With a Docket Alarm membership, you'll
get a whole lot more, including:
- Up-to-date information for this case.
- Email alerts whenever there is an update.
- Full text search for other cases.
- Get email alerts whenever a new case matches your search.
One Moment Please
The filing “” is large (MB) and is being downloaded.
Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.
Your document is on its way!
If you do not receive the document in five minutes, contact support at support@docketalarm.com.
Sealed Document
We are unable to display this document, it may be under a court ordered seal.
If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.
Access Government Site
