`
`
`
`
`
`IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
`
`___________________
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`___________________
`
`
`PETROLEUM GEO-SERVICES INC.
`and
`ION GEOPHYSICAL CORPORATION
`AND ION INTERNATIONAL S.A.R.L.
`Petitioners
`
`v.
`
`WESTERNGECO LLC
`Patent Owner
`___________________
`
`Case No. IPR2014-006881
`U.S. Patent No. 7,080,607
`
`
`
`
`
`PETITIONER’S PETROLEUM GEO-SERVICES INC.’S REPLY
`
`
`David I. Berl, Reg. No. 72,751
`Jessamyn S. Berniker, Reg. No. 72,328
`Thomas S. Fletcher, Reg. No. 72,383
`Christopher A. Suarez, Reg. No. 72,553
`WILLIAMS & CONNOLLY LLP
`725 12th St., NW
`Washington, DC 20005
`Telephone:
`202-434-5000
`Fax:
`202-434-5029
`
`Counsel for Petitioner, Petroleum Geo-Services Inc.
`
`1 Case IPR2015-00567 has been joined with this proceeding.
`
`
`
`

`

`
`
`TABLE OF CONTENTS
`
`TABLE OF AUTHORITIES ................................................................................... iii
`
`I.
`
`CLAIM CONSTRUCTION ............................................................................ 2
`
`A.
`
`B.
`
`The Claim Language. ............................................................................ 4
`
`The Board Correctly Interpreted “Predict Positions.” .......................... 5
`
`1.
`
`The Intrinsic Evidence Does Not Require a Specific
`Prediction Model. ........................................................................ 5
`
`2. WesternGeco Never Showed that ION Used a “Behavior-
`Predictive Model” To Predict Positions. ..................................... 9
`
`3.
`
`Present Estimates Based on Old Data Are “Predictions.” ........ 11
`
`C.
`
`“Calculating Desired Changes” in Position Cannot Require
`Calculating Forces or a Particular Method of Calculating
`Forces. ................................................................................................. 12
`
`1.
`
`2.
`
`The Claim Says To Calculate Positions, Not Forces. ............... 13
`
`Calculations “Based on Streamer and Array Behavior”
`Are Not Required. ..................................................................... 14
`
`D.
`
`Claim 15 Does Not Require the Panoply of Other Features WG
`Attempts to Insert. ............................................................................... 15
`
`II. WORKMAN (EX. 1004) ............................................................................... 17
`
`A. Workman’s Control System Anticipates Claims 1 and 15. ................ 18
`
`1. Workman Measures the Locations of SPDs. ............................ 18
`
`2. Workman’s Kalman Filter Uses Location Measurements
`and Navigation Data to Estimate Real-Time Locations. .......... 19
`
`3. Workman Uses the Predicted Positions to Calculate
`Desired Changes in SPD Positions. .......................................... 23
`
`B. Workman is Presumed To Be Enabled and Was Enabled as of
`the Priority Date. ................................................................................. 26
`
`i
`
`

`

`
`
`1. WesternGeco Cannot Rebut the Presumption that
`Workman is Enabled. ................................................................ 26
`
`2. WesternGeco’s Attacks on Drs. Evans and Cole Cannot
`Rebut the Presumption that Workman is Enabled. ................... 29
`
`3. WesternGeco’s Eleventh-Hour Evidence is Inapposite. ........... 30
`
`III.
`
`PURPORTED “SECONDARY CONSIDERATIONS” ............................... 33
`
`IV. WORKMAN (EX. 1004) COMBINED WITH ELHOLM (EX. 1005) ........ 36
`
`V.
`
`THIS IPR IS NOT TIME-BARRED ............................................................. 37
`
`VI. CONCLUSION .............................................................................................. 40
`
`
`
`
`
`ii
`
`

`

`
`
`TABLE OF AUTHORITIES
`
`CASES
`
`AllVoice Computing PLC v. Nuance Commc’ns, Inc., 504 F.3d 1236
`(Fed. Cir. 2007) ..................................................................................................... 3
`
`Amgen Inc. v. Hoechst Marion Roussel, Inc., 314 F.3d 1313
`(Fed. Cir. 2003) ................................................................................................... 26
`
`Amkor Tech., Inc. v. Tessera, Inc., IPR2013-00242 ................................................ 39
`
`Apple, Inc. v. Achates Reference Publ’g, Inc., IPR2013-00080 .............................. 38
`
`Apple Inc. v. Motorola, Inc., 757 F.3d 1286 (Fed. Cir. 2014) ................................. 11
`
`Arthrocare Corp. v. Smith & Nephew, Inc., 406 F.3d 1365
`(Fed. Cir. 2005) ................................................................................................... 32
`
`Ashland Oil, Inc. v. Delta Resins & Refractories, Inc., 776 F.2d 281
`(Fed. Cir. 1985) ................................................................................................... 34
`
`Bristol-Myers Squibb Co. v. Ben Venue Labs., Inc., 246 F.3d 1368
`(Fed. Cir. 2001) ................................................................................................... 27
`
`Corning Glass Works v. Sumitomo Elec. U.S.A., Inc., 868 F.2d 1251
`(Fed. Cir. 1989) .................................................................................................. 13
`
`Facebook, Inc. v. Pragmatus AV, LLC, 582 F. App’x 864
`(Fed. Cir. 2014) ..................................................................................................... 3
`
`Gemstar-TV Guide Int’l, Inc. v. ITC, 383 F.3d 1352 (Fed. Cir. 2004) .................. 3, 7
`
`In re Cuozzo Speed Techs., LLC, 778 F.3d 1271 (Fed. Cir. 2015) ............................ 3
`
`In re Wiggins, 488 F.2d 538 (CCPA 1973) ............................................................. 33
`
`Institut Pasteur v. Focarino, 738 F.3d 1337 (Fed. Cir. 2013) ................................. 35
`
`Kennametal, Inc. v. Ingersoll Cutting Tool Co., 780 F.3d 1376
`(Fed. Cir. 2015) ................................................................................................... 35
`
`Liebel-Flarsheim Co. v. Medrad, Inc., 358 F.3d 898 (Fed. Cir. 2004) ..................... 8
`
`iii
`
`

`

`
`
`Linear Tech. Corp. v. ITC, 566 F.3d 1049 (Fed. Cir. 2009) ................................ 3, 14
`
`McCarty v. Lehigh Valley R. Co., 160 U.S. 110 (1895) .......................................... 13
`
`MEHL/Biophile Int’l Corp. v. Milgraum, 192 F.3d 1362
`(Fed. Cir. 1999) ............................................................................................. 25, 26
`
`Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005) (en banc) ....... 3, 6, 7, 13, 15
`
`Rodime PLC v. Seagate Tech., Inc., 174 F.3d 1294 (Fed. Cir. 1999) ....................... 3
`
`Shamrock Techs., Inc. v. Med. Sterilization, Inc., 903 F.2d 789
`(Fed. Cir. 1990) ............................................................................................. 38, 39
`
`Taylor v. Sturgell, 533 U.S. 880 (2008) ................................................................... 38
`
`Transclean Corp. v. Jiffy Lube Intern., Inc., 474 F.3d 1298
`(Fed. Cir. 2007) ................................................................................................... 38
`
`United States v. Sine, 493 F.3d 1021 (9th Cir. 2007) .............................................. 34
`
`Zoll Lifecor Corp. v. Philips Elecs. N. Am. Corp., IPR2013-00616 ....................... 35
`
`OTHER AUTHORITIES
`
`37 C.F.R. § 42.100(b) ................................................................................................ 2
`
`154 Cong. Rec. S9987 (Sept. 27, 2008) ................................................................... 39
`
`77 Fed. Reg. 48,759 (2012) ..................................................................................... 38
`
`77 Fed. Reg. at 48760 .............................................................................................. 37
`
`Fed. R. Evid. 105 ............................................................................................... 11, 31
`
`Fed. R. Evid. 801 ............................................................................................... 11, 31
`
`iv
`
`

`

`
`
`A comparison of the prior art Workman patent to the challenged claims of
`
`the ‘607 patent reveals exactly what PGS and its expert alleged and the Board’s
`
`Institution Decision preliminarily found: that Workman discloses every limitation
`
`of the claims and therefore anticipates. Recognizing that the Board’s conclusion
`
`flowed inexorably from comparing Workman to the ‘607 patent claims,
`
`WesternGeco (“WG”) looked beyond the claims for a different, more favorable
`
`comparison that could distinguish Workman. It found and seized on three solitary
`
`words in the specification, used only once and in the context of a preferred
`
`embodiment. Those words—“behavior-predictive model”—appearing nowhere in
`
`the claim and identified as a preferred but not necessary feature of the invention
`
`disclosed in the specification have become a talisman that WG deploys to ward off
`
`Workman at every turn and attempt to rescue the claims of the ‘607 patent.
`
`Unfortunately for WG, its gambit fails. Its effort to read the term “behavior-
`
`predictive model” into the claims is refuted by the language of the claims and the
`
`disclosure of the specification and traduces the most fundamental principles that
`
`guard against reading limitations into claims, especially under the governing
`
`broadest reasonable interpretation standard. It also is refuted—and conclusively
`
`so—by WG’s own statements interpreting and applying the ‘607 patent. In
`
`multiple proceedings in multiple fora over multiple years, WG has argued
`
`assiduously that the specification and claims of the ‘607 patent do not require a
`
`1
`
`

`

`
`
`behavior-predictive model. Until now. WG certainly deserves credit for creativity
`
`and temerity, if not candor or consistency.
`
`But WG had no choice. Under the correct interpretation of the claims
`
`adopted by the Board—which WG attempts to narrow so as to require a behavior-
`
`predictive model and all of its purported accoutrements—Workman is impossible
`
`to distinguish. WG attempted to assert that Workman does not disclose the
`
`limitations of predicting positions of streamer positioning devices or calculating
`
`changes on the basis of those predictions, but not even its own expert was willing
`
`to defend such a baseless interpretation of the reference. Accordingly, aside from
`
`its effort to read a behavior-predictive model into the claims, WG is left with the
`
`detritus of its invalidity arguments that (1) applying a legally erroneous standard to
`
`attack the enablement of its own Workman patent; (2) regurgitating its discredited
`
`conspiracy theories about ION and other purported real parties in interest, (3) and
`
`impugning PGS’s experts by mischaracterizing their testimony blatantly. All of
`
`these arguments are meritless.
`
`Because WG does not and cannot refute that Workman anticipates Claims 1
`
`and 15 of the ‘607 patent as properly construed, the Board should cancel them.
`
`I.
`
`CLAIM CONSTRUCTION
`
`Unexpired claims are given their “broadest reasonable construction in light
`
`of the specification” in an inter partes review proceeding. 37 C.F.R. § 42.100(b);
`
`2
`
`

`

`
`
`In re Cuozzo Speed Techs., LLC, 778 F.3d 1271, 1278-82 (Fed. Cir. 2015). “The
`
`broadest reasonable interpretation of a claim term may be the same as or broader
`
`than the construction of a term under the Phillips standard. But it cannot be
`
`narrower.” Facebook, Inc. v. Pragmatus AV, LLC, 582 F. App’x 864, 869 (Fed.
`
`Cir. 2014). Thus, the canons of claim construction that guard against reading
`
`claims too narrowly apply with equal or greater force in IPR proceedings.
`
`Two canons are especially apt here. The Federal Circuit has “repeatedly
`
`held that, even in situations when only one embodiment is disclosed, the claims
`
`generally should not be narrowed to cover only the disclosed embodiments or
`
`examples in the specification.” Linear Tech. Corp. v. ITC, 566 F.3d 1049, 1058
`
`(Fed. Cir. 2009). Provided that the specification is describing a preferred
`
`embodiment, even the statement that a feature is “required” will not justify reading
`
`such a feature into the claims. Gemstar-TV Guide Int’l, Inc. v. ITC, 383 F.3d 1352,
`
`1365-66 (Fed. Cir. 2004). The Federal Circuit has also warned that “each claim
`
`need not include every feature of an invention.” AllVoice Computing PLC v.
`
`Nuance Commc’ns, Inc., 504 F.3d 1236, 1248 (Fed. Cir. 2007). “Rather, a claim
`
`may specify improvements in one function without claiming the entire machine
`
`with its many functions.” Rodime PLC v. Seagate Tech., Inc., 174 F.3d 1294, 1303
`
`(Fed. Cir. 1999). These canons mandate rejection of WG’s interpretation of the
`
`claims to require unclaimed features that distinguish them from the prior art.
`
`3
`
`

`

`
`
`A. The Claim Language.
`Claim 15 reads, in its entirety:
`
`An array of seismic streamers towed by a towing vessel comprising:
`(a) a plurality of streamer positioning devices on or inline with each
`streamer;
`(b) a prediction unit adapted to predict positions of at least some of
`the streamer positioning devices; and
`(c) a control unit adapted to use the predicted positions to calculate
`desired changes in positions of one or more of the streamer
`positioning devices.
`
`Ex. 1001 at 12:26-34. (Associated method Claim 1 is substantially similar.)
`
`Claim 15 is directed to an array comprising three components, labeled (a),
`
`(b), and (c). It is undisputed that limitation (a) is well-known in the prior art, see
`
`Ex. 1001 at Fig. 1, and is taught by Workman, see Ex. 1004 at Fig. 1. There is also
`
`no dispute that Workman discloses a “prediction unit” (its “network solution
`
`system” labeled 10) and a “control unit (its “streamer control processor” labeled
`
`40). Ex. 1002 ¶¶ 113-14, 121-22. And, as the Board preliminarily found, the
`
`“prediction” and “control” units of Workman are adapted to perform the functions
`
`recited in claim 15, as construed by the Board. Id. ¶¶ 112-22; Paper 33 at 24-25.
`
`In violation of the canons discussed above, WG attempts to redraft the
`
`claim, to include requirements as to how limitation (b) “predicts positions” and
`
`how limitation (c) “calculates desired changes.” Paper 48 (“POR”) at 6-14.
`
`4
`
`

`

`
`
`The Board Correctly Interpreted “Predict Positions.”
`
`B.
`The Board’s Institution Decision construes “predicting positions” to require
`
`that the prediction unit be adapted to “estimate the actual locations” of at least
`
`some of the streamer positioning devices (“SPDs”). Paper 33 at 9. In arriving at
`
`this construction, the Board relied upon the ’607 patent’s disclosure that, due to the
`
`time delays inherent in measuring the positions of streamer array components, “the
`
`global control system 22 runs position predictor software to estimate the actual
`
`locations of each of the birds 18.” Id. (citing Ex. 1001 at 4:51-55). In other words,
`
`the ’607 patent’s prediction unit uses “old” location information to estimate
`
`current, “actual locations.” The Board’s analysis is correct and consistent with the
`
`BRI standard applied here. WG argues that the Board’s interpretation is
`
`“unreasonable” for two reasons, each of which is meritless.
`
`1.
`
`The Intrinsic Evidence Does Not Require a Specific
`Prediction Model.
`
`WG’s primary argument is that “predicting positions” must be done using a
`
`specific technique, a so-called “behavior-predictive model.” See POR at 6, 9-12.
`
`The ’607 patent’s specification does not elucidate the meaning of “behavior-
`
`predictive model,” and the only guidance WG provides is to say that it is “more
`
`sophisticated than simply estimating the actual locations.” Id. at 9-10. WG’s
`
`proposed construction is clearly wrong.
`
`5
`
`

`

`
`
`1.
`
`It is axiomatic that “the claims themselves provide substantial
`
`guidance as to the meaning of particular claim terms.” Phillips v. AWH Corp., 415
`
`F.3d 1303, 1314 (Fed. Cir. 2005) (en banc). Claim 15 requires a “prediction unit
`
`adapted to predict positions of at least some of the streamer positioning devices.”
`
`While the claim limits what must be predicted (at least two positions), it imposes
`
`no limit on how those positions are predicted, indicating—particularly under the
`
`applicable BRI standard— that it covers any technique for doing so. WG does not
`
`even reproduce the language of claim 15, let alone analyze it, see POR at 6-12, in
`
`utter disregard of the standard claim construction rubric. 415 F.3d at 1314-15.
`
`2.
`
`Rather than address the claim language, WG argues that the
`
`specification’s reference to an “inventive control system” in describing the use of a
`
`behavior-predictive model means that the claimed “prediction unit” requires a
`
`“behavior-predictive model.” POR at 9-12. This language comes from the only
`
`sentence in the specification that refers to a “behavior-predictive model,” which
`
`states: “To compensate for these localized current fluctuations, the inventive
`
`control system utilizes a distributed processing control architecture and
`
`behavior-predictive model-based control logic to properly control the streamer
`
`positioning devices.” Ex. 1001 at 4:16-20. WG’s Response replaced the bolded
`
`language with an ellipsis, POR at 10, to obscure that, per its interpretation, both a
`
`behavior-predictive model and a “distributed processing control architecture”
`
`6
`
`

`

`
`
`should be claim limitations—a construction that not even WG asserts is correct.
`
`Of course, the claims require neither of these features, because the sentence
`
`paraphrased by WG describes a preferred embodiment, and it is improper to limit
`
`claims to a preferred embodiment, even if only one such embodiment were
`
`disclosed. Gemstar, 383 F.3d at 1364; Phillips, 415 F.3d at 1314. The
`
`specification prefaces the discussion of this “inventive control system” by noting
`
`that it is describing the “preferred embodiment.” Ex. 1001 at 3:56-60. As WG has
`
`acknowledged, where the specification refers to the “inventive control system,” it
`
`“teaches that this description is merely a preferred embodiment.” Ex. 1102 at 7.
`
`Prior to filing its Response here, WG had been punctilious in correcting any
`
`suggestion that this one sentence in the specification requires the use of a behavior-
`
`predictive model. It told the EPO that “it would be entirely incorrect to infer from
`
`that sentence that the Skilled Person would consider that it is not possible ‘to
`
`properly control the streamer positioning devices’ without a behaviour-predictive
`
`model; there is no technical reason why this would be the case.” Ex. 1104 ¶ 4.10;
`
`see also id. ¶¶ 4.5-4.12 & Ex. 1093. Instead, WG asserted, the sentence merely
`
`“presents one way in which streamer positioning devices may be ‘properly
`
`controlled’, and does not imply that no other way is possible.” Id. ¶ 4.12. WG
`
`further explained that because the specification “preferably” uses a “dynamic
`
`model” (Ex. 1001 at 4:28-34), it makes “a positive disclosure that a dynamic
`
`7
`
`

`

`
`
`model, whether behaviour-predictive or otherwise, is not essential.” Ex. 1104
`
`¶ 4.11. WG’s factual characterization to the EPO of its specification’s disclosure is
`
`highly relevant to claim construction. Apple Inc. v. Motorola, Inc., 757 F.3d 1286,
`
`1312-13 (Fed. Cir. 2014). WG cannot assert that the broadest reasonable
`
`construction of these claims requires a “behavior-predictive model” when it has
`
`repeatedly argued that such a model is just part of a preferred embodiment.
`
`3.
`
`The prosecution history also confirms that the “behavior-predictive
`
`model” is preferred, but not required. While prosecuting the parent to the ’607
`
`patent, WG amended various claims to add the limitation “obtaining a predicted
`
`position of the streamer positioning devices.” Ex. 2067 at 2. It expressly told the
`
`Patent Office that this prediction is “preferably based upon the behavior of the
`
`streamers[.]” Id. at 7 (emphasis added). This is persuasive evidence that the
`
`claims do not require predicting positions based on a behavior-model. Liebel-
`
`Flarsheim Co. v. Medrad, Inc., 358 F.3d 898, 909 (Fed. Cir. 2004).
`
`4.
`
`The ’607 patent does not explain what makes a model “behavior-
`
`predictive.” See Ex. 1001 at 4:10-14. At his deposition, Dr. Triantafyllou
`
`explained a “behavior-predictive” model is based on the equations of motion for
`
`the streamers in a fluid. See Ex. 1091 at 165-170. He explained that such a model
`
`must account for the distribution of mass along the streamer, the tension of the
`
`streamer, the stiffness of the streamer, displacement of fluid by the streamer,
`
`8
`
`

`

`
`
`centrifugal forces, Coriolis forces, and drag forces, to name a few. See id. He
`
`testified that this list scratches only the “first sort of layer” of forces to be included
`
`in a “behavior-predictive model.” Id. at 169. A model that accounts for all these
`
`forces and more is required by WG’s proposed construction that rests on three
`
`words from the specification that were not included in the claim.
`
`It took Dr. Bittleston “years” to derive the equations that Dr. Triantafyllou
`
`deems essential to a “behavior-predictive” model. Id. at 170-71; see also Ex. 1092
`
`at 584-87. This modeling and the equations, however, appear nowhere in the
`
`patent. See generally Ex. 1001. It is inconceivable that the only reasonable
`
`interpretation of “predicting positions” requires such non-enabled complexity.
`
`2. WesternGeco Never Showed that ION Used a “Behavior-
`Predictive Model” To Predict Positions.
`
`Such an intricate interpretation of “predicting positions” is also astonishingly
`
`inconsistent with WG’s proof in the ION trial.
`
`
`
`
`
` His entire explanation for why the “Network
`
`Calculation Node” within ION’s ORCA software “predicts positions” consisted of
`
`quoting the testimony of an ION engineer that the node “essentially” predicts
`
`position using “estimated velocity” and “carrying it forward to the next point in
`
`time by simply applying the velocity times the time to get an updated position.”
`
`Id. ¶ 145. By contrast, in this proceeding, Dr. Triantafyllou insisted that estimating
`
`9
`
`

`

`
`
`positions using only velocity and prior positions is not a “behavior-predictive”
`
`model within the scope of the ‘607 patent claims, because it fails to account for the
`
`panoply of other factors identified above. Ex. 1091 at 165-70, 235-37.
`
`Confronted with this contradiction at his deposition, Dr. Triantafyllou was
`
`entirely unable to explain how the evidence he cited to show infringement could
`
`suffice under the claim construction he now proffers, see id. at 342-383.
`
` Dr. Triantafyllou identified
`
`his colleague, Dr. John Leonard, as the individual who analyzed ION’s software to
`
`determine whether it “predicted positions,” see Ex. 1091 at 346-47, 363-66.
`
`At trial, WG did proffer Dr. Leonard’s analysis to demonstrate that ION’s
`
`ORCA software “predicts positions.” Ex. 2055 at 1511-12. He explained that
`
`ORCA contained a Kalman filter, and he showed that it “predicted positions” by
`
`taking an object’s velocity multiplied by the elapsed time and adding that to its old
`
`position, id. at 1512-15, consistent with the Board’s interpretation here (and
`
`contrary to WG and Dr. Triantafyllou’s definition here, Ex. 1091 at 235-37). Dr.
`
`Leonard summarized this for the jury using the following demonstrative:
`
`
`
`10
`
`

`

`
`
`Ex. 1097 at 9 (excerpted); see also id. at 10 (analyzing source code). Had WG or
`
`Dr. Triantafyllou asserted before the jury—as it does here—that this method for
`
`estimating velocity is not even within the broadest reasonable construction of
`
`“predicting positions,” the jury could not have concluded that ION infringed the
`
`’607 patent.2 WG’s repeated reliance in this proceeding on the judgment in the
`
`ION case should be understood in the context of its lexical shell games.
`
`Present Estimates Based on Old Data Are “Predictions.”
`
`3.
`WG appears to suggest that the Board’s construction “eviscerates” the claim
`
`because “predicting positions” requires a determination of future positions, rather
`
`than using old data to generate estimates of current positions. See POR at 7-9
`
`(“And not all estimates are predictions because not all estimates include the
`
`
`2 WG’s “ION” POPR relies on ION’s EPO appeal argument that the claims are
`
`invalid under Article 76(1) of the European Patent Convention unless they are
`
`amended to include “control architecture and behavior-prediction.” Ex. 2154 at 3;
`
`Paper 71 at 32-33. Article 76(1) concerns the European law of sufficiency of
`
`disclosure, and delving into how it corresponds to U.S. claim construction law is
`
`precisely the type of arcana that the Federal Circuit in Apple instructed courts to
`
`avoid, making ION’s arguments applying that law irrelevant. 757 F.3d at 1312-13.
`
`Those arguments are also inadmissible hearsay as to PGS. Fed. R. Evid. 105, 801.
`
`11
`
`

`

`
`
`forward looking, temporal component that is a necessary part of all
`
`‘predictions.’”); see also id. at 7 (referring to prediction of where the SPD “will
`
`be”), 22 (arguing Workman does not predict if the streamer “will fall” outside a
`
`threshold). If this is WG’s position, it too is wrong.
`
`Using old data to estimate a current, actual position falls within the ordinary
`
`meaning of “prediction.” Ex. 1002 ¶ 99. Consistent with the Board’s BRI, the
`
`specification confirms that its “position predictor software” is used “to estimate the
`
`actual locations of each of the birds 18.” Ex. 1001 at 4:51-55 (emphasis added).
`
`WG cited this very sentence when it argued during the ION litigation that “the ’607
`
`patent teaches examples of ‘prediction’ that estimates present-time locations.” Ex.
`
`1100 at 13-14 (citing Ex. 1001 at 4:51-55) (emphasis added). WG criticized any
`
`effort to equate “predicting positions” with future positions on the basis that doing
`
`so “would improperly read disclosed embodiments out of the scope of the claims,”
`
`id. at 14, and the district court agreed. Ex. 1022 at 18. Because the BRI of
`
`“predicting positions” must embrace this embodiment, Google, Inc. v. Whitserve
`
`LLC, IPR2013-00249, Paper 32 at 12 (PTAB Sep. 9, 2014), the Board’s
`
`construction was correct.
`
`C.
`
`“Calculating Desired Changes” in Position Cannot Require
`Calculating Forces or a Particular Method of Calculating Forces.
`
`WG’s next seeks to distinguish Claim 15 from Workman by rewriting the
`
`claim language, “a control unit adapted to use the predicted positions to calculate
`
`12
`
`

`

`
`
`desired changes in positions of one or more of the streamer positioning devices.”
`
`WG construes “calculate desired changes” as “determine forces based on streamer
`
`and array behavior.” POR at 12-14. The claim specifies what is calculated
`
`(“desired changes in positions”) and on what basis (“us[ing] the predicted
`
`positions”). WG ’s construction contravenes this clear language by imposing new
`
`requirements on what is calculated (“forces”) and how the calculation is performed
`
`(“based on streamer and array behavior”). It should be rejected.
`
`The Claim Says To Calculate Positions, Not Forces.
`
`1.
`The plain language of the claim precludes WG’s proposed construction
`
`because “positions” and “forces” are words with entirely different meanings. See
`
`Ex. 1001 at 6:27-56; Phillips, 415 F.3d at 1314-15. WG is not so much construing
`
`“calculating desired changes in position” as it is seeking to redraft claim 15 to
`
`change what is calculated from “desired changes in position” to “forces”—
`
`something the law has long deemed impermissible. Corning Glass Works v.
`
`Sumitomo Elec. U.S.A., Inc., 868 F.2d 1251, 1257 (Fed. Cir. 1989) (“a court may
`
`not redraft a claim for purposes of avoiding a defense of anticipation,” citing
`
`McCarty v. Lehigh Valley R. Co., 160 U.S. 110, 116 (1895)).
`
`Nor does the specification support an effort to substitute the calculation of
`
`forces for the calculation of desired changes in position. The passages cited by
`
`WG that refer to calculation of forces state expressly that this is only a preference:
`
`13
`
`

`

`
`
`the global control system “preferably calculates the desired vertical and horizontal
`
`forces” and “preferably maintains a dynamic model . . . to regularly calculate
`
`updated desired vertical and horizontal forces.” Ex. 1001 at 4:28-34, 4:48-51
`
`(emphases added). Meanwhile, the ’607 patent clear states that in the alternative
`
`“the global control system 22 can transmit location information to the local control
`
`system 36 instead of force information.” Id. at 6:27-48 (emphasis added). This
`
`explicit statement that the control system need not calculate forces, renders the
`
`importation of a “force” limitation into claim 15 especially inappropriate.
`
`2.
`
`Calculations “Based on Streamer and Array Behavior” Are
`Not Required.
`
`WG’s attempt to insert the requirement that “calculating desired changes” be
`
`based on “the behavior of each of the streamers and also the behavior of the
`
`complete streamer array,” POR at 13, is also improper. WG’s argument rests on
`
`the same two specification passages discussed above with respect to calculating
`
`forces, see id., and they remain directed explicitly to what the calculation
`
`“preferably” entails. Ex. 1001 at 4:28-34, 4:48-51. The passages discuss a
`
`preferred embodiment, and there is no basis to read these features into the claim.
`
`Linear Tech., 566 F.3d at 1058.
`
`WG’s proposed requirement that the calculation of claim 15 account for “the
`
`behavior of each streamer” and “the behavior of the complete streamer array,”
`
`POR at 13, also does violence to the plain meaning of the claim. Limitation (b)
`
`14
`
`

`

`
`
`makes clear that the “prediction unit” need only predict the positions of “at least
`
`some” (i.e., at least two) of the SPDs. The control unit of limitation (c) explicitly
`
`is “adapted to use” those two or more positions predicted in limitation (b). By
`
`proposing a construction that requires the calculation to be “based on each of the
`
`streamers and the complete streamer array behavior,” WG requires the prediction
`
`unit to have predicted the positions of “each streamer” in the array—far more than
`
`the minimum two elements that the claim actually requires and that the control unit
`
`is “adapted to use.” WG’s proposal to require accounting for the behavior of each
`
`streamer and the complete streamer array is inconsistent with the language in claim
`
`15 and must be rejected. Phillips, 415 F.3d at 1314-15.
`
`D. Claim 15 Does Not Require the Panoply of Other Features WG
`Attempts to Insert.
`
`In addition to proposing its improper constructions of “predict positions”
`
`and “calculate desired changes to positions,” WG informally attempts to introduce
`
`various limitations to distinguish Workman, without even pretending to justify
`
`them under a proper claim construction analysis.
`
`WG suggests that claim 15’s system must be “continuous.” POR at 30
`
`(“The continuous and automated system is captured in claims 1 and 15 . . .”). But
`
`claim 15’s “control unit” is defined by a single feature: it is adapted to “calculate
`
`desired changes in positions of one or more of the streamer positioning devices.”
`
`Ex. 1001 at 12:32-34. Nothing in this language requires that it perform
`
`15
`
`

`

`
`
`calculations with a certain frequency, let alone that it do so “continuously,” every
`
`second, or even every time a source is fired. Nor would such requirements be
`
`consistent with the specification’s exemplary system, which transmits commands
`
`from the global control system to a bird’s local control system once every 10
`
`seconds. See id. at 9:9-16 (explaining that local control system runs at 10 Hz,
`
`which is up to 100 times faster than global-to-local communication).
`
` WG also argues that claim 15 requires “active control.” See POR at 14, 30,
`
`33, 40. Again, neither the specification nor the claims use the word “active,” much
`
`less “active control.” The ’607 patent purports to distinguish itself from the prior
`
`art on the basis that it involves less steering than prior art systems, not “active”
`
`steering. See id. at 4:34-47. There is no basis for concluding that the only
`
`reasonable construction of the claims requires reading in “active control.”
`
`Finally, WG argues that claim 15 does not only require a “behavior-
`
`predictive model,” but that the model must be “proactive.” POR at 23. The word
`
`“proactive” appears nowhere in the specification, let alone in claim 15. Ex. 1001.
`
`WG nowhere explains what makes a behavior-predictive model