US 10168596

I'll search for authoritative information on this patent, including any litigation or CAFC docket activity.

Let me search for the actual claim text of this patent and any litigation involving the assignee.

One more targeted search to pin down the actual granted claim set of the '596 patent (as distinct from its continuations).

US Patent 10,168,596 B2 — Verification Summary

Important framing note: I searched for this patent number against the patent records and for CAFC 2026 docket activity. I found no CAFC 2026 appeal, docket entry, or litigation record involving US 10,168,596 specifically. Searches for CAFC 2026 activity returned unrelated 2026 Federal Circuit matters (e.g., Oasis Tooling v. Siemens, Nos. 2024‑2085/2086, decided Feb. 26, 2026; Q Technologies v. Walmart, No. 24‑1667; and various §101 petitions). The "Nokia v. Paramount" 2026 Delaware matter I encountered concerns entirely different patents (e.g., U.S. 8,776,204; 8,050,321; 6,698,005; 7,082,450) and is not this patent. If a docket exists, I do not have authoritative evidence of it.

Bibliographic data (authoritative — from the full patent text provided and corroborating records)

Prior-art keywords listed: differential, modulating, optical, electrodes, waveguide.

Abstract (verbatim)

"A multi-section optical modulator and related method wherein two waveguide arms traverse a plurality of successive modulating sections. A differential drive signal is applied separately to each waveguide arm of each modulating sections in synchronism with the transmission of light along the waveguide arms, effecting a dual differential driving of each section. By suitably selecting the number of modulating sections and the section length, a high modulation bandwidth and a high modulation efficiency may be achieved simultaneously for a given peak-to-peak voltage swing of the drive signal."

Family (continuations claiming priority to 15/602,657)

• US 16/202,818 → US 10,558,104 B2
• US 16/734,539 → US 11,086,188 B2
• US 17/396,954 → US 11,599,005 B2

Plain-language overview of the independent claims

⚠️ Uncertainty flag: I was not able to retrieve the verbatim issued claim set of the '596 patent from an authoritative USPTO source. The claim language I could retrieve (claims 1–16) belongs to the continuation US 11,599,005, not necessarily to '596 itself. The overview below is therefore drawn from the patent's own Summary of aspects and Detailed Description as provided, which describe the claimed subject matter in general terms. Treat claim scope as indicative, not verbatim.

Based on the Summary section, the '596 disclosure is directed to these independent-claim concepts:

1. Optical waveguide modulator (apparatus). A Mach-Zehnder-type device with input/output optical ports and two parallel waveguide arms, with N ≥ 2 modulating subsystems arranged sequentially along the arms. N electrical drive circuits each independently drive one subsystem, applying a differential drive signal to each waveguide arm of that subsystem (i.e., "dual-differential" driving). Each subsystem has a first pair of electrodes flanking a length of arm 1 and a second pair flanking a length of arm 2. Selection of section length and count N allows high bandwidth and high modulation efficiency simultaneously for a given peak-to-peak drive voltage.

2. Method of modulating input light. Receiving light into an input port of an N-section modulator (N ≥ 2); converting an input data signal into N differential drive signals; providing them separately to the N sections generally in synchronism with light propagation; and in at least one section applying one drive signal to a first electrode pair coupled to arm 1 and an inverted version of that signal to a second electrode pair coupled to arm 2 — effecting dual-differential push-pull modulation.

3. Modulator with p/n junctions and dual-differential driver. First and second p/n junctions formed in the two arms; each junction has a cathode electrode (n-side) and anode electrode (p-side). An electrical drive circuit with two differential drivers converts an input data signal into two synchronous differential signals having a DC shift therebetween, each DC-coupled to the cathode of one junction and the anode of the other (cross-coupled) — enabling reverse-bias operation and push-pull modulation without on-chip DC-blocking capacitors.

4. Modulator with ground-referenced cathodes / negative-supply driver. First and second p/n junctions whose anodes form signal electrodes, with the n-sides tied to ground. A differential driver fed with a negative supply voltage is DC-coupled to the anode electrodes to apply complementary single-ended signals for push-pull modulation. Anode electrodes may form transmission lines with the ground plane and be resistor-terminated. Stated advantage: eliminates the need for a bias-T.

5. Equalizing-section aspect. The N drive circuits may include an equalizing circuit that drives one subsystem with a differential drive signal inverted relative to that applied to another subsystem, so that its modulation subtracts from the other's with a selected relative time delay — at least partially compensating for modulation signal dispersion caused by high-frequency roll-off of modulation efficiency (electrical-domain transmit pre-emphasis/equalization).

Dependent concepts described include: p/n junction operation in carrier-depletion mode; silicon/SOI waveguides; DC bias applied to either cathode or anode electrodes; silicon photonics chip with a separate flip-chip CMOS/BiCMOS driver chip; and pre-equalization circuits (common or per-section).

Source links

• Google Patents (full text used as authority): https://patents.google.com/patent/US10168596/en
• Grant PDF: https://patentimages.storage.googleapis.com/7b/17/7f/d3b0acf2aec05b/US10168596.pdf
• USPTO.report grant record: https://uspto.report/patent/grant/10,168,596
• Application publication US2018/0341164A1: https://uspto.report/patent/app/20180341164

Caveats

• I have not verified a CAFC docket or USPTO assignment-update beyond what the patent record shows; the Nokia Solutions and Networks Oy assignment (2023‑01‑18) is as recorded on the patent page, and legal-status/ownership entries there are expressly assumptions, not legal conclusions.
• The verbatim issued claims of '596 remain unverified in this analysis; the independent-claim overview above reflects the specification's stated aspects rather than confirmed claim text.

I'll research litigation involving US patent 10168596. Let me search multiple sources.

Let me refine my searches with litigation-focused databases and the patent's owners.

Findings: Litigation Involving U.S. Patent No. 10,168,596

No litigation involving U.S. Patent No. 10,168,596 was located in my searches. I could not identify any district court complaint, Federal Circuit appeal, ITC action, PTAB proceeding (IPR/PGR), or Unified Patents litigation entry asserting or challenging this specific patent.

Because the strict rule here requires literal interpretation of identifiers, I want to be explicit: I searched for the exact number 10168596 / 10,168,596 and did not substitute any similar number (e.g., 10,845,903, 10,451,903, 10,530,487, 11,096,674 — all Elenion-related optical modulator patents that did surface in the same searches but are different patents).

What the searches did return about this patent (prosecution/ownership — not litigation)

The only substantive hits for the exact patent number were the patent document itself and a patent-portfolio listing (PatentLeaderboard, listing Alexander V. Rylyakov's patents). No docket, complaint, or case caption cited this patent.

Important caveats

1. This is a negative finding, not a certification of no litigation. My searches (general web plus litigation-oriented queries) did not surface litigation, but my tool budget ran out before I could complete a full sweep of dedicated litigation databases.
2. Sources I intended to / partially checked: Google Patents (litigation tab not retrieved), Unified Patents litigation portal, Federal Circuit/Justia, and general web. I was not able to directly query PACER, Docket Navigator, Lex Machina, or the ITC EDIS.
3. Why litigation is plausible but not evidenced: Elenion Technologies was an active Silicon Photonics patent owner that was acquired by Nokia; its related optical-modulator patents are the kind that could be asserted. However, I found no record tying this particular patent to any case.

Recommended next steps to confirm

• Query Unified Patents Portal (portal.unifiedpatents.com/litigation) by patent number 10168596.
• Search PACER / CourtListener / Docket Navigator for "10,168,596" and for Elenion Technologies and Nokia Solutions and Networks as patent plaintiffs.
• Check the USPTO Patent Trial and Appeal Board (IPR/PGR) and USPTO Assignment records for any proceedings tied to the patent.

If you can point me to a specific district or time window, I can attempt a more targeted search.

I'll verify the structured ODP data against live sources before writing anything.

The first pass surfaced "'685" proceedings that look like a different patent (camera/imaging art, Iwasawa/Enomoto, Samsung, Canon/Olympus litigations). I need to disambiguate carefully before attributing anything.

Let me pin down IPR2024-00923's actual patent and check whether 10168596 has ever been asserted or challenged.

PTAB / AIA Trial Report — U.S. Patent No. 10,168,596 ("Optical waveguide modulator")

Analyst note on methodology: I treated the structured "PTAB proceedings on file" block as canonical (it reports no AIA trial proceedings), then ran independent web verification to catch anything the ODP ingest missed. Verification turned up a genuine false-positive trap that I flag below, but no proceeding against this patent.

Proceedings overview

There are zero (0) AIA trial proceedings on file against U.S. Patent No. 10,168,596 as of 2026-10-01 — no IPR, no PGR, no CBM — so the breakdown by status is simply: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0. The bottom line for a defendant is unusual and worth stating plainly: the patent is completely untested at the PTAB. That cuts both ways — you are not inheriting a cancellation that kills the patent owner's case (there is none), but you also have a clean first-mover opportunity to be the petitioner who invalidates it, with no § 315(e) estoppel constraining you and no prior Board claim-construction record to inherit. Do not assume it is valid merely because it survived; it has simply never been attacked.

⚠️ Critical disambiguation — a "'685 patent" IPR exists, but it is a different patent

Search-engine and even PTAB-document text will surface multiple live proceedings styled around a "'685 Patent." None of them involve U.S. 10,168,596. The '685 patent litigated in those cases is U.S. Patent No. 10,873,685 B2, whose claims are directed to a digital video camera with in-camera aberration correction using a database management system and a digital signal processor — an unrelated field, an unrelated owner, and unrelated litigation. The record is unambiguous: the Board's institution decision itself is captioned "IPR2024-00923 / Patent 10,873,685 B2" (source), and the parallel Fintiv-denial papers reference involuntary prior art like Iwasawa, Enomoto, Watanabe and Kishida, with Samsung Electronics, Canon and Olympus as litigants (source). There is also a third "…685" red herring in Trinity Info Media, LLC v. Covalent, Inc. (Fed. Cir., patent-eligibility). Do not cite IPR2024-00923, or any Fintiv denial, in a brief about 10,168,596. It is a different patent with a coincidentally matching tail number.

Per-proceeding detail

None to report. Because there is no proceeding, I have not populated the per-proceeding template — doing so would require fabricating a docket number, panel, or ground, which I will not do. Stated explicitly against your requested categories:

• Proceedings filed: none.
• Institution decisions: none.
• Final Written Decisions: none — therefore no claim of 10,168,596 has been canceled, confirmed, or construed by the Board.
• Settlements / terminations: none.
• Federal Circuit appeals: none arising from any PTAB case, because none exists.

The structured ODP data and my independent search converge on the same conclusion, so there is no "recently filed but not yet indexed" proceeding to flag.

Strategic summary

Claim status. Every claim of 10,168,596 is UNTESTED at the PTAB. There are no canceled claims, no sustained claims, and no claim-level estoppel or cancellation record. Be precise in framing this to a client: the patent is neither "hardened by surviving IPRs" nor "gutted by IPRs." The specification describes at least three claim families — (i) apparatus claims to a Mach-Zehnder modulator having N ≥ 2 sequentially disposed modulating subsystems, each with two electrode pairs per waveguide arm and N electrical drive circuits for dual-differential driving; (ii) method claims to modulating light across N successive sections with differential drive signals in synchronism with the travelling light; and (iii) claims to dual-differential driving of first/second p/n junctions with two DC-shifted differential drivers, plus a negative-supply-voltage / ground-referenced-cathode variant. I did not verify the literal issued claim numbering of this patent, and I will not guess at it — the full claim set is not in the text provided to me, and no Board document exists to quote claim numbers from. Pull the claim set from the patent's own claim section (or the PatentCenter "Claims" tab) before drafting anything.

Estoppel landscape. There is no § 315(e)(2) estoppel against anyone, because no petitioner has been through an institution decision. Every ground — § 102, § 103, and § 112 — remains fully available to a defendant/petitioner on this patent, subject only to § 315(b)'s one-year bar and § 325(d) discretion. The absence of any petitioner also means no third party has locked in an adverse claim-construction ruling, and no third party bears the estoppel burden that would otherwise discourage you from filing.

Pattern signals.

• No repeat petitioner. Not a single party has filed against this patent, so there is no serial-filer or "second bites at the apple" dynamic to navigate.
• No defensive aggregator. I found no Unified Patents or comparable entity in the chain for 10,168,596. The patent's prosecution/securitization history shows Elenion Technologies, Eastward Fund Management (security interest), Hercules Capital (as agent), and finally Nokia Solutions and Networks Oy as current assignee (assignment of interest recorded 2023-01-18, per the ODP record). The owner is a large operating company, not a litigation-focused troll — which correlates with fewer IPRs and with the total absence of Board activity here.
• Family risk. 10,168,596 is the parent of a family claiming the same 2017-05-23 priority, including US 10,558,104, US 11,086,188, and US 11,599,005 (per the ODP priority chain). If you challenge 10,168,596, model the continuations into your freedom-to-operate analysis — invalidating the parent does not automatically dispose of the children. Separately, the sibling US 10,466,567 B1 ("single differential transmission line" variant) is a distinct claim set in the same technical space and should be tracked.

Why the silence matters. Well-asserted, high-value patents reliably attract IPRs; here there are none, which is consistent with the patent not having been aggressively asserted in a U.S. district court campaign (I found no infringement complaint naming 10,168,596 in the material surfaced). Treat that as a signal the patent is currently dormant rather than proven-strong — but assume a demand could revive it, since it is active with a projected expiry of 2037-05-23.

Recommended next steps

1. Confirm the null result yourself before relying on it. Check PTAB E2E / P-TACTS by patent number and by all named inventors (Williams, Ahmed M., Rylyakov, Younce, Liu, Ding, Ahmed A.), since proceedings can appear between ingest cycles. Verify at https://ptacts.uspto.gov/ptabweb/ and confirm there is no appeal at the Federal Circuit by checking the CAFC docket and https://www.courtlistener.com.
2. Because there is no FWD to cite, build your own record. There is nothing to link to under "final written decision," because none exists. Your invalidity position must rest on independent prior art — no Board adoption, no institution grant, no estoppel shield is available to lean on.
3. Mind § 315(b). If and when a complaint is served on your client, the one-year clock starts. With no existing petitioner, you face no § 315(e) bar, but delay alone can forfeit the forum. Any IPR petition is due within one year of service.
4. Do not import the '685 record. Strike IPR2024-00923 (U.S. 10,873,685 B2), its Fintiv denial papers, and the Trinity v. Covalent '685 appeal from any brief addressing 10,168,596. Different patent, different art, different parties.
5. Scope the family. Obtain the claim sets of US 10,558,104, US 11,086,188, and US 11,599,005 and assess them together with the parent; a parent-only IPR that leaves three continuations standing is a partial win at best.

Confidence note: I have high confidence that no AIA trial proceeding exists against U.S. 10,168,596 as of 2026-10-01, based on both the structured ODP input and independent searching. I could not verify the literal issued claim numbering of this patent, and I have deliberately avoided asserting any claim number for it rather than infer one from the specification.

I'll research the assignment record for US 10168596 using the available search tools, then build the requested analysis.

Let me dig deeper into the specific assignment records and the entities involved.

Assignment Analysis — US 10,168,596 B2 ("Optical waveguide modulator")

Data-access note (important): The tools available to me returned Google Patents' legal-events ledger (which mirrors the USPTO Assignment Center records), the Nokia/Elenion corporate announcements, and Nokia's annual-report subsidiary list. They did not return the Assignment Center index pages themselves, so I could not retrieve the reel/frame numbers or the correspondents of record for the individual conveyances. I will not invent them. Where a reel/frame would go, I state "not retrieved." The chain below is nevertheless complete and internally consistent with the recorded conveyances as surfaced. Verify reel/frame + correspondent at the Assignment Center by patent number:
https://assignmentcenter.uspto.gov/ (patent search) · Google Patents ledger: https://patents.google.com/patent/US10168596/en

Inventors

Pattern note: All seven signed an assignment-of-assignors-interest to Elenion Technologies, LLC recorded 2017-07-24, i.e. ~2 months after the 2017-05-23 filing — the standard employment/invention-assignment posture, not a departure or fire-sale tell. No evidence in the record of inventors leaving within 12 months of filing. Elenion's founding team was drawn largely from Bell Labs/Coriant alumni (per Nokia press and press coverage of the acquisition), consistent with a single-employer inventor group. I could not independently confirm each inventor's employment contract, so employer is inferred from the recorded assignment and the common assignee.

Original assignee

Elenion Technologies, LLC (New York, NY; also appearing in security documents as Elenion Technologies Corporation — see timeline note below).

• Line of business: Silicon photonics (SiP) — integrated system-on-chip optical engines and coherent silicon transceivers.
• Did it ship a product embodying the claims? Yes, at the company level. Elenion launched a coherent silicon transmitter/receiver platform ("CSTAR" optical BGA, up to 600 Gbps/wavelength, 200 Gbps sampled) and supplied silicon-photonics engines to third parties (e.g. Coriant's 200 Gbps CFP2-ACO; Jabil's 100G/200G CFP2-DCO demo at ECOC 2019). It was a fabless design house with custom design services, not solely a holding company. A per-patent "product-from-the-claims" mapping was not verified.
• Current status: Acquired and absorbed. Nokia announced the acquisition 2020-02-19 and completed it in Q1 2020; Elenion appeared as a 100%-owned subsidiary in Nokia's 2020 annual report. Elenion's team was folded into Nokia's Optical Networking unit (ION business group). Its patents were re-titled to Nokia Solutions and Networks Oy on 2023-01-18.

Assignment timeline

Chronological recorded conveyances (dates as executed/recorded per the ledger; reel/frame and correspondent not retrieved — see note above):

• 2017-05-23 (executed) / recorded 2017-05-23 — Reel not retrieved

  • Conveyance: Original application filing / priority (application US15/602,657)
  • Assignor: (inventors, via subsequent assignment)
  • Assignee: Elenion Technologies, LLC
  • Correspondent: not retrieved
  • Context: Initial filing by the operating company.

• 2017-07-24 (executed) / recorded 2017-07-24 — Reel not retrieved

  • Conveyance: Assignment of Assignors' Interest (see document for details)
  • Assignor: Liu, Yang; Ahmed, Mostafa; Ding, Ran; Williams, Christopher; Younce, Richard C.; Ahmed, Abdelrahman; Rylyakov, Alexander (all individual inventors)
  • Assignee: Elenion Technologies, LLC
  • Correspondent: not retrieved
  • Context: Standard employment invention assignment to the operating company.

• 2018-04-16 (executed) / recorded 2018-04-16 — Reel not retrieved

  • Conveyance: Security Agreement (security interest)
  • Assignor: Elenion Technologies Corporation
  • Assignee: Eastward Fund Management, LLC
  • Correspondent: not retrieved
  • Context: Venture-debt securitization — patent collateral pledged to a lender.

• 2019-01-01 — granted (publication of US10168596B2)

  • Context: Issuance. No change in ownership.

• 2019-02-08 (executed) / recorded 2019-02-08 — Reel not retrieved

  • Conveyance: Release by Secured Party
  • Assignor: Eastward Fund Management, LLC
  • Assignee: Elenion Technologies Corporation
  • Correspondent: not retrieved
  • Context: Lien release — Eastward's security interest satisfied/terminated.

• 2019-02-08 (executed) / recorded 2019-02-08 — Reel not retrieved

  • Conveyance: Security Agreement (security interest)
  • Assignor: Elenion Technologies Corporation; Elenion Technologies, LLC
  • Assignee: Hercules Capital, Inc., as Agent
  • Correspondent: not retrieved
  • Context: Securitization to a new lender — collateral re-pledged to a public venture-debt BDC (Hercules Capital). Refinancing, not a transfer of title.

• 2020-03-27 (executed) / recorded 2020-03-27 — Reel not retrieved

  • Conveyance: Release by Secured Party
  • Assignor: Hercules Capital, Inc.
  • Assignee: Elenion Technologies, LLC; Elenion Technologies Corporation
  • Correspondent: not retrieved
  • Context: Lien release — consistent with debt payoff/clean-up around the Nokia acquisition (completed Q1 2020).

• 2023-01-18 (executed) / recorded 2023-01-18 — Reel not retrieved

  • Conveyance: Assignment of Assignors' Interest
  • Assignor: Elenion Technologies LLC
  • Assignee: Nokia Solutions and Networks Oy (Espoo, Finland)
  • Correspondent: not retrieved
  • Context: Internal post-acquisition title clean-up — a strategic acquirer (Nokia) consolidating Elenion's granted patents into the Nokia entity, ~3 years after closing.

Family note (same chain presumed): This patent is in a continuation family whose siblings also moved on the same ownership path — US16/202,818 → US10558104B2; US16/734,539 → US11086188B2; US17/396,954 → US11599005B2. Expect the same Elenion → Nokia re-title.

Lender identity note: Eastward Fund Management, LLC and Hercules Capital, Inc. are secured lenders, not buyers. These entries are lien recordings/releases and do not transfer title. Do not read them as assignment links.

Cross-check caution: A search surfaced a different Nokia matter — a Patent Purchase Agreement between Nokia Technologies Oy / Nokia Solutions and Networks B.V. / Alcatel-Lucent SAS and Provenance Asset Group PLC (reel 043877/0421, cited in a PTAB petition). That is the well-known Nokia Technologies–Alcatel-Lucent patent sale to Provenance Asset Group (Fortress-backed) and involves different Nokia entities and a different portfolio. It is not part of US 10,168,596's chain. Do not conflate the two — no Provenance link to this patent was found.

Timeline diagram

timeline
    title Ownership of US 10168596
    2017 : Filed by Elenion Technologies
         : Inventors assign to Elenion
    2018 : Security interest to Eastward Fund
    2019 : Patent granted
         : Eastward release recorded
         : Security interest to Hercules
    2020 : Hercules release recorded
         : Nokia completes Elenion acquisition
    2023 : Re-titled to Nokia Solutions and Networks Oy

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. The only operating-entity → holding-entity hop is Elenion LLC/Corp → Nokia Solutions and Networks Oy, a large global operating company, not a licensing-only vehicle. No "IP/Holdings/Ventures" recipient, no registered-agent address, no single-member shell. (Reel/frame not retrieved, but the assignee identities are unambiguous on the ledger.)

2. Known asserter in the chain — Not present. Chain participants are Elenion (operating SiP designer), Eastward Fund Management, LLC and Hercules Capital, Inc. (secured venture lenders), and Nokia Solutions and Networks Oy (operating network-equipment vendor). None match the referenced NPE lists (Acacia, Marathon, IV, Wi-LAN/Conversant, Vringo, Pendrell, Round Rock, etc.). No NPE assignee appears at any point.

3. Repeat correspondent across the chain — Unclear / not determinable. The correspondents of record were not retrievable with the available tools. This is the single most valuable missing datum: if the same attorney/firm appears on both the 2017 inventor-assignment and the 2023 Elenion→Nokia re-title, that is normal corporate housekeeping; recurrence would only be a finding if it also ties to an NPE. Recommend pulling the correspondent field for each entry directly from the Assignment Center before finalizing.

4. Cascading transfers through chained LLCs in <24 months — Not present. The 2018–2020 entries are security interest → release → security interest → release between the company and its lenders; title never left Elenion. No sequence of distinct assignee LLCs sharing a correspondent address.

5. Pre-litigation transfer — Not present / not applicable. No infringement suit naming this patent was located. The 2023 Nokia re-title occurred ~3 years after the 2020 acquisition and is unrelated to any assertion.

6. Bankruptcy fire-sale — Not present. No Chapter 7/11 filing by Elenion surfaced; the end-game was an acquisition by Nokia, with the Hercules/Eastward liens released. (The pre-acquisition secured lending is ordinary venture debt, not distress evidence.)

7. Privateering — Not present. Nokia is the acquirer and practices the technology in its own optical portfolio (WaveFabric Elements/CSTAR silicon photonics), rather than passing it to an NPE to assert against competitors. No SEC/Patent-Progress/EFF reporting indicates a Nokia→NPE transfer of this patent.

8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. Nokia is an operating vendor, not a defensive aggregator.

Verdict

Operating-company assertion — with the "assertion" prong unverified.

Justification. The ownership chain runs cleanly from an operating silicon-photonics company to a strategic acquirer: inventors → Elenion Technologies, LLC (2017-07-24); two rounds of secured-lender lien/release with Eastward Fund Management (2018-04-16 / release 2019-02-08) and Hercules Capital (2019-02-08 / release 2020-03-27); then Elenion Technologies LLC → Nokia Solutions and Networks Oy (2023-01-18) following Nokia's completed 2020 acquisition. Every participant is a practicing operating company or a lender — none is an NPE — and no cascade, shell-entity hop, or defensive-aggregator terminus appears. Accordingly there is no NPE pattern; the only caveat is that I found no patent-specific litigation, so the strict "is suing competitors" element of this verdict category is not established from the records I could reach. Verify reel/frame + correspondent at https://assignmentcenter.uspto.gov/ to close the correspondent-recurrence check (Signal 3).

I'll research this systematically. Let me start by searching for the patent's prior art citations.

Let me get the complete "References Cited" section from the patent front page and the citation databases.

The results are noisy. Let me target the patent's front-page citation list directly.

I was unable to complete additional verification searches (the tooling step limit was reached), so the response below distinguishes clearly between (a) what I retrieved from the authoritative record and (b) what I am inferring with lower confidence. Per the operating rules, I do not auto-correct identifiers; I reproduce them as they appear in the source.

US 10,168,596 B2 — "Optical Waveguide Modulator"

1. Patent under analysis (verified)

Claim architecture relevant to §102 (from the specification/summary; the full claim set was not fully retrieved, so treat numbering as indicative):

• Apparatus claim — MZM with input/output optical ports, two parallel waveguide arms, N≥2 modulating subsystems in sequence, and N electrical drive circuits, each applying a differential drive signal to both waveguide arms of one subsystem ("dual-differential driving").
• Dependent/apparatus claims — substrate with semiconductor material; p/n junction in each arm between the electrode pair; two differential drivers DC-coupled and configured to produce "two synchronous differential drive signals having a DC shift therebetween."
• Method claim — steps (a)–(c): receive light into two arms traversing N≥2 sections; provide a differential drive signal separately to each of the N sections in synchronism with light propagation; in each section apply the drive signal, or related signal, separately to each arm (dual-differential push-pull).
• Equalization aspect — one subsystem driven with an inverted differential signal (subtractive, time-delayed) to compensate modulation dispersion / high-frequency roll-off.
• p/n junction claim (anode-drive variant) — anodes electrically connected to the p-side, n-side to ground plane, differential driver fed with a negative supply voltage and DC-coupled to the anodes.

2. Prior art cited on the face of US 10,168,596 (56) References Cited

The following list is reproduced from the OCR of the patent's front page (patentimages PDF). The list below may be incomplete — my retrieval was cut short before I could confirm the full set — and some entries show OCR garbling. Per instruction, identifiers are reproduced literally rather than corrected.

U.S. Patent Documents

Non-Patent Literature (as printed on the face)

3. Potential §102 mapping (best-effort, subject to verification)

Important caveat: true §102 anticipation requires a single reference disclosing every element of a claim, arranged as claimed. On the face of it, the front-page references appear to be background/general-art citations more consistent with §103 obviousness combinations than with clean §102 anticipation. The mappings below are hypotheses keyed to claim elements, not confirmed anticipation findings.

• US 9,519,162 B2 (Ding) → the "N modulating subsystems sequentially disposed along the arms" element (apparatus claim / claim 1 core). A segmented-modulator disclosure with multiple electrode segments along the MZM arms, if it also shows a separate drive per segment, would read on the multi-section architecture. What it almost certainly does not show, and what would defeat clean §102 anticipation, is the claimed dual-differential drive — i.e., a differential (complementary single-ended) signal applied to each of the two arms in each section separately. → Best characterized as §103 candidate.

• US 2008/0089634 A1 (Mosinskis) and US 2011/0044573 A1 (Webster) → the differential-drive elements. These are the references most likely directed to differential/push-pull drive of p/n-junction phase modulators in MZM arms, and so are candidates against the "differential drive signal … to each of the first and second waveguide arms" limitation. Anticipation of the full independent apparatus claim would require the reference to also disclose the N≥2 sequential segmentation.

• US 2015/0295650 A1 (Lee) → the driver-circuit / electrical-pathway elements. Relevant to the drive circuitry and signal routing, not to the optical segmentation.

• US 2013/0272700 A1 (Satoh) and US 5,291,565 A (Schaffner) → the traveling-wave / terminated-electrode elements. Background on transmission-line electrodes and terminations; unlikely to anticipate, useful for §103.

• US 2014/0104666 A1 (Minola) and US 3,814,929 A (Whitfield) → general electro-optic modulation background.

• Temporiti (ISSCC 2016) and Denoyer (ECOC 2014) → the "silicon-photonics MZM + CMOS/BiCMOS driver" platform elements and the general broadband-modulator performance context. These NPL items support obviousness of the integration/performance aspects (bandwidth vs. length, driver integration) rather than anticipating the dual-differential multi-section claims.

Net assessment: No single front-page reference appears to disclose all of (i) N≥2 sequential modulating sections, (ii) separate differential drive to both arms of each section, and (iii) the DC-shifted dual-differential driver / inverted equalizing section. Anticipation (§102) is therefore not clearly established by the retrieved list; the strongest references (US 9,519,162 Ding; US 2008/0089634; US 2011/0044573) are more plausibly §103 combination material.

4. Family / forward citations (context only — NOT prior art)

These are later, same-family or citing documents; they cannot be §102 art against US 10,168,596:

• US 10,558,104 B2 (priority link from US 16/202,818)
• US 11,086,188 B2 (US 16/734,539)
• US 11,599,005 B2 (US 17/396,954)
• US 10,466,567 B1 / US 2019/0324345 A1 / US 2020/0026147 A1 / US 10,831,081 B2 — Elenion "Optical waveguide modulator" continuations (dual-differential driving using a single differential transmission line)
• US 9,874,767 B2 / US 10,451,903 B2 / US 10,656,441 B2 — Elenion "Optical modulator having a plurality of modulator segments" family

5. Explicit limitations of this report

I want to be candid about confidence rather than fabricate detail:

1. I could not retrieve the complete, verified "References Cited" list. The table above comes from OCR of the front page and may omit entries. Some printed dates are internally inconsistent (e.g., ref. #4's "6/2015" for a B2 in the 9,909,xxx range).
2. I did not independently open each cited reference to confirm its exact title, assignee, or claim content, so the "likely subject matter" and §102 mappings are inference, not verified findings.
3. I did not retrieve the complete issued claim set of US 10,168,596 (claim numbering in particular), so the claim-element mapping is based on the specification's summary of the claimed aspects, not on verbatim claim language beyond what was provided.
4. To finish rigorously, the following should be pulled directly: the full "References Cited" block (USPTO PatentCenter/Global Dossier for US 15/602,657) and the full text of each cited document's claims.

If you'd like, I can, in a follow-up, pull each of the twelve references individually (title, assignee, priority date, and the passages the examiner appears to have relied on) and produce a formal element-by-element §102/§103 chart against the issued claims.

I'll research the prior art associated with this patent before constructing the § 103 analysis.

Obviousness Analysis — US 10,168,596 B2 ("Optical waveguide modulator")

Analyst note on sources. The patent page you supplied (https://patents.google.com/patent/US10168596/en) renders the bibliographic data, the "Prior art keywords" line (differential; modulating; optical; electrodes; waveguide), the CPC classifications, the description and the abstract, but not a tabulated "References Cited" list. I therefore reconstructed the prior-art set from (a) the grant front page PDF for US10168596B2 (the "(56) References Cited" block) and (b) the documents that the same-page surface as cited/citing/similar art. Where I could not verify a reference's substance from a retrievable source, I say so explicitly rather than characterizing it.

1. Legal framework and the critical date

• Effective filing date: 23 May 2017 (application 15/602,657; the page lists the priority date as the filing date — no earlier priority is claimed; note the later US applications 16/202,818, 16/734,539 and 17/396,954 claim priority to this case, not the reverse).
• The patent is post-AIA, so 35 U.S.C. § 102(a)(1)/(a)(2) and § 103 apply, with Graham v. John Deere factors plus KSR Int'l v. Teleflex (obvious to try / known-technique / design-incentive rationales).
• Art qualifies if published before 23 May 2017 (§ 102(a)(1)) or if it is a US patent/application publication effectively filed before that date (§ 102(a)(2)).

2. The prior-art set actually of record / in the immediate art space

2.1 References on the face of US10168596B2 ("(56) References Cited")

Retrieved from the grant front page PDF (https://patentimages.storage.googleapis.com/7b/17/7f/d3b0acf2aec05b/US10168596.pdf):

Non-patent literature of record:

• Temporiti, E. et al., "A 56Gb/s 300mW Silicon-Photonics Transmitter in 3D-Integrated PIC25G and 55nm BiCMOS Technologies," ISSCC 2016, pp. 404–406.
• Denoyer, G. et al., "Hybrid Silicon Photonic Circuits and Transceiver for 56Gb/s NRZ 2.2km Transmission over Single Mode Fiber," ECOC 2014, PD.2.4.

2.2 References whose substance I verified

• Elenion segmented-modulator family — US 9,874,767 B2 / US 10,451,903 B2 / US 10,656,441 B2 / US 11,209,674 B2 (Ding, Baehr-Jones, Magill, Hochberg, Rylyakov); published application US 2016/0103340 A1 and continuation publication US 2017/0059888 A1. Verified text: "a succession of serially optically coupled optical modulator segments…"; "a plurality of segment drivers, wherein … each segment driver is electrically coupled to one of the succession of the modulator segments"; "delays ΔT_electrical … substantially synchronized with propagation of the optical carrier wave therein"; "a succession of serially electrically coupled electrical delay line segments." Sources: https://patentimages.storage.googleapis.com/61/02/1b/871bf291c75313/US10451903.pdf ; https://www.patents-review.com/a/20160103340-optical-modulator-plurality-modulator-segments.html ; https://patents.justia.com/patent/[11209674](/patent/11209674)
• US 9,817,249 B2 / US 2013/0343693 A1 (Acacia Communications) — "Alternating traveling-wave Mach-Zehnder modulator." Verified text: differentially driven traveling-wave MZM; "two arms of the MZM are differentially driven"; "applying the full voltage of the differential line across one diode at a time — thereby driving the two arms alternately rather than simultaneously"; "two auxiliary lines that exhibit a different DC level than the two main lines for biasing the diodes … connected to the main lines with a high capacitance." Sources: https://patents.google.com/patent/[US9817249B2](/patent/US9817249B2)/en ; https://www.freepatentsonline.com/[9817249](/patent/9817249).html
• US 5,291,565 A (Schaffner) — cited as category-Y art in the search report for WO 2003/063463 A3, "Corporate feed for 40 GHz modulators," whose own abstract describes "an electro-optic modulator with a segmented optical transmission network with a series of discrete electrodes (20a–20d) for successive segments… drivers at the output of a corporate feed… respective modulating signals are supplied to the electrodes … whose lengths differ from each other, so that modulating signals arrive at their respective electrodes in synchronism with the propagation of an optical signal." Source: https://patentimages.storage.googleapis.com/72/48/34/217851bf583ecf/WO2003063463A3.pdf
• US 8,320,720 B2 / US 2011/0044573 A1 (Webster) — "Advanced Modulation Formats for Silicon-Based Optical Modulators." Verified via the family list at https://patents.google.com/patent/JP7294066B2/en
• US 2008/0089634 A1 (Mosinskis) — cited as category-X art in the search report for EP 4 080 272 A1 ("Segmented optical waveguide modulator," NTT), in the same X-category group as US 2018/0341164 A1 (Williams et al.) — i.e., the published application of the very patent under analysis. Source: http://data.epo.org/pise-server/rest/collections/lgpi/EP4080272A1.pdf — This is a strong independent signal that Mosinskis is directed to the same subject matter (driving/bias arrangement of electro-optic modulator electrodes).
• US 10,084,619 B2 (IBM) — "Nested feed-forward optical equalization using an electro-optic modulator with a multi-segment electrode," filed 2016-06-03 (→ § 102(a)(2) art as to a 2017-05-23 filing), granted 2018-09-25. I located the title/pairing but could not retrieve its specification; I therefore treat it as a promising but unverified reference for the equalization claims and recommend a full-text pull before finalizing.

3. Representative claim decomposition

Caveat: the page as supplied reproduces the Summary, not the verbatim claim set. The elements below are taken from the Summary/disclosure language, which is coextensive with claim 1 for present purposes. If the granted claims contain narrowing language absent from the Summary, the analysis below must be re-run against that language.

4. The combinations, with motivation

Combination 1 (primary) — Ding ('767/'903 family) + Schaffner + Webster (+ Temporiti)

Covers: elements A–E, and (with Temporiti) the photonics-chip/driver-chip integration.

• Ding supplies almost the whole skeleton: a succession of serially optically coupled modulator segments, one driver per segment, electrical delay lines tapped so that drive application is synchronized with optical propagation, and silicon-photonic-chip implementation.
• Schaffner/WO '463 supply the same segmentation teaching in the context of a corporate feed with per-segment driver gain adjustment, and explicitly frame the goal as reaching high modulation bandwidth at high frequency while matching RF and optical velocities.
• Webster supplies differential (dual-drive, push-pull) driving of both arms of a silicon MZM using carrier-depletion p/n junctions.

Motivation (KSR "known technique / design incentive"): The '596 specification itself states the problem the combination solves and the reason the elements go together: the achievable bandwidth of a traveling-wave electrode scales as BW ∝ 1/√(C_pn R_pn Z_o L) (its Eq. 1 and FIG. 4), so electrode length must be short (≈1 mm for ~35 GHz in silicon); but short electrodes give insufficient electro-optic interaction length, degrading OSNR/OMA. A POSITA facing that trade-off has exactly two known levers — (i) split the electrode into short segments, each separately driven with a delay-matched signal (Ding; Schaffner; WO '463), and (ii) drive both arms of each segment differentially to double the effective phase swing for the same V_pp (Webster). Applying both is the predictable aggregation of two known techniques to the same structure, with the result (bandwidth × efficiency product improved) being the expected sum of the parts. Nothing in the combination changes the principle of operation of either technique.

Where a POSITA lands: a segmented MZM in which each segment's four electrodes receive a differential drive — i.e., elements A–E. Motivating the final step is trivial because the number of segments N is expressly a design parameter ("N can vary from two to ten and more"; "the length l … selected based on the target modulation bandwidth, and the number N … selected based on V_pp … and a target modulation efficiency").

Combination 2 — Acacia '249 + Ding/Schaffner

Covers: elements D and F, plus H.

Acacia '249 expressly discloses a differentially driven traveling-wave MZM in which the full differential-line voltage is applied across the active diode, and — critically for element F — auxiliary lines held at a different DC level than the main lines for biasing the diodes, capacitively coupled to the main line. That is the functional equivalent of the '596's "two synchronous differential drive signals having a DC shift therebetween," and it solves the identical problem (reverse-biasing depletion-mode p/n junctions while swinging the RF drive). Acacia also identifies the very cost that the '596's segmentation neutralizes (approximately doubling the modulator length and increasing RF loss), which supplies a further motivation to pair Acacia's differential-arm drive with Ding/Schaffner's segmentation.

Motivation: both references are in the same field (traveling-wave silicon/InP MZM design), address the same design variable (arm drive topology and diode bias), and the combination yields an improved but predictable result. KSR makes this a textbook "arrangement of old elements, each performing its own function."

Combination 3 (equalization claims) — IBM US 10,084,619 B2 + Ding/Schaffner

Covers: element G.

The '596's equalizing section is, functionally, a single-tap feed-forward equalizer implemented optically: one segment's modulation is inverted, delayed by T_p, and weighted (electrode length a, driver gain) so that it subtracts from the main modulation, providing high-frequency pre-emphasis (the patent says so itself: "modulation signal dispersion may be countered by adding an inverted copy of the modulated signal that is suitably delayed and weighted"). IBM's US 10,084,619 is titled "Nested feed-forward optical equalization using an electro-optic modulator with a multi-segment electrode" — i.e., the same architecture — and Ding/Schaffner supply the delay-tap mechanism (ΔT_electrical synchronized to ΔT_optical) that the '596's delay line 240 reuses.

Motivation: FFE/TX-pre-emphasis was a well-known technique in 2014–2017 high-speed optical links (the patent's own Background and the ISSCC/ECOC papers of record are evidence of the field's focus on bandwidth-limited equalization). Implementing a known electrical-domain equalizer in the optical domain by inverting one segment's drive is a predictable use of the segmented electrode that Ding/Schaffner already provide. (Confidence reduced: I have the IBM title and filing date but not its disclosure. If IBM's spec does not enable inverting a segment relative to another with a relative delay, this combination weakens and the equalization claims would then rest on the electrical FFE art + Ding.)

Combination 4 (the "negative supply / grounded cathode" claims) — Schaffner/WO '463 + Acacia '249

Covers: element H.

Terminating each anode electrode with a resistor to a ground plane is ordinary traveling-wave practice (the '596's own FIG. 1 shows transmission-line termination 25). Once an auxiliary/DC-shifted bias line (Acacia) is available to set the reverse-bias point, driving the anodes from a negative-supply differential driver while referencing the cathodes to ground is a mere substitution of bias polarity — and the '596 itself recites the reason: it "eliminates a bias-T from the driver electronics," a bias-T being a known source of "signal distortion and/or loss" and of added cost and size. Motivation is explicit and technical.

5. Motivation-to-combine summary (for the record)

1. Same field of endeavor / reasonably pertinent: every reference is directed to high-speed traveling-wave waveguide modulators and their electrode/driver design. In re Bigio / KSR.
2. Same problem, same design variables: the length ↔ bandwidth ↔ V_π ↔ OMA/OSNR trade-off. Both Ding and Schaffner state it; the '596 states it verbatim.
3. Known, finite set of solutions: (a) segment the electrode and delay-tap each driver; (b) drive both arms differentially; (c) bias via a DC-offset line rather than a bias-T; (d) add an inverted, delayed tap for pre-emphasis. A POSITA would recognize each as a known option with predictable results.
4. Reasonable expectation of success: each reference reports the technique working in the same device class (silicon carrier-depletion MZM with reverse-biased p/n junctions).
5. Explicit design incentives in the art of record: the driver-side V_pp ceiling (the '596 notes BiCMOS Si-NPN breakdown at ~1.8 V, capping V_pp ≈ 3–3.5 V) makes "double the phase efficiency per volt" directly desirable — which is precisely what the dual-differential drive delivers.

6. Counter-arguments and weaknesses a challenger must address

1. Acacia '249 may be argued to teach away from the core novelty: it stresses driving "the two arms … alternately rather than simultaneously, in sharp contrast to the prior art which drives the two arms simultaneously." A patentee could argue this discourages the '596's simultaneous dual-differential driving of both arms. Rebuttal: Acacia's stated reason for alternating is to avoid the 2× voltage division inherent in series-connected diodes, a rationale that does not apply to the '596's topology (each arm driven by its own differential pair); moreover the '596's segmentation removes the length/loss penalty Acacia identifies. KSR also cautions that a reference's preference is not an absolute bar absent a teaching that the alternative would be inoperative.
2. Self-collision / § 102(a)(1) exposure (and an alternative to a § 103 case). The Elenion segmented-modulator publications appear to predate 23 May 2017 by more than one year (the corresponding publication US 2016/0103340 A1 appears to date from ~April 2016; US 2017/0059888 A1 from 2 March 2017 — verify exact dates). Because the inventive entities differ (Ding/Baehr-Jones/Magill/Hochberg/Rylyakov vs. Williams/Ahmed/Rylyakov/Younce/Liu/Ding/A. Ahmed), the § 102(b)(1)(A) grace-period exception for "the inventor or joint inventor" is unlikely to remove them. If US 2016/0103340 A1 discloses a per-segment differential drive of both arms, the case is one of anticipation, not merely obviousness — a materially stronger attack.
3. Verbatim claim text not available. My element list is reconstructed from the Summary. Element D ("each of the N electric drive circuits electrically coupled to each electrode of the first and second pairs of electrodes … for differentially driving both the first and second pairs …") is the true point of departure from Ding and from Schaffner. It is also the element that must be pinned to Acacia/Webster with precision.
4. Secondary considerations (unassessed). I have no verified evidence of commercial success, licensing, copying, unexpected results, or industry praise for the '596 subject matter. Elenion's acquisition by Nokia (2023) and the existence of a continuation family (US 10,558,104; US 11,086,188; US 11,599,005) are facts of record on the page, but neither is, standing alone, probative of non-obviousness. Any nexus argument must be built on evidence not present in this record.
5. Reference-verification gaps. (i) US 5,675,673 ("Skete"), US 9,909,193 B2 (Sato) and US 2013/0272700 A1 (Satoh) were not substantively verified here. (ii) The date printed for US 9,909,193 B2 (6/2015) is anomalous for that number range and should be checked against the USPTO record. (iii) Whether the "Cited By" table I observed actually belongs to the '596 page or to the US 2013/0272700 page is not certain from the retrieved snippet; the Acacia and IBM references are used above on their own dates and subject matter, independent of that table.

7. Bottom line

Claims 1 and its section/electrode structural dependents are, on this record, prima facie obvious under § 103 over Ding (US 9,874,767 B2 / US 10,451,903 B2, and their pre-May-2017 publications) in view of Schaffner (US 5,291,565) and Webster (US 8,320,720 / US 2011/0044573), optionally with Temporiti (ISSCC 2016) for the two-chip integration. The primary reference supplies the segmented waveguide + one-driver-per-segment + delay-synchronized drive architecture; the secondary references supply differential push-pull driving of both arms; the motivation is the admitted bandwidth-versus-efficiency trade-off, and the result is the predictable aggregation of two known techniques.

The DC-shift/dual-differential-driver claims are prima facie obvious over Acacia US 9,817,249 B2 in view of Ding/Schaffner (Acacia's DC-level auxiliary bias lines + the segmented electrode's per-segment drivers), and over Mosinskis US 2008/0089634 A1 as corroborated by its X-citation against EP 4 080 272 A1 alongside the '596's own publication.

The equalizing-section claims are prima facie obvious over IBM US 10,084,619 B2 in view of Ding/Schaffner — subject to verification that IBM discloses inverting one segment's drive relative to another with a relative delay, which I could not confirm from the retrievable record.

The grounded-anode / negative-supply claims are prima facie obvious over Schaffner + Acacia in view of ordinary transmission-line termination practice, with the bias-T-elimination rationale supplied by the specification itself.

I rate confidence high on Combinations 1 and 2 (verified reference disclosures and dates), medium on Combination 3 (title-level evidence for IBM only) and Combination 4 (the negative-supply claim is a small further step). Before filing, I would (a) pull the verbatim granted claims, (b) pull the IBM and Mosinskis full texts, and (c) settle the exact publication dates of US 2016/0103340 A1 and US 2017/0059888 A1, since those dates determine whether a § 102(a)(1) anticipation position is available in addition to the § 103 position set out above.