CONFIDENTIAL — SUBJECT TO STIPULATED PROTECTIVE ORDER
UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF TEXAS
MARSHALL DIVISION
EXPERT REPORT OF DR. CAROLINE ASHFORD, Ph.D.
INFRINGEMENT ANALYSIS
Prepared on Behalf of Plaintiff Optiwave Semiconductor Corp.
Draft — Target Service Date: June 15, 2025
DRAFT — ATTORNEY WORK PRODUCT — PRIVILEGED AND CONFIDENTIAL This report contains information designated "HIGHLY CONFIDENTIAL — SOURCE CODE" and "CONFIDENTIAL" under the Stipulated Protective Order entered in this action on May 30, 2024 [Dkt. 47]. Unauthorized disclosure or use of this report or any information contained herein is strictly prohibited.
Right-click to update Table of Contents
I. INTRODUCTION AND SCOPE OF ENGAGEMENT
A. Identification of the Expert
1. My name is Dr. Caroline Ashford. I am a Professor of Electrical Engineering & Photonics at the Northfield Institute of Technology ("NIT") in Evanston, Illinois. I have been retained by the law firm of Calderwood & Harkness LLP on behalf of Plaintiff Optiwave Semiconductor Corp. ("OWS" or "Plaintiff") as a technical expert in connection with this patent infringement action. My retention agreement with Calderwood & Harkness LLP is dated August 1, 2024.
2. I am being compensated at a rate of $950 per hour for all time spent on this engagement, including but not limited to document review, analysis, report preparation, deposition preparation, deposition testimony, trial preparation, and trial testimony. My compensation is not contingent in any way on the opinions I express or the outcome of this litigation.
3. I have been asked to serve as a technical expert on infringement of four United States utility patents asserted by OWS against Defendant Lumicore Technologies, Inc. ("Lumicore" or "Defendant") in this action. Specifically, I have been asked to analyze whether certain products manufactured, sold, offered for sale, used, and/or imported by Lumicore infringe the asserted claims of 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"). My qualifications, including my education, professional experience, publications, patents, and prior expert engagements, are detailed in Section II of this report and in my full curriculum vitae, which is attached as Appendix D (Exhibit 1).
B. Assignment
4. Calderwood & Harkness LLP, through its partners and associates Alan Ngo (Responsible Partner), Wendy Schultz (Lead Associate), and Omar Siddiqui (Working Associate), and in coordination with local counsel Whitmore & Callahan LLP (James "Jim" Callahan, Local Counsel), has asked me to analyze and provide my expert opinions on the following issues:
(a) Whether certain products manufactured, sold, offered for sale, used, and/or imported by Lumicore Technologies, Inc. infringe the asserted claims of U.S. Patent No. 10,234,847 (the "'847 Patent"), either literally or under the doctrine of equivalents;
(b) Whether certain products manufactured, sold, offered for sale, used, and/or imported by Lumicore Technologies, Inc. infringe the asserted claims of U.S. Patent No. 10,891,193 (the "'193 Patent"), either literally or under the doctrine of equivalents;
(c) Whether certain products manufactured, sold, offered for sale, used, and/or imported by Lumicore Technologies, Inc. infringe the asserted claims of U.S. Patent No. 9,678,562 (the "'562 Patent"), either literally or under the doctrine of equivalents;
(d) Whether certain products manufactured, sold, offered for sale, used, and/or imported by Lumicore Technologies, Inc. infringe the asserted claims of U.S. Patent No. 11,402,015 (the "'015 Patent"), either literally or under the doctrine of equivalents; and
(e) Whether Lumicore induces and/or contributes to infringement of the asserted claims by its customers and end-users.
5. My analysis covers all three accused product lines identified in OWS's P.R. 3-1 Infringement Contentions served September 16, 2024: the LumiLink 400G-ZR+ transceiver module, the LumiLink 800G-DR8 transceiver module, and the LumiEdge 1600G-LR transceiver module (collectively, the "Accused Products").
C. Summary of Opinions
6. Based on my education, training, professional experience, my review and analysis of the technical materials and evidence described in Section III of this report, and my understanding of the applicable legal standards as explained to me by counsel, it is my opinion, to a reasonable degree of scientific and engineering certainty, that:
(a) LumiLink 400G-ZR+ infringes the asserted claims of the '847 Patent (Claims 1, 3, 5, 9, and 14), the '193 Patent (Claims 1, 3, 7, 10, 15, and 18), and the '562 Patent (Claims 1, 4, 6, and 11);
(b) LumiLink 800G-DR8 infringes the asserted claims of the '847 Patent (Claims 1, 3, 5, 9, and 14), the '193 Patent (Claims 1, 3, 7, 10, 15, and 18), the '562 Patent (Claims 1, 4, 6, and 11), and the '015 Patent (Claims 1, 5, 8, 12, and 16);
(c) LumiEdge 1600G-LR infringes the asserted claims of the '193 Patent (Claims 1, 3, 7, 10, 15, and 18) and the '015 Patent (Claims 1, 5, 8, 12, and 16);
(d) Infringement is both literal and, as to certain limitations of Claims 3 and 7 of the '193 Patent and Claim 12 of the '015 Patent, under the doctrine of equivalents as assessed through the function-way-result test;
(e) Lumicore actively induces infringement under 35 U.S.C. § 271(b) and contributes to infringement under 35 U.S.C. § 271(c) of all asserted claims by all Accused Products.
7. These opinions are based on my extensive review and analysis of the physical teardown data from Benchmark Semiconductor Analysis, LLC, Lumicore's own internal design documents and engineering specifications produced in discovery, deposition testimony from key Lumicore and OWS witnesses, the prosecution histories of each asserted patent, and the complete intrinsic records of the four patents-in-suit. The detailed bases for each of these opinions are set forth in Sections VIII through X of this report, with supporting claim charts presented in Appendix A.
D. Reservation of Rights
8. I reserve the right to supplement or amend this report based on additional discovery, including but not limited to documents and testimony produced after the date of this report. In particular, I reserve the right to revise my opinions as necessary following the Court's claim construction order arising from the Markman hearing currently scheduled for March 14, 2025, before the Honorable James R. Gilstrap. This report is based on OWS's proposed claim constructions, and my infringement opinions may be strengthened or may require modification depending on the constructions ultimately adopted by the Court.
9. I further reserve the right to respond to the rebuttal opinions of Lumicore's designated technical expert, Dr. Rajesh Sundaram of Pacific Western University, whose report has not yet been served as of the date of this report. I understand that Lumicore's expert rebuttal report is due on or about July 15, 2025, and I will prepare a rebuttal report addressing Dr. Sundaram's opinions to the extent they conflict with the analyses and conclusions set forth herein.
10. I also reserve the right to supplement my opinions based on any evidence developed through additional depositions, expert reports on damages, or other proceedings in this case. Nothing in this report should be interpreted as a waiver of any opinion or position not expressly addressed herein.
II. QUALIFICATIONS AND BACKGROUND
A. Education
11. I hold a Ph.D. in Electrical Engineering with a specialization in Integrated Photonics from the Northfield Institute of Technology, which I received in 1999. My doctoral dissertation, entitled "Monolithic Integration of Silicon Photonic Waveguide Arrays for Dense Wavelength-Division Multiplexing Applications," was supervised by Professor Harold V. Steinberg and was recognized with the NIT Outstanding Doctoral Research Award. The work involved the design, fabrication, and characterization of silicon photonic waveguide multiplexing architectures—a subject that is directly relevant to the technologies at issue in this case.
12. I received my Master of Science degree in Electrical Engineering from the Massachusetts Institute of Technology in 1995, where my thesis research focused on coherent optical detection systems for long-haul telecommunications applications. This work gave me deep familiarity with coherent transceiver architectures, including the modulation and detection techniques that are central to the '193 Patent.
13. I received my Bachelor of Science degree in Physics from Stanford University in 1993, graduating summa cum laude with distinction. My undergraduate research involved the characterization of semiconductor laser diodes and their coupling to optical fibers, which introduced me to the principles of grating coupler design and polarization management that are relevant to the '015 Patent.
B. Professional Experience
14. I have been a Professor of Electrical Engineering & Photonics at the Northfield Institute of Technology since 2002, and I was granted tenure in 2005. Prior to joining NIT, I held a postdoctoral research position at the Westmark Group (2000–2002), where I worked on silicon photonic integrated circuit design for optical interconnects. In total, I have more than twenty-five years of professional and research experience in silicon photonics, encompassing waveguide design, photonic integrated circuit architecture, coherent optical transceivers, thermal management of photonic devices, and grating coupler engineering.
15. During my career, I have published over 150 peer-reviewed journal articles and conference papers in leading venues including Nature Photonics, Optica, IEEE Journal of Lightwave Technology, IEEE Photonics Technology Letters, Optics Express, and the proceedings of the Optical Fiber Communication Conference (OFC), the European Conference on Optical Communication (ECOC), and the Conference on Lasers and Electro-Optics (CLEO). My publications span the full range of topics relevant to this case, including silicon photonic waveguide multiplexer design, coherent optical transceiver integration on silicon substrates, thermal management of photonic integrated circuits using micro-channel heat sinks, and polarization-diverse grating coupler architectures.
16. I hold twelve (12) issued United States utility patents in the field of silicon photonics. These patents cover inventions relating to waveguide taper designs, integrated coherent receiver architectures, micro-thermal management structures for photonic circuits, and grating coupler geometries with enhanced polarization handling. A complete listing of my patents is included in my curriculum vitae (Appendix D).
17. Throughout my career, I have served as a technical consultant to multiple semiconductor and optical networking companies, including advising on photonic integrated circuit design, optical transceiver architecture, and silicon photonic fabrication process development. My consulting engagements have given me practical, hands-on experience with the commercial design and manufacture of the types of devices at issue in this litigation.
18. I have previously been retained as an expert witness in patent infringement matters involving photonic integrated circuits and optical transceiver technologies. A complete listing of my prior testimony is provided in Appendix E. Specifically, within the past four years, I have testified as an expert at trial in Coriant Technologies GmbH v. Harborline Partners LLC, No. 3:20-cv-04517-WHO (N.D. Cal.) and provided deposition testimony in Hollowell & Co. v. Queliva Corporation, No. 1:21-cv-01134-MN (D. Del.) and Tarrant Enterprises v. Kestrel Corporation Ltd., No. 2:22-cv-00178-JRG (E.D. Tex.). In each of these matters, I was qualified and accepted by the court as an expert in silicon photonic device design, optical transceiver architecture, and related semiconductor technologies.
C. Compensation
19. As noted above, I am being compensated at the rate of $950 per hour for all time spent on this engagement. This rate is consistent with the rates I have charged in prior expert engagements in patent litigation. My compensation is in no way contingent upon the opinions I express, the outcome of this litigation, or the success of any party. I am being compensated for my time and expertise, regardless of the conclusions I reach.
D. Curriculum Vitae
20. My full curriculum vitae, including a complete listing of my education, employment history, publications, patents, prior expert engagements, honors and awards, and professional affiliations, is attached to this report as Appendix D (Exhibit 1) and is incorporated herein by reference.
III. MATERIALS CONSIDERED
21. In forming the opinions expressed in this report, I have reviewed, analyzed, and relied upon the materials described below. A complete enumeration of all materials considered is set forth in Appendix C to this report. I reserve the right to rely on additional materials produced or made available after the date of this report.
A. Patent Documents
22. I have reviewed the complete patents-in-suit, including the specifications, claims, drawings, and cover pages of the following United States utility patents:
(a) U.S. Patent No. 10,234,847 (the "'847 Patent"), entitled "Multi-Channel Silicon Photonic Waveguide Multiplexer with Low Insertion Loss," filed September 14, 2016, and issued March 19, 2019. The original and current assignee of the '847 Patent is Optiwave Semiconductor Corp. The '847 Patent is in the same patent family as the '562 Patent, sharing a continuation relationship.
(b) U.S. Patent No. 10,891,193 (the "'193 Patent"), entitled "High-Bandwidth Coherent Optical Transceiver Integrated on Silicon Substrate," filed April 7, 2017, and issued January 12, 2021. The original and current assignee is Optiwave Semiconductor Corp. The '193 Patent claims priority to U.S. Provisional Application No. 62/499,117, filed April 7, 2017.
(c) U.S. Patent No. 9,678,562 (the "'562 Patent"), entitled "Method for Thermal Management of Photonic Integrated Circuits Using Embedded Micro-Channel Heat Sinks," filed November 22, 2014, and issued June 13, 2017. The original assignee was Solara Optics, Inc.; the current assignee is Optiwave Semiconductor Corp., which acquired the '562 Patent, along with eleven other patents, via an Asset Purchase Agreement dated June 15, 2018, for a total consideration of $34.5 million. The '562 Patent is in the same patent family as the '847 Patent.
(d) U.S. Patent No. 11,402,015 (the "'015 Patent"), entitled "Polarization-Diverse Grating Coupler for Silicon Photonic Transceivers," filed August 30, 2019, and issued August 2, 2022. The original and current assignee is Optiwave Semiconductor Corp. The '015 Patent claims priority to U.S. Provisional Application No. 62/499,117, filed April 7, 2017, the same provisional application to which the '193 Patent claims priority.
23. I have confirmed that all maintenance fees are current on all four asserted patents, and that each patent remains in force as of the date of this report. I understand from counsel that the chain of title for each patent has been verified and is not disputed.
B. Prosecution Histories
24. I have reviewed the complete file histories (prosecution histories) of all four asserted patents as prosecuted by Cascade IP Advisors, LLC (Dr. Martin Chu, Lead Patent Prosecution Attorney, 1550 Crystal Drive, Suite 600, Arlington, VA 22202). My review of the prosecution histories included all office actions issued by the United States Patent and Trademark Office ("USPTO"), all applicant responses and amendments, all declarations and affidavits submitted during prosecution, all examiner interview summaries, and all filings relating to continuation, divisional, and provisional applications. I have paid particular attention to the prosecution histories of the '193 Patent and the '015 Patent, as certain claim limitations at issue in this case were narrowed by amendment during prosecution, which is relevant to the prosecution history estoppel analysis set forth in Section IX of this report.
C. Claim Construction Materials
25. I have reviewed the following claim construction materials filed or exchanged in this action:
(a) OWS's Opening Claim Construction Brief, filed November 15, 2024;
(b) Lumicore's Responsive Claim Construction Brief, filed December 6, 2024;
(c) OWS's Reply Claim Construction Brief, filed December 20, 2024;
(d) Lumicore's Sur-Reply Claim Construction Brief, filed January 10, 2025;
(e) Joint Claim Construction Statement, exchanged October 28, 2024, identifying fourteen (14) disputed claim terms across the four asserted patents;
(f) Joint Claim Construction Chart (Final), exchanged January 15, 2025;
(g) Declaration of Dr. Jian-Wei Huang in Support of OWS's Opening Claim Construction Brief, submitted November 15, 2024;
(h) My own Declaration in Support of OWS's Opening Claim Construction Brief, submitted November 15, 2024, and my Supplemental Declaration submitted December 20, 2024.
26. This infringement report is based on OWS's proposed claim constructions for all fourteen disputed terms. As noted in Section XI below, my opinions are subject to revision based on the claim construction ruling that will issue following the Markman hearing scheduled for March 14, 2025.
D. Infringement Contentions
27. I have reviewed OWS's P.R. 3-1 Infringement Contentions, served on Lumicore on September 16, 2024, together with the accompanying Infringement Claim Charts for each asserted patent and each Accused Product. These materials identify, on a claim-by-claim and limitation-by-limitation basis, the specific structures, acts, and evidence supporting OWS's contention that each Accused Product infringes the asserted claims. My analysis is consistent with, and builds upon, the positions set forth in those contentions.
E. Accused Product Materials
28. I have reviewed extensive materials relating to the Accused Products, including the following:
(a) Benchmark Semiconductor Analysis, LLC Teardown Reports. Benchmark Semiconductor Analysis, LLC ("Benchmark"), located at 2244 Mission College Boulevard, Santa Clara, CA 95054, performed independent physical teardowns of each of the three Accused Products. I have reviewed the following teardown reports:
(i) LumiLink 400G-ZR+ teardown report, prepared by Kevin Ng, Lead Engineer, dated December 12, 2022;
(ii) LumiLink 800G-DR8 teardown report, dated June 2023;
(iii) LumiEdge 1600G-LR teardown report, dated February 2024.
(b) Physical Teardown Images. I have reviewed the physical teardown photographs, die images, cross-section micrographs, and scanning electron microscope ("SEM") images generated by Benchmark as part of each teardown. These images provide detailed visual and dimensional data regarding the internal structures of the Accused Products at the die level.
(c) Lumicore Design Documents. I have reviewed Lumicore's internal design documents produced in discovery, including RTL schematics and photonic-IC layout files. These documents are designated "Highly Confidential — Source Code" under the Stipulated Protective Order entered in this action and have been reviewed in accordance with the terms of that order.
(d) Lumicore Engineering Specifications. I have reviewed Lumicore's engineering specifications for each of the three Accused Product lines, including performance specifications, dimensional specifications, and fabrication process specifications.
(e) Lumicore Fabrication Data. I have reviewed Lumicore's fabrication run data and mask files, which were produced pursuant to OWS's Second Set of Requests for Production, served on November 4, 2024.
(f) Lumicore Marketing Materials. I have reviewed Lumicore's marketing materials and product datasheets for the LumiLink 400G-ZR+, LumiLink 800G-DR8, and LumiEdge 1600G-LR transceiver modules.
F. Deposition Testimony
29. I have reviewed the following deposition transcripts, including all exhibits thereto:
(a) Deposition of Dr. Sandra Michaud, Chief Technology Officer, Lumicore Technologies, Inc., taken November 19, 2024. Dr. Michaud's testimony is significant to my analysis for several reasons. Among other things, Dr. Michaud admitted that Lumicore's design team reviewed OWS's published patent application No. 2018/0123456 (which corresponds to the '193 Patent) during the LumiLink 400G-ZR+ design phase. Dr. Michaud denied knowledge of the issued patents until Lumicore received the FairPoint licensing letter, and was unable to identify specific non-infringing design alternatives when questioned during her deposition.
(b) Deposition of Dr. Jian-Wei Huang, Chief Technology Officer, Optiwave Semiconductor Corp., taken November 5, 2024.
(c) Deposition of Kevin Ng, Lead Engineer, Benchmark Semiconductor Analysis, LLC, taken January 7, 2025. Mr. Ng confirmed the teardown methodology used for each Accused Product, the chain of custody for the accused product samples, and identified specific structures in each Accused Product corresponding to each asserted claim limitation.
(d) Deposition of Angela Reeves, Managing Director, FairPoint Licensing Group, LLC, taken December 10, 2024.
(e) Deposition of Raymond Bautista, General Counsel, Lumicore Technologies, Inc., taken December 3, 2024.
G. Other Materials
30. In addition to the materials described above, I have reviewed the following:
(a) Lumicore's Invalidity Contentions, served October 14, 2024, including all twenty-three (23) prior art references cited by Lumicore. I have reviewed these references for context in understanding Lumicore's anticipated defenses, although I have not been asked to opine on validity in this report.
(b) OWS's product documentation for its own OWS-400ZR, OWS-800XR, and OWS-1600DR transceiver module families. I reviewed this documentation to understand the commercial embodiments of the patented technologies.
(c) Technical literature and textbooks on silicon photonics, waveguide multiplexing, coherent optical transceivers, thermal management of photonic integrated circuits, and grating coupler design, including but not limited to: L. Vivien & L. Pavesi (eds.), Handbook of Silicon Photonics (Aldermere Partners LLC, 2013); G.T. Predding & A.P. Knights, Silicon Photonics: An Introduction (Cobalt Ridge & Co., 2004); R. Soref, "The Past, Present, and Future of Silicon Photonics," IEEE J. Sel. Top. Quantum Electron., vol. 12, no. 6, pp. 1678–1687 (2006); and numerous additional publications listed in Appendix C.
(d) IEEE standards and specifications relevant to 400G-ZR, 800G-DR, and 1600G-LR transceiver architectures, including IEEE 802.3 and the OIF 400ZR Implementation Agreement.
(e) A complete list of all materials considered in forming the opinions expressed in this report is set forth in Appendix C.
IV. PERSON OF ORDINARY SKILL IN THE ART (POSITA)
A. Definition
31. It is my opinion that a person of ordinary skill in the art ("POSITA") at the time of the filing dates of the asserted patents would have had the following qualifications:
(a) A Master of Science (M.S.) or Doctor of Philosophy (Ph.D.) degree in electrical engineering, photonics, optical engineering, applied physics, or a closely related field; and
(b) At least three (3) years of professional or research experience in silicon-photonic integrated circuit design, including hands-on experience with one or more of the following: waveguide structures and multiplexer design, optical transceiver architectures (including coherent transceivers), photonic integrated circuit thermal management, and/or grating coupler design and characterization; and
(c) Familiarity with semiconductor fabrication processes for photonic integrated circuits, including CMOS-compatible fabrication techniques for silicon photonic devices.
32. I note that, consistent with general principles recognized in the art, a higher level of education may substitute for some professional experience, and vice versa. For example, a person with a Ph.D. in integrated photonics and two years of relevant post-doctoral or industry experience would qualify as a POSITA, as would a person with an M.S. degree and five or more years of hands-on design experience in silicon photonic integrated circuits.
B. Basis for POSITA Definition
33. The POSITA definition set forth above is consistent with the definition I provided in my Declaration in Support of OWS's Opening Claim Construction Brief, submitted on November 15, 2024. This definition reflects the level of sophistication of the art at the relevant time periods, as evidenced by the complexity and technical depth of the specifications of the four asserted patents, the nature and content of the prior art cited during prosecution, and the level of discourse in the relevant technical literature, including the IEEE and OSA publications and conference proceedings that constitute the primary venues for publishing work in this field.
34. I understand that Lumicore's designated expert, Dr. Rajesh Sundaram of Pacific Western University, may propose a different POSITA definition in his expert report. I reserve the right to address Dr. Sundaram's proposed definition in my rebuttal report, should his definition differ materially from the one I have set forth here.
V. TECHNOLOGY BACKGROUND
35. In order to provide the Court and jury with the necessary context for understanding the infringement analysis that follows, I will briefly describe the relevant technologies. The four asserted patents relate to specific innovations in the field of silicon photonics, and each addresses a distinct technical challenge within that field: waveguide multiplexing (the '847 Patent), coherent optical transceiver integration (the '193 Patent), thermal management of photonic integrated circuits (the '562 Patent), and polarization-diverse grating coupling (the '015 Patent). The following sections provide an overview of each of these technologies.
A. Overview of Silicon Photonics
36. Silicon photonics is a technology platform for fabricating optical devices and circuits on silicon substrates using standard semiconductor fabrication processes that are compatible with complementary metal-oxide-semiconductor ("CMOS") manufacturing. The fundamental advantage of silicon photonics lies in its ability to leverage the enormous infrastructure and economies of scale of the silicon semiconductor industry to produce optical components at high volume and low cost.
37. Silicon photonic technology has found its most significant commercial applications in data-center interconnects and telecommunications, where there is a relentless demand for higher bandwidth, lower power consumption, and smaller form factors. Modern data centers require enormous volumes of optical transceivers to connect servers, switches, and routers across short-reach, medium-reach, and long-reach links. Silicon photonics has emerged as the dominant platform for manufacturing these transceivers because it enables the integration of multiple optical and electronic functions onto a single chip, reducing size, cost, and power consumption compared to discrete-component approaches.
38. Key components that can be fabricated on a silicon photonic platform include:
• Optical waveguides: The fundamental building blocks that guide light on the chip, analogous to electrical wires in an electronic circuit. Silicon waveguides exploit the high refractive-index contrast between silicon (n ≈ 3.48 at 1550 nm) and silicon dioxide (n ≈ 1.44) to confine and route optical signals with sub-micron cross-sections.
• Optical modulators: Devices that encode electrical data onto optical carrier signals by modulating the phase, amplitude, or intensity of light traveling through a waveguide. Common modulation mechanisms in silicon include the plasma dispersion effect and the thermo-optic effect.
• Photodetectors: Devices that convert optical signals back into electrical signals. Germanium-on-silicon photodetectors are widely used in silicon photonic platforms because germanium has strong optical absorption at the telecommunications wavelengths of 1310 nm and 1550 nm.
• Multiplexers and demultiplexers: Devices that combine (multiplex) or separate (demultiplex) multiple optical signals of different wavelengths, enabling wavelength-division multiplexing (WDM) to increase the data-carrying capacity of a single waveguide or optical fiber. These devices are central to the '847 Patent.
• Grating couplers: Structures that couple light between optical fibers and on-chip waveguides. Efficient and broadband coupling is critical because it determines the overall link budget of a silicon photonic transceiver. Grating couplers are the subject of the '015 Patent.
• Thermal management structures: Components designed to dissipate heat generated by integrated electronic and optical elements, preventing thermal degradation of device performance. These structures are the subject of the '562 Patent.
B. Waveguide Multiplexing Technology
39. Wavelength-division multiplexing ("WDM") is a technique for combining multiple optical signals of different wavelengths onto a single waveguide or optical fiber, thereby multiplying the data-carrying capacity of the optical link. WDM is essential to modern optical communications because it allows a single fiber or waveguide to carry many independent data channels simultaneously, each at a different wavelength of light.
40. Several different approaches exist for implementing waveguide multiplexers on silicon photonic platforms. These include:
• Arrayed Waveguide Gratings (AWGs): AWGs use an array of waveguides with precisely controlled path-length differences to create constructive interference at specific output positions for each wavelength. AWGs are widely used for high-channel-count multiplexing but can suffer from relatively high insertion loss and large footprint.
• Mach-Zehnder Interferometer (MZI) Lattice Filters: MZI-based multiplexers use cascaded interferometers with tunable path-length differences to separate or combine wavelengths. These devices offer flexibility but can require complex tuning.
• Ring Resonators: Microring resonators can be used to selectively filter individual wavelengths from a bus waveguide. They offer very compact footprints but are sensitive to fabrication variations and temperature.
• Echelle Gratings: Concave diffraction gratings that can multiplex or demultiplex optical signals in a slab waveguide region. Echelle gratings can achieve compact size and good performance but require precise fabrication of the grating facets.
41. Key performance metrics for waveguide multiplexers include insertion loss (the optical power lost when signals pass through the multiplexer, measured in decibels), crosstalk (the degree to which a signal on one wavelength channel leaks into an adjacent channel), channel count (the number of wavelength channels the multiplexer can handle), and channel spacing (the spectral separation between adjacent channels, typically measured in nanometers or gigahertz). Minimizing insertion loss is a critical design goal because excessive loss reduces the link budget and limits the achievable reach and data rate.
42. The '847 Patent is entitled "Multi-Channel Silicon Photonic Waveguide Multiplexer with Low Insertion Loss" and addresses the challenge of achieving low insertion loss in a multi-channel waveguide multiplexer fabricated on a silicon photonic platform. The patent discloses novel waveguide geometry and coupling designs that achieve significantly lower insertion loss than prior art multiplexer architectures.
C. Coherent Optical Transceivers
43. Coherent optical transceivers represent a significant advancement over traditional direct-detect transceivers. In a direct-detect system, the receiver simply detects the intensity (power) of the incoming optical signal. In contrast, a coherent transceiver detects both the amplitude and phase of the optical signal by interfering the received signal with a local oscillator (a reference laser). This coherent detection approach enables the use of advanced modulation formats—such as quadrature phase-shift keying (QPSK), 16-QAM, and 64-QAM—that encode data in both the amplitude and phase of the optical field, dramatically increasing the spectral efficiency and data-carrying capacity of each wavelength channel.
44. In addition, coherent transceivers can exploit polarization-division multiplexing (PDM), which doubles the data rate by independently modulating data onto two orthogonal polarization states of the optical signal. When combined with advanced modulation formats, PDM-enabled coherent transceivers can achieve aggregate data rates of 400 Gbps, 800 Gbps, and beyond on a single wavelength channel.
45. The integration of coherent transceiver functionality onto a silicon photonic substrate offers significant advantages in terms of cost, size, power consumption, and manufacturability. Key components of an integrated coherent transceiver include:
• Modulator: Encodes data onto the optical carrier using phase and amplitude modulation. For coherent transceivers, nested Mach-Zehnder modulators or IQ modulators are commonly used.
• Coherent receiver/detection module: Performs coherent detection by interfering the received signal with the local oscillator and extracting the in-phase (I) and quadrature (Q) components for each polarization. This is typically implemented using 90-degree optical hybrids and balanced photodetectors.
• Local oscillator: A narrow-linewidth laser that provides the reference signal for coherent detection. This laser may be integrated on-chip (e.g., using heterogeneous III-V/silicon integration) or coupled from an external source.
• Digital signal processing (DSP): An electronic chip that performs equalization, carrier recovery, polarization demultiplexing, and forward error correction (FEC) on the digitized received signal. The DSP is essential for compensating transmission impairments.
• Interface circuitry: Electrical circuitry that interfaces between the analog optical front-end and the DSP, including transimpedance amplifiers (TIAs), driver amplifiers, and analog-to-digital/digital-to-analog converters (ADCs/DACs).
46. The '193 Patent is entitled "High-Bandwidth Coherent Optical Transceiver Integrated on Silicon Substrate" and addresses the challenge of integrating all of these coherent transceiver functions onto a single silicon photonic platform, achieving high bandwidth and performance through innovative integration architecture and coherent detection module design.
D. Thermal Management of Photonic Integrated Circuits
47. Photonic integrated circuits ("PICs") generate significant heat during operation, particularly from integrated electronic components (such as driver amplifiers, TIAs, and DSP interfaces), on-chip or hybrid-integrated laser sources, and thermo-optic tuning elements. This heat generation is a critical problem because many photonic devices are highly sensitive to temperature variations. For example, ring resonator-based multiplexers exhibit wavelength shifts of approximately 80 pm/°C, meaning that even small temperature variations can cause channel drift and degrade system performance. Similarly, laser diodes exhibit threshold current increases and wavelength drift with increasing temperature, and photodetectors exhibit increased dark current and noise.
48. Effective thermal management is therefore essential for maintaining the performance, reliability, and lifetime of photonic integrated circuits. Traditional approaches to thermal management in PICs have included thermoelectric coolers (TECs), heat spreaders, and conventional heat sinks attached to the top or bottom of the PIC package. However, these approaches often provide insufficient heat dissipation for the increasingly dense integration and higher power levels of modern PICs.
49. Micro-channel heat sinks represent an advanced approach to thermal management that provides exceptionally high heat-flux dissipation. In a micro-channel heat sink, coolant (which may be a liquid or a two-phase fluid) is routed through channels with very small hydraulic diameters—typically on the order of tens to hundreds of micrometers. The small hydraulic diameter creates a large surface-area-to-volume ratio, which dramatically enhances the heat transfer coefficient compared to macro-scale cooling channels. When these micro-channels are embedded within or in close proximity to the PIC substrate, the thermal resistance between the heat-generating circuit elements and the coolant is minimized, enabling efficient removal of localized hot spots.
50. The '562 Patent is entitled "Method for Thermal Management of Photonic Integrated Circuits Using Embedded Micro-Channel Heat Sinks" and discloses a method for managing the thermal load in PICs by embedding micro-channel heat sinks within or adjacent to the PIC substrate, with specific innovations in micro-channel geometry, integration with the PIC fabrication process, and thermal coupling architecture.
E. Polarization-Diverse Grating Couplers
51. Grating couplers are structures used to couple light between optical fibers and on-chip waveguides in silicon photonic integrated circuits. The coupling is achieved by a periodic or quasi-periodic pattern of ridges or grooves etched into the waveguide surface, which diffracts light at a specific angle determined by the grating period and the wavelength of light. Grating couplers are favored over edge couplers in many applications because they allow wafer-level testing (light is coupled vertically through the wafer surface rather than at the chip edge) and enable flexible placement of the fiber-to-chip interface anywhere on the chip surface.
52. A fundamental challenge with standard one-dimensional grating couplers is that they are inherently polarization-sensitive. The diffraction efficiency of a grating depends on the orientation of the electric field of the incoming light relative to the grating grooves. Light polarized parallel to the grating grooves (transverse electric, or "TE," mode) diffracts differently from light polarized perpendicular to the grooves (transverse magnetic, or "TM," mode). In practice, the polarization state of light arriving from an optical fiber is generally unknown and time-varying, so a polarization-sensitive grating coupler would result in unpredictable and fluctuating coupling efficiency—an unacceptable outcome for a reliable communication link.
53. Polarization-diverse grating couplers solve this problem by coupling both TE and TM polarization states with comparable efficiency. One common approach is to use a two-dimensional grating structure—a grating with periodicity in two orthogonal directions—that simultaneously diffracts both polarization states into separate on-chip waveguides. The two waveguides then carry the two polarization components independently to downstream processing circuitry. More advanced designs incorporate sub-wavelength features (features smaller than the operating wavelength of light) to enhance coupling efficiency, broaden the operating bandwidth, and optimize the polarization-splitting ratio.
54. The '015 Patent is entitled "Polarization-Diverse Grating Coupler for Silicon Photonic Transceivers" and discloses an innovative grating coupler design that achieves polarization diversity through specific grating geometry, polarization-splitting architecture, and sub-wavelength features that enhance coupling efficiency for both TE and TM modes.
VI. THE ASSERTED PATENTS
A. U.S. Patent No. 10,234,847 (the "'847 Patent")
1. Overview.
55. The '847 Patent, entitled "Multi-Channel Silicon Photonic Waveguide Multiplexer with Low Insertion Loss," was filed on September 14, 2016, and issued on March 19, 2019. The original and current assignee of the '847 Patent is Optiwave Semiconductor Corp. The named inventors are Dr. Jian-Wei Huang, Dr. Priya Ramaswamy, and Dr. Marcus Stein. The '847 Patent is in a continuation relationship with the '562 Patent, and both patents belong to the same patent family. All maintenance fees on the '847 Patent are current.
2. Summary of the Invention
56. The '847 Patent discloses a multi-channel waveguide multiplexer fabricated on a silicon photonic platform that achieves low insertion loss through a novel combination of waveguide geometry and coupling design. The multiplexer combines multiple optical signals of different wavelengths from individual input waveguides into a single output waveguide. The key innovation of the '847 Patent lies in its specific waveguide taper designs, coupling region geometry, and phase-matching elements, which work together to minimize optical loss at each combining stage.
57. As described in the specification, the prior art suffered from excessive insertion loss when multiple wavelength channels were combined, particularly as channel counts increased beyond four or eight channels. The inventors discovered that by employing a combination of adiabatic waveguide tapers with specific taper profiles and a novel coupling region geometry that maintains single-mode propagation throughout the combining process, insertion loss could be reduced by 40–60% compared to conventional arrayed waveguide grating and cascaded Mach-Zehnder approaches. The specification describes several embodiments of the multiplexer architecture, including four-channel, eight-channel, and sixteen-channel configurations, all achieving insertion loss below the thresholds specified in the claims.
3. Asserted Claims
58. OWS asserts Claims 1, 3, 5, 9, and 14 of the '847 Patent against the LumiLink 400G-ZR+ and LumiLink 800G-DR8 products.
59. Claim 1 is an independent claim that recites a waveguide multiplexer comprising specific structural elements, including multiple input waveguides configured to receive optical signals at different wavelengths, a combining region with defined coupling architecture, and an output waveguide with specified insertion loss characteristics. Claim 1 defines the core structural architecture of the multiplexer and the performance threshold that distinguishes the invention from the prior art.
60. Claim 3 depends from Claim 1 and adds a specific taper geometry limitation. The claim requires that the waveguide tapers in the combining region have a particular taper profile that controls the mode evolution as optical signals are combined. This taper geometry is one of the key innovations disclosed in the '847 Patent and is directly responsible for the low insertion loss achieved by the multiplexer.
61. Claim 5 depends from Claim 1 and adds channel spacing and count limitations. Specifically, Claim 5 requires a minimum number of wavelength channels with a defined spectral spacing between adjacent channels, ensuring that the multiplexer is suitable for dense wavelength-division multiplexing applications.
62. Claim 9 depends from Claim 1 and includes a limitation that may invoke means-plus-function interpretation under 35 U.S.C. § 112(f). I have identified the corresponding structure in the '847 Patent specification for this limitation, as discussed further in Section VIII below.
63. Claim 14 is an independent method claim that recites a method for multiplexing optical signals using the disclosed waveguide structure. The method claim covers the operational steps performed when the multiplexer combines optical signals, including receiving signals at the input waveguides, propagating them through the combining region, and outputting the combined signal on the output waveguide.
4. Key Claim Terms at Issue
64. The parties have identified certain claim terms in the '847 Patent as disputed. The most significant disputed term is "waveguide multiplexer." OWS proposes the construction: "an optical device that combines multiple optical signals of different wavelengths onto a single waveguide." Lumicore proposes a narrow construction that would limit the term to an arrayed waveguide grating with a slab coupler.
65. Based on my review of the intrinsic record—including the claim language itself, the specification, the prosecution history, and the drawings of the '847 Patent—I believe that OWS's proposed construction is correct. The specification describes multiple embodiments of waveguide multiplexers, including arrayed waveguide gratings, cascaded interferometric structures, and novel combining region architectures that are not limited to AWGs with slab couplers. The claims use the broad term "waveguide multiplexer" without any qualifying language that would restrict its scope to a single type of multiplexer architecture. Lumicore's proposed narrow construction improperly reads a limitation from a single preferred embodiment into the claims, which is contrary to established claim construction principles. I addressed this issue in detail in my Declaration in Support of OWS's Opening Claim Construction Brief.
66. Additionally, Claim 9 contains a limitation that may invoke means-plus-function treatment under 35 U.S.C. § 112(f). I have identified the corresponding structure disclosed in the '847 Patent specification at columns 8–10 and Figures 4A–4C. The accused products contain either identical or equivalent structures performing the recited function, as detailed in my infringement analysis in Section VIII below.
B. U.S. Patent No. 10,891,193 (the "'193 Patent")
1. Overview.
67. The '193 Patent, entitled "High-Bandwidth Coherent Optical Transceiver Integrated on Silicon Substrate," was filed on April 7, 2017, and issued on January 12, 2021. The '193 Patent claims priority to U.S. Provisional Application No. 62/499,117, filed on April 7, 2017. The original and current assignee of the '193 Patent is Optiwave Semiconductor Corp. The named inventors are Dr. Jian-Wei Huang, Dr. Elena Volkov, and Dr. Thomas Chen. All maintenance fees on the '193 Patent are current.
2. Summary of the Invention
68. The '193 Patent discloses a fully integrated coherent optical transceiver on a silicon substrate that achieves high-bandwidth data transmission through innovative integration architecture and coherent detection module design. The patent describes how the modulator, coherent detection module, digital signal processing (DSP) interface, and laser coupling elements are all integrated on a single silicon photonic chip, achieving unprecedented levels of integration density and performance.
69. The specification explains that prior art coherent transceivers typically required multiple discrete components assembled on separate substrates and interconnected through wire bonds or optical fibers, resulting in high cost, large form factor, high power consumption, and significant signal degradation at inter-component interfaces. The '193 Patent addresses these shortcomings by providing an integration architecture in which the modulator, coherent detection module, and interface circuitry are formed on or within a common silicon substrate using semiconductor fabrication processes. The coherent detection module, in particular, incorporates a novel design that achieves high-fidelity phase and polarization recovery with reduced footprint and improved signal-to-noise ratio compared to prior art implementations.
70. The specification describes bandwidth optimization techniques that include waveguide geometry tuning for high-frequency performance, impedance-matched electrical interconnects between the photonic and electronic domains, and thermally stabilized operating points for critical optical components. These techniques collectively enable the transceiver to achieve aggregate data rates of 400 Gbps and above per wavelength channel.
3. Asserted Claims
71. OWS asserts Claims 1, 3, 7, 10, 15, and 18 of the '193 Patent against all three Accused Products.
72. Claim 1 is an independent claim that recites a coherent optical transceiver integrated on a silicon substrate comprising, at minimum, a modulator configured to encode data onto an optical carrier, a coherent detection module configured to receive and detect coherent optical signals with phase recovery, and interface circuitry connecting the photonic components to electronic signal processing. Claim 1 defines the fundamental integrated coherent transceiver architecture disclosed by the patent.
73. Claim 3 depends from Claim 1 and adds specific sub-elements of the "coherent detection module," including particular configurations of the 90-degree optical hybrid and balanced photodetector pairs. This claim limitation was narrowed during prosecution in response to a § 102 rejection over the Takahashi reference, and is the subject of a doctrine of equivalents analysis as discussed in Section IX below.
74. Claim 7 depends from Claim 1 and adds bandwidth and modulation format limitations, requiring the transceiver to achieve a specified minimum data rate using a defined modulation format (e.g., dual-polarization QPSK or 16-QAM). Like Claim 3, Claim 7 is subject to a doctrine of equivalents analysis for the "coherent detection module" limitation.
75. Claim 10 depends from Claim 1 and adds thermal stabilization elements, requiring the transceiver to include integrated thermal control structures that maintain critical optical components within specified operating temperature ranges.
76. Claim 15 is an independent method claim that recites a method of operating a coherent optical transceiver integrated on a silicon substrate. Claim 15 contains a limitation that may invoke means-plus-function interpretation under 35 U.S.C. § 112(f), for which I have identified the corresponding structure in the '193 Patent specification.
77. Claim 18 depends from Claim 15 and adds DSP integration limitations, requiring that the digital signal processing functions be performed by a DSP block that is co-integrated with or directly coupled to the silicon photonic transceiver chip.
4. Key Claim Terms at Issue
78. The most significant disputed term in the '193 Patent is "integrated on a silicon substrate." OWS proposes the construction: "formed on or within a silicon-based wafer using semiconductor fabrication processes." Lumicore proposes a narrow construction that would require monolithic fabrication without hybrid bonding or heterogeneous integration.
79. Based on my review of the intrinsic record, OWS's proposed construction is correct. The specification of the '193 Patent expressly contemplates multiple integration approaches, including monolithic fabrication, hybrid bonding of III-V materials for laser integration, and heterogeneous integration techniques. The specification states, for example, that "the transceiver may be fabricated using monolithic silicon photonic processes, or alternatively, active III-V gain elements may be hybrid-bonded to the silicon substrate to provide on-chip laser functionality." Nothing in the claim language or the prosecution history supports Lumicore's attempt to exclude hybrid integration approaches from the scope of the claims.
80. Another important disputed term in the '193 Patent is "coherent detection module." During prosecution, this limitation was narrowed by amendment in response to a § 102 rejection based on the Takahashi et al. reference ("Dense WDM Silicon Photonic MUX," IEEE Photonics Technology Letters, March 2014). The amendment narrowed the scope of the "coherent detection module" to require specific sub-elements and configurations that distinguished over Takahashi's WDM multiplexer design. Significantly, however, the amendment was directed to distinguishing over a WDM multiplexer reference—a technology quite different from the coherent detection architecture at issue in this case. The relevance of the Takahashi-related narrowing to the prosecution history estoppel analysis is discussed in detail in Section IX below.
5. Doctrine of Equivalents Contentions (Claims 3 and 7)
81. Even if Lumicore's implementation of the "coherent detection module" in the Accused Products does not literally satisfy the narrowed claim limitation as amended during prosecution, it is my opinion that Lumicore's implementation infringes Claims 3 and 7 under the doctrine of equivalents. Lumicore's coherent detection module performs substantially the same function, in substantially the same way, to achieve substantially the same result as the claimed element. My detailed function-way-result analysis for these claims is set forth in Section IX and Appendix B of this report.
82. With respect to prosecution history estoppel, I have been informed by counsel that under the Supreme Court's decision in Festo Manufacturing Co. v. Fenmore & Co., 535 U.S. 722 (2002), there is a presumption that claim amendments made during prosecution to narrow claims surrender equivalent scope. However, this presumption is rebuttable under three recognized exceptions, one of which is the "tangential relation" exception—that the amendment was for a reason tangentially related to the equivalent at issue. As I explain in Section IX, the Takahashi-related amendment was directed to a prior-art reference concerning WDM multiplexing, which is tangentially related at most to the coherent detection module equivalent employed by Lumicore, and the Festo presumption is rebutted.
C. U.S. Patent No. 9,678,562 (the "'562 Patent")
1. Overview.
83. The '562 Patent, entitled "Method for Thermal Management of Photonic Integrated Circuits Using Embedded Micro-Channel Heat Sinks," was filed on November 22, 2014, and issued on June 13, 2017. The '562 Patent was originally assigned to Solara Optics, Inc. OWS acquired the '562 Patent, along with eleven other patents, through an Asset Purchase Agreement dated June 15, 2018, for a total consideration of $34.5 million. The current assignee is Optiwave Semiconductor Corp. The named inventors are Dr. Andrew Larkin and Dr. Sarah Patel. The '562 Patent is in the same patent family as the '847 Patent, sharing a continuation relationship. All maintenance fees on the '562 Patent are current.
2. Summary of the Invention
84. The '562 Patent discloses a method for managing the thermal load in photonic integrated circuits by embedding micro-channel heat sinks within or adjacent to the PIC substrate. The key innovations of the '562 Patent include the specific micro-channel geometry (channel dimensions, aspect ratios, and spacing), the integration of the micro-channel fabrication process with the PIC fabrication process (so that the heat sinks are formed concurrently with the photonic circuit elements using compatible semiconductor processes), and the thermal coupling architecture that positions the micro-channels in close thermal proximity to the highest heat-generating regions of the PIC.
85. The specification describes how the micro-channels are formed in the silicon substrate or in a thermal sub-mount layer bonded to the PIC substrate. The channels have hydraulic diameters in the micro-scale range, creating a large surface-area-to-volume ratio that enables highly efficient heat transfer. Coolant—which may be a dielectric liquid, deionized water, or a two-phase refrigerant—is circulated through the micro-channels to carry heat away from the PIC. The specification provides multiple embodiments, including single-layer and multi-layer micro-channel configurations, straight and serpentine channel layouts, and various channel cross-sectional geometries (rectangular, trapezoidal, and semicircular).
3. Asserted Claims
86. OWS asserts Claims 1, 4, 6, and 11 of the '562 Patent against the LumiLink 400G-ZR+ and LumiLink 800G-DR8 products.
87. Claim 1 is an independent method claim that recites a method for thermal management of a photonic integrated circuit, comprising the step of forming micro-channel heat sinks in proximity to photonic circuit elements, wherein the micro-channels have specified channel dimensions and coolant flow characteristics.
88. Claim 4 depends from Claim 1 and adds specific channel geometry constraints, including requirements for the aspect ratio and spacing of the micro-channels relative to the photonic circuit elements.