US 11326938

I'll verify against live sources for litigation and docket activity, then summarize.

US Patent 11,326,938 B2 — Analyst Summary

Verification note: I searched for litigation, PTAB, and Federal Circuit activity tied to this specific number. I found no CAFC 2026 docket entry, no IPR/PGR, and no district court case naming US 11,326,938 or its assignee as a party. I want to be explicit that this is a negative search result, not proof of absence — free web sources do not comprehensively index all federal dockets, and I could not directly query PACER or the CAFC's own docket system. Treat "no known litigation" as provisional. All bibliographic and technical content below is taken from the authoritative full text of the patent supplied to me and cross-checked against uspto.report and Justia listings, which matched.

1. Bibliographic Data

Classifications: G01J 1/0219; G01J 1/0238; G01J 1/0488; G01J 1/0492; G01J 1/1626; G01J 1/24; G01J 1/4204; G01J 2001/444; A61B 5/0059; A61B 5/742.

Related family / siblings:

• KR 10-2155085 B1 (published 2020-09-11) and KR 10-2020-0091693 A — Korean counterparts.
• US 11,305,091 B2, "Circadian rhythm management apparatus and system" — filed the same day (2019-07-29) as application 16/525,183 with the same Korean priority date but a different Korean priority number (10-2019-0008822). It cites US 2020/0232844 A1 (the pre-grant publication of the patent at hand) among its references. Separate patent, related subject matter.
• A later Kookmin filing, KR 10-2912693 B1 (2026-01-13), and a US publication 2025/0325228 A1 on a chip-integrated bio-illuminance measurement approach — these pursue a different architecture (integrated photosensor pair, no filters), so they read more as design-arounds/improvements than as continuations.

Prior art cited on the face: KR 10-2001-0090519 A; US 2011/0064098 A1; US 2016/0027282 A1; KR 10-2017-0123065 A; US 2017/0328767 A1; US 2018/0339127 A1; US 2020/0230346 A1. Non-patent literature: Korean Office Action for KR 10-2019-0008809 dated 2020-05-18; Oh et al., "Healthy, natural, efficient and tunable lighting: four-package white LEDs…," Light Sci. Appl. 3, e141 (2014).

2. Abstract (as issued)

Disclosed is a bio illuminance measuring apparatus including a circadian lambda filter passing external light along according to a circadian rhythm sensitivity curve, a visual lambda filter passing the external light along according to a visual sensitivity curve, a photo sensing portion sensing and converting the external light, which has passed through the circadian lambda filter, into a circadian wavelength signal and sensing and converting the external light, which has passed through the visual lambda filter, into a visual wavelength signal, and an illuminance calculating portion which calculates a ratio between the circadian wavelength signal and the visual wavelength signal, calculates a circadian action factor by applying the ratio between the circadian wavelength signal and the visual wavelength signal to a circadian action function which varies according to the visual wavelength signal, and calculates a bio illuminance value of the external light on the basis of the circadian action factor.

3. Independent Claim 1 — Plain-Language Overview

There is exactly one independent claim. Everything else (claims 2–12) depends from it, directly or indirectly.

The apparatus has four required elements:

1. A circadian lambda filter — a bandpass filter that transmits external light according to a "circadian rhythm sensitivity curve," i.e., the spectral sensitivity of the hormonal system that drives circadian rhythm (embodiments name melatonin or cortisol). Spec describes a 400–600 nm band with peak transmittance near 450 nm.

2. A visual lambda filter — a bandpass filter that transmits external light according to the human photopic visual sensitivity curve.

3. A photo sensor that outputs two signals:

   • a circadian wavelength signal = (circadian wavelength voltage) − (dark voltage), and
   • a visual wavelength signal = (visual wavelength voltage) − (dark voltage).

   The dark-voltage subtraction exists to cancel photodiode dark current / self-generated noise. Claim 1 expressly recites that both signals are analog signals formed as these differences.

4. An illuminance calculator that (a) takes the ratio of the circadian signal to the visual signal (the "C/V ratio"), (b) plugs that ratio into a circadian action function that varies according to the visual wavelength signal to produce a circadian action factor (CAF), and (c) computes a bio illuminance value from that CAF.

Plus a required structural limitation inside the calculator: a function storage holding a plurality of circadian action functions, each valid for a different range of the visual wavelength signal.

Plain reading: rather than using an expensive spectrometer to measure the spectrum (and hence color temperature) of ambient light and then deriving bio-illuminance from a lookup, the claim does it with two filtered photodiodes. The ratio of the two filtered readings is a proxy for color temperature; the ratio is converted to a CAF by the appropriate member of a stored family of linear functions, selected according to how bright the visual channel is; and bio-illuminance is the CAF times ordinary visual illuminance. The novelty hook is the signal-magnitude-dependent selection among multiple stored CAF-vs-C/V functions, plus the dark-current differential sensing.

Dependent claims add: a function caller to select the stored function by visual signal (2); the CAF as a dependent variable of the C/V ratio (3); proportionality of CAF to the C/V ratio (4); a visual illuminance calculator (5); Biolux = CAF × Lux (6); three-photodiode arrangement where the third is light-blocked and outputs dark current (7); first/second/third variable resistors converting currents to voltages (8); a controller that adjusts gain by varying those resistances and digitizes the analog signals (9); a display (10); the circadian sensitivity curve being a hormone-based curve peaking in the circadian band (11); and the visual sensitivity curve peaking in the visual band (12).

Spec support for the math: C = (I₁×R₁) − (I₃×R₃); V = (I₂×R₂) − (I₃×R₃); CAF = F(C/V) = A×(C/V) + B (Equation 3, where A and B vary with the visual signal V); and Biolux = CAF × Lux (Equation 4). CAF itself is defined in the spec as the ratio of circadian efficacy of radiation (CER) to luminous efficacy of radiation (LER). Note that the linear form A×(C/V) + B appears only in the specification — claim 1 does not recite it.

4. Drafting and Consistency Observations (flagged, not corrected)

Per your instruction to interpret identifiers literally, here are discrepancies that appear on the face of the document and that I am reporting rather than silently harmonizing:

1. Antecedent-basis mismatch between claim 1 and claim 8. Claim 1 recites a "photo sensor." Claim 8 recites "wherein the photo sensing portion further comprises…" while depending from claim 7, which in turn depends from claim 1. The term "photo sensing portion" is used throughout the specification but never introduced in the claims. This is a facial inconsistency that could matter to validity or construction.

2. Two different stated visual wavelength bands. The specification's Summary/Definitions portion as published states the visual wavelength band "may be 380 nm to 780 nm," while the Detailed Description states the visual wavelength band "may be 450 nm to 700 nm" and describes the visual filter as passing 450–700 nm. Both appear in the same patent. The Background separately says light of 380–780 nm has maximum sensitivity.

3. Identifier ambiguity in the figures. Figure 7 is described as containing sub-panels "7A to 7C" and representative values "V₁, V₂, V₃," but the reference numerals in the text also use "V" for the visual wavelength signal itself. A reader should not conflate the sub-figure/representative-value notation with the signal variable.

4. Claim count vs. described apparatus. The specification describes a circadian rhythm reinforcing portion, a storage portion storing reference bio-illuminance values, and a warning function on the display. None of these are claimed in US 11,326,938; some of that subject matter appears instead in the sibling US 11,305,091.

5. Bottom Line

US 11,326,938 is a Kookmin University apparatus patent (original assignee "Kookmin University," now held by its Industry Academy Cooperation Foundation) granted 2022-05-10 on a 2019-07-29 US filing claiming Korean priority to 10-2019-0008809 of 2019-01-23. It claims a two-filter, three-photodiode optical front end that derives a circadian action factor from the ratio of a circadian-channel signal to a visual-channel signal via a stored family of functions indexed by visual signal magnitude, and outputs bio-illuminance. It has a single independent claim and is currently active with an adjusted expiration of 2040-10-13. I found no evidence of any CAFC 2026 or other litigation involving this number, but I cannot certify that no such proceeding exists.

I'll search for litigation involving US patent 11326938.

Let me search more targeted litigation databases and check the patent owner's litigation history.

Let me check litigation-specific databases and dockets directly.

Litigation Search Results: US Patent No. 11,326,938

Bottom line: I found no evidence of any litigation involving US Patent No. 11,326,938.

Searches performed

I searched for the exact patent number ("11326938" / "11,326,938") combined with litigation, infringement, district court, docket, PACER/Docket Alarm, and Unified Patents portal terms, plus the patent owner's name (Kookmin University Industry Academy Cooperation Foundation). No court docket, complaint, or case record naming this patent appeared.

Case list

There are no known district-court, ITC, or Federal Circuit proceedings asserting or involving US 11,326,938 to report.

Important contextual notes (verified from the record)

• No litigation activity is reflected for this patent. The full patent text (Google Patents, fetched 2026-09-24) shows only the following relevant entries for US 11,326,938:
  • Post-grant (PTAB): none. There is no IPR, PGR, or CBM listed for this patent.
  • "Families Citing this family (3)" — i.e., later patents citing this one — not litigation.
  • "Citations (7)" — prior art cited, not litigation.
  • The only USPTO proceedings shown are ordinary prosecution events (non-final office action 2021-12-13; response 2022-03-10; notice of allowance 2022-03-29; issue 2022-05-10) plus a maintenance-fee payment (4th year, 2025-09-22). Assignee is Kookmin University Industry Academy Cooperation Foundation.
• The related, separately-numbered Kookmin patent US 11,305,091 B2 (Circadian Rhythm Management Apparatus and System, App. No. 16/525,183, KR priority 10-2019-0008822, filed 2019-07-29) and the related application US 2020/0230346 A1 appear in the same family, but those likewise show no litigation in the sources reviewed.
• A nearby IPR that is often confused with this space — Unified Patents, LLC v. Kwangwoon University Industry-Academic Collaboration Foundation, IPR2020-01048 (PTAB, filed June 9, 2020, terminated Nov. 23, 2021, final written decision) — concerns US Patent No. 8,867,854, a different patent and different owner. It should not be attributed to US 11,326,938.

Caveats on my confidence

• This is a negative finding. It is based on web searches of publicly indexed litigation sources; I was unable to complete direct, exhaustive queries against the Unified Patents litigation portal and PACER/Docket Alarm for the exact number before reaching my search limit.
• A negative result does not prove no litigation exists. Litigation can be recent/unindexed, filed under a different asserted-patent identifier typo, sealed, or not captured in the sources I could reach.
• Since the patent is active (adjusted expiration listed as 2040-10-13) and owned by a university foundation, I cannot rule out future or recently filed actions.

Recommendation: To confirm with high confidence, a direct query should be run on:

1. Unified Patents litigation caselist — https://portal.unifiedpatents.com/litigation/caselist
2. PACER / Docket Alarm / CourtListener for "11,326,938" and for "Kookmin University Industry Academy Cooperation Foundation" as plaintiff
3. USPTO Patent Trial and Appeal Board (PTAB) and USPTO Patent Assignment records

If you'd like, I can attempt those specific targeted lookups.

I'll verify the ODP's "no proceedings" finding against public sources before writing this up.

No hits so far. Let me run two more targeted checks (PTAB docket aggregators and the patent owner's name) to be thorough.

PTAB Proceedings Report — U.S. Patent No. 11,326,938 B2

Patent: Bio illuminance measuring device for determining circadian action
Patent Owner: Kookmin University Industry Academy Cooperation Foundation (Seoul, KR)
Inventors: Hyun Sun Mo; Dae Jeong Kim
App. No.: 16/525,190 | Filed: 2019-07-29 | Priority: 2019-01-23 (KR 10-2019-0008809) | Granted: 2022-05-10
Listed adjusted expiration: 2040-10-13 | Claims: 12 (3 independent: claims 1, 7, 8 are dependent; independent claim is 1, with 7, 8, 9, 10, 11, 12 variously depending — 1 is the sole broad independent apparatus claim)

Proceedings overview

Total AIA trial proceedings on US 11,326,938: zero (0). The USPTO ODP canonical proceedings block returns no IPR, PGR, or CBM for this patent, and my independent web searches surfaced no petition, institution decision, FWD, or appeal naming US 11,326,938 or its application number 16/525,190. The breakdown is therefore 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denials — and the bottom-line defensive posture for a defendant is the worst-case variant: this is a virgin, never-challenged patent whose full claim set — including independent claim 1 in its original, unamended form — is intact, and there is no prior petitioner whose estoppel or adverse FWD you can borrow.

Proceeding-by-proceeding detail

None to report. There is no proceeding number to populate, and I will not manufacture one. The per-proceeding template (type / filed / status / judge panel / grounds / institution decision / FWD / settlement / appeal / defensive value) is inapplicable.

What I checked (diligence log)

Caveat on confidence: Docket Alarm and PTAB E2E pages are not fully crawled by general web search, and the ODP block is the authoritative list here. I treat "zero proceedings" as high confidence for anything instituted, and moderate-to-high confidence that no petition has ever been filed — a recently-filed, not-yet-decided petition (2019–2022 patents saw heavy § 314(a)/Fintiv discretionary denials) is the residual risk. Verify directly at PTAB E2E and the USPTO ODP API before relying on this in a filing.

Strategic summary

Claim status: everything is untested, and that cuts against you. Nothing has been canceled, nothing has been narrowed by amendment, and no claim has been affirmed over art. All twelve claims of 11,326,938 stand as granted. If a demand letter cites claim 1 — or any of claims 7–12, which pull in the three-photodiode/dark-current differential circuit, the variable-resistor gain control, and the visual illuminance multiplier — you cannot answer with an FWD. There is no "claim 1 is dead, so the demand is sanction-bait" argument available, because no one has ever killed it.

Estoppel landscape: a blank slate, which is the one genuine advantage here. Because no IPR was ever instituted, § 315(e)(2) estoppel attaches to no one. There is no petitioner/privy bar, no "grounds raised or reasonably could have been raised" limitation, and no risk that your own art was already before a panel. You are free to file a first-chance IPR on any § 102/§ 103 ground using patents and printed publications, and to run parallel invalidity contentions in district court without the usual estoppel sequencing problem. That is a meaningful tactical benefit: your petition is a first petition, so General Plastic factor 1 (same petitioner previously filing on the same claims) is neutral, and you will not be a "me-too" follower. The trade-off: you also get no joinder vehicle and no free ride on someone else's expert work.

Pattern signals: quiet patent, live family, long runway. This is an academic patent owner (Kookmin University's technology-transfer foundation), not an NPE, and there is no Unified Patents or other defensive aggregator anywhere in the chain on this patent. That is consistent with the zero-proceeding record — university patents are infrequently the target of IPRs until they are asserted. Two forward-looking signals matter for a defendant: (i) the family is still growing — sibling US 11,305,091 B2 ("Circadian rhythm management apparatus and system," granted 2022-04-19, App. No. 16/525,183, same 2019-01-23 KR priority) claims the system-level combination of the bio-illuminance measurement of the '938 patent with a circadian-rhythm reinforcing light source, and a later Korean continuation (KR 10-2912693 B1, filed 2022-12-29) issued in 2026; and (ii) the term is listed to run to 2040-10-13 (~442 days of PTA beyond the 20-year date). A patent owner with a decade-plus of runway, a live family, and no invalidity history on the flagship measurement claim is a patent owner with every incentive to assert rather than license cheaply.

Estoppel/§ 325(d) note on prosecution art. The examiner's cited art was thin and mostly foreign or tangential: KR 10-2001-0090519 (Shimamura, the applicant's own "Related Art Document"), US 2011/0064098 (Ricoh), US 2016/0027282 (Salutron — blue-light exposure monitoring), KR 10-2017-0123065 (Kookmin's own illumination sensor), US 2017/0328767 (Apple color ambient light sensor), US 2018/0339127 (Brown University circadian lighting), and — notably — US 2020/0230346 A1, Kookmin's own co-pending sibling application cited against it. Under § 325(d) the Board may exercise discretion to reject grounds built on "the same or substantially the same prior art or arguments"; art that was before the examiner is the risk zone, and Apple/Salutron/Ricoh-based grounds sit closest to it. Prior art in the circadian-lighting literature that was not of record — e.g., the Oh et al. four-package-white-LED paper (Light Sci. Appl. 3, e141, 2014), which the examiner cited but apparently only as background, and the circadian-illuminance measurement family around KR 10-2166726 and KR 10-2082050 — is the cleaner § 325(d) path and should be the first place a searcher looks.

Recommended next steps

1. Do not wait for an FWD that does not exist. Since there is no proceeding, the only invalidity vehicles currently open are (a) a first-chance IPR petition under 35 U.S.C. §§ 311–319 on § 102/§ 103 grounds limited to patents and printed publications, (b) PGR — closed, the 9-month post-grant window lapsed on or about 2023-02-10, and (c) ex parte reexamination, which under the 2025–26 PTAB procedural tightening (record-high 607 procedural denials in 2025; reexamination requests at an all-time high) has become the practical fallback for art the Office won't institute on. Note the patent is AIA/FITF (priority 2019-01-23), so IPR is available on this patent at any time, with no § 315(b) one-year bar until you are actually served with a complaint.
2. Audit the claim-1 differential-voltage limitation as your lead theory. Claim 1 is unusually specific for an independent claim: the circadian and visual wavelength signals are each defined as a difference between a photodiode-derived voltage and a dark voltage from a third, light-blocked photodiode (claims 1, 7, 8), and the circadian action function is selected from a stored plurality of functions indexed to ranges of the visual wavelength signal (claim 1, plus the function-caller of claim 2). Those are the narrowing hooks. Pre-2019 ambient-light-sensor and colorimeter art that uses a dark/reference channel for offset correction, and any reference disclosing piecewise or range-indexed color-correction functions, is where I would search first.
3. Preserve the § 315(e)(2) clean slate. Once you file, you cannot un-ring the bell. Because no one else has petitioned, you have the rare luxury of choosing your own timeline — but if litigation is already on file, run the Fintiv-style § 314(a) analysis early, since parallel-proceeding discretionary denial remains the dominant cause of non-institution.
4. If active proceedings had existed, the milestones would be: institution decision within 6 months of the petition's filing date (35 U.S.C. § 314(b)), and the FWD within 12 months of institution (§ 316(a)(11)). With zero proceedings, there are no pendency dates to track.
5. Verify the negative directly before relying on it. Confirm at PTAB E2E and CourtListener that no petition or Federal Circuit appeal exists, and re-run this check monthly — a first petition on a university portfolio patent is exactly the kind of event that appears with no public warning.

I'll research the assignment chain for US 11,326,938, starting with the USPTO assignment records and the inventor/assignee details.

Let me search the assignment records and inventor details more specifically.

Let me look for the specific assignment reel/frame records and check inventor/assignee relationships.

Assignment & Ownership Analysis — US Patent No. 11,326,938

Patent: Bio illuminance measuring device for determining circadian action
App. No.: US 16/525,190 · Filed: 2019-07-29 · Granted: 2022-05-10
Priority: KR 10-2019-0008809 (2019-01-23) → KR 102155085B1
Status: Active; adjusted expiration 2040-10-13; 4th-yr maintenance fee paid 2025-09-22.

Sourcing note (be aware before relying on this): My direct queries to the USPTO Patent Assignment Center and PatentCenter did not return a filled record before I hit the search limit. The assignment entry below is reconstructed from the authoritative Google Patents "Legal Events" / "Assignment" block (fetched 2026-09-24), which mirrors the USPTO assignment record and exposes the reel/frame and assignor/assignee names but not the correspondent of record. Treat the reel/frame and dates as reliable; treat the correspondent field as an explicit gap.

Inventors

• Both inventors are named on the face of the patent and both appear as assignors in the single recorded assignment to Kookmin's technology-transfer foundation — the standard academic "inventors assign to the university foundation" arrangement. There is no evidence of either inventor retaining individual ownership.
• Unverified lead: patentleaderboard.com groups a person named "Hyun-Sun Mo" under a Samsung-labeled page listing 7 patents including US 11,326,938. I could not confirm this is the same individual or that any Samsung employment/assignment exists for this patent. I am flagging it only so it can be checked; it does not change the ownership chain below, which shows Kookmin as assignee.
• Unusual patterns: none of the classic red flags. No indication that all inventors departed within 12 months of filing; no evidence of a split between named inventors and assignee. The co-pending, same-priority-date Kookmin family member (US 2020/0230346 A1 / US 11,305,091 B2, "Circadian Rhythm Management Apparatus and System") suggests a single Kookmin research group filing a cluster on 2019-01-23 — worth noting as related-family context only, not as this patent's chain.

Original assignee

• Entity of record: Kookmin University Industry Academy Cooperation Foundation (Republic of Korea) — the industry-academy cooperation foundation of Kookmin University, i.e., the university's IP commercialization / technology-transfer arm. Google Patents separately lists "Kookmin University" as both current and original assignee at the institution level; the recorded assignment names the Foundation. These are functionally the same ownership (university → its TTO).
• Does it ship a product embodying the claims? No. This is a university technology-transfer entity; by its nature it holds and licenses IP rather than manufacturing. That alone is not an NPE/troll signal in the pejorative sense — university TTOs are a recognized non-practicing category distinct from litigation-funding shell LLCs.
• Primary line of business: Management and commercialization of Kookmin University research IP. Third-party trackers (OnScope/IPQwery, PlainPatent) show the foundation holds roughly 140–196 granted US patents across a very broad technology range, active since ~2007, with a large, diversified university research portfolio.
• Current status: Operating / active. The patent is live (4th-year maintenance fee paid 2025-09-22 per the USPTO legal-events record).

Assignment timeline

Only one assignment is recorded for this patent. There are no post-grant transfers, no security interests, no name changes, and no releases.

• 2019-08-01 (executed / effective) / recorded 2019-08-13 — Reel 050038/0869
  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST" (inventor → institutional owner)
  • Assignor: MO, HYUN SUN; KIM, DAE JEONG (the two named inventors)
  • Assignee: KOOKMIN UNIVERSITY INDUSTRY ACADEMY COOPERATION FOUNDATION, Korea, Republic of
  • Correspondent: Not disclosed in the retrieved record. The Google Patents assignment block renders this entry as "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)" and omits the correspondent attorney/firm and address. I could not retrieve the correspondent from the Assignment Center before reaching the search limit. Because there is only one link in the chain, the "repeat correspondent" signal cannot be assessed regardless.
  • Context: Original inventor-to-university transfer; standard institutional ownership formalization for an academic filing. (Note the assignment was executed 2019-08-01, after the 2019-07-29 US filing date — a common sequencing for university filings and not a transfer of ownership away from the inventing institution.)

Implication of a single-link chain: the original assignee (Kookmin's Foundation) still owns the patent. No fire-sale, no securitization, no transfer-to-asserter.

Timeline diagram

timeline
    title Ownership of US 11326938
    2019 : Filed by Kookmin University
         : Inventors assign to Kookmin Foundation
    2022 : Issued as US 11326938
    2025 : Fourth year maintenance fee paid

NPE / troll-pattern signals

Verdict

Insufficient data — specifically, the chain contains only the original assignment.

The entire ownership record is a single inventor-to-institution assignment executed 2019-08-01 and recorded 2019-08-13 at Reel 050038/0869, conveying US 16/525,190 from inventors Hyun Sun Mo and Dae Jeong Kim to Kookmin University Industry Academy Cooperation Foundation, with no subsequent transfers, security interests, name changes, or releases. Per the rubric, "only the original assignment" is the textbook Insufficient-data case for NPE scoring: there is no shell-entity transfer, no known asserter, no cascading transfers, and no litigation-generated pre-litigation transfer to evaluate — the current holder is the inventing university's own technology-transfer arm, an active non-practicing academic entity rather than a litigation-funding NPE. The residual uncertainty is the unretrieved correspondent of record and the negative litigation finding, neither of which supplies an NPE signal here.

Verification links:

• USPTO Assignment Center (search by patent number 11326938 / app 16/525,190): https://assignmentcenter.uspto.gov/
• Legacy USPTO assignment portal: https://assignment.uspto.gov/patent/index.html
• Google Patents legal-events source for Reel 050038/0869: https://patents.google.com/patent/US11326938/en

Recommended follow-up to close the gap: pull the recorded assignment PDF from Assignment Center for Reel 050038/0869 to capture the correspondent of record and confirm the execution/recording dates, and confirm there are no abstracts of title or name-change records filed under the Foundation's name that Google Patents did not surface.

I'll research the specific prior-art citations listed for US Patent No. 11,326,938. Let me run targeted searches on each reference.

Let me search for the remaining references.

Prior-Art Analysis — US Patent No. 11,326,938

0. Verification of the target patent (literal ID match)

I confined the search to patent number 11,326,938 exactly, as instructed. The record returned:

Because the patent has a 2019 effective filing date, it is an AIA patent: prior art is governed by 35 U.S.C. § 102(a)(1) (publicly available before the effective filing date) and § 102(a)(2) (earlier-filed U.S. applications/patents that later publish). Anything publicly published before 2019-01-23 is § 102(a)(1) art.

Scope/method caveat: This analysis works from the citation list printed on the face of US 11,326,938 (the "Citations (7)" / "Patent Citations (7)" tables) plus the abstract-level and specification-level disclosure I could retrieve for each reference. It is a provisional § 102 mapping: I did not obtain full claim sets and full text for every foreign reference, so any "anticipates" label below is a potential reading, not a legal conclusion (and anticipation requires every element of a claim in a single reference).

1. The seven cited references — overview

Note: the Google Patents listing flags references #2, #3, #5, #6, #7 with the "cited by examiner" marker ("*"); the two Korean documents (#1, #4) are listed among the cited references as well. Reference #4's family (Kookmin) and reference #1 are also echoed in the sibling patent US 11,305,091.

2. Reference-by-reference analysis and § 102 mapping

Reference 1 — KR 10-2001-0090519 A

• Full citation: KR 10-2001-0090519 A, "Apparatus for measuring illuminance, and exposure apparatus," applicant/inventor Shimamura Teruo.
• Dates: filed 2000-03-24; published 2001-10-18.
• Description: An illuminance-measuring apparatus and an associated exposure apparatus; this is the document expressly named as the "RELATED ART DOCUMENT" in the '938 specification's background. It is a classical photometry/illuminance measurement disclosure — i.e., a visual-λ (photopic) illuminance meter architecture.
• § 102 potential: As § 102(a)(1) art (published ~18 years pre-filing), it can anticipate at most a generic illuminance-measuring apparatus sub-combination (the "visual lambda filter + photo sensor + illuminance output" aspects reflected in claim 1 and claim 12). It does not disclose the circadian-λ filter, the C/V ratio, the CAF-vs-circadian-action-function computation, or the plural-function storage (claims 1, 2, 3, 4). No claim of '938 is anticipated by this reference alone.

Reference 2 — US 2011/0064098 A1

• Full citation: US 2011/0064098 A1, "Semiconductor laser control device and image forming device," Ricoh Company, Ltd.
• Dates: filed 2009-09-15; published 2011-03-17.
• Description: A semiconductor-laser control device and image-forming device (optical-power/sensor feedback for a laser). It is a photodetector/amplification-control disclosure, not a circadian or colorimetry disclosure.
• § 102 potential: § 102(a)(1) art. Its only plausible tangential relevance is to generic photo-detector drive/feedback (variable resistor / gain adjustment background that underlies claim 9's controller). It discloses none of: circadian λ filter, visual λ filter, C/V signal ratio, circadian action function, CAF, function storage. No claim of '938 is anticipated. It is best characterized as peripheral/background art.

Reference 3 — US 2016/0027282 A1 (granted as US 9,655,525 B2)

• Full citation: US 2016/0027282 A1, "User-wearable devices that monitor exposure to blue light and recommend adjustments thereto," Salutron, Inc. (inventor Yong Jin Lee); granted as US 9,655,525 B2 on 2017-05-23.
• Dates: filed 2014-07-25; published 2016-01-28.
• Description: A wearable optical sensor is filtered to emulate the spectral response of intrinsically photosensitive Retinal Ganglion Cells (ipRGCs) — peak ~480 nm, melanopsin — to detect blue light and compute a metric that is compared to a threshold to trigger an exposure-adjustment notification. The specification expressly ties the sensor response to melatonin suppression / circadian entrainment via the ipRGC→SCN pathway.
• § 102 potential: Strong § 102(a)(1) art for the "circadian lambda filter that passes external light according to a circadian rhythm sensitivity curve" element of claim 1 (and the hormone/melatonin context of claim 11). Its optical sensor "adapted to detect blue light … similar to the spectral response of ipRGCs" reads directly on the circadian-filter concept. However, it lacks the visual lambda filter, the two-signal C/V ratio, the circadian action function, the plural-function storage, and the dark-voltage differential limitation. No single claim of '938 is anticipated in full, but this is one of the two most material references for the circadian-filter element.

Reference 4 — KR 10-2017-0123065 A

• Full citation: KR 10-2017-0123065 A, "Illumination sensor and illumination control device using the same," Kookmin University Industry Academy Cooperation Foundation.
• Dates: filed 2016-04-28; published 2017-11-07.
• Description: An illumination sensor plus a lighting-control device that uses it — the applicant's own earlier sensor work. Because it predates 2019-01-23 by more than a year, it is § 102(a)(1) art notwithstanding the common ownership.
• § 102 potential: This is the most probative reference because it is the applicant's own antecedent sensor disclosure; it is the natural candidate for disclosing the sensor/circuit architecture (photo-sensing portion, gain control) later claimed in '938. I could not retrieve its full claim text in this pass, so I cannot map individual elements with confidence. Flagged as requiring full-text retrieval — if it discloses a two-filter (circadian + visual) sensing arrangement, it would be the leading anticipation candidate for claim 1.

Reference 5 — US 2017/0328767 A1 (granted as US 10,254,160 B2)

• Full citation: US 2017/0328767 A1, "Color Ambient Light Sensor Circuitry For Electronic Devices," Apple Inc.; granted as US 10,254,160 B2.
• Dates: filed 2016-05-16; published 2017-11-16.
• Description: An electronic device with a color ambient-light sensor having photodiodes of different spectral sensitivities plus "a dark current photodiode that is insensitive to light," whose output "is subtracted from the signals of the other photodiodes." The corrected photodiode signals are digitized by a set of ADCs and converted to ambient-light intensity/color.
• § 102 potential: Strong § 102(a)(1) art for the differential/dark-current-subtraction limitation of claim 1 ("circadian wavelength signal corresponding to a difference between a circadian wavelength voltage and a dark voltage"; likewise the visual signal) and for the structural features of claims 7–8 (multiple photodiodes + dark reference; light-insensitive reference detector). The "dark current … subtracted" teaching maps closely onto '938's Equations 1–2 (C = VC − VD; V = VV − VD). What it does not disclose is the circadian-λ / visual-λ filter pairing, the C/V ratio, the CAF, or the plural circadian-action-function storage. No claim of '938 is anticipated in full, but this reference alone reads on the dark-voltage-difference element that appears in claim 1 and claims 7–8.

Reference 6 — US 2018/0339127 A1 (WO 2018/218241 A1; granted as US 10,850,061 B2)

• Full citation: US 2018/0339127 A1, "Lighting System for Circadian Control and Enhanced Performance," Brown University (also published as WO 2018/218241 A1).
• Dates: filed 2017-05-26; published 2018-11-29 (parent provisional 62/511,692, 2017-05-26).
• Description: A lighting-control system that tunes spectral distribution/intensity of a bioactive band to shift or entrain the circadian clock. Materially, its specification states: "The normalized action spectrum for melatonin suppression (C(λ)) and for photopic luminosity (V(λ)) … are used to calculate the Circadian Action Factor (CAF)," and it cites the Oh et al. figures of merit CER and CAF.
• § 102 potential: Strong § 102(a)(1) art for the conceptual heart of claim 1 — the use of both a circadian action spectrum C(λ) and a photopic luminosity function V(λ) to derive a Circadian Action Factor. It also bears on claims 3 and 4 (CAF defined/derived from a C-to-V relationship; CAF proportional to that relationship). Critically, however, this is a light-emitting/control system, not a measuring apparatus: it does not disclose a circadian lambda filter and a visual lambda filter each passing external light onto a photo sensor, the analog differential signals, or the plural circadian action functions stored per range of the visual wavelength signal (claim 1's function storage; claim 2's function caller). No claim of '938 is anticipated in full, but this is the strongest art for the CAF/circadian-action-function limitation.

Reference 7 — US 2020/0230346 A1 (granted as US 11,305,091 B2)

• Full citation: US 2020/0230346 A1, "Circadian rhythm management apparatus and system," Kookmin University Industry Academy Cooperation Foundation; granted as US 11,305,091 B2.
• Dates: priority 2019-01-23 (KR 10-2019-0008822); published 2020-07-23.
• Description: A circadian-rhythm management apparatus/system — the applicant's co-filed sibling application in the same family.
• § 102 potential: Not prior art. It shares the same effective filing/priority date (2019-01-23) and the same applicant as '938, so it cannot qualify under § 102(a)(1) (not publicly available before the effective filing date) or § 102(a)(2) (same effective filing date; common ownership). It is listed only as a related-family citation and should be excluded from any anticipation/obviousness attack. No claim of '938 is anticipated.

3. Highly relevant non-patent literature on the face of '938

• Oh, J.H.; Yang, S.J.; Do, Y.R., "Healthy, natural, efficient and tunable lighting: four-package white LEDs for optimizing the circadian effect, color quality, and vision performance," Light: Science & Applications, 3, e141 (2014). This is the origin of the CER (circadian efficacy of radiation) and CAF (circadian action factor) figures of merit relied on in '938 (see the '938 specification defining CAF as CER/LER). As a 2014 printed publication it is § 102(a)(1) art and is the closest art to the CAF concept underlying claims 1, 3, and 4.
• Korean Office Action for KR 10-2019-0008809, dated 2020-05-18 (4 pages) — prosecution record of the KR priority case, not prior art itself, but useful for identifying what the Korean examiner considered relevant.

4. Bottom line on anticipation (§ 102)

No single reference cited on the face of US 11,326,938 anticipates any claim of '938 in full. Anticipation requires all elements of a claim in one reference, and the cited art is split across distinct sub-features:

The "function storage" of a plurality of circadian action functions that vary per range of the visual wavelength signal, and the specific computation chain C/V ratio → CAF via a visual-signal-selected function → Biolux = CAF × Lux, appear to be the features not disclosed by any of the seven cited references individually — which is consistent with these references supporting an obviousness (§ 103) combination rather than a § 102 anticipation.

Confidence and limits

• I am confident about the bibliographic data above (patent number, dates, assignees, titles) because it is drawn from the patent's own citation tables.
• I am less certain about element-by-element § 102 mappings for the two Korean references (#1 and #4) and for the Ricoh laser-control reference (#2), because I could not retrieve their full texts/claims in this pass. Those should be treated as provisional.
• The Salutron, Apple, and Brown characterizations are grounded in their abstracts/specification excerpts I retrieved and quoted; those read directly on the corresponding '938 limitations.
• No contradictions arose between the patent's own text and the search results. One item worth flagging for clarity: reference #7 (US 2020/0230346 A1) and the related US 11,305,091 B2 appear in the same family as the target and are the applicant's own filings — they must not be mischaracterized as prior art despite appearing in the "Citations" table.

If you'd like, I can pull full claim text for KR 10-2017-0123065 A and KR 10-2001-0090519 A to firm up the § 102 mapping on those two, which are the least-documented but potentially most probative references.

I'll analyze the obviousness of US 11,326,938 by first gathering grounding on the key prior-art references cited on the patent page.

Let me gather grounding on the remaining cited references (Apple, Ricoh, Kookmin illumination sensor) and the closely related circadian-action-factor module.

Obviousness Analysis — U.S. Patent No. 11,326,938 B2

"Bio Illuminance Measuring Device for Determining Circadian Action" (Kookmin University / 35 U.S.C. §103)

Note: This section builds on the earlier "Litigation summary" only insofar as it establishes there is no post-grant challenge (no IPR/PGR/CBM) to rely on. The prior-art set is taken from the prior-art section of the source page (Citations (7), Family Cites Families (1), Non-Patent Citations (2), plus the "Similar Documents" list), interpreted literally per the operating rules. I do not auto-correct any identifier, and I flag where a reference's date or applicability cannot be verified.

1. Effective filing date and the prior-art window

• US 16/525,190 filed 2019-07-29, claiming KR priority 10-2019-0008809 (2019-01-23). Under AIA §102, the effective filing date is 2019-01-23.
• Any reference published/issued, or filed-and-published, before 2019-01-23 is available.

Reference that cannot be used: US 2020/0230346 A1 (Kookmin, "Circadian rhythm management apparatus and system," app. 16/525,183) — same applicant, same effective filing date (2019-01-23), parent US 11,305,091 B2. It is not §102/§103 prior art against the '938 (same-day, common-ownership family member); it is listed only because it is a family/similar document. It must be excluded from any combination.

2. Level of ordinary skill (Graham factor 1)

A POSITA here would be a person with a B.S. in electrical/optical engineering (or equivalent) and 2–4 years' experience in photometry, ambient-light sensing, and human-centric/circadian lighting measurement, including: (a) shaping photodiode spectral response with optical filters, (b) dark-current/offset compensation and ADC front-ends for photodiodes, and (c) familiarity with photopic V(λ) and the melatonin/circadian action spectrum C(λ) and the derived figures of merit (CER, CAF). All the cited art is squarely in this person's field.

3. Scope of the claims

Claim 1 (as issued) requires: (i) a circadian lambda filter; (ii) a visual lambda filter; (iii) a photo sensor generating a circadian signal = (circadian wavelength voltage − dark voltage) and a visual signal = (visual wavelength voltage − dark voltage); (iv) an illuminance calculator computing a C/V ratio, applying that ratio to a circadian action function that varies with the visual wavelength signal to obtain CAF, and computing bio illuminance from CAF; (v) a function storage holding multiple circadian action functions, one per range of the visual wavelength signal; and (vi) both signals being analog signals (voltage differences). Dependent claims add: function caller (2); CAF as dependent variable of the ratio (3); proportionality (4); visual-illuminance calculator (5); Biolux = CAF × Lux (6); three photodiodes incl. a dark photodiode (7); three variable resistors (8); controller varying the resistors' resistance and A/D-converting (9); display (10); melatonin/cortisol & 400–600 nm "circadian band" (11); human-eye visual band (12).

Notably, the file history shows a non-final office action (2021-12-13) followed by an amendment/response (2022-03-10) and allowance (2022-03-29). That sequence is consistent with the broad concept having been met by art, with allowance turning on the added dark-voltage-difference and function-storage recitations — both of which, as shown below, are themselves known.

4. Element-by-element mapping

5. Grounds of rejection under §103

Ground A — Salutron '282 in view of Apple '767, further in view of Oh (2014) and Brown '127

Claim 1. Salutron teaches the core sensor concept: a photodiode behind a filter shaped to the ipRGC (melanopsin) action spectrum — a circadian lambda filter — and it expressly contemplates additional channels ("using the optical sensor, or a further optical sensor, to detect one or more additional colors… besides blue light"). Adding a visual V(λ) channel is the natural companion channel (Oh/Brown define V(λ) as the reference photopic channel). Apple supplies the exact claimed signal conditioning: multiple spectrally-different photodiodes plus a dark-current photodiode whose signal is subtracted, digitized by ADCs. Oh supplies the CAF/CER figures of merit and the C(λ)/V(λ) framework; Brown expressly states CAF is calculated from C(λ) and V(λ). Thus the C/V ratio, the CAF derived therefrom, and Biolux = CAF × Lux are the predictable application of known figures of merit to two known filtered channels. The V-segmented function storage is met by Apple's segmented-calibration teaching (selecting a stored function/matrix based on light-source/CCT class).

Motivation to combine (KSR):

• Same field, same problem: all three address quantifying/acting on circadian-effective light received by a person.
• Salutron itself suggests the combination by proposing a further sensor for additional wavelengths and by using a display to tell the user about light exposure.
• Apple solves an error problem the '938 admits needing: dark-current subtraction for multi-photodiode sensors — the '938's own specification says the dark-voltage subtraction "remove[s] self-generated noise inherent in a photo diode." A POSITA seeking accuracy would apply Apple's known dark-current subtraction.
• Predictable results: combining a C(λ) channel with a V(λ) channel yields, by definition of CER/LER, a signal ratio monotonically related to CAF; no new structural or functional result arises.

Ground B — Brown '127 (with Oh) as the primary reference, in view of Apple '767 and Salutron '282

Use Brown as the primary because it names C(λ), V(λ) and CAF and states they are used to calculate the CAF, and because it "receive[s] sensed light values." A POSITA implementing a measuring (not control) device from Brown would (a) provide the two action-spectrum filters (Salutron/Oh) and (b) use Apple's dark-current-corrected multi-photodiode front end. This is an obvious substitution of a compact filter-plus-photodiode implementation for the reference spectrometer Brown/Oh use, motivated by the express design pressure the '938 recites (avoid "an expensive reference spectrometer… limitation in miniaturization").

Dependent-claim grounds

• Claims 2–4 (function caller; CAF dependent on the ratio; proportionality): Applying the known linear/log relationship between a circadian/visual channel ratio and CAF (Oh; Brown) and storing/selecting the appropriate relationship is routine implementation. Apple's selection among stored calibration matrices per light-source class supplies "calling the function corresponding to the visual wavelength signal."
• Claims 5–6 (visual illuminance calculator; Biolux = CAF × Lux): Visual illuminance from the V(λ) channel is the classic photometer function (Shimamura; Apple; Oh). Biolux = CAF × Lux is CAF = CER/LER applied directly.
• Claims 7–8 (three photodiodes incl. dark; variable resistors): Apple '767 (dark photodiode, subtraction) + Ricoh '098 / conventional transimpedance-to-variable-resistor I–V conversion.
• Claim 9 (controller varies resistances for gain, converts to digital): Apple's ADC digitization + Ricoh's gain/variable-resistor teaching; the purpose (keeping signals within a fixed processing range across widely varying illuminance) is a recognized engineering necessity, not an inventive contribution.
• Claim 10 (display): Salutron.
• Claims 11–12 (hormone/400–600 nm; visual band): Salutron's ipRGC/melanopsin rationale and Oh/Brown's C(λ) and V(λ) action spectra.

6. Anticipated counterarguments and where the case is strongest/weakest

Applicant's likely rebuttal: No cited reference discloses or suggests deriving CAF from the ratio of two broadband-filtered channel signals; Oh/Brown compute CAF from a full spectral power distribution (spectrometer), so the C/V-ratio shortcut is a novel algorithmic insight.

Rebuttal analysis: This is the weakest point for the examiner and the strongest for the applicant, because the CAF↔C/V-ratio relationship is the central limitation of independent claim 1. However, it is a mathematical consequence: with a detector whose responsivity is C(λ) and one whose responsivity is V(λ), the ratio of the two weighted integrals is proportional to CER/LER = CAF (up to per-channel gain). A POSITA armed with Oh's definitions and Brown's statement that CAF is calculated from C(λ) and V(λ) would recognize the ratio approach as an obvious implementation — precisely the "simple method… omitting an additional component [spectrometer]" the '938 touts as its advantage. Under KSR, such predictable, definition-driven computation is obvious; the allowance appears to have rested on the narrower dark-voltage-difference and V-segmented function-storage recitations, both of which are independently taught (Apple dark-current subtraction; Apple segmented calibration by light-source type). Strengthen the record by citing Oh for the CAF definition and the Apple '767 publication/parent for the dark-current photodiode and segmented calibration.

Secondary considerations (Graham): None of record. The earlier litigation review found no litigation, license, or PTAB proceeding on the '938, so there is no nexus-bearing evidence of unexpected results or commercial success to rebut the prima facie case.

7. Conclusion

• Claim 1 is likely obvious over Salutron '282 + Apple '767 + Oh (2014), and alternatively over Brown '127 (+ Oh) + Apple '767 + Salutron '282. Motivation: same field; Salutron's own suggestion of an additional sensor channel; Apple's dark-current subtraction solves an admitted accuracy problem; Oh/Brown supply the CAF/C(λ)/V(λ) framework yielding predictable results; and a strong KSR design incentive to replace an expensive spectrometer with filtered photodiodes.
• Claims 2–12 are obvious as set out in §5, each adding only known, predictable details (function selection/segmented calibration, photopic illuminance computation, Biolux = CAF × Lux, dark photodiode, variable resistors, gain control + ADC, display, and the known melatonin/hormone and photopic action spectra).
• Excluded: US 2020/0230346 A1 / US 11,305,091 B2 (same-family, same-day — not §103 prior art).
• Unverified documents to check before relying on them: the "Similar Documents" items KR 10-2166726 B1 ("Module For Measuring Circadian Action Factor And Bio Illuminance Measuring Device") and KR 10-2082050 B1 ("Apparatus For Measuring Circadian Illuminance") appear highly on-point, but I could not verify their priority/publication dates or assignees from the fetched material, so I do not rely on them here. If either predates 2019-01-23 and is third-party, it would materially strengthen (and in some claim charts, simplify) the rejection; if it is a later Kookmin filing, it cannot be prior art.

Confidence: High on the disclosure content of the cited references and on the legal framework; moderate on ultimate outcome for claim 1 because the C/V-ratio central limitation is not literally taught (it requires the KSR "obvious computation" step), while claims 7–9 (dark photodiode / variable resistors / gain + ADC) and claims 10–12 are strongly obvious.