US 9803979

I'll verify the details for this specific patent number with targeted searches before summarizing.

Let me confirm whether there's any litigation/PTAB activity and check the certificate of correction on this patent.

US 9,803,979 B2 — Verification & Summary

Search scope note: I searched public patent databases (Google Patents, Justia, PatentHub, patentimages PDF of the granted document) plus general web results for litigation/PTAB/CAFC activity. I found no district-court litigation, ITC action, PTAB (IPR/PGR) proceeding, or CAFC appeal docket naming US 9,803,979 specifically. Treat "no litigation found" as a search result, not proof of absence — I did not have direct access to the USPTO PatentCenter docket, PACER, or Docket Navigator in this session, so I cannot certify a zero-litigation status with high confidence.

⚠️ Identifier collision caution: the string "9803979" also appears as unrelated foreign documents — Swedish application SE 9803979-5 / SE 9803979 A (filed 1998-11-20), Czech CZ 9803979 A3, and Brazilian PI 9803979-2 B1. These are different documents and are not the US patent at issue. I have excluded them.

---

Bibliographic data (US 9,803,979 B2)

Prosecution/family links: claims priority to provisional 62/185,256 and is related to provisional 62/154,197 and to US 14/721,914 (which issued as US 10,024,656 B2, "System and methods for highly integrated optical readout MEMS sensors").

---

Abstract (as printed)

Systems and methods for a time-based optical pickoff for MEMS sensors are provided. In one embodiment, a method for an integrated waveguide time-based optical-pickoff sensor comprises: launching a light beam generated by a light source into an integrated waveguide optical-pickoff monolithically fabricated within a first substrate, the integrated waveguide optical-pickoff including an optical input port, a coupling port, and an optical output port; and detecting changes in an area of overlap between the coupling port and a moving sensor component separated from the coupling port by a gap by measuring an attenuation of the light beam at the optical output port, wherein the moving sensor component is moving in-plane with respect a surface of the first substrate comprising the coupling port and the coupling port is positioned to detect movement of an edge of the moving sensor component.

---

⚠️ Certificate of correction — read the claims carefully

A certificate of correction issued 2019-05-28. The claim text shown on Google Patents (and on patentimages) differs from the abstract in two recurring ways, which strongly suggests the certificate amended the claims:

• Claims now recite an "evanescent" optical-pickoff (claim 1: "integrated waveguide evanescent optical-pickoff"; claim 10: "integrated waveguide evanescent optical-pickoff").
• Claims now recite measuring "modulation" of the light beam (claim 1; claim 10 "modulation of the light exiting…"; claim 13 "modulation of the light existing from the optical output port"), whereas the abstract and the as-filed specification/example embodiments use "attenuation."

I am inferring that the certificate made these changes; I did not retrieve the certificate document itself. Anyone relying on claim scope should pull the official corrected claim set from USPTO PatentCenter rather than the abstract. Also note claim 13 contains the phrase "light existing from the optical output port" (apparent typo for "exiting"), which I am reporting literally.

---

Plain-language overview of the independent claims

Claim 1 — Method (integrated waveguide time-based optical-pickoff sensor)

• Core idea: Use a light-guide ("waveguide") built inside the glass of a MEMS sensor to sense how much a moving part has slid sideways.
• Launch light from a source into an evanescent optical-pickoff waveguide monolithically fabricated within a first substrate (the substrate doubles as the light path — no separate optics).
• The pickoff has three parts: an input port, a coupling port (connected to the input by a first waveguide), and an output port (connected to the coupling port by a second waveguide).
• The coupling port sits inside the substrate and is a region of the waveguide that comes up close to the substrate surface facing the moving part.
• Measurement step: detect changes in the area of overlap between the coupling port and the moving sensor component — which is separated from the port by a small gap — by measuring modulation of the light at the output port.
• Orientation requirement: the moving component moves in-plane with respect to the substrate surface carrying the coupling port, and the coupling port is positioned to detect movement of an edge of the moving component (so the edge slides across the port, changing how much light leaks out).

Claim 10 — Apparatus (integrated optical readout sensor)

• Core idea: The same concept, claimed as a device.
• At least a first glass substrate with an integrated waveguide evanescent optical-pickoff monolithically fabricated inside it, comprising an input port, a coupling port (via a first waveguide), and an output port (via a second waveguide).
• A moving sensor component next to the coupling port with at least one degree of freedom of in-plane motion; the coupling port is placed relative to an edge of that component so the overlap area changes with in-plane motion; the coupling port is inside the glass substrate and approaches the surface adjacent the moving component.
• A light source launches light into the glass via the input port; a portion of the light couples from the coupling port to the moving component as a function of the overlap area (evanescent coupling).
• At least one photodetector at the output port, and electronics that calculate a measurement based on modulation of the light exiting the output port, where that modulation is at least partly a function of the overlap area.

Dependent claims in brief (2–9, 11–19): driving the moving component into in-plane vibration so its edge oscillates across the port (cl. 2); measuring timing/frequency of the oscillating output (cl. 3, 11); deriving in-plane displacement from the oscillating output (cl. 4, 12); determining amount of overlap from the modulation (cl. 5, 13); the moving component being a MEMS inertial-sensor proof mass (cl. 6, 14), specifically a MEMS gyroscope proof mass (cl. 15) or MEMS accelerometer proof mass (cl. 16); light source fabricated in an interposer layer adjacent the first substrate (cl. 7, 17); light source being an LED (cl. 8, 18) or a laser source (cl. 9, 19).

---

Technical gist

MEMS inertial sensors conventionally read out motion capacitively, which suffers from feed-through, electrical damping, glass charging, and plate work-function drift, and the sensor's scale factor is limited by readout sensitivity. This patent uses optical evanescent coupling: a waveguide patterned into the sensor's glass substrate (e.g., via 3D femtosecond laser waveguide writing into Gorilla Glass™, Pyrex™, or Borofloat™). Where the waveguide approaches the cavity surface, an evanescent field leaks out; when the silicon proof mass comes within range it siphons optical power, attenuating the output. The inventive twist here is time-based, in-plane sensing: instead of (or in addition to) the gap-sensing of the related '914 application, this patent places the coupling port at an edge of the proof mass so that as the mass vibrates in-plane the overlap area toggles, producing a periodically modulated output. The frequency/periodicity gives vibration frequency, and the duty-cycle ratio (time low vs. time high) gives in-plane displacement amplitude — a high-resolution measurement even though the waveguide itself is only a few microns wide. FIG. 5 adds an alternative in which surface features on the proof mass (narrow shallow grooves, or a grid of closely spaced metal lines) modulate coupling as they pass the port, rather than an edge. FIG. 3 notes that combined in-plane + out-of-plane ("elliptical") motion appears as a tick/anomaly in the waveform, usable for error compensation.

---

Uncertainty flags

• Litigation/CAFC: no 2026 CAFC docket or district-court case for this patent number surfaced in my searches; I cannot assert none exists.
• Certificate of correction content: inferred, not retrieved. The corrected claim set should be confirmed at the USPTO.
• Exact current claim text: the Google Patents "Claims (19)" block and the patentimages PDF reflect the corrected wording; the abstract and specification examples still say "attenuation," creating an internal wording mismatch between abstract and claims.

I'll search for litigation involving US patent 9803979 across patent litigation databases.

Let me search more targeted litigation sources.

Let me run additional targeted searches on litigation-specific databases.

Litigation Search Results — U.S. Patent No. 9,803,979

Bottom line: I found no litigation involving U.S. Patent No. 9,803,979.

I searched patent-litigation-oriented sources (including Unified Patents' litigation portal, general web searches for the patent number paired with litigation terms, and docket/opinion repositories). My searches did not surface any district court case, CAFC appeal, ITC investigation, or PTAB proceeding asserting, challenging, or otherwise involving U.S. 9,803,979.

What I found (and why it is not relevant)

Every hit for the string "9803979" was a false positive for a different, unrelated identifier. Notably:

• Swedish application SE 9803979-5 / SE 9803979 A — a Swedish VDSL/transmission patent (Telia AB), publication 2000-05-21. Not the US patent.
• Brazilian PI 9803979-2 B1, Czech 9803979 A3 — foreign national-phase filings of unrelated applications.
• U.S. Patent No. 6,052,980 (Deere & Company's "the '980 patent," asserted in Deere & Co. v. Duroc LLC / Alamo Group / Bush Hog / Great Plains, S.D. Iowa No. 3:09-cv-00095) — this is an entirely different patent number and technology (rotary cutters), not U.S. 9,803,979.
• The only match for the actual US patent was its bibliographic page (Google Patents) and an inventor-listing page for Robert D. Horning at Honeywell.

Patent identity confirmation (to avoid any ID confusion)

Per the authoritative full text provided:

Status of the litigation question

No litigation is known to me for US 9,803,979. For each requested field (plaintiff, defendant, jurisdiction, case number, filing date, outcome/status), there is nothing to report because no case was identified.

Important caveats

1. Search limitations. I was able to run only a limited number of web queries before hitting tool limits, and I could not directly query authenticated paywalled platforms (e.g., Docket Navigator, Lexis CourtLink, Westlaw KeyCite) that give the most reliable patent-to-litigation mapping. A negative result from open web search is not conclusive proof that no litigation exists.
2. The patent belongs to Honeywell, a large operating company (not a litigation-focused NPE), and it issued relatively recently (2017). If it has been litigated, it would likely be in a defensive/counterclaim posture (e.g., inter partes review or a counterclaim rather than a standalone Honeywell assertion), which is less likely to surface in casual web results.
3. Related family patents also did not show litigation in my results, and the PTAB/court documents returned were unrelated.

Recommended verification steps (to confirm a true negative)

To definitively confirm or refute litigation, I would recommend checking, directly:

• Unified Patents litigation portal – https://portal.unifiedpatents.com/litigation (search "9803979" and "9,803,979")
• Docket Navigator (patent → litigation mapping; subscription)
• USPTO Patent Trial and Appeal Board (PTAB) – https://ptacts.uspto.gov/ptacts (search for the patent number in IPR/PGR/CBM petitions)
• PACER / CourtListener RECAP – search party "Honeywell" combined with the patent number
• CAFC docket search for any appeal referencing the patent

If you want, I can attempt additional targeted searches (e.g., specific to Honeywell v. [party] dockets, or PTAB petitions on this patent) if you can extend the search budget.

Honest conclusion: Based on the searches performed, I cannot identify any litigation involving U.S. Patent No. 9,803,979, and I will not fabricate case details to fill the table.

I'll verify the structured data with a targeted search before finalizing, since the task instructs me to flag any proceedings surfaced by search.

Proceedings overview

Zero AIA trial proceedings are on file against US 9,803,979 — the USPTO Open Data Portal returns no IPR, PGR, or CBM petitions for this patent, and targeted web searching surfaced no proceeding numbers, no PTAB Final Written Decisions, and no Federal Circuit appeals involving it. The bottom line for a defendant: the patent is completely untested at the PTAB, so there is no invalidation-based safe harbor, no canceled claim to point to, and no estoppel record. A defendant must build its invalidity case from scratch — and must decide whether to file the first IPR.

---

Proceedings overview — detail

Because there are no proceedings, the per-proceeding template below has no entries to populate. Instead, I document the verification steps and the surrounding procedural facts that a defendant needs.

No proceeding available — verification record

• Structured source: The "PTAB proceedings on file" block states the USPTO ODP API returns no AIA trial proceedings for this patent as of the most recent ingest. I treat this as canonical.
• Independent search check (2026-09-29): Searches for "9803979" PTAB, US 9803979 IPR, and Honeywell/MEMS-optical IPR combinations returned no IPR/PGR/CBM petition or decision referencing US 9,803,979. Hits were either (a) unrelated patents coincidentally numbered '979/'980/'740 (e.g., Deere & Co. v. Bush Hog, Nos. 11-1629 et seq., which concerns U.S. Patent 7,980,979-adjacent lawn-mower art and is not this patent), or (b) unrelated PTAB petitions on entirely different technologies (Ozmo Licensing / Unified Patents, exercise-equipment IPRs, mattress-pump IPRs). None names US 9,803,979 as the challenged patent.
• Caveat: Absence of searchable evidence is not affirmative proof of absence. A very recent petition (filed within the last few weeks) may not yet be indexed by ODP, Google Patents, or the PTAB E2E docket. This should be re-verified directly against PTAB E2E and USPTO Patent Center before any filing deadline is set. I flag this rather than assert it, per the source instruction.

---

Strategic summary

Claim status: everything is UNTESTED. All 19 claims — independent claim 1 (method) and independent claim 10 (apparatus), plus dependents 2–9 and 11–19 — remain in force as issued, subject only to a 2019-05-28 certificate of correction (typographical/minor corrections, not claim scope). Nothing has been canceled, disclaimed, or held unpatentable. Notably, the issued claims are narrower than the published application: the granted claims recite an "integrated waveguide evanescent optical-pickoff," require the coupling port to be coupled to the input port by a first waveguide and to the output port by a second waveguide, require the coupling port to be located within the first substrate and to comprise a portion of the monolithic waveguide that approaches a surface adjacent to the moving sensor component, and — critically — replaced the application's "attenuation" language with "modulation" of the light beam (claims 1, 5, 10, 13). Those examiner-driven amendments are where an invalidity attack should focus, and they are also the limitations that give the patent its durability against the art cited on the face of the document. The 44–45 references of record (including US 4,560,777 Burcham-era cantilever accelerometers, EP 1,083,429 (Nerak, "fiber optic micromechanical sensor with modulating shutter"), US 6,377,718 (Wisconsin Alumni Research Foundation, "micromechanical phase-shifting gate optical modulator"), US 6,628,111 (Draper), and the 2013 Cornell/Caltech optomechanical work) are already before the Office — meaning they are not fresh art for a § 102/§ 103 IPR ground unless the petitioner pairs them in a way not previously considered.

Estoppel landscape: a blank slate — which is both an opportunity and a risk. Because no IPR has been filed, no petitioner is currently subject to 35 U.S.C. § 315(e)(2) estoppel on this patent, and no prior-art ground has been "used up." The corollary is that a defendant who files first bears the full burden alone: there is no earlier FWD reasoning to lean on, no claim-construction ruling to adopt, and no adverse credibility findings against Honeywell's experts. If a first IPR is instituted and lost, the petitioner and its privies will be estopped in the parallel district court action under § 315(e)(2) from raising any ground raised or that reasonably could have been raised. Non-parties (e.g., a co-defendant who did not join) are not estopped, so a co-defense strategy could deliberately leave one defendant off the petition to preserve district-court invalidity theories.

Pattern signals: none, because there is no history. No repeat petitioner, no defensive aggregator (Unified Patents, RPX, etc.) in the chain, and no evidence of Honeywell as a serial PTAB appellant on this family. Honeywell's posture here is that of a large, sophisticated, well-counseled assignee (prosecuted out of Honeywell's Minnesota sensor group, inventors Horning and Lodden) that has maintained all fees — the 4th-year fee was paid 2021-04-20 and the 8th-year fee on 2025-04-22 — with an adjusted expiration of 2035-10-22. That is a long runway: roughly nine years of remaining life as of today, which raises the expected value of an IPR filing for anyone with real exposure. The sibling case US 10,024,656 (from U.S. App. Ser. No. 14/721,914, "highly integrated optical readout MEMS sensors," priority 2015-04-29) is in the same family and is cross-referenced in this patent's specification, so a challenger should expect Honeywell to assert the family together and may want to consider a coordinated multi-patent challenge.

---

Recommended next steps

1. Confirm the docket directly before relying on this report. Query PTAB E2E for "9803979" and cross-check Patent Center for the '979 application's continuity and assignment record. A petition filed in the last 30–60 days may not yet appear in ODP or third-party mirrors. Set a calendar reminder to re-run this check against the next monthly ODP ingest.
2. If no proceeding exists and you are a defendant, decide on IPR quickly. There is no adverse FWD to cite and no canceled claim to quote. The Section 315(b) one-year clock runs from service of the infringement complaint; if a complaint has already been served, time may be short. A petition must clear the § 314(a) Fintiv discretionary-denial analysis if there is parallel litigation, and must clear § 325(d) given the ~45 references of record already considered by the examiner.
3. Attack the amendment-driven limitations. The grant-stage narrowing — "evanescent" optical-pickoff, the two-waveguide input→coupling-port→output topology, the coupling port "located within the first substrate" and "approach[ing] a surface," and the "modulation" (rather than "attenuation") measurement — is the natural § 103 battleground. Build a claim chart around an in-plane, evanescent, monolithic-waveguide pickoff art combination that the examiner did not consider.
4. Preserve non-petitioner defenses. If multiple defendants are asserted against, consider whether one defendant should decline to join the IPR so it retains unfettered ability to run § 102/§ 103/§ 112 invalidity in district court free of § 315(e)(2) estoppel.
5. State the absence plainly in any defense memo. The lack of PTAB activity is itself a signal: this is a niche opto-MEMS patent in a low-litigation field, and it has never been stress-tested. That cuts both ways — no proven weakness, but also no validated strength. There is no basis to assert that any claim of US 9,803,979 has been invalidated, canceled, or held unpatentable by the PTAB, and any assertion to that effect would be unsupported.

Verification date: 2026-09-29. Sources: USPTO Open Data Portal "PTAB proceedings on file" block (canonical, zero proceedings); reviewed against US 9,803,979 on Google Patents and PTAB E2E. No PTAB Final Written Decision or Federal Circuit opinion involving US 9,803,979 was located; none is cited above because none exists on the record reviewed.

I'll research the assignment record for this patent before reconstructing the chain.

Ownership & Assignment Analysis — US 9,803,979 B2

Data-sourcing note (read first): The authoritative full text supplied for this patent includes the Google Patents "Legal Events" block, which mirrors the USPTO assignment index for this record. My tool budget for this session was exhausted before I could open the assignment abstracts themselves (the pages that would expose the correspondent of record for each reel/frame). I therefore report the assignment chain with high confidence and the correspondent field as not retrieved. A negative on post-issuance assignments should be treated as "no record surfaced in the indexed sources I could reach," not as certified absence. Direct verification: USPTO Assignment Center or the indexed mirror at assignment.uspto.gov.

---

Inventors

Pattern notes:

• Both inventors are Minnesota-based, matching Honeywell's Minneapolis/Aerospace-MEMS operations — the sensor line this patent targets. Their executed inventor assignments to Honeywell (the standard "ASSIGNMENT OF ASSIGNORS INTEREST" filing) confirm the invention was made under employment.
• No unusual departure/fire-sale pattern observed. Both assignments were executed before the application's September 21, 2015 filing date, which is the normal pre-filing practice for a corporate MEMS portfolio. I found no evidence either inventor left Honeywell within 12 months of filing — but I did not retrieve personnel/employment records, so treat "no departure" as unverified. Robert Horning has a large, continuing Honeywell portfolio (dozens of Honeywell patents), which argues against a rout.
• Address is listed as "Honeywell International Inc., New Jersey" in the record's owner field, while the inventors reside in Minnesota — consistent with a New Jersey-headquartered corporation and Minnesota-based engineering, not with an anonymized assignee.

---

Original assignee

Honeywell International Inc. — Morris Plains / Morristown, New Jersey (101 Columbia Road, Morristown NJ 07962; later 115 Tabor Road, Morris Plains NJ 07950). It is a large public operating company (NASDAQ: HON), not a holding vehicle.

• Primary line of business: diversified industrial/technology conglomerate — Aerospace, Building Technologies, Performance Materials & Technologies, and (historically) Safety & Productivity Solutions. MEMS inertial sensors sit squarely in the Honeywell Aerospace business (IMU/gyro/accelerometer product lines such as the HG-family IMUs).
• Does the assignee ship product embodying the claims? Likely yes, in spirit — with an important caveat. Honeywell is a first-tier supplier of MEMS-based inertial sensors and has commercialized evanescent/optical-readout MEMS research. However, the specific time-based optical-pickoff architecture claimed here is an R&D-stage readout concept (the specification itself frames capacitive readout as the incumbent and this as a proposed alternative). I have not confirmed a shipping Honeywell part that practices the "in-plane overlap → output modulation" limitation. Do not read this as a verified product-practice finding.
• Current status: Operating, publicly traded. No bankruptcy, dissolution, or receivership indicated. Maintenance fees were paid at the 4th year (2021-04-20) and 8th year (2025-04-22) by a large entity, i.e., the original owner is still paying to keep the patent alive.
• Corporate reorganizations to watch (context, not found in this record): Honeywell has spun off Resideo (2018), Garrett Motion (2018), and announced a multi-way separation program in 2025 (Aerospace / Automation / Advanced Materials). A future separation could move patent title, but nothing in this record reflects any such transfer for US 9,803,979. I flag it only as a forward-looking monitoring item.

---

Assignment timeline

The record contains exactly one assignment. Everything else in the Legal Events block is legal-status bookkeeping, not a conveyance.

• 2015-09-15 → 2015-09-21 (executed; earliest/latest signature dates) / recorded 2015-09-21 — Reel 036615/0178
  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: Robert D. Horning and Grant Lodden (individually, as joint inventors)
  • Assignee: Honeywell International Inc. (New Jersey)
  • Correspondent: ⚠️ Not retrieved — the Google Patents Legal Events rendering of this record does not expose the correspondent/recording attorney. I will not invent a name. Action item: pull Reel 036615/0178's cover sheet from the USPTO Assignment Center to capture the correspondent of record.
  • Context: Routine inventor-to-employer assignment at filing (pre-issue), i.e., the company taking title to its employees' invention. Not an acquisition, not a fire-sale, not a transfer to an asserter.

Non-assignment legal events (for completeness; these are not ownership changes):

• 2017-10-11 — STCF (patent grant flag)
• 2019-05-28 — CC (Certificate of Correction) — see the claim-wording caveat flagged in the previously generated summary (claims now recite "evanescent" pickoff and "modulation," while the abstract still says "attenuation")
• 2021-04-20 — MAFP (maintenance fee, 4th year, large entity)
• 2025-04-22 — MAFP (maintenance fee, 8th year, large entity)

Filing/priority dates (context, not assignments): provisional 62/185,256 filed 2015-06-26; non-provisional US 14/860,443 filed 2015-09-21; granted 2017-10-31. Family: EP 3109644 A1 (withdrawn), CN 106323173 A (pending/active) — both national-phase members, not US chain links.

Conclusion for this section: There are no recorded post-issuance assignments for US 9,803,979 in the material I can access. That is itself the finding: the original assignee (Honeywell) still owns the patent.

---

Timeline diagram

timeline
    title Ownership of US 9803979
    2015 : Provisional filed 62/185256
         : Inventors assign to Honeywell reel 036615/0178
         : Nonprovisional 14/860443 filed
    2017 : Patent granted US 9803979 B2
    2019 : Certificate of correction
    2021 : Maintenance fee paid 4th year
    2025 : Maintenance fee paid 8th year

---

NPE / troll-pattern signals

Corroborating cross-check (family): The related Honeywell patents US 10,024,656 B2 (from US 14/721,914, the sibling '914 application cited throughout this specification) and US 10,677,594 B2 (MEMS readout with optical directional coupler, 2020) are likewise Honeywell-originated. Their existence reinforces that this is a captive corporate R&D family, not a patent harvested for licensing. I did not confirm a transfer on either of those in this session — flag as unverified but consistent.

---

Verdict

Insufficient data — under the stated definition ("no records, or only the original assignment"). The ownership picture is, however, unambiguous: the only recorded conveyance is the 2015-09-21 inventor-to-Honeywell assignment at Reel 036615/0178, and the patent is still maintained by the large-entity original owner as of the 8th-year fee paid 2025-04-22. There is no NPE signal on any of the eight tests — no shell transfer, no asserter, no cascading LLC chain, no pre-suit transfer, no bankruptcy sale, no privateering, and no defensive-aggregator destination — so this reads as a normal captive operating-company patent held for defensive/product purposes, not a monetization asset. I stop at "Insufficient data" rather than "Operating-company assertion" solely because no assertion (suit or demand) was identified, which that verdict category requires.

To upgrade confidence, verify directly:

1. Reel 036615/0178 cover sheet — capture the correspondent of record (the one field I could not retrieve). → https://assignmentcenter.uspto.gov/
2. Assignment search for patent 9,803,979 and application 14/860,443 to confirm no later reel/frame exists. → https://assignment.uspto.gov/patent/index.html
3. PTAB (IPR/PGR) and PACER/CourtListener searches for "Honeywell" + this patent to confirm the no-assertion premise. → https://ptacts.uspto.gov/ptacts

Consistency check vs. prior sections: This analysis is consistent with the earlier litigation review (no case identified) and the claim-wording caveat (2019 certificate of correction; "evanescent"/"modulation" in claims vs. "attenuation" in abstract). No contradiction found. One correction to emphasize, carried forward: the certificate of correction (2019-05-28) is a claims/typo event, not an assignment — do not mistake the "CC" entry for a change of ownership.

I'll verify key prior-art references against live sources, then build the §102 analysis on the authoritative citation list from the patent.

Let me verify a few more of the strongest §102 candidates.

Let me verify the remaining high-relevance references.

Prior-Art Analysis — US 9,803,979 B2

Scope and method

I worked from the authoritative full text supplied (Google Patents "Citations (44)/Patent Citations (45)" lists, the Non-Patent Citations list, and the granted/corrected claim set). I then ran targeted live searches to verify the highest-relevance references; three additional verification queries on US 4,420,259, US 2015/0020590 and US 9,069,004 were cut off by a tool limit, so for those three I rely on the patent's own citation data plus your training-level knowledge and flag confidence accordingly.

Identifier discipline: I analyzed only the references literally cited on the face of US 9,803,979. Consistent with the earlier "no auto-correct" rule and the identifier-collision warning in the prior section, I did not import same-looking foreign numbers (e.g., SE 9803979, CZ 9803979, BR PI 9803979 — none appear as citations here anyway).

---

Governing law and critical date

• Effective filing / critical date: priority to provisional US 62/185,256 filed 2015-06-26; non-provisional US 14/860,443 filed 2015-09-21. The application is post-AIA (filed after 2013-03-16), so AIA 35 U.S.C. §102 governs.
• Two §102 pathways matter here:
  • §102(a)(1) — references published/patented before 2015-06-26.
  • §102(a)(2) — U.S. patents/applications effectively filed before 2015-06-26 but published later (e.g., US 9,395,177; US 2016/0320180; US 2016/0349283).
• Claim wording note (from the prior section): the granted/corrected claims recite an "integrated waveguide evanescent optical-pickoff" and detection of "modulation" of the light (claim 1 and claim 10), whereas the abstract/spec use "attenuation." My anticipation analysis is against the claim language, i.e., an evanescent pickoff.

Bottom line up front: No single cited reference appears to disclose every element of independent claim 1 or claim 10. The cited art is overwhelmingly §103 (obviousness) art or background. The closest references are the evanescent-coupling MEMS sensor family — WO 2015/088738 A1 (Intel), US 2014/0363119 A1 (Freescale), US 2015/0020590 A1 (Caltech), US 9,069,004 B2 / US 2014/0283601 A1 (Cornell), EP 1 083 429 A2 (Nerak), and US 2006/0103851 A1 (Ramot) — none of which shows the claimed substrate-monolithic evanescent pickoff whose coupling port approaches the substrate surface and is positioned to detect an edge of an in-plane-moving proof mass.

---

Tier 1 — Closest references (highest §102/§103 relevance)

1. WO 2015/088738 A1 — "Opto-mechanical inertial sensor" (Intel Corporation)

• Filing/priority: PCT filed 2014-11-20 (PCT/US2014/066702); priority 2013-12-13; published 2015-06-18. §102(a)(1) art (published before 2015-06-26).
• Disclosure (verified): A MEMS inertial sensing device with two evanescently coupled waveguides (102/104). At least a section (106) of one waveguide is movable ("shearing") substantially parallel to the fixed waveguide; the gap provides evanescent coupling, and relative shear changes transmitted intensity/phase read at detectors. Configured as accelerometer or gyroscope. Sibling US 9,285,391 B2 (Mach-Zehnder arm version).
• §102 analysis: Discloses evanescent coupling modulated by MEMS relative motion and optical detection of the resulting signal — the closest prior art to the evanescent coupling + in-plane relative motion concept. However, the moving element is a waveguide arm, not a silicon proof mass moving past a fixed pickoff port; the coupling is waveguide-to-waveguide, not a coupling port that approaches the substrate surface adjacent the moving component; and there is no edge-overlap detection. Does not anticipate claims 1 or 10 (missing at least the "monolithically fabricated within a first substrate … coupling port compris[ing] a portion of the … waveguide that approaches a surface … adjacent to the moving sensor component" and "positioned to detect movement of an edge" limitations). Strong §103 art.

2. US 2014/0363119 A1 — "Integration of a MEMS Beam with Optical Waveguide and Deflection in Two Dimensions" (Freescale/NXP)

• Filing/publication: filed 2013-06-10; published 2014-12-11; issued as US 9,435,952 B2. §102(a)(1) art.
• Disclosure (verified): A deflectable MEMS optical beam waveguide formed in an IC (optical beam structure encapsulated by a waveguide beam structure), extending into a deflection cavity, with deflection electrodes providing two-dimensional deflection control — used for die-to-die optical communication, not sensing.
• §102 analysis: Discloses a MEMS movable waveguide with integrated waveguiding structure — structural overlap with the "moving component / integrated waveguide" idea. But there is no proof mass, no inertial sensing, no edge-overlap coupling port, no evanescent pickoff. Does not anticipate. §103 art for "MEMS beam + integrated optical waveguide."

3. US 2015/0020590 A1 — "Optomechanical accelerometer" (California Institute of Technology)

• Filing/publication: filed 2012-03-02; published 2015-01-22. §102(a)(1) art.
• Disclosure: An accelerometer in which a mechanical (proof-mass) displacement is transduced optically via near-field/evanescent coupling between a mechanical resonator and an optical element/cavity.
• §102 analysis: Relevant accelerometer with optical, near-field readout. It reads displacement by near-field coupling, but the architecture is cavity/resonator near-field coupling, not a substrate-monolithic waveguide pickoff with input/coupling/output ports and a coupling port approaching the substrate surface to detect an edge overlap of an in-plane-moving mass. Does not anticipate claims 1/10. §103 art.

4. US 9,069,004 B2 / US 2014/0283601 A1 — "Optomechanical sensors based on coupling between two optical cavities" (Cornell University)

• Filing/publication: priority 2011-10-08; US 2014/0283601 A1 published 2014-09-25; US 9,069,004 B2 issued 2015-06-30. §102(a)(1) as to the 2014 publication.
• Disclosure: Optomechanical sensor in which a mechanical element changes coupling between two optical cavities, producing an optically readable displacement signal.
• §102 analysis: Optomechanical readout of a MEMS element — relevant background. But cavity-to-cavity coupling, not an evanescent waveguide pickoff port over a proof-mass edge. Does not anticipate.

5. EP 1 083 429 A2 — "Fiber optic micromechanical sensor with modulating shutter" (Nerak SA; inventor Guldimann)

• Filing/publication: filed 2000-08-01 (EP 00306552.1); priority 1999-09-10; published 2001-03-14 (A3 2002-01-23). §102(a)(1) art.
• Disclosure (verified): Micromechanical (silicon) sensor with a movable mass carrying a shutter at its free end; the shutter is positioned relative to an optical fiber in a groove and blocks/reflects/attenuates light as the mass displaces; output intensity gives acceleration/force/displacement (half-blocked at quiescence).
• §102 analysis: This is the clearest "moving edge modulates a light beam; intensity indicates displacement" reference. It discloses (i) a MEMS inertial sensor, (ii) an edge/shutter of the moving mass positioned relative to a fixed light conduit, and (iii) intensity-based measurement. It is missing the monolithically integrated evanescent waveguide in the substrate and the coupling port that approaches the substrate surface (it uses free-space fiber-to-fiber coupling, not evanescent coupling). Does not anticipate claims 1/10 (fails the "integrated waveguide evanescent optical-pickoff monolithically fabricated within a first substrate" limitations). Notably strong §103 art, especially combined with a waveguide-integration reference.

6. US 2006/0103851 A1 — "Displacement sensor based on photonic crystal waveguides" (Ramot at Tel Aviv University)

• Filing/publication: priority 2004-11-17 (prov. 60/628,113); published 2006-05-18; issued as US 7,466,424. §102(a)(1) art.
• Disclosure (verified): Displacement sensor with at least two co-planar photonic-crystal waveguide (PCWG) segments separated by a gap, one segment movable; relative in-plane displacement changes the transmitted/coupled light, detected optically; includes accelerometer embodiments.
• §102 analysis: Discloses in-plane displacement sensed optically via waveguide-coupling change across a gap — conceptually near the inventive idea. But it is PCWG-to-PCWG coupling between two waveguide segments, with no monolithic glass-substrate evanescent pickoff over a silicon proof-mass edge, no input/coupling/output port structure, and no "port approaches the substrate surface." Does not anticipate, but is meaningful §103 art for the "in-plane waveguide-coupling displacement sensing" concept.

---

Full citation-by-citation table (all 45)

Legend: [A] = appears to anticipate an independent claim (none found); [C] = closest art / meaningful §103; [B] = background/§103-support only; [N] = not relevant to the pickoff claims. "Relevant claims" = claims the reference most bears on.

---

Non-patent citations (18) — brief §102/§103 note

Several NPL entries are arguably more on-point than the patents, and they were examiner-cited:

• Burcham et al., "Micromachined Silicon Cantilever Beam Accelerometer Incorporating an Integrated Optical Waveguide," Proc. SPIE 1793 (1992) — cantilever accelerometer with integrated optical waveguide readout. [C] Strong §103 art for "integrated optical waveguide measuring a MEMS mechanical element."
• Culshaw, "Fibre optic sensor: integration with micromachined devices," Sensors & Actuators A 47 (1995) — fiber-optic + micromachined integration. [B]
• Dong et al., "An On-Chip Opto-Mechanical Accelerometer," IEEE MEMS 2013 (Jan. 2013) — on-chip optomechanical accelerometer. [C] Relevant §103 art (on-chip optical MEMS acceleration sensing).
• Guldimann et al., "Fiber-Optic Accelerometer With Micro-Optical Shutter Modulation and Integrated Damping," IEEE/LEOS Optical MEMS 2000, pp. 141–142 — the paper counterpart of EP 1 083 429; shutter-modulated fiber-optic accelerometer. [C] Strong §103 art (edge/shutter modulates intensity; notes horizontal detection possible).
• Kalenik et al., "A cantilever optical-fiber accelerometer," Sensors & Actuators A 68 (1998) — cantilever + fiber-optic accelerometer. [B/C]
• Rogers, "Evanescent Wave Coupling Using Different Subwavelength Gratings for a MEMS Accelerometer," Univ. South Florida (2011) — evanescent coupling to a MEMS accelerometer element. [C] Notably on-point for evanescent coupling + MEMS accelerometer; still not a substrate-monolithic pickoff with a coupling port over a proof-mass edge.
• Lodden et al., "System and Methods for Highly Integrated Optical Readout MEMS Sensors," U.S. Appl. No. 14/721,914 (filed 2015-05-26) — the related '914 application (applicant's own; issued as US 10,024,656 B2). [B] Same-family; directed to vertical/gap sensing. Not prior art to the same inventive entity in the relevant sense.
• Fraunhofer HHI, "Femtosecond Laser Processing of Miniaturized Optical Systems and Sensors in Glass Substrates" (2015) — supports the 3-D femtosecond laser waveguide-writing-in-glass fabrication technique recited in the spec. [B] Enablement/§103 background for "monolithically fabricated within a glass substrate."
• Osellame et al. (2009); Sauter et al., "Making Optical MEMS Sensors more compact using Organic Light Sources and Detectors," IEEE ETFA 2014; Jiang et al., "Optical actuation of silicon cantilevers" (2013) — fabrication/actuation background. [N/B]
• EPO ESRs (EP 16166097.2 and EP 16174671.4), and the USPTO Restriction Requirement / Office Action / Final Office Action / Notice of Allowance for 14/721,914 — prosecution documents, not prior art; [N] (useful only to understand how the related case was examined).

---

Consolidated conclusion

• No cited reference — patent or NPL — appears to anticipate (i.e., disclose every element of) independent claim 1 or claim 10 as granted/corrected. In particular, none of them shows all of: (1) a waveguide evanescent optical-pickoff monolithically fabricated within a glass substrate; (2) a coupling port comprising a portion of that waveguide that approaches the substrate surface adjacent the moving part; (3) detection of a change in area of overlap between the port and the moving part via the modulation at an output port; and (4) the moving part moving in-plane, with the port positioned to detect movement of an edge of it.
• The most relevant references are, in order: WO 2015/088738 A1 (Intel), EP 1 083 429 A2 (Nerak), the Guldimann 2000 and Rogers 2011 NPL items, US 2006/0103851 A1 (Ramot), US 2015/0020590 A1 (Caltech), US 9,069,004 B2 / US 2014/0283601 A1 (Cornell) and US 2014/0363119 A1 (Freescale). These bracket the concept (evanescent coupling readout of MEMS motion; in-plane displacement via coupling changes; edge/shutter intensity modulation) but each lacks at least the monolithic-substrate-evanescent-pickoff-with-surface-approaching-coupling-port / edge-overlap combination.
• Dependent claims (2–9, 11–19) add functionality already common in the art in isolation (driving a proof mass into in-plane vibration; measuring oscillation timing/frequency; deriving displacement from duty cycle; MEMS gyroscope/accelerometer proof masses; LED/laser sources; interposer-mounted sources). These would be evaluated individually for §103 in view of the above art, but none is anticipated by a single reference on the current record.

Confidence and caveats

• High confidence on the bibliographic dates and descriptions that I verified by live search (Intel WO 2015/088738; Freescale US 2014/0363119; EP 1 083 429; US 2006/0103851; US 4,567,771; plus the Google Patents face-of-patent citation list).
• Medium confidence on US 4,420,259 A, US 2015/0020590 A1, and US 9,069,004 B2 — my additional verification searches were cut off by a tool limit, so those descriptions rest on the patent's citation data and general knowledge. Flagged accordingly.
• The certificate of correction (2019-05-28) changed "optical-pickoff" → "evanescent optical-pickoff" and "attenuation" → "modulation" in the claims; my §102 mapping is against the corrected claim language. If you need to rely on this for an invalidity/validity position, confirm the corrected claim set at USPTO PatentCenter and run the above Tier-1 references against those exact words.

I'll ground the obviousness analysis by verifying the key prior-art references that appear on this patent's face (the "Citations (45)" and "Non-Patent Citations (18)" lists).

Obviousness Analysis — U.S. Patent No. 9,803,979 B2 under 35 U.S.C. § 103

Builds on: the verified bibliographic data, claim overview, and litigation/correction findings already generated. I do not repeat them. Two prior findings are load-bearing here and I re-flag them below (Claim–spec wording mismatch; prior-art dating).

Nature of this analysis: This is an examiner-style / invalidity-defense-style opinion grounded in the references that appear on the face of the patent (the "Citations (45)" and "Non-Patent Citations (18)" lists). It is an analytical prediction, not a legal conclusion, and it depends heavily on claim construction.

---

0. Two threshold flags that change the § 103 analysis

(a) The corrective-amendment wording mismatch is dispositive to framing.
As previously flagged, the issued claims recite an "integrated waveguide evanescent optical-pickoff" and measuring "modulation" of the light, while the abstract/specification examples say "attenuation." This matters for § 103 because:

• If "modulation" is given its ordinary meaning (any change in transmitted optical intensity — up or down), the shutter-type references (EP 1 083 429, Guldimann, Kalenik) and the evanescent references (Rogers, Culshaw, Burcham) all read directly on the detecting step.
• If the claims are construed narrowly to require evanescent attenuation, the aperture/shutter references (which block/transmit light in free space) are weaker on the "evanescent" limitation but still strong on "modulation," and the evanescent references (Rogers, Culshaw, Burcham) carry the "evanescent" limitation.
• Either way, a § 103 combination is available — but the primary reference changes with the construction. I address both constructions.

(b) Prior-art dating under the AIA (effective filing date = 2015-06-26, the '256 provisional).
The '979 is an AIA patent. Therefore, only references publicly available before 2015-06-26, or US application publications "effectively filed" before that date, qualify. Applying that filter to the patent's own citation list:

Consequence: the combinable universe is wide and, critically, every core element (evanescent MEMS readout; in-plane edge-overlap modulation; MEMS + integrated waveguide; glass-substrate femtosecond-written waveguide) was independently in the public domain years before 2015-06-26. That is a classic KSR fact pattern.

---

1. Level of ordinary skill / claim construction posture

POSITA: a person with a graduate degree (or equivalent experience) in MEMS/optoelectronics with at least ~2–3 years designing MEMS inertial sensors and/or integrated-optics readouts. The patent itself demonstrates this level (its background presumes familiarity with capacitive pickoff error mechanisms and evanescent coupling physics).

Key terms and their effect on scope:

---

2. Element-by-element mapping of the prior art to the independent claims

Reading of the table: no single reference teaches every limitation. But each independent claim is a small number of known elements arranged in a combination whose parts were individually known and whose interaction (overlap → intensity modulation) was expressly known. That is the § 103 posture.

---

3. Obviousness grounds (combinations and motivations)

I present the strongest grounds first.

Ground 1 — EP 1 083 429 (Nerak/Guldimann) ⊕ Rogers 2011 ⊕ Fraunhofer/Osellame

Covers claims 1–9 (and, applied to claim 10, the apparatus).

• EP 1 083 429 teaches a MEMS sensor in which a movable mass carries a shutter whose edge is displaced horizontally, in-plane, across the end faces of input/output optical fibers, blocking/transmitting light so the detected intensity is a function of the mass displacement (see abstract: "horizontally displaced such that the shutter … is capable of influencing the light … used for measuring parameters such as acceleration, force, or displacement"). This supplies the in-plane edge-overlap → intensity-modulation core (claim elements (c), (d), (e)).

• Rogers 2011 teaches the evanescent version of the same measurement: evanescent coupling between a glass and a silicon structure whose relative position changes the output intensity read by a photodiode — i.e., the "integrated waveguide evanescent optical-pickoff."

  "As the platform is displaced, the grating separation between the two gratings changes and causes the output intensity to change. Using the coupled subwavelength grating simulations as a reference to the output intensity change with respect to gap, the mechanical and coupling sensitivity properties … is presented." — Rogers abstract (verified, USF Digital Commons / ADS).

• Fraunhofer HHI 2015 / Osellame 2009 teach that such waveguides are made monolithically inside a glass substrate by femtosecond laser writing — supplying "monolithically fabricated within a first substrate" and claim 10's "first glass substrate."

• Motivation to combine: The '979's own background frames the problem: free-space/fiber-coupled optical readouts are "not rugged enough to handle the harsh environmental factors," and capacitive readouts suffer electrostatic errors; the patent asserts that monolithic integration is what makes the optical approach stable. A POSITA seeking a rugged, alignment-tolerant optical MEMS readout would therefore (i) replace EP 1 083 429's free-space fiber/shutter gap with (ii) an evanescent coupler written into the glass per Rogers/Fraunhofer, eliminating the alignment-sensitive air gap. This is exactly KSR's "known technique to improve similar devices in the same way." Reasonable expectation of success is supported because Rogers empirically demonstrates intensity-vs-position evanescent readout and Burcham/Culshaw (below) show integration of waveguides with micromachined silicon a decade earlier.

• Dependent claims 2–5, 6: in-plane drive (the gyroscope/accelerometer drive mode), measuring the frequency/period of the oscillating output, and deriving displacement from duty cycle are the ordinary signal-processing use of a modulated optical output — Guldimann/Kalenik's fiber-optic accelerometers already output a time-varying intensity from which acceleration/displacement is derived. Claims 8–9 (LED/laser): Rogers uses a 1.55 µm semiconductor laser; LEDs were a routine design substitute.

Ground 2 — Freescale US 2014/0363119 ⊕ Rogers 2011 ⊕ Fraunhofer/Osellame

Covers claims 1 and 10 (and dependents).

• Freescale discloses "a deflectable MEMS optical beam waveguide … formed with an optical beam structure … encapsulated by a waveguide beam structure … surrounded by a plurality of deflection electrodes … to provide two-dimensional deflection control." That is a moving MEMS structure integrated with an optical waveguide and deliberately moved in-plane (lateral) and out-of-plane (vertical) — supplying the substrate-integrated-waveguide and in-plane-MEMS-motion hooks (elements (a), (d), and part of (f)).
• Rogers supplies the evanescent coupling mechanism and the "overlap/position → optical intensity" detection.
• Fraunhofer/Osellame supplies the glass monolithic fabrication.
• Motivation to combine: Freescale's own field is "optical MEMS circuits and devices," and its stated purpose is alignment feedback measured from output optical intensity. Repurposing the same MEMS-beam + integrated-waveguide architecture to sense inertial displacement (rather than to communicate/align) is a use of a known device for its natural, analogous function — the classic KSR rationale ("if a technique has been used to improve one device, and a person of ordinary skill … would recognize that it would improve similar devices in the same way, using the technique is obvious"). Freescale even contemplates "sensor communication systems, automotive sensor and control systems," signaling the MEMS-sensor adjacency.

Ground 3 — Intel WO 2015/088738 (or its US family US 9,285,391 / US 2015/0168442) ⊕ EP 1 083 429

Covers claims 1 and 10; also supports claims 6, 14–16 (inertial proof mass) and 9/19 (laser).

• Intel discloses a MEMS inertial sensor with a laser, waveguide(s), a movable proof-mass waveguide, and a detector, where "the movement of the first waveguide relative to the second waveguide … results in a change in light intensity of the … light beam," and the detected intensity change yields the inertial measure. This supplies elements (a)–(c) and the inertial-proof-mass identity (claims 6, 14–16) with a laser source (claims 9/19).
• Intel's coupling is endface shearing / Mach-Zehnder, not "edge overlap of a coupling port." EP 1 083 429 supplies precisely the missing edge-overlap, in-plane shutter mechanism and the "area of overlap" concept.
• Motivation to combine: Both references solve the same problem (optical MEMS inertial readout that avoids electrostatic-capacitive limitations), and both detect inertial motion from a change in transmitted optical intensity. Substituting EP 1 083 429's overlap-vs-blocked-area mechanism for Intel's shearing-coupling mechanism is an obvious alternative that a POSITA would try to trade off sensitivity vs. dynamic range. Note the date: WO 2015/088738 published 2015-06-18, eight days before the '979 priority date — it is § 102(a)(1) art and, being so close in time, is powerful evidence of contemporaneous knowledge.

Ground 4 — Caltech US 2015/0020590 ⊕ EP 1 083 429 / Kalenik

Alternative primary for claims 1 and 10.

• Caltech discloses an optomechanical accelerometer: a proof-mass-deformed waveguide arm whose deformation changes optical path length and thus output intensity detected at a photodetector, from which acceleration is determined. This supplies the waveguide-plus-proof-mass-plus-intensity-detection architecture and the accelerometer proof mass (claims 6, 16).
• EP 1 083 429 / Kalenik supply the in-plane edge-overlap modulation concept.
• Motivation: same field, same problem, same measured quantity (optical intensity as a function of proof-mass displacement); combining a proven optical readout with a proven in-plane edge-modulation mechanism is a predictable aggregation.

---

4. Why the specific limitations the patent emphasizes are not savings

The '979's ostensible point of novelty is time-based measurement — deriving frequency and duty-cycle (time-low/time-high) ratio from a periodically modulated output (spec, col. discussion of FIGS. 2 and 5):

"the amplitude of the lateral or in-plane displacement of the proof mass … can be determined from the ratio of the time that the output intensity is low … to the time that the intensity is high."

A POSITA would find this obvious because:

1. The modulated waveform is inherent to the prior art. Any edge-overlap or shutter mechanism produces a periodic intensity waveform once the mass is driven/oscillating — EP 1 083 429, Guldimann (2000), and Kalenik (1998) all produce exactly this.
2. Extracting frequency and duty cycle is elementary signal processing within the ordinary skill of the art, and the claims recite only measuring "timing," "in-plane displacement," and "amount of overlap" from the oscillating output — results, not novel algorithms.
3. The evidentiary weight of the claims' "time-based" framing is low because the specification supplies no comparative data showing the time-based approach yields unexpected sensitivity over simply measuring the analog intensity (Rogers' approach). Alleged advantages recited in a specification, unsupported by data, do not rebut a prima facie case.

The FIG. 5 surface-feature embodiment (grooves / metal-line grid on the proof mass) is not separately claimed, but if it were, it is squarely taught by Rogers (subwavelength grating structures on glass and silicon, whose relative motion modulates coupling) and by US 2006/0103851 A1 (photonic-crystal displacement sensor) — both on the face of the patent.

---

5. Rebuttal considerations (what could defeat or blunt the prima facie case)

Stated evenhandedly, because they matter:

1. "Evanescent" + "edge overlap" as a combination. Intel/Freescale use endface/shearing or beam-deflection coupling, not a surface-adjacent "coupling port" overlapped by an in-plane edge. A patentee could argue the combination of evanescent surface coupling with in-plane edge overlap was non-obvious. Counter: Rogers and Culshaw/Burcham place evanescent coupling adjacent to a moving micromachined element, and EP 1 083 429 puts the moving element's edge in the optical path — the motivation (ruggedness/alignment, expressly in the '979 background) unites them.
2. Claim construction of "monolithically fabricated within a first substrate." If construed to require femtosecond-laser-written waveguides in glass, Fraunhofer/Osellame makes it routine. If instead construed to require the pickoff be integrated with the sensor's own glass substrate (vs. an attached substrate), the argument narrows to a design choice.
3. Objective indicia (secondary considerations). None is documented on the record I can verify. As previously reported, no litigation is known for the '979, and the earlier analysis noted Honeywell is a large operating company, not a litigating NPE. There is no known evidence of unexpected results, long-felt need, copying, or licensing attributable to these claims. Absent such evidence, the KSR/Deere factors favor obviousness on these facts. (This is an absence-of-evidence statement, not proof.)
4. The "modulation" vs. "attenuation" ambiguity could cut either way and should be resolved before any final obviousness conclusion — but neither construction escapes Grounds 1–3.

---

6. Conclusion and confidence

Overall: On the references on the face of US 9,803,979, each independent claim is a combination of known elements (evanescent MEMS coupling; in-plane edge/shutter overlap modulation; substrate-integrated waveguide; glass femtosecond-written waveguide) performing their known functions, with a clear articulated motivation rooted in the patent's own stated problem (ruggedness and alignment). Under KSR, this is a strong prima facie § 103 case.

Confidence: Moderate-to-high that claim 1 (under the broad "modulation" construction) and claim 10 are obvious over Ground 1 or Ground 2. Moderate under a narrow "evanescent-attenuation + in-plane-edge-overlap" construction, where Ground 1 (Rogers + EP 1 083 429) remains the best vehicle.

Confidence limiters (explicit):

• I did not retrieve the certificate of correction itself, so the operative claim text — and thus the exact limitation set mapped above — should be confirmed at USPTO PatentCenter before relying on this.
• I cannot certify the full prosecution history or applicant's actual/examiner's arguments; the file wrapper could contain statements (e.g., a narrowing construction of "evanescent" or "coupling port") that alter the mapping.
• Prior-art status of the Fraunhofer citation and of US 2016/0349283 (Bramhavar) needs a date check (102(a)(1) vs. 102(a)(2)) and the § 102(b)(2) exceptions analysis for Bramhavar.
• If you want, I can attempt targeted lookups for (i) the certificate-of-correction text and (ii) the '979 file-wrapper rejections, to firm up which grounds the examiner actually relied on.