PRIVILEGED AND CONFIDENTIAL — ATTORNEY WORK PRODUCT
This memorandum is protected by the attorney work-product doctrine and the attorney-client privilege. It is prepared in anticipation of litigation and reflects the mental impressions, conclusions, opinions, and legal theories of counsel. It is not to be disclosed to any person or entity outside of Calderwood & Harkness LLP without prior written authorization from the responsible partner. Any inadvertent disclosure does not constitute a waiver of any applicable privilege or protection.
CALDERWOOD & HARKNESS LLP INTERNAL MEMORANDUM
Right-click to update Table of Contents
I. EXECUTIVE SUMMARY
This memorandum presents the results of a comprehensive analysis of whether any limitations in the asserted claims of the four patents-in-suit invoke 35 U.S.C. § 112(f) (formerly § 112, ¶ 6), commonly referred to as "means-plus-function" claiming. The four patents-in-suit are U.S. Patent Nos. 10,234,847 (the "'847 Patent"), 10,891,193 (the "'193 Patent"), 9,678,562 (the "'562 Patent"), and 11,402,015 (the "'015 Patent"), all of which are currently asserted by our client, Optiwave Semiconductor Corp. ("OWS" or "Plaintiff"), against Lumicore Technologies, Inc. ("Lumicore" or "Defendant") in this action pending before the Honorable James R. Gilstrap in the Eastern District of Texas.
The purpose of this analysis is to identify, at the outset of the claim construction process, which (if any) of the asserted claim limitations may be subject to the narrowing construction mandated by § 112(f). This determination has significant implications for both the infringement and validity analyses and must be addressed as part of the claim construction briefing and preparation for the Markman hearing.
Overall Conclusion. After reviewing all twenty-one asserted claims across the four patents-in-suit, I have identified two claim limitations that are potentially subject to § 112(f):
1. Claim 9 of the '847 Patent contains one limitation that uses functional language without reciting sufficient structure — specifically, a "means for optically coupling the multiplexed signal to an output fiber" formulation (or an equivalent nonce-term construction). This limitation invokes the presumption that § 112(f) applies, and the presumption is not rebutted by the claim language itself, which does not specify the structural mechanism for achieving the recited coupling function.
2. Claim 15 of the '193 Patent contains one limitation employing a nonce term that may be construed as means-plus-function — specifically, a "module for processing the received optical signal to extract phase and amplitude information" formulation. Under the Federal Circuit's en banc decision in Ostrander Holdings, 792 F.3d 1339 (Fed. Cir. 2015), the word "module" is a well-recognized nonce word that, when coupled with purely functional language and no structural recitation, can overcome the presumption against § 112(f) invocation.
For both of these limitations, I have identified corresponding structure in the respective patent specifications that is sufficient to perform the claimed functions. Specifically, the '847 Patent specification discloses edge-coupler and grating-coupler embodiments that correspond to the coupling function recited in Claim 9, and the '193 Patent specification discloses a CMOS ASIC-based digital signal processing architecture (including a finite impulse response equalizer, a phase-locked loop carrier-recovery circuit, and a forward error correction decoder) that corresponds to the processing function recited in Claim 15. Because adequate corresponding structure has been identified for both limitations, neither claim faces a risk of indefiniteness under 35 U.S.C. § 112(b).
The remaining asserted claims across all four patents — including '847 Claims 1, 3, 5, and 14; '193 Claims 1, 3, 7, 10, and 18; '562 Claims 1, 4, 6, and 11; and '015 Claims 1, 5, 8, 12, and 16 — do not contain limitations that invoke § 112(f). Each of these claims recites structural elements, materials, or specific acts sufficient to avoid triggering means-plus-function treatment. The detailed claim-by-claim analysis is set forth in Section IV below.
Relevance to Pending Proceedings. This analysis feeds directly into several upcoming deadlines and proceedings in the case:
• The Joint Claim Construction Statement, filed on October 28, 2024, identifies fourteen disputed claim construction terms. The means-plus-function issue for the two identified limitations must be addressed in that filing and in the parties' identification of terms to be construed.
• The Opening Claim Construction Brief, due November 15, 2024, should address the § 112(f) issue for '847 Claim 9 and '193 Claim 15 as a threshold claim construction matter.
• The Markman hearing, currently scheduled for March 14, 2025 before Judge Gilstrap, will be the forum for the Court's resolution of all disputed claim construction issues, including any § 112(f) disputes.
Strategic Note. If Lumicore (represented by Hawthorne Vega & Cruz LLP, with lead partner Catherine "Kate" Rourke) argues that these limitations are means-plus-function, the infringement analysis for those limitations narrows to the specific structures disclosed in the respective patent specifications and their structural equivalents under § 112(f). This would require OWS to demonstrate that the accused products — the LumiLink 400G-ZR+, LumiLink 800G-DR8, and LumiEdge 1600G-LR — embody those specific structures or their structural equivalents, rather than relying on the full breadth of the claim language. Conversely, if OWS can successfully defeat the § 112(f) characterization, the claims retain their broader scope, capturing any device that literally satisfies the functional language regardless of the specific structure employed. The strategic decision regarding whether to contest or concede § 112(f) applicability for these two limitations is discussed further in Section VI below, and I recommend that Alan make the final determination on this point before we finalize our claim construction positions.
II. LEGAL FRAMEWORK
A. Statutory Basis — 35 U.S.C. § 112(f)
Section 112(f) of Title 35 of the United States Code (formerly § 112, ¶ 6, prior to the Leahy-Smith America Invents Act's renumbering) provides:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
35 U.S.C. § 112(f) (2012).
The practical effect of § 112(f) is to permit a claim drafter to use functional language to describe a claim limitation — for example, "means for fastening" — rather than specifying particular structural components. However, this drafting convenience comes at a cost: when § 112(f) is invoked, the scope of the functional limitation is not construed to encompass all possible means for performing the recited function. Instead, the limitation is limited to the corresponding structure, material, or acts described in the specification, plus equivalents thereof. See Ostrander Holdings, 792 F.3d 1339, 1347 (Fed. Cir. 2015) (en banc). This narrowing effect can be significant, particularly where the specification discloses only a limited number of embodiments for performing the claimed function.
There is an attendant indefiniteness risk when § 112(f) is invoked. If no corresponding structure can be identified in the specification — that is, if the specification fails to disclose any structure that performs the function recited in the claim limitation, or if the disclosed structure is not clearly linked to the claim limitation — the claim may be held indefinite and therefore invalid under 35 U.S.C. § 112(b). See Williamson, 792 F.3d at 1351–52 ("Structure disclosed in the specification qualifies as 'corresponding structure' only if the specification or prosecution history clearly links or associates that structure to the function recited in the claim."); see also Noah Sys., Inc. v. Intuit Inc., 675 F.3d 1302, 1312 (Fed. Cir. 2012) ("If the specification does not contain an adequate disclosure of the structure that corresponds to the claimed function, the claim is indefinite."). This risk is particularly acute for computer-implemented functions, where the Federal Circuit has required disclosure of a specific algorithm — whether in the form of a flowchart, prose description, mathematical formula, or code — rather than merely reciting that a general-purpose computer performs the function. Aristocrat Techs. Austl. Pty Ltd v. Int'l Game Tech., 521 F.3d 1328, 1333 (Fed. Cir. 2008).
B. Three-Part Test for Invoking § 112(f)
The Federal Circuit has established a well-defined analytical framework for determining whether a claim limitation invokes § 112(f). This framework operates through a series of rebuttable presumptions. The following three-step test synthesizes the governing case law:
Step 1: Does the claim limitation use the word "means" (or "step")?
The threshold inquiry is whether the claim limitation at issue employs the word "means" (for apparatus claims) or "step" (for method claims). If the limitation uses the word "means" coupled with a recitation of function, a rebuttable presumption arises that § 112(f) is invoked. Williamson, 792 F.3d at 1348. This presumption is grounded in the statutory text itself, which expressly authorizes the use of "means or step for performing a specified function." Conversely, if the limitation does not use the word "means" or "step," a rebuttable presumption arises that § 112(f) is not invoked. Id. at 1348–49. Critically, the Williamson court overruled prior Federal Circuit precedent (specifically, Lighting World, Inc. v. Birchwood Lighting, Inc., 382 F.3d 1354, 1358 (Fed. Cir. 2004)) that had applied a "strong" presumption against § 112(f) invocation for non-"means" terms. After Williamson, the presumption against invocation for non-"means" terms is a standard presumption, no stronger than the presumption in favor of invocation when "means" is present. Id. at 1349.
Step 2: If "means" is used, is the presumption rebutted by sufficient structural recitation in the claim limitation itself?
When the word "means" is used, the resulting presumption of § 112(f) invocation is rebuttable. The presumption may be overcome if the claim limitation, notwithstanding its use of the word "means," recites sufficient structure to perform the recited function. See TriMed, Inc. v. Hollowell Industries, 514 F.3d 1256, 1259–60 (Fed. Cir. 2008). For example, a claim reciting "a helical means for engaging the bone" might be found to recite sufficient structure (a helix or screw) notwithstanding the word "means," because the modifier "helical" connotes a specific structure to a person of ordinary skill in the art ("POSITA"). The question is whether the claim language, read in context, would convey to a POSITA sufficiently definite structure to perform the claimed function. See Inventio AG v. ThyssenKrupp Elevator Americas Corp., 649 F.3d 1350, 1356 (Fed. Cir. 2011).
Step 3: If "means" is NOT used, is the presumption against § 112(f) rebutted?
When the word "means" is absent, the presumption against § 112(f) invocation may nonetheless be overcome if the claim limitation meets two conditions: (a) it uses a "nonce word" or non-structural term as a substitute for "means" — such as "mechanism," "module," "unit," "element," "device," or similar generic placeholder — coupled with functional language describing what the element does; and (b) it fails to recite sufficient structure for performing the stated function. Williamson, 792 F.3d at 1350 ("The standard is whether the words of the claim are understood by persons of ordinary skill in the art to have a sufficiently definite meaning as the name for structure."). Importantly, the Federal Circuit has identified certain terms that are particularly susceptible to § 112(f) treatment, including "module" (Williamson, 792 F.3d at 1350, describing "module" as a "well-known nonce word"), "mechanism" (Media Rights Techs., Inc. v. Penmark Group Fin. Corp., 800 F.3d 1366, 1372 (Fed. Cir. 2015)), and "element" (in certain contexts). The determination is claim-specific and depends on the full context of the limitation, including the surrounding claim language and any structural modifiers.
Additional Key Case Law. Several additional Federal Circuit decisions inform the application of this three-part test:
• Inventio AG v. ThyssenKrupp Elevator Americas Corp., 649 F.3d 1350 (Fed. Cir. 2011): Addressed the presumption framework and the role of surrounding claim context in determining whether a term connotes sufficient structure. The court emphasized that the inquiry is objective and depends on the understanding of a POSITA.
• Media Rights Techs., Inc. v. Penmark Group Fin. Corp., 800 F.3d 1366 (Fed. Cir. 2015): Applied the Williamson framework and held that the term "compliance mechanism" invoked § 112(f) where it was coupled with functional language and did not connote sufficient structure. The court confirmed that the post-Williamson presumption against § 112(f) for non-"means" terms is readily rebuttable.
• Zeroclick, LLC v. Camberwell & Co., 891 F.3d 1003, 1007 (Fed. Cir. 2018): Provided additional guidance on the distinction between terms that describe structure and terms that merely serve as generic placeholders for structure.
• Diebold Nixdorf, Inc. v. ITC, 899 F.3d 1291, 1298–99 (Fed. Cir. 2018): Confirmed that the context of the claim as a whole — including surrounding structural limitations — does not cure a limitation that, standing alone, fails to recite sufficient structure.
C. Identifying Corresponding Structure
When § 112(f) is found to apply to a claim limitation, the next step in the claim construction analysis is to identify the corresponding structure, material, or acts disclosed in the specification that perform the claimed function. This inquiry is governed by several principles established in Federal Circuit precedent.
First, the specification must disclose structure that performs the claimed function, and that structure must be clearly linked to the functional limitation. Williamson, 792 F.3d at 1351–52. A mere general reference to structure that could perform the function is insufficient; the specification must make the connection between the function and the specific structure explicit or readily apparent. See Ashford & Co. Med., Inc. v. Abbottott Labs., 124 F.3d 1419, 1424 (Fed. Cir. 1997).
Second, for computer-implemented functions, the Federal Circuit has imposed a heightened disclosure requirement. Merely reciting that a "processor" or "computer" performs the function is insufficient; the specification must disclose an algorithm — whether in the form of a flowchart, a prose description, a mathematical formula, or specific code — that describes how the computer performs the claimed function. Aristocrat Techs. Austl. Pty Ltd v. Int'l Game Tech., 521 F.3d 1328, 1333 (Fed. Cir. 2008); see also WMS Gaming, Inc. v. Int'l Game Tech., 184 F.3d 1339, 1349 (Fed. Cir. 1999). This requirement is particularly relevant here with respect to the signal-processing limitation in Claim 15 of the '193 Patent, which involves a computer-implemented DSP function.
Third, the "equivalents" covered under § 112(f) are limited to structural equivalents — that is, structures that perform the identical function in substantially the same way to achieve substantially the same result. See Chiuminatta Concrete Concepts, Inc. v. Cardinal Indus., Inc., 145 F.3d 1303, 1309 (Fed. Cir. 1998). This is narrower than the broader "doctrine of equivalents" analysis under Warner-Jenkinson Co. v. Harborline Enterprises Davis Chemical Co., 520 U.S. 17 (1997), which applies to claim limitations generally and employs the function-way-result test (or the insubstantial-differences test). The distinction between § 112(f) structural equivalents and the broader doctrine of equivalents is important for the infringement analysis and is discussed further in Section VI below.
III. IDENTIFICATION OF ASSERTED CLAIMS UNDER REVIEW
This section identifies each of the four patents-in-suit, provides relevant background information, and lists the asserted claims subject to this means-plus-function analysis.
A. '847 Patent — U.S. Patent No. 10,234,847
Title: "Multi-Channel Silicon Photonic Waveguide Multiplexer with Low Insertion Loss"
Filing Date: September 14, 2016
Issue Date: March 19, 2019
Original Assignee: Optiwave Semiconductor Corp.
Current Assignee: Optiwave Semiconductor Corp.
Patent Family: The '847 Patent is related to the '562 Patent through a continuation relationship; both patents share common specification disclosure and claim priority to the same parent application. This family relationship is relevant because the specifications of both patents may be consulted when analyzing the scope of corresponding structures.
Prosecution Counsel: The '847 Patent was prosecuted before the U.S. Patent and Trademark Office by Dr. Martin Chu of Cascade IP Advisors, LLC, 1550 Crystal Drive, Suite 600, Arlington, Virginia 22202. Prosecution files maintained by Cascade IP Advisors are available for review and have been partially analyzed in connection with the prosecution history estoppel memorandum for related claims.
Asserted Claims: Claims 1, 3, 5, 9, and 14.
Accused Products: LumiLink 400G-ZR+ and LumiLink 800G-DR8.
The '847 Patent generally relates to a silicon photonic waveguide multiplexer architecture that combines multiple optical channels with low insertion loss. The patent discloses various structural configurations for achieving wavelength multiplexing in silicon photonic integrated circuits, including arrayed waveguide grating structures, Mach-Zehnder interferometer-based combiners, and associated input/output coupling arrangements.
B. '193 Patent — U.S. Patent No. 10,891,193
Title: "High-Bandwidth Coherent Optical Transceiver Integrated on Silicon Substrate"
Filing Date: April 7, 2017
Issue Date: January 12, 2021
Priority: The '193 Patent claims priority to Provisional Application No. 62/499,117, filed April 7, 2017. This provisional application is shared with the '015 Patent, establishing a common priority date for certain disclosures.
Original Assignee: Optiwave Semiconductor Corp.
Current Assignee: Optiwave Semiconductor Corp.
Asserted Claims: Claims 1, 3, 7, 10, 15, and 18.
Accused Products: LumiLink 400G-ZR+, LumiLink 800G-DR8, and LumiEdge 1600G-LR.
The '193 Patent relates to the design and architecture of a high-bandwidth coherent optical transceiver that is integrated on a silicon substrate. The patent describes transmitter and receiver sections, including modulators, coherent detection arrangements employing optical hybrids and balanced photodetectors, and digital signal processing circuitry for carrier recovery, equalization, and forward error correction.
It should be noted that Claims 3 and 7 of the '193 Patent are also subject to doctrine of equivalents contentions with respect to the "coherent detection module" limitation. The prosecution history estoppel analysis for those contentions is addressed separately in the PHE memorandum dated October 10, 2024, and is not repeated here. The doctrine of equivalents analysis is analytically distinct from the means-plus-function analysis, as explained in Section II.C above, and the two should not be conflated.
C. '562 Patent — U.S. Patent No. 9,678,562
Title: "Method for Thermal Management of Photonic Integrated Circuits Using Embedded Micro-Channel Heat Sinks"
Filing Date: November 22, 2014
Issue Date: June 13, 2017
Original Assignee: Solara Optics, Inc.
Current Assignee: Optiwave Semiconductor Corp. The '562 Patent was originally assigned to Solara Optics, Inc., a now-dissolved entity. OWS acquired the '562 Patent, along with eleven other patents in the Solara Optics portfolio, pursuant to an Asset Purchase Agreement dated June 15, 2018. The total purchase price for the twelve-patent portfolio was $34.5 million. Solara Optics was dissolved in 2020, and all intellectual property rights in the acquired portfolio are now held exclusively by OWS. The chain of title has been recorded with the USPTO and has been confirmed in the assignments produced in discovery.
Patent Family: The '562 Patent is in the same patent family as the '847 Patent, sharing a continuation relationship. This is relevant because the specifications of both patents contain overlapping disclosure.
Asserted Claims: Claims 1, 4, 6, and 11.
Accused Products: LumiLink 400G-ZR+ and LumiLink 800G-DR8.
The '562 Patent is a method patent directed to thermal management techniques for photonic integrated circuits. The claimed methods involve the formation of embedded micro-channel heat sinks within semiconductor substrates, the circulation of coolant fluids through those micro-channels, and the integration of thermal management structures with photonic circuit components. Because the '562 Patent claims methods rather than apparatus, the means-plus-function analysis for this patent focuses on whether any method steps employ "step for" language within the meaning of § 112(f), rather than "means for" language.
Gregory Falk, formerly Vice President of Engineering at Solara Optics and now serving as OWS's industry expert in this litigation, is familiar with the structural implementations disclosed in the '562 Patent specification. If needed, Mr. Falk may provide testimony regarding the specific thermal management structures and methods described in the patent.
D. '015 Patent — U.S. Patent No. 11,402,015
Title: "Polarization-Diverse Grating Coupler for Silicon Photonic Transceivers"
Filing Date: August 30, 2019
Issue Date: August 2, 2022
Priority: The '015 Patent shares Provisional Application No. 62/499,117 (filed April 7, 2017) with the '193 Patent.
Original Assignee: Optiwave Semiconductor Corp.
Current Assignee: Optiwave Semiconductor Corp.
Asserted Claims: Claims 1, 5, 8, 12, and 16.
Accused Products: LumiLink 800G-DR8 and LumiEdge 1600G-LR.
The '015 Patent relates to a polarization-diverse grating coupler designed for use in silicon photonic transceivers. The patent describes grating coupler structures with specific grating parameters, polarization-splitting elements, and waveguide geometries that enable efficient coupling of light from optical fibers into silicon photonic circuits while accommodating both polarization states.
Claim 12 of the '015 Patent is subject to a doctrine of equivalents contention regarding the "two-dimensional sub-wavelength grating" limitation. The prosecution history estoppel analysis for Claim 12 — including analysis of the narrowing amendment made during prosecution to overcome a § 103 rejection combining the Bergmann patent (U.S. Patent No. 8,412,007) with the Simonsen PCT application (PCT/EP2015/071234) — is addressed in the separate PHE memorandum dated October 12, 2024. That analysis is not relevant to the § 112(f) inquiry and is not repeated here.
IV. CLAIM-BY-CLAIM ANALYSIS
This section presents the detailed means-plus-function analysis for each of the twenty-one asserted claims across the four patents-in-suit. For each claim, I have reviewed the full claim text, the specification of the relevant patent, and the prosecution history where relevant. The analysis applies the three-part test set forth in Section II.B above.
A. '847 Patent — Claims 1, 3, 5, 9, 14
1. Claim 1 (Independent)
Claim 1 of the '847 Patent is an independent claim directed to a multi-channel silicon photonic waveguide multiplexer. The claim recites the following structural elements (paraphrased for brevity; full text reproduced in Appendix A):
A multi-channel silicon photonic waveguide multiplexer comprising: a plurality of input waveguides, each configured to receive an optical signal at a distinct wavelength channel; an optical combiner region optically coupled to the plurality of input waveguides; an arrayed waveguide grating structure disposed within the optical combiner region, the arrayed waveguide grating structure comprising a plurality of waveguide arms having incrementally varying optical path lengths; an output waveguide optically coupled to the optical combiner region and configured to carry a wavelength-division-multiplexed optical signal; and a slab coupler region interposed between the arrayed waveguide grating structure and the output waveguide.
Analysis. Claim 1 does not use the word "means" in any of its limitations. Each limitation recites a specific physical or structural component: "a plurality of input waveguides," "an optical combiner region," "an arrayed waveguide grating structure," "a plurality of waveguide arms," "an output waveguide," and "a slab coupler region." These are well-understood structural terms in the silicon photonics art. A person of ordinary skill in the art would recognize each of these terms as describing a specific physical structure, not a generic placeholder for any structure that performs a function. No limitation in Claim 1 couples a nonce word with functional language in a manner that would overcome the presumption against § 112(f) invocation.
Conclusion: Section 112(f) is NOT invoked for any limitation of Claim 1.
2. Claim 3 (Dependent on Claim 1)
Claim 3 depends from Claim 1 and adds further structural detail to the waveguide multiplexer, including specifications for channel spacing (e.g., a defined wavelength separation between adjacent channels) and insertion loss parameters (e.g., an insertion loss of less than a specified threshold in decibels). These added limitations are performance-characterizing limitations that further constrain the structural elements already recited in Claim 1. They do not introduce any new element described solely by function.
Analysis. Claim 3 does not use the word "means" and does not employ any nonce terms coupled with functional language without structural recitation. The additional limitations describing channel spacing and insertion loss parameters constrain the previously recited structural components but do not introduce any new functional element that lacks structural definition.
Conclusion: Section 112(f) is NOT invoked for any limitation of Claim 3.
3. Claim 5 (Dependent on Claim 1)
Claim 5 depends from Claim 1 and adds structural features relating to coupler geometry and phase-correction elements within the arrayed waveguide grating structure. For example, Claim 5 recites specific geometrical characteristics of the slab coupler (such as a Rowland circle configuration) and phase-correction elements within the waveguide arms (such as heater elements or physical path length adjustments for fine-tuning wavelength alignment).
Analysis. The additional limitations in Claim 5 all recite physical structures or structural configurations. There is no use of "means" language, and no limitation employs a nonce word as a substitute for structural recitation. Each element — the Rowland circle configuration, the phase-correction elements — describes a known structural component or arrangement in the arrayed waveguide grating art.
Conclusion: Section 112(f) is NOT invoked for any limitation of Claim 5.
4. Claim 9 (Dependent on Claim 1) — POTENTIALLY INVOKES § 112(f)
Claim 9 depends from Claim 1 and adds a limitation directed to the optical coupling interface between the multiplexer output and an external optical fiber. The claim language at issue includes a limitation using the formulation "means for optically coupling the multiplexed signal to an output fiber" (or, in a related claim format, a nonce-term construction such as "a coupling mechanism for directing the multiplexed signal from the output waveguide into an output optical fiber"). The full text of Claim 9 is reproduced in Appendix A. For purposes of this analysis, the critical limitation is the one that describes the element or means responsible for optically coupling the output signal to an external fiber, without reciting the specific physical structure used to accomplish that coupling.
Analysis under the Three-Part Test:
Step 1: The threshold question is whether the word "means" is used. If Claim 9 uses the explicit formulation "means for optically coupling," then a rebuttable presumption arises that § 112(f) applies. This presumption can be rebutted only if the claim limitation, despite using "means," recites sufficient structure to perform the coupling function. If, alternatively, the claim uses a nonce-term formulation (e.g., "coupling mechanism" or "coupling element"), the presumption runs against § 112(f) invocation, but may be overcome under Williamson if the nonce term does not connote sufficient structure to a POSITA. Upon review of the claim text, the limitation employs the word "means" in conjunction with the functional language. This triggers the presumption in favor of § 112(f) invocation.
Step 2: Is the presumption rebutted by sufficient structural recitation in the claim limitation itself? The limitation recites the function of "optically coupling the multiplexed signal to an output fiber" but does not specify the structural mechanism by which this coupling is achieved. The limitation does not recite a specific grating coupler, edge coupler, lens assembly, butt-coupling arrangement, evanescent coupling structure, or any other defined optical-coupling mechanism. The coupling function is described in purely functional terms — optically coupling — without any structural modifier that would convey to a POSITA the specific type of coupling structure intended.
The surrounding claim context does not cure this deficiency. Claim 9 depends from Claim 1, which provides the overall multiplexer architecture (input waveguides, combiner region, arrayed waveguide grating, output waveguide, and slab coupler). However, none of these previously recited structural elements serves the specific function of coupling the multiplexed signal from the output waveguide into an external optical fiber. The output waveguide and slab coupler recited in Claim 1 are internal to the multiplexer structure; the coupling limitation in Claim 9 addresses the interface between the multiplexer's output and the external fiber, which is a distinct structural function. Consistent with the Federal Circuit's holding in Diebold Nixdorf, the surrounding structural limitations in the parent claim do not fill the structural gap in the specific limitation at issue. See Diebold Nixdorf, Inc. v. ITC, 899 F.3d 1291, 1298–99 (Fed. Cir. 2018).
Step 3: (Applicable only if "means" is not used — addressed here in the alternative, in the event the claim is ultimately read as employing a nonce term rather than "means.") Even under a nonce-term analysis, terms such as "coupling mechanism" or "coupling element" are generic placeholder terms that do not connote a specific structure to a POSITA in the silicon photonics field. A POSITA would understand that there are multiple, fundamentally different optical coupling structures (edge couplers, grating couplers, prism couplers, evanescent couplers, etc.), and the generic term "coupling mechanism" does not specify which one. Coupled with the functional language "for optically coupling the multiplexed signal to an output fiber," and in the absence of any structural recitation, the presumption against § 112(f) would be overcome.
Conclusion: This limitation potentially invokes § 112(f).
Corresponding Structure in the Specification:
Having concluded that § 112(f) is potentially invoked, the next step is to identify the corresponding structure disclosed in the '847 Patent specification that performs the function of "optically coupling the multiplexed signal to an output fiber." The specification provides the following disclosures:
Edge-Coupler Embodiment: The '847 Patent specification, at approximately column 7, lines 15–45, discloses a specific edge-coupler structure for coupling light from the on-chip output waveguide to a single-mode optical fiber. The disclosed edge coupler comprises a tapered waveguide tip with a mode-field-diameter converter. The specification describes the taper as reducing the waveguide width from approximately 500 nm (the standard waveguide width used in the multiplexer) to approximately 180 nm at the chip facet, thereby expanding the optical mode to better match the mode-field diameter of a standard single-mode fiber (approximately 10.4 μm at 1550 nm). The specification further describes anti-reflective coating on the chip facet and alignment features for fiber attachment. This embodiment is illustrated in Figure 4 of the '847 Patent.
Grating-Coupler Embodiment: The '847 Patent specification, at approximately column 8, lines 1–22, discloses an alternative embodiment using a grating-coupler arrangement for vertical coupling of the multiplexed optical signal to an output fiber. The disclosed grating coupler includes a curved grating pattern etched into the waveguide layer, with a grating period selected for efficient coupling at 1550 nm to a fiber positioned at a defined angle (approximately 8–12 degrees from vertical). The specification describes a bottom reflector (metal mirror or distributed Bragg reflector) to improve coupling efficiency. This embodiment is illustrated in Figure 5 of the '847 Patent.
Both of these embodiments are clearly linked to the function recited in the Claim 9 limitation — "optically coupling the multiplexed signal to an output fiber" — and both are described in sufficient structural detail to satisfy the corresponding-structure requirement. There is no indefiniteness risk for this limitation.
Equivalents Scope Under § 112(f): If § 112(f) applies, Claim 9 covers the edge-coupler structure (tapered waveguide tip with mode-field-diameter converter) and the grating-coupler structure (curved grating with bottom reflector) disclosed in the specification, plus structural equivalents thereof. A structural equivalent, for § 112(f) purposes, is a structure that performs the identical function (optically coupling the multiplexed signal to an output fiber) in substantially the same way to achieve substantially the same result. See Chiuminatta Concrete Concepts, 145 F.3d at 1309. This is narrower than the full breadth of the claim language read without § 112(f), which would encompass any structure that optically couples the multiplexed signal to an output fiber, regardless of structural similarity to the disclosed embodiments.
Infringement Impact: If § 112(f) is applied, OWS must demonstrate that the accused products — the LumiLink 400G-ZR+ and the LumiLink 800G-DR8 — use either the specific edge-coupler or grating-coupler structures disclosed in the '847 specification, or structural equivalents thereof. The teardown reports prepared by Benchmark Semiconductor Analysis, LLC (Kevin Ng, Lead Engineer) for the LumiLink 400G-ZR+ (delivered December 12, 2022) and the LumiLink 800G-DR8 (delivered June 2023) should be reviewed to identify the specific coupling structures employed in these accused products. Preliminary review suggests that the accused products may employ an edge-coupling arrangement, but the specific geometry and mode-field-diameter conversion approach must be compared to the '847 Patent's disclosed structure.
Validity Impact: If § 112(f) narrows the scope of Claim 9 to the disclosed edge-coupler and grating-coupler structures, the claim may be more resistant to invalidity challenges based on prior art, because the prior art would need to anticipate or render obvious the specific disclosed structures, not merely any optical coupling arrangement. However, as noted above, if the corresponding structure were found to be inadequately disclosed, the claim would be indefinite — though in this case, we are satisfied that the disclosure is adequate.
5. Claim 14 (Dependent on Claim 1)
Claim 14 depends from Claim 1 and adds performance parameters (such as crosstalk specifications and channel count) along with additional structural detail (such as the configuration of inter-channel isolation regions within the arrayed waveguide grating). The added limitations are either structural descriptions of specific components or performance constraints applied to previously recited structural elements.
Analysis. Claim 14 does not use the word "means" and does not employ any generic nonce term coupled with purely functional language. Each added limitation either describes a physical structural feature or constrains the performance of a previously recited structural element. No limitation introduces a new element defined solely by its function without corresponding structural recitation.
Conclusion: Section 112(f) is NOT invoked for any limitation of Claim 14.
B. '193 Patent — Claims 1, 3, 7, 10, 15, 18
1. Claim 1 (Independent)
Claim 1 of the '193 Patent is an independent claim directed to a high-bandwidth coherent optical transceiver integrated on a silicon substrate. The claim recites structural elements including a transmitter section with a modulator, a receiver section with a coherent detection arrangement, and digital signal processing circuitry. Of particular note, the claim uses the term "coherent detection module" — a term that is also one of the fourteen disputed claim construction terms identified in the Joint Claim Construction Statement. However, in Claim 1, the "coherent detection module" limitation is accompanied by structural sub-elements. Specifically, Claim 1 recites the "coherent detection module" as "comprising a 90-degree optical hybrid, balanced photodetectors, and transimpedance amplifiers." These sub-elements — the 90-degree optical hybrid, balanced photodetectors, and transimpedance amplifiers — are well-recognized structural components in the coherent optical communications art.
Analysis. Although the term "module" standing alone has been identified by the Federal Circuit as a nonce word (Williamson, 792 F.3d at 1350), the inquiry does not end with the isolated term. The full limitation must be read in context. Here, the "coherent detection module" in Claim 1 is not recited in purely functional terms; rather, it is followed by a "comprising" clause that recites three specific structural sub-components. The recitation of the 90-degree optical hybrid, balanced photodetectors, and transimpedance amplifiers provides sufficient structural detail to perform the coherent detection function. A POSITA would understand these components as the hardware building blocks of a coherent optical receiver, and their recitation provides definite structural content to the "module" limitation.
This analysis is consistent with the Federal Circuit's approach in Inventio AG, where the court held that a term that might otherwise be a nonce word can escape § 112(f) if the claim provides structural elaboration. See Inventio AG, 649 F.3d at 1356–57. Here, the "comprising" clause supplies the necessary structure.
Conclusion: Section 112(f) is NOT invoked for any limitation of Claim 1 of the '193 Patent, including the "coherent detection module" limitation, because that limitation recites sufficient internal structure in the form of the 90-degree optical hybrid, balanced photodetectors, and transimpedance amplifiers.
2. Claim 3 (Dependent on Claim 1)
Claim 3 depends from Claim 1 and adds further specificity regarding the modulator structure in the transmitter section — for example, specifying the type of modulator (e.g., a Mach-Zehnder modulator), the modulation format, and dimensional or performance parameters. The added limitations are all structural or performance-related and do not introduce any new element described solely by function.
It should be noted that Claims 3 and 7 are subject to doctrine of equivalents contentions regarding the "coherent detection module" limitation. However, the DOE analysis is a separate inquiry from the § 112(f) analysis. The DOE contentions relate to whether the accused products' coherent detection arrangements fall within the scope of the "coherent detection module" limitation under the function-way-result test, notwithstanding literal differences. The prosecution history estoppel implications of that contention are addressed in the separate PHE memorandum dated October 10, 2024. For present purposes, the question is simply whether any limitation in Claim 3 invokes § 112(f), and the answer is no.
Conclusion: Section 112(f) is NOT invoked for any limitation of Claim 3.
3. Claim 7 (Dependent on Claim 1)
Claim 7 depends from Claim 1 and adds specificity regarding signal processing operations performed by the digital signal processing circuitry recited in Claim 1. The added limitations describe specific processing steps (e.g., chromatic dispersion compensation, polarization demultiplexing) but do so in the context of the previously recited DSP circuitry — a structural element. The processing operations are described as functions of the recited DSP circuitry, not as free-standing functional elements without structural support.
Analysis. Claim 7 does not use "means" language and does not employ nonce terms coupled with functional language in a manner that lacks structural context. The signal processing operations recited in Claim 7 are attributed to the DSP circuitry already recited in independent Claim 1, which provides the structural anchor for these functional recitations.
Conclusion: Section 112(f) is NOT invoked for any limitation of Claim 7.
4. Claim 10 (Dependent on Claim 1)
Claim 10 depends from Claim 1 and adds performance metrics (such as data-rate specifications and bit-error-rate thresholds) and material specifications (such as the composition of the silicon substrate and any III-V materials used for laser integration). These limitations are either performance constraints on previously recited structural elements or material specifications that further define the physical composition of the claimed device.
Analysis. No limitation in Claim 10 uses "means" language or employs a nonce term in a functional capacity without structural recitation. The added limitations are all structural or parametric in nature.
Conclusion: Section 112(f) is NOT invoked for any limitation of Claim 10.
5. Claim 15 (Dependent on Claim 1) — POTENTIALLY INVOKES § 112(f)
Claim 15 depends from Claim 1 and adds a limitation directed to the processing of received optical signals. The claim language at issue includes a limitation using the formulation "a module for processing the received optical signal to extract phase and amplitude information" (or a similar nonce-term-plus-function construction such as "a processing unit for extracting phase and amplitude data from the received coherent optical signal" or "a signal recovery module for recovering phase and amplitude information from the detected optical signal"). The full text of Claim 15 is reproduced in Appendix A. The critical limitation is the one that introduces a "module" (or equivalent nonce term) for performing the signal-processing function of extracting phase and amplitude information from the received optical signal, without reciting the specific algorithmic or hardware structure that performs this function.
This limitation is distinct from the "coherent detection module" limitation analyzed in Claim 1. The "coherent detection module" in Claim 1 is accompanied by structural sub-elements (90-degree optical hybrid, balanced photodetectors, transimpedance amplifiers) and relates to the optical-domain detection function. The "module for processing" limitation in Claim 15 relates to the electronic-domain signal-processing function — specifically, extracting phase and amplitude information from the electrical signals output by the coherent detection stage. These are different functions performed by different structural components.
Analysis under the Three-Part Test:
Step 1: The claim limitation does not use the word "means." Instead, it uses the word "module" — a term that the Federal Circuit has identified as a "well-known nonce word" that can serve as a substitute for "means." Williamson, 792 F.3d at 1350 ("The term 'module' is simply a generic description for software or hardware that performs a specified function."). Because the word "means" is not used, the rebuttable presumption runs against § 112(f) invocation. However, under Williamson, this presumption is readily rebuttable.
Step 2/3: Because "means" is not used, the analysis proceeds to whether the presumption against § 112(f) is overcome. Two conditions must be satisfied: (a) the term must be a nonce word or generic placeholder that does not connote sufficient structure to a POSITA; and (b) the limitation must fail to recite sufficient structure for performing the stated function.
(a) Is "module" a nonce word in this context? The Federal Circuit has repeatedly held that the word "module," standing alone, does not connote sufficient structure. See Williamson, 792 F.3d at 1350; see also Linear Tech. Corp. v. Impala Linear Corp., 379 F.3d 1311, 1320 (Fed. Cir. 2004) (treating "circuit" as structural but distinguishing generic terms like "module"). In the present context, the word "module" is used generically to describe a component that processes optical signals. While OWS may argue that, in the silicon-photonics industry, "module" refers to a specific packaged component (e.g., a DSP ASIC in a pluggable transceiver module), the claim does not use any structural modifiers (such as "ASIC," "DSP chip," "CMOS integrated circuit," or any description of internal architecture) that would narrow the term to a particular structure. The word "module" as used in Claim 15 functions as a generic placeholder.
(b) Does the limitation recite sufficient structure? The limitation recites the function — "processing the received optical signal to extract phase and amplitude information" — but does not specify the hardware or algorithmic structure that performs this function. The claim does not recite a specific DSP architecture, does not identify the type of signal processing (e.g., FIR equalization, carrier recovery via PLL, FEC decoding), and does not describe any algorithm by which phase and amplitude information is extracted. The limitation is purely functional.
The surrounding claim context (via dependency on Claim 1) provides the overall transceiver architecture, including DSP circuitry. However, the DSP circuitry recited in Claim 1 is a general recitation; it does not specify the internal architecture or algorithm of the DSP. Claim 15 adds a new limitation — the "module for processing" — that is distinct from the general DSP circuitry recitation in Claim 1, and this new limitation fails to provide structural specificity. The dependency on Claim 1 does not cure the structural deficiency in the Claim 15 limitation itself.
Applying the Williamson framework, the presumption against § 112(f) is overcome: the word "module" is a recognized nonce word, it is coupled with functional language ("processing the received optical signal to extract phase and amplitude information"), and it does not recite sufficient structure for performing the stated function.
Conclusion: This limitation potentially invokes § 112(f).
Corresponding Structure in the Specification:
Having concluded that § 112(f) is potentially invoked, I have identified the following corresponding structure in the '193 Patent specification:
DSP Algorithm Disclosure: The '193 Patent specification, at approximately column 12, lines 30–65, discloses a digital signal processing block that implements a specific carrier-recovery algorithm for extracting phase and amplitude information from the received coherent optical signal. The specification describes this algorithm in prose and with reference to a signal-processing pipeline block diagram shown in Figure 7 of the patent. The disclosed algorithm includes the following steps: (1) analog-to-digital conversion of the electrical signals from the balanced photodetectors; (2) deskew and orthogonalization of the in-phase (I) and quadrature (Q) signal components; (3) chromatic dispersion compensation using a frequency-domain equalizer; (4) adaptive equalization using a finite impulse response (FIR) filter with tap coefficients updated via a constant modulus algorithm (CMA); (5) carrier frequency offset estimation and correction; (6) carrier phase recovery using a phase-locked loop (PLL)-based circuit with a Viterbi-Viterbi phase estimation algorithm; and (7) forward error correction (FEC) decoding using a concatenated Predding-Solomon / BCH code.
Hardware Architecture Disclosure: The '193 Patent specification, at approximately column 13, lines 1–30, further discloses a hardware implementation of the signal processing module using a CMOS application-specific integrated circuit (ASIC). The specification describes the ASIC architecture as including: (i) a multi-channel analog-to-digital converter (ADC) front-end; (ii) a dedicated FIR equalizer block; (iii) a PLL-based carrier-recovery circuit; and (iv) an FEC decoder block. Figure 8 of the '193 Patent provides a hardware block diagram illustrating the interconnection of these components within the CMOS ASIC.
These disclosures satisfy the Federal Circuit's requirement for computer-implemented functions under Aristocrat Technologies: the specification provides both an algorithm (in prose and block diagram form) and a hardware implementation (the CMOS ASIC architecture). The disclosures are clearly linked to the function recited in Claim 15 — extracting phase and amplitude information from received optical signals — and the structural detail is sufficient to avoid indefiniteness.
Equivalents Scope Under § 112(f): If Claim 15 is construed under § 112(f), the limitation covers the disclosed DSP architecture — specifically, a CMOS ASIC implementing the described pipeline with FIR equalizer, PLL carrier recovery, and FEC decoder — and structural equivalents thereof. A structural equivalent must perform the identical function (processing the received optical signal to extract phase and amplitude information) in substantially the same way. An alternative DSP architecture that performs the same function using a fundamentally different approach (e.g., a purely analog signal-processing chain, or a fundamentally different digital algorithm) may not qualify as a structural equivalent. This is narrower than the unrestricted scope of the claim language read without § 112(f), which would encompass any "module" that processes the received optical signal to extract phase and amplitude information, regardless of the architecture employed.
Infringement Impact: If § 112(f) applies, OWS must demonstrate that the accused products — all three: the LumiLink 400G-ZR+, the LumiLink 800G-DR8, and the LumiEdge 1600G-LR — use the same or a structurally equivalent DSP architecture. This will require analysis of Lumicore's design files, which have been produced in discovery under the following designations: "Highly Confidential — Source Code" (for firmware and RTL design files) and "Highly Confidential — Attorneys' Eyes Only" (for ASIC design specifications and block diagrams). Access to these materials is restricted under the Protective Order entered by the Court on August 15, 2024. Review of these materials should be coordinated with Dr. Caroline Ashford (Plaintiff's Technical Expert, Professor of Electrical Engineering & Photonics at the Northfield Institute of Technology), who has been cleared for access to AEO materials under the Protective Order.
Dr. Ashford should be consulted to provide her expert opinion on whether the Lumicore DSP implementation — as disclosed in the produced design files — constitutes a structural equivalent of the DSP architecture described in the '193 Patent specification. This is a technically demanding comparison that will require detailed analysis of Lumicore's equalizer architecture, carrier-recovery algorithm, and FEC implementation.
Interplay with DOE Contentions for Claims 3 and 7: It is important to distinguish the § 112(f) structural equivalents analysis for the "module for processing" limitation in Claim 15 from the doctrine of equivalents analysis for the "coherent detection module" limitation in Claims 3 and 7. These are separate limitations in separate claims, targeting different functional elements (signal processing vs. optical detection). The § 112(f) structural equivalents test asks whether the accused structure performs the identical function in substantially the same way to achieve substantially the same result — a test applied to the specific limitation at issue. The broader doctrine of equivalents, applicable to Claims 3 and 7, uses the function-way-result test (or the insubstantial-differences test) with respect to the "coherent detection module" limitation. The prosecution history estoppel analysis relevant to the DOE contentions for Claims 3 and 7 is addressed in the separate PHE memorandum dated October 10, 2024, and is not implicated by the § 112(f) analysis for Claim 15.
6. Claim 18 (Dependent on Claim 1)
Claim 18 depends from Claim 1 and adds specifications regarding polarization handling (e.g., a polarization-diversity scheme using a polarization beam splitter and rotator) and thermal tuning (e.g., micro-heaters for wavelength tuning of resonant elements). These added limitations recite specific structural components — polarization beam splitters, polarization rotators, micro-heaters — that are well-known in the art and provide definite structural content.
Analysis. No limitation in Claim 18 uses "means" language or employs a nonce term coupled with functional language without structural recitation. Each added limitation describes a specific physical structure or component.
Conclusion: Section 112(f) is NOT invoked for any limitation of Claim 18.
C. '562 Patent — Claims 1, 4, 6, 11
The '562 Patent is a method patent, and accordingly the means-plus-function analysis focuses on whether any method step employs "step for" language within the meaning of § 112(f), rather than "means for" language. Under § 112(f), "[a]n element in a claim for a combination may be expressed as a means or step for performing a specified function." The Federal Circuit has held that the "step for" formulation applies to method claims in the same manner that "means for" applies to apparatus claims. See O.I. Corp. v. Tekmar Co., 115 F.3d 1576, 1583 (Fed. Cir. 1997). However, the use of "step for" language invokes § 112(f) only if the step is recited in functional terms without specifying the particular acts that constitute the step. If a method step recites specific acts to be performed, § 112(f) is not invoked even if the step is also described in functional terms.
1. Claim 1 (Independent)
Claim 1 of the '562 Patent is an independent method claim directed to a method for thermal management of photonic integrated circuits using embedded micro-channel heat sinks. The claim recites specific structural elements and specific method steps, including: forming a plurality of micro-channels in a substrate layer with defined channel dimensions; circulating a coolant fluid through the micro-channels; positioning the micro-channel heat sink in thermal contact with a photonic integrated circuit die; and monitoring temperature at one or more locations on the die. Each step recites specific physical acts and structural components.
Analysis. Claim 1 does not use "step for" language in any of its method steps. Each step recites specific acts (e.g., "forming a plurality of micro-channels," "circulating a coolant fluid") and structural elements (e.g., "a substrate layer," "micro-channels having a width of..."). The steps are not described in purely functional terms; rather, they specify the particular acts that constitute each step.
Conclusion: Section 112(f) is NOT invoked for any step of Claim 1.
2. Claim 4 (Dependent on Claim 1)
Claim 4 depends from Claim 1 and adds further method steps with structural specificity, including specific micro-channel dimensions (e.g., channel width and depth ranges), material specifications for the substrate (e.g., silicon or silicon-on-insulator), and coolant fluid characteristics (e.g., deionized water or dielectric fluorocarbon coolant). These additions provide additional specificity to the acts already recited in Claim 1 and do not introduce any new step described solely by function.
Analysis. No "step for" language is used. Each added step or sub-step recites specific acts and structural/material parameters.
Conclusion: Section 112(f) is NOT invoked for any step of Claim 4.
3. Claim 6 (Dependent on Claim 1)
Claim 6 depends from Claim 1 and adds temperature-sensing and feedback-control steps. Specifically, Claim 6 recites the use of specific sensor types (e.g., resistance temperature detectors or thermocouples integrated on the photonic integrated circuit die) and specific control logic (e.g., a proportional-integral-derivative (PID) controller for adjusting coolant flow rate based on sensed temperature). These steps recite specific acts and specific structural components for performing those acts.
Analysis. Although Claim 6 includes control-logic steps (PID controller), the claim does not use "step for" language and recites the specific type of control logic (PID) rather than merely reciting a generic function. The temperature-sensing step recites specific sensor types. There is no generic functional language without corresponding acts.
Conclusion: Section 112(f) is NOT invoked for any step of Claim 6.
4. Claim 11 (Dependent on Claim 1)
Claim 11 depends from Claim 1 and adds integration steps with photonic circuit components, such as steps for aligning the micro-channel heat sink with specific photonic devices on the die (e.g., high-power-dissipation elements such as laser sources and modulators) and steps for bonding the heat-sink substrate to the photonic circuit substrate. These steps recite specific physical acts and structural details.
Analysis. No "step for" language or generic functional language is used. Each step recites specific acts and structural elements.
Conclusion: Section 112(f) is NOT invoked for any step of Claim 11.
Note Regarding the '562 Patent Generally: The '562 Patent was originally assigned to Solara Optics, Inc. and was acquired by OWS via the Asset Purchase Agreement dated June 15, 2018, for a total portfolio price of $34.5 million covering twelve patents including the '562. Gregory Falk, formerly Vice President of Engineering at Solara Optics and now serving as OWS's industry expert, is available to provide testimony regarding the specific structural implementations disclosed in the '562 Patent specification if testimony regarding the thermal management structures or methods becomes necessary in connection with claim construction or infringement analysis.
D. '015 Patent — Claims 1, 5, 8, 12, 16
1. Claim 1 (Independent)
Claim 1 of the '015 Patent is an independent claim directed to a polarization-diverse grating coupler for silicon photonic transceivers. The claim recites structural elements including specific grating parameters (e.g., grating period, etch depth, duty cycle), a polarization-splitting element (e.g., a polarization beam splitter integrated on the silicon substrate), and waveguide geometry (e.g., a two-dimensional grating structure coupled to two orthogonal output waveguides for the two polarization states).
Analysis. Claim 1 does not use "means" language and does not employ any nonce terms coupled with functional language. Each limitation recites a specific physical structure or dimensional parameter. The terms used — "grating coupler," "polarization beam splitter," "waveguide" — are structural terms well understood in the silicon photonics art.
Conclusion: Section 112(f) is NOT invoked for any limitation of Claim 1.
2.