JEFFREY COLE, Magistrate Judge.
Pursuant to Federal Rule of Civil Procedure 50, Defendants Tellabs, Inc., Tellabs Operations, Inc. and Tellabs North America, Inc. (collectively, "Defendants") hereby move the Court to grant judgment as a matter of law that claims 6-8 of U.S. Patent No. 7,227,681 ("the '681 Patent") are not infringed for the reasons set forth below.
Judgment as a matter of law is appropriate as to a particular issue when "a reasonable jury would not have a legally sufficient evidentiary basis to find for the party on that issue." Fed. R. Civ. P. 50(a). "[A] mere scintilla of supporting evidence will not suffice." Millbrook v. IBP, Inc., 280 F.3d 1169, 1173 (7th Cir. 2002) (quotation omitted). In deciding a motion for judgment as a matter of law, "the court should review all of the evidence in the record" and "should give credence to the evidence favoring the nonmovant as well as that evidence supporting the moving party that is uncontradicted and unimpeached." Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 151 (2000) (quotations omitted).
Fujitsu Limited asserts that the MIAM product infringes claims 6-8 of the '681 Patent. The claims read as follows:
The specification of the '681 Patent discloses, and claims 6-8 require, an optical amplifier with a specialized controller that keeps the gain constant when if the number of channels in a WDM optical signal changes:
Joint Ex. 2000, 2:38-47 (emphasis added).
When the number of channels (or wavelengths) is not changing, this specialized controller operates in a mode in which the total power of the optical signal is kept at a constant level. See Joint Ex. 2000, Fig. 4 of the '681 Patent (below). This mode is called "automatic level control" or "ALC" for short. Figures 4(A) and (B) disclose this control methodology. See, Joint Ex. 2000, Fig. 4 of the '681 Patent. In particular, the controller operates in automatic level control or "ALC" mode, before time t1 and after time t3. That is, before time t1 and after time t3 the controller ensures that the total power of the optical signal stays at a constant level. Transcript of Proceedings — Trial ("Trial Tr."), dated 8/28/2012, at 424:8-25; 426:2-5.
Joint Ex. 2000, Fig. 4 of the '681 Patent. Dr. Willner repeatedly explained that the claimed controller is in AGC mode during the variation of the number of channels — that this is the '681 invention. Trial Tr. at 394:21-395:11 ("This whole paragraph, it says `objects of the present invention,' and that means that it's this invention. It's the '681 invention."); 402:11-15; 405:10-15; 406:8-12; 407:8-3; 398:5-10 ("While the number of channels is being varied, either through adding or dropping of the different color data channels, the controller is going to hold the gain constant or approximately constant.").
Between times t1 and t3, however, the function of the controller temporarily changes. During this time period, the controller ignores the constant power objective and instead operates to ensure that there is an approximately constant gain. Id.; see also, Trial Tr. at 425:2-14; 426:11-18. This second mode, in which the focus is on keeping the gain approximately constant, is known as the automatic gain control or "AGC" mode. Id. In Figure 4, the disclosed controller is faced with a change in the number of channels occurring at time t2, between times t1 and t3. Id.
At the heart of the controller disclosed in the '681 Patent is a monitor signal processing circuit. See, e.g., Joint Ex. 2000, Fig. 10, element 70. This monitor signal processing circuit receives a "control signal, which warns of a variation in the number of channels" in the WDM optical signal. Joint Ex. 2000 at 7:55-59; see also, 433:9-17. Upon receiving this control signal, which occurs at time t1 in Figure 4, the controller changes from its automatic level control mode to its automatic gain control mode. Joint Ex. 2000, Fig. 4; col. 9:36-45. The controller reverts to its automatic level control mode at time t3, which may occur automatically a fixed time after t1, or it may occur upon receipt of a second control signal that indicates the change in channels is complete. Joint Ex. 2000, Fig. 4; col. 9:46-52. This control signal is critical to the "proper" operation of the controller disclosed in the specification of the '681 Patent — it allows the controller to be in AGC mode and hold the gain approximately constant during the actual variation of channels. See, e.g., Joint Ex. 2000, col. 7:55-10:19.
The asserted claims of the '681 Patent specifically require "a controller which controls the gain to be approximately constant." See, Joint Ex. 2000, claim 6. The only way to ensure that the gain is "approximately constant during variation of the number of channels," as taught and required by the '681 Patent, is to be certain the controller is in its automatic gain, or AGC, control mode. See, e.g., Joint Ex. 2000, col. 7:55-10:19; Trial Tr. at 394:21-395:11; 398:5-10; 402:11-15; 405:10-15; 406:8-12; 407:8-3. The disclosed method of ensuring this is to use a control signal received prior to the actual change in the number of channels. Joint Ex. 2000 at 7:55-59.
To prove infringement of a claim of the '681 Patent, Fujitsu Limited needed to establish that a MIAM product fulfilled each and every limitation of the claim. See BMC Res., Inc. v. Paymentech, L.P., 498 F.3d 1373, 1378 (Fed. Cir. 2007). Accordingly, where Fujitsu Limited failed to provide evidence of the presence of a limitation in independent claim 6 of the '681 Patent, judgment of non-infringement as a matter of law should be granted as to that claim and all the claims that depend from it.
The MIAM product does not infringe claim 6 because Fujitsu Limited has failed to show that it meets the claim element "a controller that controls the gain to be approximately constant." While the claims of the '681 Patent are directed to a controller that keeps the gain approximately constant, the Red-C optical gain block controller in the MIAM product is designed to keep the output power constant over time. PX 2013, RED-C Optical Networks Ltd.: Tellabs Amplifier — Software Specifications, Rev. 0, 12/30/02 (RED-C Page 582-606) at PX 2013-8-9. Because the input power can and does change over time, the controller monitors the output power and intentionally changes the gain as needed to keep the output power constant. PX 2013-8 (description of the control algorithm for the Red-C OGB in the MIAM module).
Specifically, the Red-C controller in the MIAM product checks the power level twenty times a second and, based upon those checks, determines what the gain needs to be to achieve the target output power. If the gain needs to be changed, the controller changes the gain, which may be as often as once every second. PX 2013-8.
Claim 6 requires "a controller which controls the gain to be approximately constant." A gain that is constantly monitored and potentially adjusted once every second can hardly be called "constant gain." Common sense dictates that anything that might change as often as 3600 times an hour would not be considered approximately constant. As no reasonable jury could determine that this claim limitation is met, judgment of a matter of law should be granted. The MIAM product does not infringe claim 6 or the claims that depend from it.
Additionally, the MIAM product does not include a controller that controls the gain to be approximately constant at the time required by the '681 Patent. The specification of the '681 Patent discusses controlling the gain to be approximately constant over a time interval. That time interval may be different for various claims, but it always includes the point in time during a variation of the number of channels in the WDM optical signal. See, e.g., Joint Ex. 2000, Fig. 4 of the '681 Patent. This requirement was confirmed by Dr. Willner and counsel for Fujitsu Limited. Trial Tr. at 378:7-19; 394:21-395:11 ("This whole paragraph, it says `objects of the present invention,' and that means that it's this invention. It's the '681 invention."); 397:5-19; 402:11-15; 405:10-15; 406:8-12; 407:8-3; 398:5-10 ("While the number of channels is being varied, either through adding or dropping of the different color data channels, the controller is going to hold the gain constant or approximately constant."); see also, Joint Ex. 2000 at 2:38-47. To infringe claim 6, therefore, the Red-C controller must control the gain to be approximately constant during the variation of the number of channels in the WDM optical signal.
The Red-C controller in the MIAM product can be operated in one of two primary modes. As shown in RED-C Optical Networks Ltd.: Tellabs Amplifier — Software Specifications, the Red-C controller operates in either an automatic gain control ("AGC") mode or an automatic power control ("APC") mode. PX 2013-8-9. In AGC mode, the controller attempts to keep the
The initial mode for the Red-C controller is the automatic power control mode, in which it attempts to keep the power approximately constant. PX 2013-8 ("the power per channel at the output of the EDFA [erbium-doped fiber amplifier] is kept constant."). The Red-C controller checks every second to see if the input power has changed enough that the change exceeds a threshold value called STAB_LEVEL or the stability level. PX 2013-9 at Figure 5 (below); see also, Trial Tr. at 456:25-457:19. If the change in input power exceeds the stability level, the Red-C controller assumes this large a change in the input power is due to a change in channel count. PX 2013-9 at Figure 5 (below); see also, Trial Tr. at 456:25-457:19. When the change in input power exceeds the stability level, then the Red-C controller changes from its standard automatic power control mode into its automatic gain control mode. PX 2013-9 at Figure 5 (below); see also, Trial Tr. at 456:25-457:19.
However, because the Red-C controller used by the accused MIAM product only determines that the change in input power exceeds the stability level
PX 2013-9 at Fig. 5 (Control algorithm flow chart for the Red-C OGB in the MIAM module).
Based on Fujitsu Limited's failure to provide any proof that the MIAM product includes a controller that controls the gain to be approximately constant
As noted above, the evidence clearly shows that the MIAM product does not include a controller that controls the gain to be approximately constant
When multiple channels are in steady state, the individual channels are likely to be off of the target gain by a different amount because each wavelength receives a slightly different gain from the Er-doped fiber. The Tellabs' specification permits such different wavelengths to have different gains by as much as ± 0.75 dB from the steady gain. PX 2012-15, Tellabs 7100 Electrically Controllable Optical Gain Blocks for Amplifier Modules, Rev E, 07/21/03 at Table 1 (below). For example, if the steady gain is to be 6 dB, individual wavelengths will be considered steady at 5.25 dB to 6.75 dB.
The Tellabs specification further shows that the gain of each individual channel should stay within ± 0.25 dB of this steady gain in order to remain at steady state (when the input power is constant). PX 2012-15 at Table 1 (below). For example, if the target gain is to be 6 dB and a channel is at 6.75 dB, the gain of an individual channel would satisfy the specification (and remain in steady state) as long as it stays within 6.5 dB to 7.0 dB.
However, the gain of each wavelength only varies by ± 0.25 dB over time as long as the number of channels does not change. During the variation of the number of channels, the Tellabs specification for the optical gain block in the MIAM product permits a maximum excursion of +/- 0.8 dB for that wavelength —
Figure 4 from the Tellabs' specification, PX 2012-15, shows how the gain of an individual wavelength changes as the number of channels is added or dropped. PX 2012-15. This maximum excursion actually occurs during the variation in the number of channels, as shown the annotated version of Figure 4 from the Tellabs specification. PX 2012-14 at Fig. 4. In other words, the excursion or deviation is greatest, and thus the controller is keeping the gain least constant,
This maximum excursion from the setting gain — during the change in the number of channels — is therefore more than 300% greater than that permitted when the input power is constant. While the ± 0.25 dB variation over time may be permissible, a gain variation of three times the steady gain (± 0.8 dB) is clearly not steady gain. Such a change in the gain during the variation in the number of channels clearly
Fujitsu Limited failed to offer any evidence to establish that the optical amplifier in the accused MIAM product amplifies a WDM optical signal having a variable number of channels. At trial, Dr. Willner simply stated that the MIAM amplifies "this sort of signal" and generally referenced Dr. Ghera's testimony and various documentation. See, e.g., Trial Tr. at 8/28/12 at 451:4-8. No reasonable jury could find that the MIAM product satisfies this required claim element based on such unsupported opinion testimony.
Neither Fujitsu Limited nor Dr. Willner offered any evidence that a Tellabs' customer that purchased one of the accused MIAM products operated that MIAM product in a system in which there were intentional variations in the number of channels. The MIAM product can be used in a system in which there are no intentional variations in the number of channels — a system that would not meet this required element of claim 6. This is true, at least in part, because the Tellabs 7100 system, Feature Pack 2.x, which included the MIAM product, did not allow for the easy reconfiguration of channels. See, e.g., DX 240; 9/1/2003 Tellabs 7100 Optical Transport System General Description Feature Pack 2.0, produced at TLTX0043134-3247. At that time, adding or dropping channels required sending an optical technician out into the field to each location where a new channel was to be added or dropped. Given the complexity of this task, it is likely that a number of these systems had a fixed number of channels. This is especially true for the accused MIAM products, as the accused MIAM products may have been replacement amplifiers going into systems that, by 2008, where already at their maximum capacity of 32 channels. As such, these MIAM products
Given Fujitsu Limited's failure of proof regarding whether the accused MIAM products operated in a system with a WDM optical signal having a variable number of channels, judgment as a matter of law should be granted because no reasonable jury could conclude that claim 6 and all claims that depend from it are infringed.
"It is axiomatic that dependent claims cannot be found infringed unless the claims from which they depend have been found to have been infringed." Wahpeton Canvas Co. v. Frontier, Inc., 870 F.2d 1546, 1553 (Fed. Cir. 1989). For the reasons stated above, no reasonable jury could conclude that there is infringement of independent claim 6 of the '681 Patent. As such, there can be no infringement of claim 7 of the '681 Patent.
As explained above, the Red-C controller only determines that the change in input power exceeds the stability level
As noted above, dependent claims cannot be infringed unless the claims from which they depend have been found to be infringed. Wahpeton Canvas Co. v. Frontier, Inc., 870 F.2d 1546, 1553 (Fed. Cir. 1989). For the reasons stated above, no reasonable jury could conclude that there is infringement of independent claim 6 of the '681 Patent. As such, there can be no infringement of claim 8.
For all the foregoing reasons, the Court should grant judgment as a matter of law that the claims 6-8 of the '681 Patent are not infringed by the MIAM product.
