US 10371570

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

Let me search more specifically for litigation and CAFC docket activity involving this patent.

Let me do a final targeted check for any PTAB/IPR or district court activity on this specific patent.

US Patent 10,371,570 ("the '570 patent")

Important identifier note: I interpreted "10371570" literally as US 10,371,570 B2. One caution: during searching I encountered an unrelated identifier — application/serial No. 10/371,570, which belongs to US 7,194,879 ("Security container with linked primary and secondary security features," filed Feb. 21, 2003). That is a different number (an application serial, not the granted patent you asked about) and I have not merged the two. Everything below relates to US 10,371,570 B2 only.

Bibliographic data

Abstract (verbatim)

"A method and apparatus for protecting an optical sensor is disclosed. A fixed filter having a fixed passband for light transmission is placed in front of the optical sensor. A programmable filter having a variable passband for light transmission is placed in front of the fixed filter. A controllable voltage source controls a voltage at the programmable filter that shifts the passband of the programmable filter from a first state in which the passband of the programmable filter is substantially the same as the passband of the fixed filter and a second state in which the passband of the programmable filter is different than the passband of the fixed filter."

Plain-language overview

The patent addresses the problem of optical sensors being damaged ("burnout") by intense light (sunlight, laser jammers/dazzlers, glint). Because some sensors refresh at up to 10,000 frames/sec, protection must act in under ~1 microsecond. The solution is a fast, tunable solid-state "shutter" built from two stacked bandpass filters:

• A fixed filter whose passband is centered on the wavelength of interest (exemplary 1550 nm, ~2 nm bandwidth).
• A programmable (tunable) filter electrically biased to shift its passband by up to ~10 nm (Δλ). With no bias, its passband coincides with the fixed filter's (shutter ON, light passes). With bias, the passbands are shifted apart so they no longer overlap (shutter OFF, light blocked).

The tunable filter is realized with semiconductor interference structures whose refractive index changes with applied bias via carrier-depletion effects (e.g., alternating n/p-type quarter-wavelength stacks such as GaAs/AlGaAs/InGaAs, or a photonic crystal of n-type bars in a p-type matrix). Because the device is basically a capacitor, needing only a small voltage/charge change to switch means fast, low-loss operation for narrowband sources. It can also be operated as a variable attenuator by tuning only partly off-passband. Several control schemes are described: (FIG. 9) a separate wide-area control sensor; (FIG. 10) a beam-splitter tapping a portion of light to a less-sensitive control detector; and (FIG. 11) feedback from the main sensor itself, with periodic re-alignment to test whether the light level has dropped.

Independent claims (plain language)

Claim 1 — apparatus. An apparatus for protecting an optical sensor comprising:

1. a fixed filter with a fixed passband;
2. a programmable filter with a variable passband, which includes at least two quarter-wavelength layers — the first having alternating n-type and p-type layers of a first material, and the second having alternating n-type and p-type layers of a second material; and
3. a controllable voltage source that shifts the programmable filter's passband between (a) a first state substantially matching the fixed filter's passband and (b) a second state different from the fixed filter's passband.

(Note: the "at least two quarter-wavelength layers with n/p alternating layers of different materials" limitation was added during prosecution and is what distinguishes the granted claim 1 from the broader disclosure in the 2018 pre-grant publication, whose claim 1 recited only the fixed filter, programmable filter, and voltage source.)

Claim 11 — method. A method of protecting an optical sensor comprising:

1. placing a fixed filter with a fixed passband in front of the optical sensor;
2. placing a programmable filter with a variable passband in front of the fixed filter, the programmable filter likewise including at least two quarter-wavelength layers — first with alternating n/p layers of a first material, second with alternating n/p layers of a second material; and
3. controlling a voltage at the programmable filter to shift its variable passband relative to the fixed passband.

The remaining 18 claims are dependent: e.g., control-device and threshold-intensity features (claims 3–7, 13–16), beam splitter (6, 15), bilayer of n- and p-type semiconductor (8, 18), half-wavelength central layer with stacked quarter-wave bilayers (9, 19), photonic crystal (10, 20), and fixed central wavelength ≈1.55 µm (17).

Litigation / CAFC 2026 check — not found

I searched for district-court, PTAB/IPR, and Federal Circuit activity tied specifically to US 10,371,570 / application 15/420,438 and found no authoritative record of any such proceeding, and specifically none on the CAFC 2026 dockets. What the searches did surface was unrelated Raytheon litigation (e.g., Raytheon v. Cray, GE v. Raytheon / Raytheon Technologies v. General Electric, WirelessWerx IP v. Raytheon, Aperture Net v. Raytheon) — none of which involves this optical-sensor patent. I could not verify the negative with a primary database (PACER/Docket Navigator/RPX) in this session.

Confidence and caveats

• High confidence in bibliographic data, abstract, and claim text — these come from the authoritative Google Patents full text of US10371570B2 (https://patents.google.com/patent/US10371570/en) and corroborating sources (Justia patent page, JP 6902103 B2 family page).
• Medium/low confidence on legal-status details (e.g., the exact 2037-08-02 adjusted expiration), which Google Patents labels as an assumption, not a legal conclusion.
• No findings on litigation: treat as "nothing found in my searches," not as proof that none exists. To be conclusive, check PACER, the CAFC docket, PTAB (IPR/PGR), and commercial litigation databases directly.

Sources: https://patents.google.com/patent/US10371570/en · https://patents.justia.com/patent/10371570 · https://patents.google.com/patent/JP6902103B2/en

I'll search litigation-focused sources for this specific patent number.

The initial searches surfaced nothing relevant. Let me target patent-litigation aggregators and docket databases directly.

Litigation search for US 10,371,570 — result: no litigation found

I searched litigation-oriented sources (Unified Patents portal references, Justia patent/docket pages, CourtListener, Dockets.Justia, PTAB/PTABCTS, and general web queries) for the specific number 10,371,570 / application 15/420,438 / title "Burnout protection for optical sensors" (Raytheon Company). I found no record of any district court, PTAB/IPR, ITC, or Federal Circuit proceeding involving this patent.

What the searches actually returned (and why each is a false positive)

To be transparent about the near-misses, none of these involve US 10,371,570:

• Westlaw citation "2013 WL 10371570" — a New Mexico consumer-lending case (State of New Mexico ex rel. King). This is a Westlaw database accession number, not the patent number. Coincidental digit overlap only.
• US 6,249,570 ("the '570 patent") — asserted in NICE Systems v. Witness Systems (D. Del., ~2007). This is a different patent (a different '570, owned by NICE, audio-logging technology). Not related.
• Raytheon v. Cray / GE v. Raytheon Technologies / Raytheon Technologies v. General Electric — unrelated Raytheon patent disputes concerning computer hardware and gas-turbine engines. None asserts the '570 optical-sensor patent.
• Miscellaneous PTAB IPR text, drug-patent litigation, and unrelated NPE suits surfaced by keyword matching — none cite 10,371,570.

Interpretation and caveats

• Treat this as "nothing found in my searches," not as affirmative proof that no litigation exists. My session was limited and I could not query primary databases directly.
• The patent is still in force (4th-year maintenance fee paid Jan. 21, 2023; listed Active with an adjusted expiration of Aug. 2, 2037). It is a Raytheon defense/optics asset, and the large majority of Raytheon patents of this type are never asserted — zeros are expected for a portfolio patent like this.
• To make the negative conclusive, check directly: PACER, the CAFC docket, USPTO PTAB (IPR/PGR) via Patent Center/PTABCTS, and commercial dockets (Docket Navigator, RPX, Unified Patents Litigation portal at portal.unifiedpatents.com/litigation). Note the Unified Patents litigation caselist is not reliably indexed by general web search, so its absence from my results is weak evidence.

Date note (flagged, per instructions)

There is a date inconsistency between the current task framing (April 26, 2026) and my operating context (Sept. 30, 2026), and the prior generated section referenced a "CAFC 2026" check. I did not find CAFC activity on either date basis. This does not affect the "no litigation found" conclusion, but I flag it because it bears on how current any negative result can be.

Sources searched/consulted: https://patents.justia.com/patent/10371570 · https://portal.unifiedpatents.com/litigation/caselist (referenced) · https://www.docketalarm.com (PTAB) · storage.courtlistener.com · dockets.justia.com · general web search on "10371570," "10,371,570," and the patent title.

I'll verify the structured data against live sources before producing the report.

Let me run two more targeted checks to be thorough about any AIA trial or appeal.

Proceedings overview

Total AIA trial proceedings on file for US 10,371,570: 0. The structured USPTO Open Data Portal (ODP) block supplied with this analysis returns no IPR, PGR, or CBM proceedings, and independent web searches corroborated that — no proceedings to break down by active / invalidated / sustained / settled / denied, and therefore no institution denials, no final written decisions, no terminations, no settlement, and no Federal Circuit appeal to report. For a defendant, the bottom line is the mirror image of "the patent has been hardened": the patent has never been tested at the Board at all, so all 20 claims — including the two independent claims — remain presumptively valid, un-estopped, and available to the patent owner as assertion targets, while every prior-art ground remains equally available to you.

Proceedings

None. No proceeding numbers exist to report, and this section is intentionally not populated with a per-proceeding template because doing so would require inventing docket numbers — which the task instructions forbid. I am not aware of any AIA trial involving US 10,371,570, and I found no record of one.

What I checked, and what it returned:

Negative-result caveat: the ODP block is authoritative for what the Office has published, but it is an ingest snapshot. The absence of proceedings in my searches should be read as "nothing found," not as proof of non-existence. This is, however, a strong signal: this patent issued 2019-08-06 and has been active with a 4th-year maintenance fee paid 2023-01-21, so an IPR petition filed any time from 2019-08-06 forward would ordinarily be visible in ODP by 2026-09-30.

Strategic summary

Claim status: all 20 claims UNTESTED. No claim of US 10,371,570 has been canceled, narrowed, or sustained in any AIA trial. Independent claim 1 (apparatus) and independent claim 11 (method) stand exactly as granted, with dependent claims 2–10 and 12–20 untouched. There is no FWD to quote and no claim-level disposition to cite — if you are looking for a "claims 1–5 were canceled, so the demand letter is hollow" defense, it does not exist here.

Estoppel landscape: clean slate. Because no IPR was ever instituted, 35 U.S.C. § 315(e)(2) estoppel is inapplicable to every party. No defendant or petitioner is barred from raising any ground — § 102, § 103, or (in district court, since Samsung Elecs. Am., Inc. v. Prisua Eng'g Corp., 948 F.3d 1342 (Fed. Cir. 2020), forecloses § 112 in IPR) § 112 indefiniteness/written-description challenges. Two practical filters, however, will shape a petition:

• § 325(d) / Advanced Bionics + Becton, Dickinson: The face of the patent shows seven references and three non-patent citations already of record — notably US 3,699,347 (Variable electro-optic filter), US 4,724,311 (Variably transmissive filter), US 5,027,178 (Electrically tunable interference filters), US 6,785,032 (Laser countermeasure system and method), US 8,754,658 (Lockheed Martin, Variable transmission optical filter), and US 2014/0239157 (Bright source protection for low light imaging sensors), plus Bennett et al. (1990) and Mehdi et al. (1989). Any petition built on these will draw a discretionary-denial argument that the Office already considered the art. The winning petitions in this space will run on art the examiner never saw — and, given the crowded field of electrically tunable interference filters, that requires real prior-art scouting rather than re-packaging the cited references.
• § 315(b): a one-year clock runs from service of a complaint alleging infringement. No complaint involving this patent was found in the prior section of this analysis; if a demand letter (as opposed to a served complaint) is all you have, the bar has not started, and a pre-emptive IPR is a live option.

Pattern signals: no PTAB history on this patent, but Raytheon is a Board-tested patent owner. The Raytheon PTAB hits that surface in searches belong to other patents — e.g., the ’678 patent litigated in Raytheon Co. v. Sony Corp., Appeal No. 2017-1554 (Fed. Cir. 2018-04-02, nonprecedential), where the Board's obviousness findings on claims 5, 13–14, and 16–18 were affirmed even though the court found the Board's construction of "second substrate" erroneous (rendered moot by the affirmance). See https://cases.justia.com/federal/appellate-courts/cafc/17-1554/17-1554-2018-04-02.pdf and https://www.courtlistener.com/opinion/[4482890](/patent/4482890)/raytheon-company-v-sony-corporation/. Separately, General Electric petitioned against Raytheon Technologies in IPR2020-00405 (U.S. Pat. No. 10,066,548). The takeaway is not that IPRs threaten this patent, but that Raytheon is an experienced, well-resourced patent owner that defends at the Board and appeals adverse outcomes — do not expect a cheap default. There is no signal of a defensive aggregator (e.g., Unified Patents) ever touching this patent, and no serial petitioner.

An additional, patent-specific wrinkle worth pricing into any challenge: claim 1 was narrowed during prosecution to require "at least two quarter wavelength layers, the first quarter wavelength layer including alternating n-type layers and p-type layers of a first material and the second quarter wavelength layer including alternating n-type layers and p-type layers of a second material" — language absent from the 2018 pre-grant publication's claim 1. That amendment cuts both ways: it is a validity floor (the applicant conceded the broader genus), but it also creates prosecution-history-estoppel and written-description attack surface (the spec describes "quarter wavelength stacks" 302a–302d, and the claim's "first material"/"second material" distinction elides the specification's express requirement that the two materials differ in refractive index). Those are district-court arguments, not IPR arguments.

Recommended next steps

1. Treat "no PTAB activity" as the operative finding, and say it plainly in any client advice. There is no FWD, no cancellation, no estoppel, and no appellate disposition to leverage. Any assertion that claims of this patent have been invalidated at the Board would be false.
2. Verify the negative before relying on it. Before committing to a strategy premised on zero prior challenges, confirm directly: USPTO PTAB E2E / Patent Trial and Appeal Board End-to-End (https://ptab.uspto.gov), the patent's PTAB "Proceedings" tab on Google Patents, Docket Navigator, and PACER for any co-pending litigation that would fix the § 315(b) bar date. I was unable to reach those primary databases in this session, so my negative is search-based, not registry-based.
3. If you intend to challenge, file an IPR — and file on fresh art. PGR is time-barred (the 9-month post-grant window closed 2020-05-06), and CBM is unavailable (the patent is not a covered business method, and the CBM program sunset for new petitions on 2020-09-16). IPR under § 102/§ 103 is the only AIA vehicle. Build the petition around references not among the seven cited on the patent's face to blunt § 325(d).
4. Watch the § 315(b) clock. No complaint involving this patent was located; if that changes, the one-year bar date is service date + 12 months, and a defensive IPR (or a General Plastic-mindful single-petition strategy, since a second bite is hard to get) should be calendared immediately.
5. Nothing to monitor at the PTAB. There is no institution-decision deadline, no oral hearing, and no statutory 1-year FWD due date to track, because no trial exists. The only PTAB-side date to watch would be the filing of a first petition against this patent — at which point the 6-month institution deadline and the 12-month FWD deadline under § 316(a)(11) would begin to run.

If you want me to go further: tell me whether you are the patent owner or a target, and whether there is a served complaint (and its date). That determines whether the § 315(b) bar is live and whether a pre-emptive IPR is worth building.

I'll research the assignment record for US 10,371,570. Let me search authoritative sources.

Let me search for the specific assignment record and check Google Patents legal events.

Let me check the USPTO assignment database and Raytheon's corporate ownership status.

Assignment & Ownership Analysis — US 10,371,570 B2

Verification caveat (read first): I was unable to open assignmentcenter.uspto.gov / assignment.uspto.gov interactively in this session, so I could not read the raw reel/frame cover sheets directly. The assignment data below is drawn from the Google Patents "Legal Events" mirror of the USPTO assignment record embedded in the authoritative patent text you supplied, plus corroborating sources. I flag every place where I am relying on that mirror rather than a primary cover sheet. The single recorded assignment is fully consistent across both the Google Patents legal-events block and the bibliographic section, but you should confirm the correspondent field directly in Assignment Center.

Inventors

Departure-pattern check — NEGATIVE (good sign for the original assignee). Trent and Frazier did not leave Raytheon around this filing. They appear as co-inventors on a companion Raytheon application, US 15/617,163, filed June 8, 2017 — five months after this one — which issued as US 10,151,635 B1 ("Real time correction of optical window thermal gradients," issued Dec. 11, 2018) and lists a third Raytheon inventor, David J. Knapp. Continued co-filing, and the fact that the assignment to Raytheon was executed before the filing date (see below), are inconsistent with the "inventors depart and portfolio is fire-sold" precursor pattern. Sources: https://wiki.golden.com/wiki/US_Patent_10151635_Real_time_correction_of_optical_window_thermal_gradients-BYM9MJK ; https://www.patenthub.cn/cpc/patent-[166988](/patent/166988)-[US10151635B1](/patent/US10151635B1)-707eb210fe948e022699e5ac9bcafed5.html

Original assignee

Raytheon Company (Waltham, MA / Arlington, VA) — a large, publicly traded U.S. defense prime (NYSE: RTN at the time). Primary line of business: defense electronics, sensors, radars, missile systems, and space/airborne systems. Product-embodiment: Raytheon's Space & Airborne Systems and Integrated Defense Systems segments build exactly the class of product this patent implicates — laser/EO sensor systems requiring "burnout" protection against high-intensity illumination (the specification's own examples are laser jammers, glint, sun). This is an operating-company R&D asset, not a licensing vehicle.

Current status of the original assignee: Raytheon Company no longer exists as an independent public entity. It merged with United Technologies Corporation, effective April 3, 2020, to form Raytheon Technologies Corporation, which was renamed RTX Corporation effective July 2023. Raytheon Company is a wholly owned subsidiary name still used commercially (e.g., Raytheon, an RTX business). No new assignment was recorded against this patent to reflect the merger — which is normal and is not evidence of a transfer. Fla. Sunbiz record for "RAYTHEON COMPANY" shows status ACTIVE with a 2026 principal-address update: https://search.sunbiz.org/Inquiry/CorporationSearch/SearchResultDetail?...searchTerm=Marina%20Nikolic

Assignment timeline

2017-01-16 (executed, effective date) / recorded 2017-01-31 — Reel 041133 / Frame 0946

• Conveyance: Assignment (ASSIGNMENT OF ASSIGNORS INTEREST)
• Assignor: Catherine Trent; Gary A. Frazier (individually)
• Assignee: Raytheon Company (Massachusetts)
• Correspondent: Not determinable from the sources available to me — the Google Patents legal-events mirror reproduces the reel/frame, assignors, assignee, and effective date but not the attorney/agent of record. I will not guess this field. (Raytheon assignments of this era were commonly recorded by Raytheon in-house IP counsel or outside firms, but naming a specific correspondent without the cover sheet would be fabrication.)
• Context: Ordinary employment/employer assignment — the inventors conveyed their rights to their employer at filing. Note the effective date (Jan 16, 2017) precedes the application filing date (Jan 31, 2017), i.e., assigned at/around filing, the hallmark of a routine employee invention assignment, not a post-hoc transfer.

No further recorded assignments. The only other post-filing events in the legal record are administrative/annuity events, not conveyances:

• 2019-03-25 — Notice of Allowance mailed
• 2019-07-17 — Patent grant
• 2023-01-21 — 4th-year maintenance fee paid (large entity)

There is no security agreement, merger recording, license, name-change, or release recorded against this patent. That is itself a finding: the patent sits on the original operating-company's books (now under RTX) and has never entered an NPE chain.

Timeline diagram

timeline
    title Ownership of US 10371570
    2017 : Inventors assign to Raytheon Company
         : Assignment recorded reel 041133 frame 0946
         : Application 15 420 438 filed Jan 31
    2019 : Patent granted Aug 6
    2020 : Raytheon merges into Raytheon Technologies
    2023 : Parent renamed RTX Corporation

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. No transfer to any IP/Licensing/Holdings/Ventures LLC appears on this patent. The only assignee of record is Raytheon Company (reel 041133/0946), an operating defense prime at a real corporate address — no registered-agent-service address, no single-purpose entity.

2. Known asserter in the chain — NOT PRESENT. The chain (Trent/Frazier → Raytheon Company → no further recording), as reflected in the Google Patents legal events, contains no assignee matching any public NPE list (Acacia, Marathon, IV, Wi-LAN/Conversant, Pendrell, Round Rock, etc.). I searched and found no Unified Patents / RPX asserter directory entry tying this patent to such an entity.

3. Repeat correspondent across the chain — NOT PRESENT / UNVERIFIABLE. With only one recorded assignment, there is no chain across which a correspondent could recur. I could not read the correspondent field from the cover sheet, so I make no finding either way.

4. Cascading transfers — NOT PRESENT. There is exactly one conveyance (2017), with no subsequent transfers within 24 months or ever. No chained-LLC pattern exists.

5. Pre-litigation transfer — NOT PRESENT. No infringement suit naming this patent was found in my earlier analysis, so there is no pre-suit assignment to date. The sole assignment predates issuance by more than two years.

6. Bankruptcy fire-sale — NOT PRESENT. No Chapter 7/11 sale or court-supervised transfer is recorded; Raytheon/RTX is a solvent going concern.

7. Privateering — NOT PRESENT. No transfer from Raytheon to any asserting entity has been recorded, so there is no evidence the portfolio was pushed to a proxy to sue competitors. (Raytheon is regularly a defendant in unrelated suits — e.g., Raytheon v. Cray, WirelessWerx IP v. Raytheon — but that is others suing Raytheon, not Raytheon privateering this patent.)

8. Defensive aggregator (anti-NPE) — NOT PRESENT. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at the original operating company / its merger successor.

Verdict

Insufficient data — but read that as "no NPE chain exists," not "unknown owner." The literal trigger for this category is met: the Assignment Center record contains only the original assignment (inventors → Raytheon Company, executed 2017-01-16, recorded 2017-01-31, reel 041133/0946), with zero post-issuance conveyances. Every NPE signal above is not-present on the evidence available, and the patent remains held by an operating defense contractor (Raytheon Company, now within RTX Corporation). There is no asserter in the chain and no assertion activity, so no affirmative "NPE" or "operating-company-assertion" verdict is supportable on this record.

To make this conclusive, two direct lookups are still needed that I could not complete here: (1) the correspondent/attorney-of-record field on reel 041133/0946 (readable only on the Assignment Center cover sheet), and (2) confirmation that no merger/name-change assignment for this patent has been recorded post-2020 under RTX. Verify at: https://assignmentcenter.uspto.gov/ (search patent 10371570) — corroborated by https://patents.google.com/patent/US10371570/en (Legal Events) and https://patents.justia.com/patent/10371570.

I'll research each cited reference for US 10,371,570. Let me start with targeted searches on the primary patent citations.

Let me research the remaining cited references.

Prior Art Analysis — US 10,371,570 B2 ("Burnout protection for optical sensors")

Patent searched and confirmed (USPTO/Google Patents full text): US 10,371,570 B2, application 15/420,438, filed Jan. 31, 2017, granted Aug. 6, 2019, Raytheon Company, inventors Catherine Trent & Gary A. Frazier. All identifiers below are reproduced literally; no similar numbers were merged.

Methodology note: The references below are the "Citations (7)" printed on the face of the '570 patent (plus the foreign family-cites and the non-patent literature listed in the same record). The examiner-cited references are those carrying an asterisk in the Google Patents citation table (US3699347A and US4724311A); the rest were cited in the file/IDS. I did not invent extra references.

Legal standard applied (35 U.S.C. § 102): Anticipation requires a single reference disclosing every element of a claim, arranged as recited. Under that standard — and this is the key structural point — every one of the '570 patent's claims is anchored to claim 1 or claim 11, each of which now requires the programmable filter to include "at least two quarter wavelength layers, the first quarter wavelength layer including alternating n-type layers and p-type layers of a first material and the second quarter wavelength layer including alternating n-type layers and p-type layers of a second material." That limitation was added during prosecution (it is absent from the 2018 pre-grant publication's claim 1). Consequently, no cited reference can anticipate any surviving claim unless it discloses both (a) the fixed-filter + programmable-filter combination and (b) that specific alternating n/p quarter-wave multilayer. As shown below, no single reference does both — which is why the claim issued over these references.

A. The seven U.S. patent citations

Foreign family citations (listed as "Family Cites Families")

Non-patent literature

B. Claim-by-claim § 102 potential-anticipation mapping

Because the § 102 question is claim-specific, the table below states, for each claim, whether any single cited reference arguably discloses all its elements. Where a reference discloses only some elements, I label it § 103 (obviousness) material rather than § 102 anticipation.

C. Bottom line

• No cited reference anticipates any claim of US 10,371,570 under § 102. The claims' distinguishing limitation — a programmable filter of at least two quarter-wave layers of alternating n/p layers of two different materials, in combination with a fixed filter and a two-state voltage source — is split across the cited art: the tunable-semiconductor-stack element is disclosed by US 4,947,223 and US 5,027,178, while the fixed-filter + tunable-filter + intensity-control combination is disclosed by US 8,754,658, and the bright-source/threshold control by US 2014/0239157. That partitioning is a classic § 103 situation, not § 102.
• Most relevant prior art (ranked):
  1. US 8,754,658 B1 (fixed-filter + tunable-filter combination, amplitude control, feedback loop) — closest to the claimed architecture and control claims.
  2. US 4,947,223 A (electrically tunable alternating n/p quarter-wave multilayer) — closest to the programmable-filter structural limitation and to claims 8 and 9.
  3. US 5,027,178 A (electrically tunable semiconductor multilayer interference filter via pn-junction carrier effects) — reinforces the tunable-filter limitation.
  4. US 2014/0239157 A1 (bright-source protection, control circuitry responsive to intensity) — relevant to the control/threshold claims (4, 5, 7, 14, 16).
  5. US 3,699,347 A and US 4,724,311 A — background "variable/tunable filter" art only.
  6. US 6,785,032 B1 — field/context (laser countermeasures) only.
• The Bennett et al. (1990) and Mehdi et al. (1989) NPL supply the carrier-induced index-change physics underpinning the specification; they are supporting/mechanism references, not apparatus anticipations.
• The PCT Invitation (May 18, 2018) is not prior art and should be disregarded for § 102 purposes.

Caveats and confidence: High confidence in the citation list, dates, and claim text (taken verbatim from the Google Patents full text of US10371570B2, https://patents.google.com/patent/US10371570/en). Medium confidence on the substance of US 4,724,311 A and US 6,785,032 B1, for which my retrieval returned only partial descriptions; those two are peripheral to the anticipation analysis in any event. The open question whether US 4,947,223 + US 8,754,658 (or these plus Bennett/Mehdi) would support a § 103 rejection of claim 1 is a merits judgment that would require full claim-charting against both references' complete disclosures — I have flagged the combinations rather than asserting a legal conclusion.

I'll analyze obviousness using the prior art cited on the patent's own face. Let me pull details on the key references so my combinations are grounded in what each actually discloses.

Let me get details on the remaining cited references to complete the combination analysis.

Obviousness Analysis — US 10,371,570 B2 under 35 U.S.C. § 103

Building on the claim mapping and bibliographic work in the earlier sections (which I do not repeat here), this section takes the prior art as it appears on the face of the '570 patent itself — the examiner-cited "Patent Citations (7)" and "Non-Patent Citations (3)," plus the four foreign "Family Cites Families" — and tests whether the granted claims would have been obvious.

1. The references in play (all cited by/against the '570 patent)

All of these are § 102(a)(1)/(a)(2) prior art relative to the Jan. 31, 2017 priority date (the '570 patent is post-AIA, so § 103 as amended / KSR governs).

2. Person of ordinary skill in the art (PHOSITA)

A graduate-level (M.S./Ph.D.) engineer or physicist with ~2–5 years' experience in infrared/photonic sensor systems, thin-film and semiconductor interference filters, and electro-optic modulators; familiar with III-V epitaxy (MBE/MOCVD), quarter-wave stack design, pn-junction/quantum-confined Stark-effect tuning, and the free-carrier index-tuning literature (e.g., Bennett). Critically, this person knew how to design a quarter-wave stack for 1550 nm and how to voltage-tune it — this is the routine toolkit of the field, as reflected by the references themselves.

3. Claim 1 — element chart and where each element is found

Conclusion: Every element of granted claim 1 is disclosed across a small, tightly related set of references. The grant's distinguishing limitation (added in prosecution — the "at least two quarter-wavelength layers with alternating n/p layers of first and second materials") is squarely met by US 5,027,178 and US 4,947,223.

4. Combinations that render the claims obvious

Combination A (primary): US 8,754,658 + US 5,027,178 → claims 1, 8, 11, 18

• US 8,754,658 supplies the architecture and purpose: a tunable filter in series with a fixed filter, with the amplitude of incoming light controlled "by tuning the peak attenuation wavelength of the tunable filter," in real time, by voltage, with no moving parts — and expressly notes the tunable element can be a photonic crystal whose peak attenuation wavelength changes with applied voltage.
• US 5,027,178 supplies the missing hardware: a semiconductor multilayer interference filter in which each layer contains both n- and p-doped portions forming a series-connected pn junction, and layers alternate in composition, biased via electrical contacts to tune the transmission/reflection wavelength.

Motivation to combine (KSR): Both references are in the same field (voltage-tunable optical filters), address the same problem (shifting a filter's passband electrically without moving parts), and are "reasonably pertinent" to each other. US 8,754,658's "tunable filter" is a functional block awaiting a concrete implementation; US 5,027,178's device is a drop-in implementation of exactly that block. A PHOSITA seeking a fast, robust, solid-state tunable filter for US 8,754,658's system would naturally look to the semiconductor electrically tunable interference filters of US 5,027,178. Reasonable expectation of success is high because both operate on the identical physical principle — changing a multilayer's effective index to shift its passband.

Combination B (alternative primary): US 8,754,658 + US 4,947,223 → claims 1, 8, 9, 11, 18, 19

• US 4,947,223 expressly discloses (i) quarter-wave stacks of alternating n- and p-layers with electrical contacts and an applied potential that changes optical properties, (ii) layers of different indices of refraction, and (iii) a Fabry–Perot resonator with two quarter-wave HR stacks separated by a spacer — which anticipates the "half-wavelength central layer with bilayers stacked above and below" of claims 9/19 (the half-wave "absentee" spacer being the classic Fabry–Perot design, with the spacer at an integral multiple of half a wavelength).
• Motivation is the same as Combination A, and is reinforced because US 4,947,223 is itself cited on the '570 face and concerns the same monolithic tunable-multilayer technology.

Note: US 5,027,178 and US 4,947,223 are highly combinable with each other too (the '178 patent's background expressly builds on GaAs/AlGaAs multilayer stacks and cites the 1982 Burnham DBR patent), so a single obviousness combination can rest on US 8,754,658 + US 5,027,178 + US 4,947,223.

Combination C (photonic-crystal claims 10, 20): US 8,754,658 + US 3,699,347

• US 8,754,658 describes the tunable filter as a photonic-crystal / metamaterial structure whose peak attenuation wavelength changes with voltage by changing effective permittivity.
• US 3,699,347 supplies a voltage-shifted Fabry–Perot interference structure with a beam splitter.
• Motivation: US 8,754,658 already frames the tunable filter as generically implementable by photonic-crystal or liquid-crystal means; using an n-bar/p-matrix photonic crystal (the '570's FIG. 8) is a predictable substitution of one known voltage-tunable index-modulation structure for another, with predictable results.

Combination D (sensor-protection and control claims 3–7, 13–16): + US 2014/0239157 and/or US 6,785,032

• US 2014/0239157 discloses protecting an imaging sensor from bright sources using control circuitry that senses high illumination and applies a protective voltage above a threshold (claims 3, 4, 13, 14), field-wide/spatial sensing (claim 5), and an electro-optic protection measure located in front of the sensor (claims 2, 12).
• US 6,785,032 supplies the laser-countermeasure context (sensor threatened by a laser), i.e., the exact motivation in the '570 specification.
• Motivation: US 2014/0239157's stated problem ("bright sources are a significant problem … provide such systems with electro-optic protection measures") is the same problem the '570 patent addresses (sensor burnout). Combining its threshold-based control circuitry with the tunable-filter shutter is the natural, predictable engineering step; KSR permits combining a known sensing/control technique with a known optical element to achieve the known benefit.

Combination E (beam splitter, claim 6/15): + US 3,699,347

• US 3,699,347 (and US 6,785,032's countermeasure optics) use a beam splitter / half-silvered mirror to divert a portion of the input beam to a detector. Tapping a fraction of the beam to a monitor detector is a notoriously well-known technique in optical-sensor/AGC design (KSR "predictable variation"), so claim 6 is obvious over the base combination plus this reference or the general knowledge.

Combination F (wavelength, claim 17): routine design choice

• Reciting a "fixed central wavelength … about 1.55 micrometers" is the selection of a result-effective variable within the PHOSITA's routine design latitude (telecom/IR standard wavelength; US 5,027,178's materials discussion and Bennett's InP/GaAs/InGaAsP index data are expressly framed around this regime). Under In re Applied Materials / KSR, such a selection, absent unexpected results, does not confer patentability.

5. Why the "shutter" insight is not non-obvious

The '570 patent's asserted advance is that shifting two narrowband passbands apart can block light with a small voltage and thus fast, low-loss switching. But:

• The two-filter (tunable + fixed) series architecture is US 8,754,658's express teaching.
• The small-voltage-wavelength-shift tuning mechanism is US 5,027,178 (and US 4,947,223), and the carrier-induced index change enabling it is Bennett et al.
• The blocking/protection application is US 2014/0239157 and US 6,785,032.
• The article's own FIG. 7 shows only a 0.5 % passband shift for a 1 % index increase — an incremental, predictable result consistent with well-understood thin-film physics (Δλ/λ ≈ Δn/n), not an unexpected one.

Nothing in the granted claims recites the numerical parameters underpinning the "small voltage ⇒ fast, low loss" narrative (no claim recites a voltage magnitude, a <1 µs response, a 2 nm bandwidth, or narrowband-only operation). The specification's discussion of narrowband advantage is therefore not reflected in the claims and cannot supply patentability to them.

6. Claim-by-claim obviousness verdict (summary)

7. Anticipation vs. obviousness (a flag)

Combinations A/B are strong enough that a § 102 anticipation argument is worth testing as an alternative for claim 1: US 5,027,178 claim 1 already recites a multilayer interference filter with alternating-composition sublayers, each layer having n- and p-doped portions forming a series-biased pn junction. The only shortfalls are that the '178 patent does not, on its face as retrieved, (i) expressly say "quarter wavelength" for the layers it claims or (ii) place a separate fixed filter in series. Those gaps are what the '658 reference (series fixed + tunable filter) and US 4,947,223 (express "quarter-wave" stacks) close — making claim 1 at minimum obvious even if not strictly anticipated.

8. Rebuttal considerations (secondary considerations)

I found no evidence of unexpected results, long-felt unmet need, industry praise, or copying in the record available to me, and the modest 0.5 % tuning range (FIG. 7) argues against unexpected performance. A patentee might argue that US 5,027,178's quantum-confined-Stark-effect tuning needs ~100 V while the '570's carrier-depletion approach needs only a small bias — but (a) US 5,027,178 itself discloses the pn-junction/carrier route as an alternative to the large-voltage Stark route, and (b) the granted claims do not recite voltage magnitude, so this distinction is not captured by the claims. Absent claim limitations or objective evidence, this does not rebut obviousness.

9. Confidence and caveats

• High confidence for the content of US 8,754,658 B1, US 5,027,178 A, US 4,947,223 A, US 3,699,347 A, and US 2014/0239157 A1, which I retrieved or verified from Google Patents/Justia text (see sources).
• Lower confidence for US 6,785,032 B1 and US 4,724,311 A — my retrieval attempts for these two were cut off by search-step limits, so I rely on their titles/assignees as cited on the '570 face and general knowledge; I have not verified their specific disclosures. Their role in the combinations above is secondary (motivation/context) and the analysis does not depend on their internal detail.
• Confidence in Bennett et al. is based on the citation as it appears on the '570 face plus the title's express subject matter (carrier-induced refractive-index change in InP/GaAs/InGaAsP); I did not retrieve the paper itself.
• This is a technical analysis of publicly available document text, not a legal opinion; claim construction and infringement/validity determinations would require the prosecution history and full reference texts. I also flag the previously noted date inconsistency (task framing April 26, 2026 vs. prior operating context Sept. 30, 2026); it does not affect this art-based analysis, since all references predate the 2017 priority date.

Sources: https://patents.google.com/patent/[US8754658B1](/patent/US8754658B1)/en · https://patents.google.com/patent/[US5027178A](/patent/US5027178A)/en · https://patents.google.com/patent/[US4947223A](/patent/US4947223A)/en · https://patents.google.com/patent/[US3699347A](/patent/US3699347A)/en · https://patents.google.com/patent/US20140239157A1/en · https://patents.google.com/patent/US10371570/en (citations list) · Bennett et al., IEEE J. Quantum Electron. 26(1):113–122 (1990); Mehdi et al., Superlattices & Microstructures 5(3):443–449 (1989) (as cited on the '570 face).