DCT
2:19-cv-00225
Ramot At Tel Aviv University Ltd v. Cisco Systems Inc
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: Ramot at Tel Aviv University Ltd. (Israel)
- Defendant: Cisco Systems, Inc. (California)
- Plaintiff’s Counsel: Bunsow De Mory LLP
- Case Identification: 2:19-cv-00225, E.D. Tex., 12/12/2019
- Venue Allegations: Plaintiff alleges venue is proper because Defendant maintains permanent and continuous places of business in the Eastern District of Texas, specifically facilities in Richardson and Allen, and has committed acts of infringement within the district.
- Core Dispute: Plaintiff alleges that Defendant’s high-speed optical networking equipment, including its 100G, 200G, and 400G optical transceivers, infringes patents related to methods for linearizing optical digital-to-analog modulators.
- Technical Context: The technology concerns converting high-speed digital data into modulated optical signals for transmission over fiber-optic networks, a fundamental process for modern data centers and telecommunications infrastructure.
- Key Procedural History: The complaint alleges that Defendant was aware of the patented technology due to a prior lawsuit filed by Plaintiff against Defendant on November 5, 2014, involving two parent patents of the patents-in-suit; that action was later voluntarily dismissed without prejudice.
Case Timeline
| Date | Event |
|---|---|
| 2007-06-13 | Earliest Priority Date for Asserted Patents |
| 2014-11-05 | Plaintiff filed prior suit against Defendant on parent patents |
| 2018-07-24 | U.S. Patent No. 10,033,465 Issued |
| 2019-04-23 | U.S. Patent No. 10,270,535 Issued |
| 2019-10-29 | U.S. Patent No. 10,461,866 Issued |
| 2019-12-12 | Complaint Filing Date |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 10,270,535 - "Linearized Optical Digital-to-Analog Modulator"
Issued April 23, 2019
The Invention Explained
- Problem Addressed: The patent addresses the "inherent non-linear response" of common optical modulators, such as the Mach-Zehnder Interferometer (MZI), when used for analog applications (Compl. ¶19; ’465 Patent, col. 1:55-58). This non-linearity can distort the signal, limiting performance in high-bandwidth, multi-GHz communication systems (Compl. ¶17).
- The Patented Solution: The invention proposes a method where an N-bit digital input is not mapped directly to N controls. Instead, it is converted by a digital-to-digital converter into a larger, M-bit actuation vector (where M>N) that drives M electrodes of an optical modulator (’535 Patent, Abstract; Compl. ¶17). By selecting a specific M-bit pattern from a pool of available patterns, the system can generate an optical output that more closely approximates a desired linear or other functional response, thereby compensating for the modulator's natural non-linearity (’465 Patent, col. 8:15-24).
- Technical Importance: This approach allows for the use of advanced, multi-level modulation schemes like Quadrature Amplitude Modulation (QAM) and Pulse Amplitude Modulation (PAM) with improved linearity, enabling higher data rates and greater signal clarity in fiber-optic communications (Compl. ¶17, ¶19).
Key Claims at a Glance
- The complaint asserts independent claims 1 and 2.
- Independent Claim 1 (Pulse Modulation): A method comprising:
- inputting N bits of digital data (N>1) into an optical modulator;
- converting the N bits to M drive voltage values, where M>N;
- coupling the M drive voltages to enable pulse modulation of an unmodulated optical signal; and
- transmitting the resulting pulse modulated signal over an optical fiber.
- Independent Claim 2 (QAM Modulation): A method comprising:
- inputting N bits of digital data (N>1) into an optical modulator;
- converting the N bits to M drive voltage values, where M>N;
- coupling the M drive voltages to enable QAM modulation of an unmodulated optical signal; and
- transmitting the resulting QAM modulated signal over an optical fiber.
U.S. Patent No. 10,033,465 - "Linearized Optical Digital-to-Analog Modulator"
Issued July 24, 2018
The Invention Explained
- Problem Addressed: The patent targets the same technical problem of non-linear response in optical modulators used for converting high-speed digital data into modulated optical signals (Compl. ¶19; ’465 Patent, col. 1:55-58).
- The Patented Solution: The solution is a method of mapping an N-bit digital input to a vector of M voltage values, where M is equal to or larger than N (’465 Patent, Claim 1). This vector of voltages drives at least M electrodes of the modulator to produce a pulse modulated or QAM modulated output with improved linearity (’465 Patent, col. 7:41-54). The core concept, as with the ’535 patent, is using a non-direct mapping to correct for the modulator's non-linear behavior (’465 Patent, col. 8:15-24).
- Technical Importance: The invention offers a way to achieve "improved linearity of response without sacrificing efficiency or dynamic range" in high-speed optical communications, enabling more effective use of advanced modulation techniques (Compl. ¶19).
Key Claims at a Glance
- The complaint asserts independent claims 1 and 4.
- Independent Claim 1 (Pulse Modulation): A method comprising:
- inputting N bits of digital data (N>1) into an optical modulator;
- mapping the N input values to a vector of M voltage values, where M≥N; and
- driving at least M electrodes to pulse modulate an input optical stream.
- Independent Claim 4 (QAM Modulation): A method comprising:
- inputting N bits of digital data (N>1) into an optical modulator;
- mapping the N input values to a vector of M voltage values, where M≥N; and
- driving at least M electrodes to modulate by QAM an input optical stream.
U.S. Patent No. 10,461,866 - "Linearized Optical Digital-to-Analog Modulator"
Issued October 29, 2019
Technology Synopsis
The ’866 patent describes a modulation system to provide for the "linearization of a modulator device which inherently has a non-linear response" (Compl. ¶21). It discloses using a digital-to-digital converter to map an N-bit input data word to a digital drive vector of M voltage values, where M=N, to produce a pulse modulated output signal with improved linear response (Compl. ¶21, ¶64).
Asserted Claims
The complaint asserts independent claim 19 (Compl. ¶61).
Accused Features
The accused features are Cisco’s products that use advanced PAM4 modulation, including various modules and associated digital signal processing components that allegedly implement the claimed digital-to-digital conversion and modulation (Compl. ¶61, ¶64).
III. The Accused Instrumentality
Product Identification
The complaint identifies Cisco’s networking equipment, including routers, switches, and associated line cards and optical transceiver modules that provide for high-speed (100G, 200G, and 400G) optical communications (Compl. ¶2, ¶23). Specific product families named include the Cisco Network Convergence System (NCS) 4000 Series and Cisco Nexus 400G products, along with pluggable modules such as CFP2, QSFP56, and QSFP-DD (Compl. ¶23, ¶24, ¶30). The complaint collectively refers to these as the "Accused Products" (Compl. ¶30).
Functionality and Market Context
The Accused Products are alleged to employ advanced modulation techniques, including 16-QAM and 4-level Pulse Amplitude Modulation (PAM4), to achieve high data transport speeds (Compl. ¶27). The complaint alleges that these products contain Digital Signal Processors (DSPs) and/or ASICs that perform mapping of digital symbols to correct for modulator non-linearity, which is necessary to meet industry signal quality standards at high speeds (Compl. ¶29). The complaint provides an image of a Cisco NCS 4000 line card, which it alleges provides digital signal processing for optical ports (Compl. ¶24, ¶34, p. 15). Cisco is alleged to have publicly endorsed the adoption of techniques like PAM4 to achieve data rates toward 400G and beyond (Compl. ¶28).
IV. Analysis of Infringement Allegations
'535 Patent Infringement Allegations
| Claim Element (from Independent Claim 2) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| inputting the N bits of digital data into an optical modulator... | The Accused Products' line cards and modules receive multi-bit digital words or symbols that are to be converted into optical signals. | ¶34 | col. 17:54-56 |
| converting the N bits of digital data to M drive voltage values, where M>N and N>1 | Cisco's CFP2 modules allegedly process digital symbols via mapping and encoding modules in an ASIC that convert the data into corresponding voltage values for driving modulator electrodes, where M>N. The complaint provides a diagram from a technical paper showing an ASIC with symbol mapping and encoder blocks to support this. | ¶37, ¶21 | col. 18:13-16 |
| coupling the M drive voltage values to the unmodulated optical signal, said coupling enabling modulation of the unmodulated optical signal by QAM... | The generated voltage values are coupled via drivers to the branches of a Mach-Zehnder modulator to generate a QAM modulated optical signal. A block diagram showing the driver inputs to the modulator arms is provided as evidence. | ¶38, ¶22 | col. 18:16-19 |
| transmitting the QAM modulated optical signal over an optical fiber. | The Accused Products have optical fiber ports for transmitting the modulated optical signals over various types of optical fiber. | ¶39 | col. 18:20-21 |
'465 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| inputting into an optical modulator N bits of digital data in parallel, N being larger than 1 | The Accused Products' modules and signal processing components receive multiple-bit inputs of digital data for conversion. | ¶50, ¶51 | col. 17:5-8 |
| mapping a set of N input values corresponding to said N bits of digital data to a vector of M voltage values where M is equal to or larger than N | Cisco's PAM4 modules allegedly use signal processing components, such as FEC and PAM encoder modules, to map input digital symbols to corresponding voltage values for driving the modulator. The complaint presents a block diagram for a segmented MZI PAM-4 transmitter showing these encoder modules. | ¶51, ¶20 | col. 17:9-12 |
| driving at least M electrodes of the optical modulator, enabled to pulse modulate at least an input optical stream... to provide at least a pulse modulated output optical stream. | The generated voltage values are used to drive the electrodes of the optical modulator to produce a pulse modulated optical signal. | ¶52 | col. 17:13-17 |
Identified Points of Contention
- Scope Questions: A central question will be whether the "mapping" and "encoding" functions performed by the accused DSPs and ASICs, which are often designed to comply with industry standards, fall within the scope of the claimed "converting" or "mapping" steps. The defense may argue that the patents describe a specific, proprietary linearization technique distinct from standard-compliant symbol encoding.
- Technical Questions: A key factual dispute may arise over the exact relationship between N (input bits) and M (drive voltage values or electrodes) in the accused products. The asserted patents have distinct N-to-M requirements (M>N for ’535, M≥N for ’465, and M=N for ’866), meaning infringement will depend on a precise technical finding for each product and each patent. The complaint alleges these different relationships are met, but does not provide direct evidence from the products themselves.
V. Key Claim Terms for Construction
The Term: "converting the N bits of digital data to M drive voltage values" ('535 Patent) and "mapping a set of N input values... to a vector of M voltage values" ('465 Patent)
- Context and Importance: These terms define the core inventive step. The case will likely hinge on whether the functionality of the accused DSPs—performing symbol mapping and encoding per industry standards—constitutes the specific non-trivial conversion/mapping process taught in the patents for the purpose of linearization.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The claim language itself is relatively broad and does not explicitly limit the "converting" or "mapping" to a specific algorithm.
- Evidence for a Narrower Interpretation: The specification describes the process as choosing an electrode actuation pattern that "best approximates a desired ideal output for the given input" from a set of possibilities, a process designed to "offer improved linearity of response" (’465 Patent, Abstract; col. 1:27-29). The defense may argue this limits the term to a specific optimization-based mapping, not just any encoding scheme.
The Term: "digital to digital converter" ('866 Patent)
- Context and Importance: Asserted claim 19 of the ’866 patent requires the mapping step to be performed by a "digital to digital converter." The construction of this term is critical, as the complaint identifies components like a Broadcom ASIC as the infringing structure (Compl. p. 29).
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: A defendant might argue for a plain and ordinary meaning, where any digital circuit that transforms one set of digital bits into another could qualify.
- Evidence for a Narrower Interpretation: The shared specification explicitly defines the term as a "non-trivial converter in which there is typically not a one-to-one mapping between bits of the input data and bits of the output data" (’465 Patent, col. 4:62-68). Plaintiff may leverage this definition to argue the accused ASICs, which convert, for example, 4x25G traffic into 2x50G PAM4, perform such a non-trivial mapping (Compl. p. 29).
VI. Other Allegations
Indirect Infringement
The complaint alleges Cisco induces infringement by providing customers with datasheets, brochures, and other documents that instruct on how to use the Accused Products in an infringing manner (Compl. ¶42, ¶55, ¶68). It also alleges contributory infringement, stating the products are especially made for infringing use and are not staple articles of commerce (Compl. ¶43, ¶56, ¶69).
Willful Infringement
The complaint alleges that Cisco's infringement has been willful. This allegation is primarily based on a prior lawsuit filed by Ramot against Cisco in 2014 concerning parent patents, which allegedly made Cisco aware of the patented technology and the infringing nature of its activities (Compl. ¶44, ¶57, ¶70). The complaint further alleges that a parent patent was cited by the examiner during the prosecution of one of Cisco's own patents (Compl. ¶44, ¶57, ¶70).
VII. Analyst’s Conclusion: Key Questions for the Case
- A core issue will be one of functional equivalence: Does the symbol mapping performed by the accused standard-compliant DSPs and ASICs constitute the specific, non-trivial "mapping" for linearization as taught by the patents, or is it a fundamentally different, standard-encoding process that falls outside the claims' scope?
- A second key issue will be one of evidentiary proof: Can Plaintiff demonstrate that the accused products meet the specific N-to-M bit/voltage relationships required by the different asserted patents (i.e., M>N for the ’535 patent, M≥N for the ’465 patent, and M=N for the ’866 patent)? The outcome of this factual determination will be critical for assessing infringement on a claim-by-claim basis.
- Finally, the case will present a significant question of willfulness: Given the allegation of a prior lawsuit on parent patents, a central question for damages will be whether Cisco's continued activities after being put on notice constitute willful infringement, potentially leading to enhanced damages.