435221US
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`———————————
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`———————————
`
`PETROLEUM GEO-SERVICES INC.
`Petitioner
`
`v.
`
`WESTERNGECO LLC
`Patent Owner
`
`———————————
`
`Case IPR2014-00688
`U.S. Patent No. 7,080,607
`
`———————————
`
`
`
`PATENT OWNER RESPONSE
`
`Pursuant to 37 C.F.R. § 42.120, Patent Owner, WesternGeco L.L.C
`
`(“WesternGeco” or “Patent Owner”), submits this Response to the Petition for
`
`Inter Partes Review (“Petition”) of U.S. Patent No. 7,080,607 (the “’607 patent”)
`
`filed by Petitioner, Petroleum Geo-Services, Inc. (“PGS” or “Petitioner”).
`
`

`

`
`
`I.
`
`TABLE OF CONTENTS
`
`THE ’607 PATENT CLAIMS PREDICTIVE STEERING OF
`STREAMER ARRAYS ................................................................................... 1
`
`II.
`
`CLAIM CONSTRUCTION ............................................................................ 6
`
`A.
`
`B.
`
`Predicting Positions of at Least Some of the Streamer
`Positioning Devices ............................................................................... 6
`
`Using the Predicted Positions to Calculate Desired Changes ............. 12
`
`III.
`
`PRIOR ART ................................................................................................... 14
`
`A. Workman ............................................................................................. 14
`
`B.
`
`Elholm ................................................................................................. 19
`
`IV. THE ’607 PATENT IS NOT ANTICIPATED OR OBVIOUS .................... 20
`
`A.
`
`Claim 1 and 15 Are Not Anticipated By Workman ............................ 20
`
`i. Workman Does Not Disclose The Predicting Positions
`Limitations of Claims 1 and 15 ................................................. 21
`
`ii. Workman Does Not Disclose Utilizing The Predicted
`Positions Of The Streamer Positioning Devices To
`Calculate Desired Changes In Position ..................................... 27
`
`B.
`
`C.
`
`iii. Workman Is Not Enabled .......................................................... 30
`
`Claims 1 And 15 Of The ’607 Patent Are Not Obvious Over
`Workman ............................................................................................. 34
`
`Claims 1 and 15 Are Not Obvious In View Of Workman and
`Elholm. ................................................................................................ 40
`
`V.
`
`PETITIONER’S EXPERTS ARE NOT CREDIBLE ................................... 43
`
`A. Dr. Evans ............................................................................................. 43
`
`i.
`
`Dr. Evans Made Fundamental Errors In His Analysis ............. 43
`
`B.
`
`Dr. Cole ............................................................................................... 46
`
`i
`
`

`

`
`
`VI. OBJECTIVE INDICIA OF NONOBVIOUSNESS ...................................... 47
`
`VII. THE PETITION IS TIME-BARRED UNDER 35 U.S.C. § 315(b) ............. 50
`
`A.
`
`ION is an Unnamed RPI ...................................................................... 50
`
`i.
`
`ii.
`
`ION’s Involvement with Petitioner ........................................... 52
`
`.............................................. 50
`
`iii.
`
`ION is an RPI Under the Guidelines ........................................ 54
`
`iv. Additional Discovery was Prejudicially Denied ....................... 56
`
`B. Multi Klient Invest AS is an RPI ........................................................ 57
`
`C.
`
`Service ................................................................................................. 59
`
`VIII. CONCLUSION .............................................................................................. 60
`
`
`
`
`
`ii
`
`
`

`

`
`
`I.
`
`THE ’607 PATENT CLAIMS PREDICTIVE STEERING OF
`STREAMER ARRAYS
`The ’607 patent covers methods and apparatus for using predicted positions
`
`of streamer positioning devices to calculate steering commands in order to steer
`
`miles-long streamer arrays despite limited location data in
`
`order to better image geological structures, improve the
`
`streamers’ effectiveness, repeat surveys over time to manage
`
`resource recovery, and more safely and rapidly deploy and
`
`turn the arrays.
`
`Early streamer positioning involved rudimentary
`
`devices such as deflectors and tail buoys. (Ex. 1001, 3:34-39;
`
`Fig. 1 elements (16) and (20, respectively)1. Deflectors were
`
`associated with the front end of the streamer and used to
`
`
`1 Although Figure 1 of the ’607 patent is captioned as “prior art,” one of ordinary
`
`skill would recognize that much of that figure was in fact not prior art, but
`
`instead constituted inventive contributions to the state of the art, such as the
`
`global control system, its functionality (e.g., predictive analysis, control modes,
`
`streamer positioning device control, etc.), and the distributed processing control
`
`architecture. (Ex. 2042, ¶ 67).
`
`1
`
`

`

`
`
`horizontally spread the end of the streamer nearest the seismic survey vessel. (Id.
`
`at 1:34-41.) Tail buoys were associated with ropes or cables secured to the end of
`
`the streamer furthest from the seismic survey vessel, and created drag on that end
`
`of the streamer. (Id. at 1:39-41; 3:37-39.) The tension created on the seismic
`
`streamer by the deflector and tail buoy resulted in a roughly linear shape. (Id. at
`
`1:34-41.) Tail buoys floated at the surface and could rely on GPS to determine
`
`their positions. Deflectors attached to the front of the array and created fixed
`
`spacing through tension at front of the system. No steering or lateral forces were
`
`provided for the miles of length along the streamer.
`
`Streamer positioning devices are generally spaced every 200 to 400 meters
`
`along the length of a streamer. (Ex. 1001, 1:48-49.) For a modest streamer array,
`
`this means hundreds of separate streamer positioning devices are deployed on a
`
`given array. Simultaneously controlling this multitude of independent positioning
`
`devices is no easy feat. While it is easy to set a target depth and little risk exists if
`
`that depth is overshot, lateral steering requires considerations of the dynamical
`
`movement of neighboring streamers and obstructions along miles of cable
`
`deployed in the ever-changing open-water environment of the deep seas. Unless
`
`properly controlled, lateral steering can make streamer arrays more dangerous than
`
`no steering at all.
`
`The complexity of these streamer arrays led to several widely
`
`2
`
`
`

`

`
`
`acknowledged, decades-old problems, including the risk of tangling, a potentially
`
`catastrophic and dangerous failure. (Ex. 1001, 4:5-7.) Movement of the streamers
`
`relative to each other during surveys can lead to gaps in coverage, requiring
`
`repeated passes, or “in-fill,” over the same section of water. And turning such long
`
`arrays in the water can take significant time and effort, and likewise increases the
`
`risk of tangling. Despite a well-known need for the ability to accurately steer these
`
`arrays, the complex nature of the problem prevented a workable solution from
`
`being developed for many years.
`
`Compounding these difficulties were the challenges of even knowing where
`
`the streamers were during surveys. These miles-long cables towed underwater in
`
`the harsh conditions of open deep-water had minimal sensory equipment and long
`
`signal time delay in communicating data to the towing vessel. While GPS could be
`
`relied upon to determine the positions of tail buoys or deflectors floating on the
`
`ocean surface, it could do little to assess the position of streamers, which are towed
`
`underwater. Trying to steer these streamers during a survey would essentially be
`
`“driving blind,” and attempts to steer could prove more dangerous than towing
`
`with no steering at all. (See, e.g., Ex. 1001, 4:5-7 (“[i]f the birds 18 are not
`
`properly controlled, horizontal steering can increase, rather than decrease, the
`
`likelihood of tangling adjacent streamers”).) As taught in Petitioner’s art, prior to
`
`Patentee’s inventions, determining positions accurately could only be
`
`3
`
`
`

`

`
`
`accomplished long after a survey was completed, using intensive computational
`
`processes.
`
`Further, as explained in the background section of the ’607 patent, seismic
`
`arrays typically allow for the determinations of horizontal positions of the
`
`streamers only every 5 to 10 seconds. (Ex. 1001, 2:35-38.) And because complex
`
`data processing is often involved, there may be an additional 5-second delay
`
`between taking of the measurements and the determination of actual streamer
`
`positions. Id. That means the information provided to the control system is not
`
`where the streamer positioning device currently is, but where the streamer
`
`positioning device was at some time in the past. And this location data was
`
`typically obtained from compass measurements, which were aggregated to try to
`
`determine the rough shape of a streamer.
`
`This latency problem associated with lateral positioning was not solved—or
`
`even appreciated—by the prior art. (See, e.g., Ex. 1001, 2:40-43 (“the delay period
`
`and the relatively long cycle time between position measurements prevents this
`
`type of [prior art] control system from rapidly and efficiently controlling the
`
`horizontal position of the bird”).) Rather, precise position information was only
`
`needed long after the survey was completed and during on-shore data processing,
`
`to interpret and locate the reflection energy that was measured during the survey,
`
`not during the acquisition itself.
`
`4
`
`
`

`

`
`
`Thus, while the goal of streamer steering was long-known, it went
`
`unresolved for decades. WesternGeco’s patents solved the problems associated
`
`with prior art methodologies by (1) developing predictive lateral steering (as
`
`opposed to reactionary control systems such as that disclosed by Petitioner’s prior
`
`art); (2) using distributed control to apportion intelligence between ship-board and
`
`local control systems; and (3) using global steering modes rather than simply
`
`setting control targets for individual streamers.
`
`In particular, the solution described in the ’607 patent is to use a behavior-
`
`predictive control system to remedy the delays inherent in positional measurement
`
`and data acquisition, and to dynamically steer the seismic streamer array:
`
`“Due to the relatively low sample rate and time delay associated with
`the horizontal position determination system, the global control
`system 22 runs position predictor software to estimate the actual
`locations of each of the birds 18.” (Ex. 1001, 4:51-55 (emphasis
`added).)
`
`“[T]he inventive control system utilizes . . . behavior-predictive
`model-based control logic to properly control the streamer positioning
`devices.” (Ex. 1001, 4:10-14 (emphasis added).)
`
`“The global control system 22 preferably calculates the desired
`vertical and horizontal forces based on the behavior of each streamer
`and also takes into account the behavior of the complete streamer
`array.” (Ex. 1001, 4:48-51 (emphasis added).)
`
`5
`
`
`

`

`
`
`“The global control system 22 preferably maintains a dynamic model
`of each of the seismic streamers 12 and utilizes the desired and actual
`positions of the birds 18 to regularly calculate updated desired
`vertical and horizontal forces the birds should impart on the seismic
`streamers 12 to move them from their actual positions to their desired
`positions.” (Ex. 1001, 4:28-34 (emphasis added).)
`
`II. CLAIM CONSTRUCTION
`Consistent with the statute and legislative history of the Leahy-Smith
`
`America Invents Act, Pub. L. No. 112-29, 125 Stat. 284 (2011) (“AIA”), the Board
`
`interprets claims using the “broadest reasonable construction in light of the
`
`specification of the patent in which [they] appear[].” 37 C.F.R. § 42.100(b); see
`
`also Office Patent Trial Practice Guide, 77 Fed. Reg. 48,756, 48,766 (Aug. 14,
`
`2012). There is a “heavy presumption” that a claim term carries its ordinary and
`
`customary meaning. CCS Fitness, Inc. v. Brunswick Corp., 288 F.3d 1359, 1366
`
`(Fed. Cir. 2002). However, a “claim term will not receive its ordinary meaning if
`
`the patentee acted as his own lexicographer and clearly set forth a definition of the
`
`disputed claim term in either the specification or prosecution history.” Id.
`
`A.
`
`Predicting Positions of at Least Some of the Streamer Positioning
`Devices
`The broadest reasonable construction of the term “predicting positions” of at
`
`least some of the streamer positioning devices is “determining positions using a
`
`behavior-predictive model” of at least some of the streamer positioning devices.
`
`6
`
`
`

`

`
`
`The intrinsic evidence requires that “predicting positions” addresses the time
`
`lag between positional measurements and steering commands arriving at the
`
`streamer positioning device, as well as the forces acting on the streamer. (Ex.
`
`1001, 4:8-14, 4:48-55, 5:4-16; Ex. 2042, ¶¶ 40, 57, 64-65, 87-88.) The Board
`
`preliminarily construed “predicting positions” as “estimating the actual locations,”
`
`which is improper because it ignores the temporal aspect of a “prediction” and
`
`because it ignores the intrinsic evidence regarding behavior prediction. For
`
`example, the plain and ordinary meaning of the word “predicting” or “prediction,”
`
`requires a temporal aspect: “an inference regarding a future event based on
`
`probability theory” which the preliminary claim construction does not address.
`
`(Ex. 2066.)
`
`Predictive control as recited in claims 1 and 15 requires a prediction as to (1)
`
`where the streamer positioning device will be at the time when commands are
`
`received at the device and (2) taking into consideration the forces acting on the
`
`streamer. (Ex. 2042, ¶¶ 40, 64-65, 87-88.) Trying to steer streamers with simple
`
`positional estimates, rather than time-adjusted positional data considering the
`
`behavior of the system as a whole, could prove more dangerous than towing with
`
`no steering at all. (See, e.g., Ex. 1001, 4:5-7.) And not all estimates are
`
`predictions because not all estimates include the forward looking, temporal
`
`component that is a necessary part of all “predictions.” By simply equating the
`
`7
`
`
`

`

`
`
`term “predicting positions of at least some of the streamer positioning devices” to
`
`estimated actual locations, the construction ignores the time component that is
`
`associated with the term and almost completely eviscerates the word “predicting.”
`
`(Ex. 2042, ¶ 65.)
`
`Time Lag. As discussed in the background section of the ’607 patent,
`
`seismic arrays typically allow for the determinations of horizontal positions of the
`
`streamers only every 5 to 10 seconds. (Ex. 1001, 2:35-38.) And because complex
`
`data processing is often involved, there may be an additional 5-second delay
`
`between taking of the measurements and the determination of actual streamer
`
`positions. Id. That means the information provided to the control system on the
`
`streamer positioning device is not where that device currently is, but where the
`
`device was at some time in the past. Patent Owner recognized this problem and
`
`solved it using the prediction methodology disclosed and claimed in the ’607
`
`patent to predict where the streamer positioning device will be when it receives the
`
`control signals. (See, e.g., Ex. 1001, 4:51-55.)
`
`The prosecution history confirms that the term “prediction” addresses time
`
`delays and was added to the claims to distinguish prior art on that basis. For
`
`example, during prosecution of the parent application to the ’607 patent (U.S.
`
`Patent No. 6,932,017 (“the ’017 patent”)), and in response to a rejection based
`
`8
`
`
`

`

`
`
`upon Elholm,2 Patent Owner amended the claims of the ’017 patent to recite
`
`“obtaining a predicted position of the streamer positioning devices.” (Ex. 2067, at
`
`2.) Further, Patent Owner’s response indicated that Elholm failed to disclose or
`
`suggest “predicting the position of streamer positioning devices to avoid delays
`
`inherent in actual position measurements,” and using predicted positions to
`
`calculate desired changes in the streamer positioning devices. (Ex. 2067, at 8.) A
`
`notice of allowance followed.
`
`Accordingly, the intrinsic evidence clearly indicates that the term
`
`“predicting positions” or “predicted positions” includes a future looking time
`
`component that is meant to address the “delays inherent in actual position
`
`measurements,” and is not simply an estimated position. See, e.g., Microsoft Corp.
`
`v. Multi-Tech Sys., Inc., 357 F.3d 1340, 1349-50 (Fed. Cir. 2004) (“the prosecution
`
`history of one patent is relevant to an understanding of the scope of a common
`
`term in a second patent stemming from the same parent application”) (citation
`
`omitted).
`
`Dynamic Forces. The specification also makes clear that the invention
`
`requires a certain kind of prediction: “behavior prediction.” Behavior prediction is
`
`
`2 European Patent No. EP 613025, which is the European counterpart to Elholm.
`
`9
`
`
`

`

`
`
`more sophisticated than simply estimating the actual locations. The Board
`
`correctly cited to the ’607 patent in recognizing in the institution decision:
`
`the
`influence
`Localized current fluctuations can dramatically
`magnitude of the side control required to properly position the
`streamers. To compensate for these localized current fluctuations, the
`inventive control system utilizes . . . behavior-predictive model-based
`control logic to properly control the streamer positioning devices.
`
`(Ex. 1001, 4:11-14, emphasis added.) That the ’607 patent’s “prediction” is
`
`“behavior prediction” is well supported by the specification. (Ex. 1001, 4:28-33
`
`(“The global control system 22 preferably maintains a dynamic model of each of
`
`the seismic streamers 12 and utilizes the desired and actual positions of the birds
`
`18 to regularly calculate updated desired vertical and horizontal forces the bird
`
`should impart on the seismic streamers 12 to move them from their actual positions
`
`to their desired positions.”)); (Ex. 1001, 4:48-51 (“The global control system 22
`
`preferably calculates the desired vertical and horizontal forces based on the
`
`behavior of each streamer and also takes into account the behavior of the complete
`
`streamer array.”)) (Ex. 2042, ¶¶ 59-65.)
`
`Thus, the ’607 patent explains that a “behavior-predictive model-based”
`
`control logic is necessary to properly control the streamers through various
`
`dynamic forces, such as localized current fluctuations. However, localized current
`
`fluctuations are not the only dynamic forces acting upon the streamer. When a
`
`10
`
`
`

`

`
`
`streamer positioning device is steered, it imparts a force on the streamer that
`
`propagates along the streamer towards the tail. (Ex. 2042, ¶¶ 36, 64-65.) If these
`
`traveling disturbances are not considered, streamer steering can be wildly
`
`inaccurate, resulting in more harm than good. (Id.) The claimed “predict[ed]
`
`positions” solves this problem by incorporating a behavior-predictive model-based
`
`control logic that predicts the effects of natural forces and that predicts the effects
`
`of the forces resulting from operation of the system itself, such as the effects that
`
`controlling one streaming positioning device will have on others. (Ex. 1001, 4:10-
`
`14.) (Ex. 2042, ¶¶ 59-65, 87-88.)
`
`Claims 1 and 15 of the ’607 patent must be read in this context, particularly
`
`because the specification makes clear that “the inventive control system”
`
`necessarily includes behavior prediction. (Ex. 1001, 4:11-14.) See Honeywell
`
`Int'l, Inc. v. ITT Indus., Inc., 452 F.3d 1312, 1318 (Fed. Cir. 2006) (explaining the
`
`importance of reading the claims in the context of the specification, especially
`
`when the specification describes “this invention” or “the present invention”);
`
`Verizon Servs. Corp. v. Vonage Holdings Corp., 503 F.3d 1312, 1318 (Fed. Cir.
`
`2006) (“when a patent thus describes the features of the ‘present invention’ as a
`
`whole, this description limits the scope of the invention”); Chimie v. PPG Indus.,
`
`11
`
`
`

`

`
`
`Inc., 402 F.3d 1371, 1379 (Fed. Cir. 2005).3
`
`Therefore, the broadest reasonable construction of “predicting positions” of
`
`at least some of the streamer positioning devices is “determining positions using a
`
`behavior-predictive model” of at least some of the streamer positioning devices.
`
`(Ex. 2042, ¶¶ 87-88.)
`
`B. Using the Predicted Positions to Calculate Desired Changes
`The broadest reasonable construction of the term “calculate desired changes”
`
`in position of one or more of the streamer positioning devices (claims 1 and 15),
`
`requires the desired changes to take into account not only the streamer on which
`
`the streamer positioning device is located, but also the complete streamer array.
`
`Specifically, the broadest reasonable construction of “calculate desired changes” is
`
`“determine forces based on streamer and array behavior.”
`
`The “predict[ed] positions” limitation of claims 1 and 15 (element (b))
`
`considers the dynamic forces acting upon the streamer, and when calculating the
`
`
`3 The ’607 patent also maintains a dynamic model of the streamer, i.e., one that
`
`changes over time, further suggesting a temporal component to the proper
`
`construction of “predicted positions” and further suggesting behavioral-based
`
`predictions. (Ex. 1001, 4:28-29 (“The global control system 22 preferably
`
`maintains a dynamic model of each of the seismic streamers…”).)
`
`12
`
`
`

`

`
`
`desired horizontal forces and vertical forces to effectuate those predicted positions,
`
`the “calculating desired changes” step of element (c) considers both the streamer
`
`behavior as well as the behavior of the complete streamer array. (Ex. 2042, ¶ 89.)
`
`The ’607 patent plainly states this requirement: The global control system 22
`
`preferably calculates the desired vertical and horizontal forces based on the
`
`behavior of each streamer and also takes into account the behavior of the
`
`complete streamer array. (Ex. 1001, 4:48-51 (emphasis added); see also 4:29-34
`
`(a “dynamic mode of each of the seismic streamers 12” is utilized to “calculate
`
`updated desired vertical and horizontal forces the birds should impart on the
`
`seismic streamers 12 to move them from their actual positions to their desired
`
`positions”); Ex. 1001, 4:10-14.) In other words, after predicting where the
`
`streamer positioning device will be when a command signal is sent based on
`
`streamer behavior (element (b)), any calculated change to that streamer positioning
`
`device must not only be based upon the behavior of the streamer, but also the
`
`behavior of the entire streamer array (element (c)). (Ex. 2042, ¶¶ 89-90.)
`
`Moreover, this construction is dictated by the language of claims 1 and 15.
`
`Specifically, element (b) requires “predicting positions of at least some of the
`
`streamer positioning devices” and then element (c) uses those predicted positions
`
`to calculate desired changes in one or more of the streamer positioning devices. In
`
`other words, where element (b) predicts positions of some, element (c) calculates
`
`13
`
`
`

`

`
`
`the desired changes to possibly one, showing the predicted positions of more than
`
`one streamer positioning device is considered when calculating the desired changes
`
`to a single streamer positioning device. Accordingly, the desired changes of one
`
`streamer positioning device is calculated based on the behavior of the streamer on
`
`which it resides as well as the behavior of the complete streamer array.
`
`III. PRIOR ART
`A. Workman
`Workman was considered during prosecution of the ’607 patent, and claims
`
`1 and 15 were determined valid during the ION litigation in view of this reference.
`
`Thus, this very Office as well as an Article III court have found that Workman
`
`does not render the challenged claims unpatentable. This Board should not reach a
`
`contrary result. In re Baxter Int’l, Inc., 678 F.3d 1357, 1365 (Fed. Cir. 2012)
`
`(“When a party who has lost in a court proceeding challenging a patent, from
`
`which no additional appeal is possible, provokes a reexamination in the PTO, using
`
`the same presentations and arguments, even with a more lenient standard of proof,
`
`the PTO ideally should not arrive at a different conclusion.”)
`
`In any event, Workman teaches away from the claimed invention. Workman
`
`is not concerned with active or behavior-predictive positioning of streamers or
`
`streamer positioning devices. Rather, Workman focuses on the noise caused by
`
`devices on seismic streamers (Ex. 1004, 1:63 – 2:9) and reactively repositioning
`
`the devices only after certain “threshold parameters” are violated. (Ex. 1004, 3:58
`14
`
`
`

`

`
`
`– 4:8.) Workman does not teach a behavior-predictive positioning system and is
`
`unconcerned with calculating a desired position for any devices based upon where
`
`the device will be, or is assumed to be, at the time of the command as taught and
`
`claimed by the ’607 patent. (Ex. 2042, ¶¶ 98, 104-109.)
`
`Specifically, Workman uses a “threshold parameter” to provide for a limited
`
`amount of device control when confronted with an emergency situation. However,
`
`if the threshold parameter is not exceeded, e.g., the streamers are not in immediate
`
`risk of tangling or collision, then the system restarts and no commands are sent to
`
`any devices in the system. (Ex. 1004, 4:22-35, Fig. 3.) In other words, the system
`
`disclosed by Workman does nothing in this scenario. (Ex. 1004, 4:31-35.)
`
`If a threshold parameter is exceeded, a correction command will be sent to a
`
`device if (a) the towed streamer cable encounters an “at risk” situation, or (b)
`
`hydrophone noise level is sufficiently low. (Ex. 1004, 4:36-58, Fig. 3.) If the
`
`streamers are not “at risk,” and the hydrophone noise level exceeds the maximum
`
`allowable noise threshold, the system disclosed by Workman does nothing, despite
`
`a threshold parameter being exceeded. (Ex. 1004, 5:14-30, Fig. 3.). In other
`
`words, where the ’607 patent chooses precise streamer positioning to reduce noise,
`
`Workman chooses the opposite: to reduce noise by not repositioning streamer
`
`positioning devices. (Ex. 1001, 4:34-55; Ex. 1004, Fig. 3.)
`
`15
`
`
`

`

`
`
`Unlike the claims of the ’607 patent that require behavior-predictive
`
`positioning, the system disclosed by Workman is entirely reactionary. Workman
`
`does not predict the shape of a streamer and whether that predicted shape is outside
`
`of the described threshold parameters. Instead, Workman’s control system only
`
`determines if a correction needs to be made after comparing the “real time”
`
`positional data sent from the location sensing devices (i.e., the last signal sent from
`
`location sensing devices). (Ex. 1004, 3:46-62.) There simply is no prediction of
`
`streamer positioning device locations, let alone any behavior-predictive model in
`
`Workman. (Ex. 2042, ¶¶ 98, 104-109.)
`
`The distinction between the claimed predictive control system and
`
`Workman’s reactionary system is best understood by way of an example. As
`
`explained above, by the time a corrective control signal is sent to an out-of-place
`
`streamer positioning device, 10 to 15 seconds (or more) may have elapsed since
`
`the last signal transmission from the streamer positioning device. (Ex. 1001, 2:35-
`
`38.) Multiple external forces are also acting on the streamer during this time. The
`
`’607 patent solves this problem by sending a corrective control signal to the
`
`streamer positioning device based upon where the system predicts the device will
`
`be, not where the device was 10 to 15 seconds ago, and considering the dynamic
`
`forces acting on the streamer. (Ex. 2042, ¶¶ 59-65.)
`
`16
`
`
`

`

`
`
`However, in Workman, the system only sends a corrective control signal if a
`
`threshold value was already exceeded (assuming that noise is sufficiently low,
`
`otherwise Workman does nothing). Any corrective signal sent to the streamer
`
`positioning device will be based upon positional data that is 10 to 15 seconds old.
`
`(Ex. 2042, ¶ 98.) Thus, Workman may have to go through multiple rounds of
`
`trying to correct a threshold violation, and may significantly overshoot the
`
`threshold in the other direction, causing more corrections and increasing noise.
`
`But this is all that was required to satisfy the goals addressed by Workman, i.e.,
`
`avoiding emergencies such as tangles and collisions.4 There is also no indication
`
`in Workman that corrective signals are based upon the behavior of the system,
`
`taking into account the forces acting on the streamer. (Ex. 2042, ¶¶ 98, 104-109.)
`
`Further still, Workman is inapposite because it focuses on the location of
`
`streamers and sources, but not on the location of streamer positioning devices.
`
`This is consistent with the goals of Workman—avoiding the violation of threshold
`
`parameters requires only general approximations of where the streamers are,
`
`whereas the steering of the ’607 patent requires knowledge of where each steering
`
`device is exerting a force because the force will ripple down the length of the
`
`4 The ’607 patent, on the other hand, specifically avoids overshooting intended
`
`positions. (Ex. 1001, 4:34-47.)
`
`17
`
`
`

`

`
`
`streamer. (Ex. 2042, ¶¶ 36, 64-65.) There is no indication that Workman’s
`
`location sensors 15 are associated with the streamer positioning devices 14 in
`
`Workman.
`
`
`
`As shown in the excerpted image of Fig. 1, above, label 13 represents the seismic
`
`cable, label 14 depicts Workman’s streamer positioning devices, and label 15
`
`represents location sensing devices. (Ex. 1004, 2:64 – 3:29.) The reference gives
`
`no indication that the location sensing devices 15 are associated with Workman’s
`
`streamer positioning devices, each of which are merely a few feet long and
`
`positioned along a cable measuring three to eight kilometers. (Ex. 1004, 1:20-24.)
`
`Indeed, Workman’s own description of Figure 1 confirms its approximate nature:
`
`“FIG. 1 shows a generalized schematic of a marine seismic survey system.” (Ex.
`
`1004, 2:56-57.) (Ex. 2042, ¶ 145.)
`
`Dr. Evans also admitted this drawing was not to scale. (Ex. 2039 at 147:6-
`
`21 (“It can’t be drawn to scale. I assume it is a schematic, rather than being drawn
`
`to scale.”).) Dr. Evans’ analysis is the product of improper hindsight bias; he can
`
`only assume that the location sensing devices 15 are directly connected to
`
`Workman’s streamer positioning devices 14 in light of the ’607 patent’s disclosure.
`
`The Board’s reliance on this figure as disclosing the co-location of streamer
`18
`
`
`

`

`
`
`positioning devices and location sensing devices was in error, as even PGS’ expert
`
`concedes. See also Nystrom v. TREX Co., 424 F.3d 1136, 1149 (Fed. Cir. 2005)
`
`(“Under the principles set forth in our prior cases, the speculative modeling
`
`premised on unstated assumptions in prior art patent drawings cannot be the basis
`
`for challenging the validity of claims reciting specific dimensions not disclosed
`
`directly in such prior art.”).
`
`Elholm
`
`B.
`As with Workman, Elholm was considered during prosecution of the ’607
`
`patent. And, as discussed above with respect to the construction of “predicting
`
`positions,” Elholm was distinguished during prosecution of the ’017 patent by
`
`specifically adding limitations directed to predicted positions.
`
`Elholm discloses a device that can be used to position seismic equipment by
`
`creating tension at the front end of an array (using a spreading device, also known
`
`as a paravane) without having a connection to the surface. Elholm recognized that
`
`it is sometimes desirable to increase the width between seismic streamers at the
`
`head of the array as they are towed. (Ex. 1005, 1:49-53.) Elholm also recognized
`
`that there were already several solutions to creating tension or increasing the width
`
`between streamers at the head of the towed array. (Ex. 1005, 1:59-2:6.) Lastly,
`
`Elholm recognized that these prior art front-end devices all had a connection with
`
`the surface of the water (Ex. 1005, 2:7-9), which created several problems, for
`
`19
`
`
`

`

`
`
`example, debris damage (Ex. 1005, 2:9-16), friction (Ex. 1005, 2:17-24), noise
`
`(Ex. 1005, 2:25-29), and wear and tear (Ex. 1005, 2:35-38).
`
`Elholm discloses a positioning unit that attempted to solve the problems
`
`associated w