US 9031417

I'll search for authoritative information on this patent number.

Now let me check for litigation and any CAFC 2026 docket activity.

Let me do a final check for any Federal Circuit 2026 appeal activity specifically involving this patent.

US Patent 9,031,417 B2 — Summary

Important scoping note first: I found no evidence that U.S. Patent No. 9,031,417 itself appears in any Federal Circuit 2026 docket. The Federal Circuit activity in 2026 involving this patent family (see below) concerns a different, later-issued family member. Details and the distinction are given at the end.

Bibliographic data (per the authoritative full text provided and corroborating sources)

Continuity chain (verbatim from the record): This application is a continuation of Ser. No. 13/280,371 (filed Oct. 25, 2011, now U.S. Pat. No. 8,797,198), which is a continuation of Ser. No. 12/636,805 (filed Dec. 14, 2009, now U.S. Pat. No. 8,044,835), which is a continuation-in-part of PCT/IL2008/000805 (filed June 12, 2008), claiming priority of U.S. Provisional 60/943,559 (filed June 13, 2007). The '417 patent is the parent of a long continuation family (e.g., US 9,203,425; 9,479,191; 10,005,527; 10,033,465; 10,205,527/10,270,535 variants; 10,461,866; 11,133,872; 11,342,998; 11,716,148; 12,191,912; 12,388,533).

A security-interest assignment to PBLM RTAU LLC was recorded Oct. 24, 2014; an assignment to Ramot at Tel-Aviv University Ltd. was recorded Mar. 9, 2015.

Abstract (as issued)

"A modulator device for converting digital data into modulation of an optical signal includes an electronic input for receiving an input data word of N bits and an electrically controllable modulator for modulating the intensity of an optical signal, the modulator including M actuating electrodes where M≧N. An electrode actuating device, most preferably a digital-to-digital converter, operates actuating electrodes so that at least one electrode is actuated as a function of values of more than one bit of the input data word. According to an alternative, or supplementary, aspect of the invention, the set of electrodes includes at least one electrode having an effective area which is not interrelated to others of the set by factors of two. In one preferred implementation, a Mach-Zehnder modulator also provides phase modulation to give QAM functionality. Another implementation employs a semiconductor laser."

Plain-language overview of the independent claims

• Claim 1 (system for converting digital data into a modulated signal). An electrically controllable device with M actuating electrodes produces an optical signal modulated by binary voltages on those electrodes. An electrode-actuating device has an input for N-bit digital data words and at least one digital-to-digital converter (DDC) that maps each input word to a selected binary actuation vector of M bits (M > N), where the vector is chosen from a plurality of possible M-bit vectors. Crucially, at least one electrode is actuated in a way that differs from any one-to-one correspondence between input bits and electrodes.

• Claim 11 (system with optical output port). Similar structure: an electrically controllable device with M electrodes and an optical output port, modulated by binary electrode-actuation vectors applied as actuating voltages. A DDC maps N-bit input words to M-bit binary actuation vectors (M > N) and supplies corresponding actuation voltages; the DDC selects an M-bit vector to represent each N-bit input word.

• Claim 19 (optical communication system). An electronic input receives an N-bit digital word; an electrode-actuating device supplies M-bit binary electrode actuation vectors to M actuating electrodes of a Mach-Zehnder modulator (M > N). The MZM produces an optical signal modulated in response to the actuation voltages, and an optical output outputs the modulated signal.

• Claim 27 (system for converting an N-bit digital input). An electrically controllable device with M actuating electrodes produces an optical output modulated by voltages applied digitally to the electrodes; an electrode-actuating device provides M-bit binary actuation vectors, each representing an N-bit digital input word, where M > N.

Representative dependent claims: semiconductor modulator (2, 12); modulated light generation device (3); optical-to-electrical converter (4, 13, 18, 20); response shaped to PAM or QAM (5, 14, 21, 28, 29); full-dynamic-range linearization despite intrinsic non-linearity (6, 16, 22); laser light source at input (7, 15, 23, 26); more-linear-than-one-to-one actuation (8, 24); actuation as a function of more than one bit (9); kit with O/E receiver (10); electrodes of different effective areas, at least one not related by a factor of two (17); phase-modulated output (25); differing functional relationship of an actuation vector to an input bit (30).

Litigation / CAFC 2026 — what actually exists

1. Delaware case cited on the patent page: 1:21-cv-00295, Ramot at Tel Aviv University, Ltd. v. Acacia Communications, Inc. (D. Del., filed Feb. 26, 2021, Judge Leonard P. Stark). Caution: the Rule 3 Report to the Commissioner in that docket lists the asserted patents as 10,270,535 and 10,033,465 — not 9,031,417. 9,031,417 appears to be family background, not an asserted patent there. (Cisco later acquired Acacia; Cisco's 10-K describes Ramot asserting against Cisco and Acacia in E.D. Tex. (June 12, 2019) and D. Del. (Feb. 26, 2021).)

2. Federal Circuit 2026 activity — appeal nos. 24-1726, 24-1727, 24-1728: Cisco Systems, Inc. v. Ramot at Tel Aviv University Ltd., argued April 10, 2026, Rule 36 judgment issued April 13, 2026 (per curiam: Moore, Chief Judge; Mayer and Lourie), AFFIRMED, from PTAB IPR2022-00575, IPR2022-00576, IPR2022-01283. Third-party commentary (PatSnap) links this appeal to US 11,133,872 B2, a descendant of the '417 patent in the same priority family — not to 9,031,417 itself.

3. Related 2024 Delaware ruling: Cisco Systems, Inc. v. Ramot at Tel Aviv University Ltd. (D. Del., Judge Gregory B. Williams) — claims of US 11,342,998 held partly indefinite (MLex, Nov. 12, 2024), again a different family member.

Confidence and uncertainties

• High confidence on title, inventors, assignee, application number, filing date (Jul. 8, 2014), issue date (May 12, 2015), priority date (Jun. 13, 2007), and claim text — these come from the patent's own front matter/claims as fetched.
• Explicit uncertainty: I could not locate any IPR, district-court docket, or Federal Circuit 2026 appeal in which 9,031,417 itself (as opposed to a continuation/descendant such as US 11,133,872 or US 11,342,998) is the patent-in-suit. If a specific 2026 docket number for the '417 patent exists, I do not have authoritative confirmation of it, and I will not invent one. The 2026 docket I did confirm is 24-1726, which concerns a different patent in this family.
• The Google Patents "family has litigation" Delaware link (1:21-cv-00295) is a family-level litigation flag; it does not establish that '417 was asserted in that case, and the docket's own patent list contradicts such an inference.

Sources: patent full text at patents.google.com/patent/US9031417B2; CourtListener/dockets.justia.com docket 24-1726; cafc.uscourts.gov April 10, 2026 and Rule 36 judgment (filed Apr. 13, 2026); D. Del. docket 1:21-cv-00295 (dockets.justia.com); Cisco 10-K; MLex report on Cisco v. Ramot ('998 patent).

I'll research litigation involving US patent 9031417 using the sources you specified.

Let me dig deeper to confirm which cases specifically assert US 9,031,417.

Litigation Involving U.S. Patent No. 9,031,417

Important caveats up front. My dedicated searches for the literal string "9,031,417" as an asserted patent were cut off by the search tool before I could fully exhaust them. What follows is grounded in (a) the Google Patents litigation flags on the '417 patent family page itself, and (b) court/PTAB documents I did retrieve. Because U.S. 9,031,417 is the founding member of a large continuation family all titled "Linearized Optical Digital-to-Analog Modulator" (including U.S. 9,203,425; 9,479,191; 10,033,465; 10,205,527; 10,270,535; 10,461,866; 11,133,872; and 11,342,998), the litigation is a single multi-case campaign by the same patent owner, Ramot at Tel Aviv University Ltd., against Cisco Systems and Acacia Communications. Some of the specific case-to-patent mappings below are for sibling patents in the same family, and I flag where that is the case.

Per the Google Patents page for US 9,031,417, the record states: "Family has litigation — US case filed in Delaware District Court," linking to Unified Patents Portal entry for D. Del. case 1:21-cv-00295, and noting a "first worldwide family litigation filed" (Darts-IP family 39798167). This is the anchor data point tying the '417 patent family to the Delaware campaign.

Cases identified

Related but distinct proceedings (same family/parties)

• PTAB IPR (NHK/Fintiv denial) on U.S. 10,270,535 (the '535 patent, a family member), filed by Cisco; institution denied under the NHK/Fintiv rule. Cisco appealed to the Federal Circuit (Nos. 2020-2047, 2020-2049), arguing the rule’s unlawfulness (see Cisco's briefing at patentlyo.com). Ramot's Preliminary Response in that IPR expressly cites the Ramot v. Cisco, 2:19-cv-00225-JRG district-court case and lists the '417 Patent File History (U.S. App. No. 14/325,486, issued as U.S. 9,031,417) as an exhibit — corroborating that the '417 patent was in play in the Eastern District of Texas litigation.

What I could not verify

• I was unable to confirm a final judgment or settlement in the E.D. Tex. case (2:19-cv-00225-JRG) or in D. Del. 1:21-cv-00295.
• I could not confirm whether U.S. 9,031,417 was itself named as an asserted patent (versus cited in invalidity contentions) in any specific complaint; the strongest direct link is the Google Patents "family litigation" pointer to D. Del. 1:21-cv-00295.
• No ITC (Section 337) proceeding involving the '417 patent was found. (All ITC hits in my searches concerned unrelated patents, e.g., Brita's U.S. 8,167,141.)
• No IPR directly challenging U.S. 9,031,417 was found; the identified IPRs target sibling patents ('535 and '872).

Bottom line

The '417 patent is part of the Ramot-at-Tel-Aviv "linearized optical DAC" family that has been litigated by Ramot (patent owner) against Cisco Systems and Acacia Communications in E.D. Tex. (2:19-cv-00225) and D. Del. (1:21-cv-00295, 1:21-cv-01365, 1:22-cv-00674), with parallel PTAB IPRs (IPR2022-00575/-00576/-01283, affirmed by the Federal Circuit on April 13, 2026, and an earlier NHK/Fintiv-denied IPR on the '535 patent, Fed. Cir. Nos. 20-2047/20-2049). The single Delaware case specifically flagged on the '417 patent's Google Patents page is 1:21-cv-00295 (D. Del., filed Feb. 26, 2021).

If you need the precise asserted-patent list and docket-level outcomes, the authoritative sources to pull next are the PACER dockets for 2:19-cv-00225-JRG and 1:21-cv-00295-GBW, and the Unified Patents Portal page referenced by the patent (portal.unifiedpatents.com/litigation/Delaware%20District%20Court/case/1:21-cv-00295).

PTAB / AIA Trial Landscape for US 9,031,417 ("Linearized Optical Digital-to-Analog Modulator")

Bottom line up front: I found no AIA trial proceeding — IPR, PGR, or CBM — ever filed against US 9,031,417 itself. This matches the structured "PTAB proceedings on file" block, which reports the USPTO Open Data Portal returns no AIA trial proceedings for this patent as of the most recent ingest. The patent is untested at the PTAB and stands as issued (claims 1–30 intact). Its family, however, has been litigated and reexamined heavily, and those proceedings are a roadmap for anyone facing the '417. Details and a clear separation between "'417" and "family-adjacent" below.

Proceedings overview

Total AIA trial proceedings on US 9,031,417: 0. Breakdown by status: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0. Defensive posture: no PTAB-based invalidation has ever been obtained against this patent, and there is no IPR estoppel to borrow from and no FWD to point a court to. The patent is not "hardened" by surviving a PTAB challenge — it has simply never been challenged there. That cuts both ways for a defendant: you get a clean slate on § 102/§ 103 art, but you also get no free kill shot, and the patent owner has a decade of practice defending this specification in reexamination and district court.

⚠️ Flag before you rely on anything below: web search surfaced only sibling patents in the same family under attack. I could not confirm any filing, institution, or decision directed at the '417. Proceeding numbers below are real and verified; do not attribute them to the '417.

Proceedings on the '417 patent

None. Neither the structured ODP block nor web search disclosed an IPR/PGR/CBM petition naming U.S. Patent No. 9,031,417 as the challenged patent. There is therefore:

• No Final Written Decision to cite.
• No claim of the '417 canceled or confirmed.
• No § 315(e)(2) estoppel running against anyone on this patent.
• No Federal Circuit appeal from a PTAB decision on this patent.

If you are a defendant, your invalidity case against the '417 is unwritten. That is a genuine advantage (unconstrained art selection) but also means zero institutional shortcut.

Family-adjacent AIA proceedings (NOT on the '417 — context only)

The '417 is the fourth patent in a chain rooted at priority date 2007-06-13 (PCT/IL2008/000805): '417 → 9,203,425 → 9,479,191 → 10,033,465 / 10,205,527 / 10,270,535 / 10,461,866 → 11,133,872 / 11,342,998 → 11,716,148 / 12,191,912 / 12,388,533. The contested activity sits on the later continuations. All share the same specification, so the art and claim-construction fights transfer.

IPR2020-00123 — Cisco Systems, Inc. v. Ramot at Tel Aviv University Ltd.

• Type: Inter Partes Review
• Patent at issue: U.S. 10,270,535 (a '417 continuation) — not the '417
• Filed: 2019-11-05 (petition date per file-history index)
• Status: Institution DENIED under 35 U.S.C. § 314(a) — discretionary denial
• Judge panel: Christopher L. Crumbley, Monica S. Ullagaddi, Jason M. Repko. Opinion by APJ Repko; dissent by APJ Crumbley
• Petition grounds: obviousness of claims 1 and 2 over the "Roberts" reference as primary, alone or with Ho/Burchfiel
• Institution decision: Denied on § 314(a) discretionary grounds tied to the parallel E.D. Tex. case (Ramot v. Cisco, 2:19-cv-00225) with an earlier trial date. The panel also weighed § 325(d) cumulativeness (Roberts is a divisional of art already before the examiner).
• Settlement / termination: none
• Appeal: Cisco filed a Notice of Appeal on 2020-07-16 seeking Federal Circuit review of the denial
• Defensive value (for the '417): Low direct value — the denial is a discretionary call, not a merits win for Ramot, and it is not estoppel-generating. It does show the Board was willing to let the district court handle this family.

IPR2020-00122 — Cisco Systems, Inc. v. Ramot at Tel Aviv University Ltd.

• Type: Inter Partes Review
• Patent at issue: U.S. 10,033,465 (a '417 continuation) — not the '417
• Filed: 2019-11-05
• Status: I could not confirm the institution outcome from available sources. Do not represent its result without pulling the E2E record.
• Defensive value: An entire IPR against a sibling of the '417 was filed and (per the litigation record) the sibling was also pushed into ex parte reexam. Useful as an art-collection lead.

IPR2020-00484 — Cisco Systems, Inc. v. Ramot at Tel Aviv University Ltd.

• Type: Inter Partes Review
• Patent at issue: U.S. 10,461,866 (a '417 continuation) — not the '417
• Filed: 2020-01-31
• Status: A merits document in this case (Paper 10) addressed the parties' obviousness arguments — Cisco relied on Roberts' "non-linear compensator" as the claimed digital-to-digital converter; Ramot's preliminary response argued Roberts discloses only a change of representation, not "converting." I could not confirm the final institution/FWD disposition. Verify before citing.
• Settlement / termination: none confirmed
• Defensive value: The record shows the central claim-construction battleground for this family — whether a broad "digital signal processor / non-linear compensator" reads on the claimed digital-to-digital converter that maps N-bit words to M-bit actuation vectors (M > N). That argument is directly transplantable to the '417, whose independent claim 1 contains the identical M>N mapping limitation.

Ex parte reexaminations on family members (Office proceedings, not AIA trials)

Ramot and Cisco jointly reported these in the E.D. Tex. case. None targets the '417; they cover the '465, '535 and '866:

• 90/014,528 ('465), filed 2020-06-09 — reexam certificate issued 2023-10-30; all challenged claims amended
• 90/014,607 ('465), filed 2020-11-10 — merged into 90/014,528
• 90/014,728 ('465), filed 2021-04-13 — merged into 90/014,528
• 90/019,333 ('465), filed 2023-12-13 — on all challenged claims; examiner's answer 2025-04-07; appeal hearing held 2025-11-04
• 90/019,420 ('535) — instituted; all challenged claims rejected in a final office action on 2025-03-31; Ramot notice of appeal 2025-05-30; appeal brief 2025-07-30
• 90/015,280 ('866) — instituted; examiner's answer 2025-02-11; Ramot reply brief 2025-04-11; appeal hearing 2025-11-04

These are examiner-side proceedings. They do not generate § 315(e)(2) IPR estoppel, but they materially weaken the family — "all challenged claims rejected" against the '535 is a meaningful signal about how the shared specification fares under prior art.

Strategic summary

Claim status on the '417: nothing canceled, nothing sustained, everything untested. Claims 1–30 are as issued. The narrowing that has happened in this family happened to other patents — most notably the '465, where every challenged claim was amended in reexam (a de facto narrowing), and the '535, where all challenged claims drew a final rejection now on appeal. None of that automatically limits the '417's claim scope, but a district court construing the '417's "electrode actuating device / digital-to-digital converter" terms will find the family's reexam amendments and the IPR2020-00484 Roberts fight highly probative of what the inventors meant.

Estoppel: essentially none running in your favor, and a time-bar trap you must clear first. Because no IPR was instituted on the '417, no petitioner is estopped under § 315(e)(2) from raising any § 102/§ 103 ground against it, and you inherit no estoppel-driven gaps to exploit. The real constraint is § 315(b): an IPR petition is barred if filed more than one year after the petitioner (or a real party in interest / privy) was served with a complaint alleging infringement of the '417. The '417 has been part of asserted-family litigation — e.g., Ramot at Tel Aviv University, Ltd. v. Acacia Communications, Inc., D. Del. 1:21-cv-00295 (filed 2021-02-26), the case the Google Patents litigation banner links to. I could not verify from the available sources which patents were asserted in 1:21-cv-00295. You must pull that complaint before anything else: if the '417 was asserted there and you (or a privy) were served, your IPR window against the '417 may already be closed, and your only PTAB-adjacent option is ex parte reexam (§ 302) or the district-court route. Conversely, if the '417 was never asserted there, your window is open.

Pattern signals. One petitioner — Cisco Systems, Inc. (together with subsidiary Acacia Communications) — is driving all the AIA and reexam activity against this family, having filed IPR2020-00122, IPR2020-00123 and IPR2020-00484 within a three-month window, plus parallel ex parte reexams. There is no defensive aggregator (e.g., Unified Patents) in the chain on the '417 that I could find; the Google Patents litigation banner shows a Unified Patents/Darts-IP "family litigation" link, but that is a litigation-database pointer, not evidence of aggregator participation. Ramot is litigating aggressively: it has pursued examiner's answers and appealed the '535 reexam rejection; Cisco, for its part, filed a DJ action the same day the '998 patent issued (1:22-cv-00674), and a Rule 36 affirmance is reported in Cisco Systems, Inc. v. Ramot at Tel Aviv University Ltd., Fed. Cir. No. 24-1726 (concerning U.S. 11,338,872) — treat the PatSnap characterization of that appeal as unverified and confirm against the CourtListener docket before relying on it.

Recommended next steps

1. Confirm the negative on the '417 directly. The ODP ingest says no AIA trials, but ingest lags. Re-run a PTABIS/E2E search on "9,031,417" and on application 14/325,486 before you certify "no PTAB activity" in any brief.
2. Pull the 1:21-cv-00295 docket and complaint (D. Del., filed 2021-02-26) and the Cisco/Acacia DJ dockets (1:21-cv-01365; 1:22-cv-00674) plus Ramot v. Cisco, E.D. Tex. 2:19-cv-00225. Determine whether the '417 specifically was ever asserted and, if so, when you/your privies were served — this fixes your § 315(b) clock.
3. Mine IPR2020-00484's record on the "digital-to-digital converter" limitation. The Roberts-as-DDC fight is the family's core § 103 vulnerability and the argument transfers directly to '417 claim 1, which requires a DDC that maps each N-bit input word to a selected M-bit actuation vector with M > N.
4. Exploit the reexam rejections. The 2025-03-31 final rejection of all challenged claims of the sibling '535 (reexam 90/019,420) is strong, fresh evidence that the shared specification's DDC-mapping claims are vulnerable. If you file, do it before the '535 appeal resolves so you can use the examiner's reasoning while it is on the record.
5. If the § 315(b) window is open, file an IPR now. With no prior FWD on the '417, no estoppel, and a shared specification already rejected in reexam on a sibling, the cost/benefit is favorable — a single-institution IPR gives you a § 315(e)(2) estoppel shield for the litigation and a two-year path to a claim-cancellation ruling.

Sources / links

• Patent: https://patents.google.com/patent/US9031417/en
• PTAB E2E / Patent Trial and Appeal Board (public decisions & petitions): https://ptacts.uspto.gov/ptabs/
• IPR2020-00123 institution decision (denial, panel Crumbley/Ullagaddi/Repko) and Notice of Appeal: https://insight.rpxcorp.com/federal_circuit/[877257](/patent/877257)/rpx-ptab/IPR2020-00123/Notice_of_Appeal_IPR2020_00123_12b2eda0d716d456244685928fd0b630.pdf
• IPR2020-00484 merits record (Roberts/"non-linear compensator" DDC dispute): https://bdiplaw.com/wp-content/uploads/2021/03/Ramot-v-Cisco-PTAB-IPR2020-00484-10.pdf
• Joint status report on family reexaminations ('465/'535/'866): https://archive.org/download/gov.uscourts.txed.[190224](/patent/190224)/gov.uscourts.txed.190224.255.0.pdf
• Delaware litigation 1:21-cv-00295: https://litigation.rpxcorp.com/litigation/dedce-74725-ramot-at-tel-aviv-university-v-acacia-communications
• CourtListener (Cisco v. Ramot claim construction / DJ actions): https://www.courtlistener.com/opinion/[10272738](/patent/10272738)/cisco-systems-inc-v-ramot-at-tel-aviv-university-ltd/

Confidence notes. High confidence: no AIA trial on the '417; identity and filing dates of IPR2020-00122/-00123/-00484; IPR2020-00123's § 314(a) denial and panel. Medium confidence: the full disposition of IPR2020-00122 and IPR2020-00484 (unconfirmed — verify on E2E). Low confidence / do not cite without verification: which patents were asserted in D. Del. 1:21-cv-00295, and the Fed. Cir. 24-1726 disposition.

I'll research the assignment record for US 9,031,417, focusing on the PBLM RTAU LLC security interest and the correspondents of record.

Let me dig into the specific PBLM RTAU LLC security interest and Ramot's enforcement history.

Ownership & Assignment Analysis — US 9,031,417 B2

Data limitation up front (read before the timeline). The authoritative full text provided for this patent surfaces its recorded post-issuance legal events (assignee names, conveyance types, and recording dates) but not the USPTO reel/frame numbers or the correspondent-of-record for each recording. I was unable to complete a live query of the USPTO Assignment Center (assignmentcenter.uspto.gov) within this session, so I am not asserting reel/frame numbers or correspondent names — doing so would be fabrication. Every reel/frame value below is marked [not retrieved]. This is the single largest gap in this analysis and the first thing to close on verification. (Assignment Center direct search: https://assignmentcenter.uspto.gov/ · alternate index: https://assignment.uspto.gov/patent/index.html — search "9031417" or application 14/325,486.)

Inventors

Unusual-pattern check: not present. The classic "all inventors departed the assignee within 12 months of filing" fire-sale precursor does not appear here. All three inventors were TAU-affiliated at the priority date, and the rights were channeled to TAU's technology-transfer company (Ramot) by ordinary university policy. Ehrlichman's later move into industry (2013 onward) is a normal academic career trajectory and occurred 6+ years after filing — it is not evidence of a portfolio transfer.

Original assignee

Ramot at Tel-Aviv University Ltd. (Tel Aviv, Israel) — original and current assignee.

• Primary line of business: Ramot is the "Business Engagement Center" / technology-transfer company of Tel Aviv University — it protects and commercializes TAU inventions, holds ≈1,000 granted patents worldwide, and has helped found 150+ startups (per Ramot's own description as quoted in the E.D. Texas complaint in Ramot v. Cisco).
• Did it ship a product embodying the claims? No. Ramot is a licensing/tech-transfer entity; it does not manufacture or sell optical modulators. This is the factual basis for treating it as a non-practicing owner.
• Current status: Operating (active TTO). Not acquired, not dissolved, not in bankruptcy.
• Continuity note: Every later member of this family — US 9,203,425; 9,479,191; 10,005,527; 10,033,465; 10,205,527 / 10,270,535; 10,461,866; 11,133,872; 11,342,998; 11,716,148; 12,191,912; 12,388,533 — remains Ramot-owned. There is no re-assignment of this family to an outside NPE anywhere on the record.

Assignment timeline

Records retrieved (from the patent's own legal-events data) show two post-filing recordings against this patent. Present in recorded-date order; note the logical ordering problem flagged under entry 2.

1. 2014-10-24 (recorded) — Reel/frame [not retrieved]

• Conveyance: Security Interest (see document for details)
• Assignor: Ramot at Tel Aviv University Ltd.
• Assignee: PBLM RTAU LLC
• Correspondent of record: not retrieved — could not be captured this session; cannot be assessed for the "repeat correspondent" signal.
• Context: Securitization / collateralization — this is a lien, not an ownership transfer. Ramot pledged patent rights as collateral to a secured party. (Naming observation: the secured party's name embeds Ramot's own acronym "RTAU," which is characteristic of a single-purpose collateral vehicle tied to the debtor rather than of a generic patent-aggregation shell. I flag this as a naming observation only; the conveyance type — "Security Interest" — is the operative evidence.)

2. 2015-03-09 (recorded) — Reel/frame [not retrieved]

• Conveyance: Assignment of Interest (see document for details)
• Assignors: Shlomo Ruschin; Ofer Amrani; Yossef Ehrlichman (all three named inventors)
• Assignee: Ramot at Tel-Aviv University Ltd.
• Correspondent of record: not retrieved — same gap as above.
• Context: Internal — the inventors' assignment of rights to the university's tech-transfer company. Recorded ~2 months before issuance (patent issued 2015-05-12), which is the standard practice of recording near issue so the assignee prints on the front page.

Ordering caveat (important): The 2015-03-09 inventor→Ramot recording carries a recorded date later than the 2014-10-24 security interest. Logically the inventor assignment must have been executed earlier — Ramot could not pledge rights it did not yet hold. The 2015-03-09 date is therefore almost certainly a recording date, not an execution date. The true execution date of the inventor assignment is [not retrieved].

No further recorded assignments exist. The chain terminates at Ramot (subject to the PBLM RTAU LLC security interest). There is no transfer to any licensing-only owner.

Litigation context (from the Delaware record and CAFC)

• 2014-11-05 — Ramot sues Cisco, E.D. Tex. (C.A. 14-1018-JRG); voluntarily dismissed 2015-02-24. The specific patent asserted in this 2014 action is not confirmed in the sources retrieved; the '417 patent did not issue until May 2015, so this action necessarily asserted an earlier family member.
• 2019-06-12 — Ramot sues Cisco, E.D. Tex. (asserting the '465 and '535 patents).
• 2021-02-26 — Ramot sues Acacia alone, D. Del. (1:21-cv-00295) — same two patents.
• 2021-09-28 — Cisco/Acacia file DJ action on the newly issued '872 patent (1:21-cv-01365); Ramot counterclaims.
• 2022-05-24 — '998 patent issues; Cisco/Acacia file DJ action at 12:01 a.m.
• 2024–2026 — PTAB invalidates the '872 patent; Fed. Cir. Nos. 24-1726/24-1727/24-1728, Cisco v. Ramot, Rule 36 AFFIRMED, Apr. 13, 2026.

Critical scoping point (consistent with the previously generated summary — no contradiction found): US 9,031,417 itself is NOT the patent-in-suit in any of these actions. The '417 patent is family background; the asserted patents are the later continuations ('465, '535, '872, '998).

Timeline diagram

timeline
    title Ownership of US 9031417
    2007 : Priority filing by TAU inventors
    2008 : PCT application filed
    2009 : Continuation in part filed
    2014 : Security interest recorded to PBLM RTAU LLC
         : Ramot sues Cisco in E D Texas
    2015 : Inventors assign rights to Ramot
         : Patent issues as US 9031417
    2019 : Ramot sues Cisco again
    2021 : Ramot sues Acacia in Delaware
    2022 : Ramot asserts newer family patents

NPE / troll-pattern signals

Verdict

NPE — moderate confidence.

Justification: The ownership chain terminates at Ramot at Tel-Aviv University Ltd., the original assignee, which neither manufactures nor sells any product embodying the claims and which is a documented serial non-practicing asserter of this family — E.D. Tex. 2014-11-05 and 2019-06-12, and D. Del. 2021-02-26 (1:21-cv-00295) and 2021-09-28 (1:21-cv-01365), with the counterclaims running through the 2026 Federal Circuit appeals. That is one strong signal (assertion without practicing), which caps confidence at moderate. The only non-Ramot party in the recorded chain, PBLM RTAU LLC, holds a Security Interest recorded 2014-10-24 — a lien, not ownership — so it does not convert this into an NPE-transfer chain, and the classic troll indicia (shell-LLC ownership transfer, cascading LLC transfers, enumerated NPE-list assignee, repeat recording correspondent, bankruptcy, privateering) are all absent. In short: this is a university tech-transfer office enforcing its own patents — non-practicing and serial, but not a classic troll structure. If the analysis were confined strictly to the eight troll-pattern indicia, every one would read "not present."

Two caveats that could move this verdict:

1. If the reel/frame and correspondent for the two recordings show a recurring recording attorney across a wider Ramot/NPE family, signal 3 would fire and could push this toward high confidence.
2. If PBLM RTAU LLC is confirmed to be a third-party patent-finance/assertion vehicle rather than a Ramot-linked collateral SPV, signal 1 could flip — the recorded date (2014-10-24) immediately preceding the first Cisco suit makes that the highest-value item to verify.

Verify at: https://assignmentcenter.uspto.gov/ (search patent 9031417 / application 14/325,486) — retrieve reel/frame and correspondent for both recordings, then re-run signals 1 and 3.

I'll research the prior art cited in US 9,031,417. Let me start with several targeted searches.

I have the core reference list. Let me get the complete citations and details on the key references.

Let me get the full citation list with dates and details on the key US references.

I've hit the point where I need to synthesize. Below is the prior-art analysis for US 9,031,417 B2 based on the authoritative full text you provided plus the citation record I retrieved.

A. Confirmation of the target

• US 9,031,417 B2 ("Linearized optical digital-to-analog modulator"), application 14/325,486, filed 2014-07-08, granted 2015-05-12, priority 2007-06-13.
• I searched only this identifier. I did not substitute adjacent numbers (9,031,415 / 9,031,416 / 9,031,418). All statements below relate to 9,031,417.
• Scope caveat stated up front: I could not open the USPTO PatentCenter/file-wrapper "References Cited" tab for '417 in this session. The citation list below is taken from the FreePatentsOnline rendering of the '417 front page and cross-checked against Google Patents/Justia pages for the cited documents. One caution flagged in §B: the FPO rendering appears to merge the "U.S. Patent Documents" list with the applicant's own parent documents, which must be handled separately for §102 purposes.

B. The patent-citation record for US 9,031,417

U.S. patent documents

(Note the date asymmetry: several references — #12, 13, 14, 15 — issued after the 2007-06-13 priority date. They can only be §102(e) art if their own filing dates precede 2007-06-13. I could verify that only for some; the §102(e) dates should be checked before relying on them.)

Foreign patent documents

Non-patent literature (applicant-cited / of record)

• Yacoubian et al., "Digital-to-analog conversion using electrooptic modulators," IEEE Photonics Technology Letters, vol. 15, pp. 117–119, Jan. 2003 — one MZI modulator per bit; one-to-one; no anticipation of the M>N/DDC claims.
• Leven et al., "A 12.5 GSample/s optical digital-to-analog converter with 3.8 effective bits," LEOS 2004, vol. 1, pp. 270–271, Nov. 2004 — per the '417 background, uses one modulator per 2 bits, highly nonlinear (3.8 effective bits for a 6-bit design); no anticipation.
• Zumbahlen, "Basic Linear Design: Digital-to-Analog Converter Architectures: Intentionally Nonlinear DACs," Analog Devices Ltd., §6.1, pp. 6.37–6.39, 2007 — general DAC architectures; acknowledged background.
• EPO prosecution papers for EP 08763563.7 (Art. 94(3) communications; oral proceedings Sep. 2014) — of-record prosecution material, not prior art per se.

C. Which references actually threaten the issued claims

Applying the §102 "every element" standard to the issued independent claims (1, 11, 19, 27 — each requiring M > N, a digital-to-digital converter mapping N-bit words → M-bit binary actuation vectors, and actuation differing from any one-to-one bit↔electrode correspondence):

• No cited reference anticipates claims 1, 11, 19, or 27. Every cited document uses a one-to-one correspondence (n electrodes/phase-shifters for n bits) or is not an electrode-mapping architecture at all. This is consistent with the claims having been allowed.
• No cited reference anticipates the dependent claims either, because each dependent claim incorporates the M>N/DDC/non-one-to-one limitations of its independent claim. References such as US 5,724,178 (EAM), US 7,978,390 (multilevel/QAM modulation) and US 7,483,597 (multi-level signaling) are directed only to the subject matter of individual dependent features (modulator genus, PAM/QAM response, etc.) and would at most support §103 combinations.

The two references with genuine §102 exposure to note are:

1. US 4,288,785 (Papuchon) — the acknowledged "closest prior art." It supplies nearly every structural element of the independent claims (multi-electrode MZI, binary common voltages, weighted electrodes, analog optical output, photodetector-based DAC). Its fatal gap is the mapping: n electrodes for n bits, one-to-one, power-of-2 lengths. The '417 claims were written precisely to exclude this (M>N + DDC + non-one-to-one).
2. US 2004/0208614 A1 (Price) — the only cited reference that the USPTO actually invoked under §102 against this family (in the later continuation Ser. No. 15/298/327), and which applicants successfully traversed. Its asserted §102 rejection is the best evidence of where the novelty was contended.

D. Ranked "most relevant prior art" for US 9,031,417

1. US 4,288,785 (Papuchon) / EP 0020216 — multi-electrode, binary-driven MZI digital-to-analog converter. Closest overall; §103 anchor; explicitly distinguished in the '417 specification.
2. US 2004/0208614 A1 (Price) — examiner-asserted §102 art against the family; traversed on the bits→voltage-conversion point.
3. US 7,061,414 (Chen/Leven/Tu) and US 7,403,711 (Chen et al.) — Lucent optical DACs (one modulator/phase-shifter per bit or per 2 bits); highly relevant background, no M>N.
4. Non-patent: Yacoubian (2003); Leven (2004) — one-modulator-per-bit and one-modulator-per-2-bits architectures.
5. Secondary structural art: US 5,724,178 (EAM drive), US 5,694,504 (semiconductor modulator π-shift), US 5,137,359 / EP 1036302 / WO 2004/074914 (digital phase modulators), US 7,203,552 / US 7,212,292 (binary-weighted actuator arrays / calibration), US 7,483,597 and US 7,978,390 (multi-level/QAM optical modulation), US 4,613,204 (electro-optic weighting), US 6,781,537 (high-speed DAC).

E. Confidence and explicit uncertainties

• High confidence in the citation list (references + dates) and in the claim limitations requiring M>N / DDC / non-one-to-one actuation — these come from the patent's own front page and claims as fetched.
• High confidence that no cited reference anticipates the issued independent claims, and that Papuchon is the closest art and Price the only examiner-asserted §102 reference in the family record.
• Medium confidence on the descriptions of US 6,781,537 (Taraschuk) and US 7,483,597 (Shastri) — I could not verify their full texts this session; their §102 relevance should be re-checked against the actual documents, particularly Shastri given the "multi-level signaling" subject matter.
• Unverified: the exact filing dates of the post-2007-issued references (items 12–15), which govern whether they are available as §102(e) art against a 2007-06-13 priority.
• Record caveat: the FreePatentsOnline rendering mixes the applicant's own parent documents (US 8,044,835; US 2010/0156679) into the "U.S. Patent Documents" table; these are same-family documents and are not prior art under §102. A file-wrapper check for application 14/325,486 at USPTO PatentCenter would settle which references were cited by the Examiner versus merely of record.

Sources: patent full text at patents.google.com/patent/US9031417B2; FreePatentsOnline.com/9031417.html (references-cited table); patents.google.com/patent/US4288785 and US7061414; patentimages/US7403711; Justia pages for US 10,270,535 (family references list); USPTO PTAB/petition document 1530722 (Ser. No. 15/298,327 prosecution — Price/Pinguet rejections); Justia US20240430011 and US11133872 (family continuity and references).

I'll ground this obviousness analysis in the actual prior-art record for this patent and its family, then build the combination theory.

Two search results are pivotal (the inventors' own IEEEI/IEEE 2008 publication and the Bell Labs/Lucent art). Let me verify the IPR record for this family and the additional prior-art references cited against it.

Obviousness analysis of US 9,031,417 B2 under 35 U.S.C. § 103

Building on the prior summary — the bibliographic record, the continuity chain, and the litigation picture are taken as established and are not repeated here. The one cross-reference that matters for § 103 is the priority statement: the '417 issued from an application filed 8 Jul 2014, but claims benefit back to 13 Jun 2007 via PCT/IL2008/000805 → 12/636,805 → 13/280,371. That chain controls which § 102/§ 103 regime applies.

1. Governing framework and the claim-construction anchors that constrain the § 103 case

Two claim-construction consequences worth stating up front: (i) Because "mapping" requires only that the input and output sets differ, a plain binary-to-thermometer decode (3 bits in, 7 electrodes out) satisfies the limitation literally — the set {0…7} is not the set {0,1,3,7,15,31,63,127}. (ii) Because claim 1 requires only that at least one electrode be actuated "in a manner differing from any one-to-one correspondence of the bits … to the actuating electrodes," any M>N architecture satisfies it by arithmetic necessity — there is no bijection from N bits to M>N electrodes. The '417 independent claims are therefore broader and more vulnerable than the later family members ('465, '535, '866), which the record shows were narrowed in response to IPR/reexamination.

2. Prior-art inventory actually on the record for this patent

3. No single reference anticipates — the case is a combination case

• Papuchon alone fails: n electrodes for n bits (M = N) and an identity bit-to-electrode map (bit i → electrode i, power-of-two lengths) — the opposite of claim 1(d)/claim 11. Papuchon is the FIG. 14 device the '417 disparages.
• Roberts alone is the closest § 102 candidate for claims 1, 11, 19 and 27 (M > N; binary actuation vectors; optical output; DSP-computed mapping to cure non-ideal transfer functions; "nominally equal drive voltages" = binary drive). It is plausible anticipation of claim 27 and strong § 102(e)/§ 103 art for claims 1/11/19 depending on its § 102 date. Treat it as the primary reference rather than a secondary one.
• Yacoubian, Leven/Chen, Saida, Vawter are the "problem statement" art and element-supply art (O/E conversion, laser source, high-speed context).

4. Grounds of rejection

Ground 1 (primary): Papuchon + Roberts — claims 1, 9, 11, 16, 19, 22, 24, 27

Motivation to combine (MPEP 2143 (A), (C), (D), (F)):

1. Same field, same device, same problem. Papuchon expressly recognises that the power-of-two sequence "did not solve the non-linearity problem" and prescribes the palliative of reducing the modulation range around the inflection point. Roberts expressly targets the identical defect — "non-ideal electrical to optical transfer functions from the electrodes." Both address the same MZ multi-electrode structure;
2. "Improve a similar device in the same way" (rationale C). Papuchon already discloses two ways of weighting a multi-electrode MZ device (unequal lengths; equal lengths + binary voltage dividers). Replacing Papuchon's static, hard-wired weighting with Roberts' computed, reconfigurable electrode-activation pattern is a substitution of one known keying mechanism for another in an otherwise unchanged device, with a predictable result (better fit to the target transfer curve);
3. Design incentive / "obvious to try" (rationale F, and KSR's "design incentives; market pressures"). Papuchon's own teaching that linearity is purchased only by sacrificing range creates a direct incentive to obtain linearity without range sacrifice; the claimed solution (spend digital logic, not analog range) is the cheapest available route and uses a DSP that Roberts shows was already co-packaged with the modulator;
4. Reasonable expectation of success. The patent itself frames the invention as a numerical optimisation over two known parameter sets — control matrix B and length vector L — with a closed-form solution (its own Eq. (7)/(9)). Post-KSR, optimising known parameters of a known device to obtain a predictable improvement ("the results of ordinary innovation") is not patentable subject matter.

Ground 2: Papuchon + Yacoubian + Roberts — adds claims 5/14/21/28 (PAM) and the "more linear than one-to-one" claims 8/24

Yacoubian supplies the explicit problem statement for the single multi-electrode device (quoted above) and the explicit design lever ("extending driving electrode lengths"); Roberts supplies the digital remapping. The combination is a textbook "identify the deficiency of the primary reference and cure it as the secondary reference teaches."

Ground 3: Papuchon (or Leven/Chen) + Roberts + Ho — claims 5, 14, 21, 29, 25 (QAM / phase modulation)

Optical field phase modulation by driving the two MZM arms differently was routine before 2007 (Papuchon's geometry itself applies opposite field directions on the two arms; Ho's 2005 textbook is a documentary anchor). Where the claims are supported only by CIP matter (the 16-/64-/256-QAM Example IV), this is where the priority defect identified in § 1 becomes outcome-determinative, because the '417's QAM content was demonstrably added in the 14 Dec 2009 CIP.

Ground 4: Roberts + the inventors' own 2008 IEEEI paper (contingent)

If any claim loses the 2007/2008 priority date (see § 1), the IEEEI 2008 paper — combined with Papuchon — makes claims 1, 6, 9, 16, 17, 22, 24, 28 prima facie obvious (indeed nearly anticipatory): it discloses M = N+1, the DDC look-up table, the optimal non-identity mapping, non-power-of-two electrode lengths, full-dynamic-range operation, and the EAM/directly-modulated-laser extensions (claims 2, 3, 12).

5. Dependent claims — grouping and support

6. Where the § 103 case is contested, and where it is weakest

1. "Is Roberts' DSP a 'digital-to-digital converter'?" This is precisely the issue on which the Board was unpersuaded in the family IPRs. Patent Owner argued (and the E.D. Tex. court quoted approvingly in its construction) that "Roberts Table 1 merely discloses an identical, one-to-one translation between binary formats … As such, it does not teach or suggest 'that a binary input data vector D_i has to be mapped to a control vector B_i, yet B_i ≠ D_i.'" The counter to that argument — and it is a strong one for the '417 specifically — is that the construed test is set-level ("decimal values of the output set and input set are not identical"), and {0,1,3,7,15,31,63,127} ≠ {0,…,7}. The Board's scepticism was aimed at the later, narrower "mapping a set of N input values … to a digital drive vector where M > N" formulation; the '417's independent claims do not contain that formulation.
2. Priority of CIP-added matter (QAM, and any later-added linearity framing). This is a threshold, binary gate: if the CIP-added matter is real, the Dec 2008 IEEEI paper and the Dec 2008 WO 2008/152642 publication become § 102(a)/(b) art for those claims on a date computation with an 11-day margin. It should be nailed down by comparing WO 2008/152642's disclosure against the '417 specification.
3. Teaching away (Ramot's best rebuttal). Papuchon teaches reducing the modulation range to buy linearity, and Yacoubian criticises the single-device multi-electrode approach as "limited by the device linearity and its extinction ratio." Post-KSR this is weak teaching-away: Yacoubian's statement is a criticism of an approach it simultaneously proposes to improve ("tune Vπ … extending driving electrode lengths"), and a reference's identification of a known deficiency with a known fix is a motivation to combine, not away.
4. Secondary considerations. None is documented in the retrieved record as a nexus-bearing objective indicium (no unexpected-results data, no licensing-due-to-validity evidence, no industry praise). Commercial success of Cisco's PAM4 modules does not run to Ramot. Conversely, Ramot's own infringement theory cuts against it: the Delaware/Texas pleadings allege infringement by Cisco's "digital to digital mapping" of "digital symbols … to other digital symbols corresponding to voltage values for driving modulators" in standard PAM4 modules — i.e., that a conventional PAM4 DSP symbol map reads on the claim. Symmetry between "conventional symbol mapping infringes" and "conventional symbol mapping is obvious" is the central vulnerability of the '417 independent claims.
5. Procedural context to weigh. The record shows (a) Cisco as appellant in Fed. Cir. Nos. 24‑1726/1727/1728 with a Rule 36 AFFIRMED on 13 Apr 2026 (from IPR2022‑00575/00576/01283), which is consistent with the Board having declined to cancel the challenged claims, but Rule 36 affirmances carry no reasoning and I do not have the final written decisions' holdings — I will not characterise them; and (b) the '465, '535 and '866 patents are in reexamination with claims amended and still rejected per Cisco's '872 petition. Those are family-level, not '417-level, data points, but they indicate that the DDC-mapping claim space in this family is under sustained § 102/§ 103 pressure.

7. Bottom line

• Most vulnerable: claims 1, 9, 11, 16, 19, 22, 24, 27, 28, 30 — over Roberts (M>N digital actuation-vector mapping in a co-packaged MZ modulator driver expressly to cure non-ideal electrode transfer functions), alone or in combination with Papuchon (binary common-voltage multi-electrode MZI DAC) and Yacoubian (explicit statement that the factor-of-two multi-electrode single device is linearity/extinction-limited, plus the electrode-length lever). Under the E.D. Tex. construction of "mapping," a thermometric/decoded input-to-electrode map satisfies the non-identity requirement.
• Vulnerable, but need one more reference or a priority ruling: claims 2, 3, 12 (semiconductor/modulated-light-generation device — one would want an EAM or directly-modulated-laser linearisation reference dated before 13 Jun 2007) and claim 17 (non-power-of-two effective areas — best evidence is the inventors' own Dec 2008 papers, which requires the CIP priority to break).
• Weakest § 103 targets: claims 5, 14, 21, 25, 29 to the extent they depend on the QAM/phase-modulation CIP matter — these turn on the priority question, not on the art.
• The single most useful next step for a rigorous § 103 case: (i) obtain Roberts' (US 7,277,603) filing date to fix its § 102(e) date; (ii) obtain the full citation and date of Taraschuk (6-bit→8-bit optical-modulator lineariser, relied on by Cisco — identifier unverified here); and (iii) diff WO 2008/152642 against the '417 specification to determine exactly which claim limitations (notably QAM) are CIP-added matter with a 14 Dec 2009 constructive date.

Citations for grounding: patents.google.com/patent/US9031417B2 (full text and claims); patents.google.com/patent/US4288785; ieeexplore.ieee.org/abstract/document/1159081 (Yacoubian & Das 2003); patents.google.com/patent/US7061414; US 7,277,603 (Roberts) via Cisco IPR exhibit record; IPR petition and institution materials at ptacts.uspto.gov and bdiplaw.com/wp-content/uploads/2021/03/Ramot-v-Cisco-PTAB-IPR2020-00484-10.pdf; E.D. Tex. claim construction reproduced at g.casetext.com/case/ramot-at-tel-aviv-univ-ltd-v-cisco-sys; IEEEI 2008 paper "Photonic Digital to Analog Conversion"; Fed. Cir. Rule 36 judgment, cafc.uscourts.gov/opinions-orders/24-1726.RULE_36_JUDGMENT.4-13-2026_2674975.pdf.