US 11422382

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

Let me run more targeted searches on litigation and the specific patent record.

US 11,422,382 B2 — Analyst Summary

1. Bibliographic identification (as published)

Source note: the above is taken from the full text of the patent as fetched from https://patents.google.com/patent/US11422382/en, which is the authoritative record supplied to me and takes precedence over any search result.

2. Abstract (as published)

"An alignment module includes a polarized dichroic mirror, a dichroic mirror, a phase retarding component and several light sources. The polarized dichroic mirror reflects specific beams and is passed by other beams. The dichroic mirror is disposed on a side of the polarized dichroic mirror, and reflects specific beams and is passed by other beams. The phase retarding component is disposed by the dichroic mirror and adapted to transform a polarization state of a beam passing the phase retarding component. The light sources are respectively disposed on different sides of the polarized dichroic mirror, and respectively output a beam toward the polarized dichroic mirror. The polarized dichroic mirror can transmit some beams toward a light pipe, and further transmit other beams toward the dichroic mirror and the phase retarding component for transforming the polarization state and then transmit the transformed beams toward the light pipe."

3. Independent claims — plain-language overview

Claim 1 (first embodiment, module 10A) — "reflect‑type" layout.
A polarized dichroic mirror reflects a first beam and a second beam that is in a first polarization state, while transmitting the second beam when it is in a second polarization state, and also transmitting a third beam. A (non‑polarizing) dichroic mirror sits on one side of the polarized dichroic mirror, reflects the second beam and transmits the third beam. A phase‑retarding component (quarter‑wave plate per claim 3) sits between the two mirrors and flips the polarization state of the second beam as it passes. Three light sources: a first source on the far side of the polarized dichroic mirror outputting the first beam; a second source on the opposite side outputting the second beam in the first polarization state; and a third source positioned so that the phase retarder lies between it and the polarized dichroic mirror, outputting the third beam that reaches the polarized dichroic mirror by passing through the dichroic mirror and the phase retarder. In the worked example the first beam is red (reflected straight to the light pipe), the second is blue (routed around a loop, polarization‑flipped on the double pass through the retarder, then transmitted), and the third is green (passes straight through both mirrors and the retarder).

Claim 5 (second embodiment, module 10B) — "dual polarized dichroic mirror" layout.
Same general architecture but rearranged so that the first beam (e.g., red) is the one that is polarization‑recycled rather than the second. A first polarized dichroic mirror reflects the first beam in the first polarization state and transmits it in the second polarization state, plus transmits the second and third beams. A dichroic mirror reflects the first beam and transmits the second and third beams, with a first phase retarder between them. A second polarized dichroic mirror is placed on the opposite side of that dichroic mirror and reflects the second beam and the third beam when they are in the first polarization state, while transmitting them in the second polarization state. Claim 5 is written in "at least one" language for all three light sources, so it covers the symmetric two‑source‑per‑color arrangement of FIG. 2 as well as a single‑source‑per‑color variant. The third light source is expressly recited as being disposed adjacent to the second light source.

Claim 15 (fourth embodiment, module 10D) — two‑stage cascaded version.
This is a cascade of two of the claim‑5‑type stages. It adds a second dichroic mirror on the far side of the second polarized dichroic mirror, plus a second phase‑retarding component between them. The first stage recycles the first beam (reflect when in first polarization state / transmit when in second); the second stage recycles the second beam (the second polarized dichroic mirror reflects the second beam in the first polarization state and transmits the third beam and the second beam in the second polarization state; the second dichroic mirror reflects the second beam and transmits the third beam; the second retarder flips the second beam's polarization on the double pass). Again "at least one" for the first and second light sources; the third light source sits on the side of the second dichroic mirror opposite the second polarized dichroic mirror.

Claim 18 (fifth embodiment, module 10E) — inverted "transmit‑first" logic.
Mirror roles are swapped relative to claim 1. The polarized dichroic mirror transmits the first beam and the second beam in the second polarization state, and reflects the second beam in the first polarization state and the third beam. The dichroic mirror transmits the first beam and the third beam and reflects the second beam, with a phase retarder between them. The second light source outputs the second beam so that it arrives at the polarized dichroic mirror in the second polarization state (i.e., it starts on the transmit path, is reflected back through the retarder by the dichroic mirror, and is thereby converted to the first polarization state so the polarized dichroic mirror can re‑direct it to the output). The third light source is again positioned with the phase retarder between it and the polarized dichroic mirror.

4. Dependent-claim map

• On claim 1: cl. 2 optical equalizer (light pipe) for receiving/equalizing all three beams; cl. 3 quarter‑wave plate; cl. 4 the 45° normal‑vector angles between the polarized dichroic mirror and each of the dichroic mirror and the phase retarder, with the specification allowing ~10% tolerance ("equal to or similar to forty‑five degrees").
• On claim 5: cl. 6 optical equalizer; cl. 7 quarter‑wave plate; cl. 8 two first light sources on opposite sides of the first polarized dichroic mirror; cl. 9 + cl. 10 and cl. 12 + cl. 13 a second phase‑retarding component (half‑wave plate or two quarter‑wave plates) placed between the second polarized dichroic mirror and the second (or third) light source; cl. 11 two second light sources (one first‑polarization, one second‑polarization); cl. 14 two third light sources (one first‑polarization, one second‑polarization). Clauses 9–14 are what cover the third embodiment (10C), which is not given its own independent claim.
• On claim 15: cl. 16 optical equalizer; cl. 17 both phase‑retarding components as quarter‑wave plates.
• On claim 18: cl. 19 optical equalizer; cl. 20 quarter‑wave plate.

5. Prosecution, family and citation profile

• Prosecution was fast: non‑final office action 2022‑01‑28; response 2022‑04‑20; notice of allowance 2022‑04‑29; issue fee paid 2022‑07‑11; granted 2022‑08‑23. Only one substantive office action of record.
• Nine cited references, all examiner citations, none of them obviously close art for a polarization‑recycling combiner: US 4,995,718 A (Honeywell); US 5,903,388 A (Sedlmayr); US 2002/0176054 A1 (Mihalakis); US 2006/0164607 A1 (Morejon); US 2006/0244871 A1 (Tzong); KR 20080004990 A (Samsung); US 2011/0235175 A1 (Poon — "Polarization converting color combiner"); WO 2013/016163 A2 (3M); US 2015/0168817 A1 (Corning).
• One third‑party family citing this family: TWI 808722 B (明基電通股份有限公司 / BenQ), "Light mixing module and laser projector having the same," priority 2022‑04‑13, published 2023‑07‑11.
• Sibling/related Qisda filings in the "similar documents" cluster include US 11,360,371 B2 ("Laser mixing module"), US 11,531,261 B2 ("Projector with preferred lighting uniformity"), US 11,199,763 B2, and US 12,001,129 B2 ("Light mixing module and laser projector having the same"). I mention these only as context; they are different patents and I have not verified their claim scope.

6. Litigation / CAFC 2026 docket check — negative result, with stated limits

I found no litigation, and specifically no CAFC 2026 appeal, involving U.S. Patent No. 11,422,382. My searches for the number and for the patentee returned:

• a Netherlands traffic‑fine appeal (ECLI:NL:RBZWB:2025:7716) whose case number happens to be 11422382 — unrelated;
• Chinese corporate‑penalty and Hong Kong exchange filing records containing "11422382" as an internal row/filing identifier — unrelated; and
• decades‑old Qisda/BenQ party records from the Thomson Licensing ITC investigation (Inv. Nos. 337‑TA‑741/749) and D. Del. 1:10‑cv‑00789 — unrelated to this patent.

In other words, the string "11422382" is not unique, and I did not find a single authoritative hit tying this patent number to a court or Commission proceeding. I was not able to query USPTO PatentCenter, the CAFC opinions/orders database, or Docket Navigator directly from this environment; I relied on open web search. Treat "no litigation found" as an absence of evidence in the sources I could reach, not as a verified statement that no suit, IPR, or appeal exists. If you need a definitive clearance, the searches to run are: CAFC docket by patent number, PACER/Docket Navigator for the patent number, and the USPTO PTAB API for IPR/PGR petitions naming 11,422,382.

7. Explicit uncertainty flags

1. Litigation status: unverified (see §6). No CAFC 2026 activity found; do not treat as confirmed none.
2. Title spelling: "Aligement" is reproduced literally from the publication; it is a typographical error in the patent document, not in my transcription.
3. Summary-of-invention drafting quirk: the "second" aspect in the Summary section (and claim 15 predecessor text) recites "a first polarized dichroic mirror, a first polarized dichroic mirror, a first phase retarding component, a second polarized dichroic mirror, a second dichroic mirror…" — the duplicate "first polarized dichroic mirror" is in the source text. The claims themselves are internally consistent.
4. CN 114200752 B is the Chinese family member; I have its grant date (2024‑10‑18) from the Google Patents family table but have not inspected its claims, so I cannot say whether the Chinese and US claim sets are coextensive.
5. No reissue, reexamination, disclaimer, or terminal disclaimer appears in the legal‑events list supplied; I did not independently verify the absence of later‑recorded documents after the 2026‑02‑11 maintenance‑fee entry.

Bottom line: US 11,422,382 B2 is a Qisda Corporation projector‑optics patent (filed 2020‑12‑07, granted 2022‑08‑23, in force through at least 2040‑12‑07) claiming several hardware arrangements that mix red, green, and blue laser beams by using one or more polarization‑selective dichroic mirrors together with a non‑polarizing dichroic mirror and a phase retarder, so that one color is routed on a folded double‑pass path through the retarder to flip its polarization and be recombined with the other colors at the output light pipe. The four independent claims correspond to four distinct optical topologies disclosed in FIGS. 1, 2, 4, and 5.

I'll search for litigation involving US patent 11422382 across patent litigation databases.

Initial searches didn't surface litigation for this patent. Let me search litigation-specific databases more directly.

Let me do a few more targeted checks before concluding.

Litigation Search Results: US Patent No. 11,422,382

No litigation involving U.S. Patent No. 11,422,382 was found.

I searched patent-litigation–oriented sources (including court docket aggregators like CourtListener/RECAP, Justia Dockets, Stanford NPE Litigation Database, and general web searches keyed to the exact patent number and to the assignee) and found no district court complaints, ITC Section 337 investigations, PTAB proceedings (IPR/PGR), or CAFC appeals in which this patent is the patent-in-suit or otherwise at issue.

Because I could not locate any case, I cannot provide the requested case-by-case fields (plaintiff, defendant, jurisdiction, case number, filing date, status) — doing so would require fabrication.

What the searches did surface (and why it is not this patent)

To be transparent about the noise and near-misses, the following results are not litigation over US 11,422,382:

• A Dutch traffic-fine appeal, ECLI:NL:RBZWB:2025:7716, Rechtbank Zeeland-West-Brabant, dated 01-08-2025. The string "11422382" appears here only as a Dutch zaaknummer/CIJB case ID in a criminal/administrative fine matter — unrelated to any patent.

• A Chinese corporate-penalty record listing "11422382" as an internal penalty ID — unrelated.

• Qisda/BenQ litigations involving other patents (Qisda Corporation is the assignee of US 11,422,382, so these appear in searches):

  • Thomson Licensing SAS et al. v. Qisda Corporation et al., No. 1:10-cv-00789 (D. Del., filed Sept. 16, 2010) — asserted the '674, '063, '006, '556, '941 patents; also ITC Inv. No. 337-TA-… (Certain Liquid Crystal Display Devices…), Commission determination June 14, 2012.
  • Alpine Technology Holdings Limited et al. v. BenQ Corporation et al., No. 3:24-cv-00403 (N.D. Tex., filed Feb. 20, 2024) — a declaratory-judgment action against BenQ/Qisda; the patents-in-suit were not identified as US 11,422,382 in the sources retrieved.

  None of these identify US 11,422,382 as a patent-in-suit.

Context on the patent itself

• US 11,422,382 B2, "Alignment module with preferred illumination uniformity applied for a projector" (note: the granted title contains the typo "Aligement"). Inventors: Jia-Ming Zhang, Ching-Tze Huang, Tsung-Hsun Wu. Assignee: Qisda Corporation. App. No. 17/114,434, filed Dec. 7, 2020; priority CN 202010910530.5 (Sept. 2, 2020); granted Aug. 23, 2022. Status: Active (4th-year maintenance fee paid per the record; anticipated expiration 2040-12-07).
• The family includes Chinese counterpart CN114200752B (granted 2024-10-18).
• Public records show this patent has not been asserted; only one later patent (TWI808722B, assigned to 明基電通股份有限公司 / BenQ) cites the family.

Caveats

My search relied on publicly indexed litigation databases and web search; it is not a substitute for a full PACER/Docket Navigator/Unified Patents Portal docket query or a Chinese litigation database check on the CN family member. It is possible a very recently filed complaint or a Chinese action is not yet indexed. If you need certainty, I recommend running the patent number directly through PACER, Docket Navigator, the Unified Patents litigation portal, and (for the CN counterpart) China's court/judgment databases. Based on the evidence available today, however, the answer is that no litigation is known for US 11,422,382.

PTAB / AIA Trial Activity — US 11,422,382 B2

Proceedings overview

Total AIA trial proceedings on file: 0 — no IPRs, no PGRs, no CBM reviews, i.e. zero active, zero invalidated, zero sustained, zero settled, zero institution denials. Per the structured "PTAB proceedings on file" block, the USPTO Open Data Portal returns no AIA trial proceedings for US 11,422,382 as of the most recent ingest, and my independent web searches surfaced nothing to the contrary. Bottom-line defensive posture: the patent is entirely untested at the PTAB — all 20 claims stand intact and unadjudicated. For a defendant, that cuts both ways: nothing is canceled, but there is also no Board precedent, no construed claim terms, and no petitioner-side estoppel tying anyone's hands. If you are being asserted against, the full statutory menu of validity challenges is still on the table — with the important caveat that the PGR window has closed.

(Ordering note: the template calls for one section per proceeding, most-impactful first. There are no proceedings to enumerate, so the sections below cover the absence itself, what that forecloses, and the surrounding posture.)

(none) — No petitioner v. Qisda Corporation

• Type: N/A — no Inter Partes Review, Post-Grant Review, or Covered Business Method review has been instituted or filed on US 11,422,382.
• Filed: N/A
• Status: N/A (the patent itself is Active per ODP; 4th-year maintenance fee paid 2026-02-11, anticipated expiration 2040-12-07)
• Judge panel: N/A — no APJ panel has been assigned.
• Petition grounds: N/A — no claims have been challenged under § 102, § 103, or § 112.
• Institution decision: N/A
• Final Written Decision: N/A — no claim of this patent has ever been canceled, confirmed, or even construed by the Board.
• Settlement / termination: N/A
• Appeal: N/A — no FWD exists to appeal, and no CAFC docket number attaches to this patent.
• Defensive value: Neutral-to-favorable for a defendant. There is no Board ruling you must work around and no adverse claim construction that binds anyone, but you also get no free "claim 1 is dead" argument. Any IPR you file would be a first impression on this specification, before a panel that has never seen the art.

Verification notes / caveats. Searches for the patent number combined with "IPR," "PTAB," "Qisda," and "alignment module" returned only unrelated documents (a Dutch traffic-fine decision, a Ninth Circuit prisoner case, and a PUBMED ID that happens to be 11422382 — none of which relate to this patent). I found no evidence of district-court assertion of this patent either. I cannot rule out a very recently filed petition that has not yet been indexed by ODP; if a petition were filed in the last few weeks, it would appear on PTAB E2E (https://ptacts.uspto.gov/ptabweb/) before it propagated elsewhere. I am not aware of any such filing, and I will not invent a docket number to fill the gap.

Strategic summary

What is canceled vs. sustained vs. untested. Nothing is canceled; nothing is sustained by adjudication. All 20 claims are untested. The claim set maps cleanly onto the five disclosed embodiments: claims 1–4 (FIG. 1, module 10A — single polarized dichroic mirror 12 / dichroic mirror 14 / quarter-wave component 16); claims 5–14 (FIG. 2 and FIG. 3, modules 10B/10C — first polarized dichroic mirror 26, dichroic mirror 28, first phase retarding component 30, second polarized dichroic mirror 32, with claim 9 adding a second phase retarding component and claim 10 specifying it as a half-wave plate or two quarter-wave plates); claims 15–17 (FIG. 4, module 10D — adding second dichroic mirror 52 and second phase retarding component 54); and claims 18–20 (FIG. 5, module 10E — the mirror-image geometry where the polarized dichroic mirror 64 passes the first beam and reflects the second/third beams). The four independent claims are 1, 5, 15, and 18. Note that issued claim 1 is narrower than the summary-of-invention text in the same document: the granted claim requires that "the phase retarding component being disposed between the third light source and the polarized dichroic mirror" and that the third beam reach the polarized dichroic mirror "by passing through the dichroic mirror and the phase retarding component" (rather than the looser "by means of" phrasing used for claim 1's ancestor in the Summary). That narrowing is a prosecution artifact, not a Board artifact.

Estoppel landscape. Because no IPR/PGR was ever filed, § 315(e)(2) estoppel does not exist on this patent. No petitioner, real party in interest, or privy is barred from raising anything. A defendant today faces no estoppel shadow at all and can raise § 102/§ 103/§ 112 grounds at the PTAB or in district court without restriction — subject only to the ordinary statutory bars: (i) § 315(b)'s one-year clock from service of a complaint alleging infringement (which would bar an IPR, not a § 282 district-court defense), and (ii) § 325(a) if you were the first-named inventor or an assignee/obligee, which is inapplicable here. The more consequential timing fact is the PGR window: the patent issued 2022-08-23, so the nine-month post-grant review window closed 2023-05-23, and no § 112 or § 101-based PGR is available to you anymore. Your AIA-trial tools are IPR-only (§ 102/§ 103 on patents and printed publications). CBM is unavailable as a matter of law — the CBM program's sunset was 2020-09-16 and this is a projector optical-engine patent, not a financial-services data-processing patent.

Pattern signals. There is no pattern, because there is no activity. No repeat petitioner, no serial petitioner, no defensive aggregator. That in itself is informative: US 11,422,382 is a manufacturer's patent, not a monetization patent — assignee Qisda Corporation (Taoyuan, TW) is one of the world's largest projector ODMs, and the disclosure is core product-design subject matter, not a licensing vehicle. Well-asserted troll portfolios reliably attract IPRs within 12–24 months of a campaign; this one has attracted none, which is consistent with it never having been asserted against anyone. One forward-looking signal worth watching: the similar-documents list for this family includes TWI808722B ("Light mixing module and laser projector having the same," 明基電通股份有限公司 / BenQ) and US 12,001,129 B2, plus thick citation traffic between Qisda's and BenQ's projector portfolios. If this patent ever becomes the tip of a competitor-to-competitor assertion (Qisda or a BenQ-side entity), that competitor's own portfolio is where the invalidating art and the petition are most likely to come from. Treat that as a hypothesis, not a finding.

Recommended next steps

• If you are a defendant being asserted against: there is no FWD to cite and no claim to describe as canceled — be explicit about that with your client and with the court. Do not represent that any claim of US 11,422,382 has been invalidated by the PTAB. Any val

I'll search the assignment and ownership records for this patent before writing anything up.

Let me try more targeted queries on the assignment record and Qisda's ownership.

Let me try to reach the assignment record's bibliographic detail and check for any post-issuance transfers.

Inventors

Pattern notes:

• All three inventors signed the assignment between 2020-11-19 and 2020-11-23 — i.e., all three executed within the same ~5-day window, roughly two-and-a-half months after the CN priority filing (2020-09-02) and about two weeks before the US application was filed (2020-12-07). This is the ordinary "assignment-before-filing" housekeeping pattern at a Taiwanese ODM, not a pre-liquidation departure.
• There is no evidence of inventor departure from Qisda. No subsequent assignment, quitclaim, or nunc-pro-tunc confirmation naming these inventors appears in the record, which is what a departure-driven cleanup usually generates.
• Caveat: the assignment records do not state an employer field; "Qisda Corporation" is an inference from the assignment chain, not a verified employment record. I did not locate a Qisda employment directory, inventor declaration, or PEDS inventor-residence record confirming employment titles.

Original assignee

Qisda Corporation (佳世達科技股份有限公司), No. 157, Shanying Rd., Guishan Dist., Taoyuan City, Taiwan. Listed on the Taiwan Stock Exchange (ticker 2352).

• Primary line of business: ODM/EMS contract manufacturing. Qisda is the former BenQ Corporation, renamed Qisda on 2007-09-01 when the BenQ brand business was spun off into a separately named BenQ Corporation (wholly owned by Qisda). Qisda's own corporate communications describe it as an ODM/EMS provider and a top-tier manufacturer in LCD monitors, projectors, and multifunction printers; its annual reports state its projector ODM business held a top-2 worldwide position (2012 report) and it was "the only manufacturer in Taiwan … experienced in LCD projector mass production," with laser projectors explicitly listed among its 2012–2013 development roadmap items.
• Product embodying the claims: the patent is directed to a projector light-mixing/alignment module (polarized dichroic mirror + dichroic mirror + retarder mixing RGB laser beams into a light pipe). Qisda manufactures laser projectors of exactly this architecture class as an ODM; the specification's stated motivation ("trends of miniature form and high brightness," RGB laser sources with wide color gamut) tracks Qisda's commercial projector line. I did not find a public teardown or part-level identification tying a specific Qisda/BenQ model to claims 1–20, so this is corroborating context rather than a documented product mapping.
• Current status: operating, publicly traded, not dissolved, not in bankruptcy. Qisda is an active filer and remains the assignee of record; the 4th-year maintenance fee on this patent was paid (due date window 2026-02-11 per the record), so the patent is being actively maintained at cost.

Assignment timeline

The USPTO assignment record for US 11,422,382 contains exactly one recorded assignment. There is no post-issuance transfer of any kind — no security agreement, no merger, no change of name, no license recordation, and no transfer to a licensing entity.

• 2020-11-19 to 2020-11-23 (executed; three signing dates within that window) / recorded 2020-12-07 — Reel 054569 / Frame 0878
  • Conveyance: Assignment (original inventors → applicant)
  • Assignor: Zhang, Jia-Ming; Huang, Ching-Tze; Wu, Tsung-Hsun (three individual inventors)
  • Assignee: QISDA CORPORATION, Taiwan
  • Correspondent: not determinable from the sources I could reach. The assignment's free-format text as indexed reads: "ASSIGNMENT OF ASSIGNORS INTEREST; ASSIGNORS: ZHANG, JIA-MING; HUANG, CHING-TZE; WU, TSUNG-HSUN; SIGNING DATES FROM 20201119 TO 20201123; REEL/FRAME: 054569/0878." No correspondent attorney or firm name is exposed in the indexed text, and I was unable to open the Assignment Center record's correspondent field directly. I am not going to guess a name. Note this is a single-record chain, so the "repeat correspondent" tell is structurally inapplicable regardless.
  • Context: initial acquisition — inventors assign to their employer-of-record before/around US filing. Not a fire-sale, not a reorg, not a securitization.

Everything after that is maintenance and prosecution events, not assignments: application filed 2020-12-07; non-final office action 2022-01-28; response 2022-04-20; notice of allowance 2022-04-29; issue fee 2022-07-11; patent granted 2022-08-23; 4th-year maintenance fee paid (event dated 2026-02-11 in the record). Pre-grant publication US2022/0066226A1 (2022-03-03).

Family note (not an assignment): the CN priority application CN202010910530.5 issued as CN114200752B on 2024-10-18. Qisda is the CN applicant; no transfer of the CN member surfaced. The only third-party citation of the family is TWI808722B (granted 2023-07-11, assignee 明基電通股份有限公司 / BenQ) — a citation by a different corporate entity, not an ownership link, and BenQ is an affiliate/brand spin-off rather than an arm's-length acquirer.

Timeline diagram

timeline
    title Ownership of US 11422382
    2020 : Inventors assign to Qisda Corporation
         : Recorded at reel 054569 frame 0878
         : US application filed
    2022 : Patent granted
    2024 : Chinese counterpart CN114200752B granted
    2026 : Fourth year maintenance fee paid

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The sole assignee is Qisda Corporation, a listed operating manufacturer (TWSE 2352), not an "IP/Licensing/Holdings/Ventures" entity. No registered-agent-service address appears; the recorded assignee address is Qisda's Taoyuan corporate address. There is no second link in the chain, so there is no shell to transfer to. Reel 054569/0878, recorded 2020-12-07.

2. Known asserter in the chain — not present. Qisda appears on no NPE list. Neither the assignor (three individuals) nor the assignee matches Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or Spangenberg entities. Qisda's litigation posture is defensive, not assertive: it was a defendant in Thomson Licensing SAS et al. v. Qisda Corporation et al., No. 1:10-cv-00789 (D. Del., filed 2010-09-16) and in the related ITC investigation (Commission determination 2012-06-14), and a defendant in Alpine Technology Holdings Limited et al. v. BenQ Corporation et al., No. 3:24-cv-00403 (N.D. Tex., filed 2024-02-20). No case naming US 11,422,382 as a patent-in-suit was located.

3. Repeat correspondent across the chain — not present / not applicable. The chain has one record; recurrence cannot exist. I could not retrieve the correspondent name for reel 054569/0878, so I cannot cross-check it against Unified Patents / RPX correspondent lists either. Stated plainly: unknown, and unverifiable from the sources I reached.

4. Cascading transfers — not present. Zero post-issuance assignments. The only transfer is the initial inventor→company assignment, executed 2020-11-19/23 and recorded 2020-12-07, and nothing since (as of the 2026-02-11 maintenance-fee event in the record).

5. Pre-litigation transfer — not present, and structurally inapplicable. There is no infringement suit naming this patent, so there is no litigation date to measure the 6-month window against; the single assignment predates the patent's grant (2022-08-23) by ~21 months and predates any conceivable assertion.

6. Bankruptcy fire-sale — not present. Qisda is an operating, publicly traded company; no Chapter 7/11 filing, no patent sale in bankruptcy proceedings, and no 363-sale docket involving this patent.

7. Privateering — not present. No transfer to an NPE; no SEC 10-K/8-K disclosure of a patent-monetization vehicle; no Patent Progress/EFF coverage of Qisda privateering. Qisda's disclosed patent activity is prosecution and defensive participation, not outbound assertion.

8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. Qisda remains the owner of record. (Corollary: the patent has not been neutralized by a defensive aggregator — it simply has never been asserted.)

Verdict

Insufficient data — under the rubric's own definition this is the correct pick, because the record contains only the original assignment.

Justification: the entire recorded chain is a single instrument — three inventors assigning to Qisda Corporation, executed 2020-11-19/23, recorded 2020-12-07 at Reel 054569 / Frame 0878. No further reel/frame entries exist, so none of signals 1–8 can be affirmatively found; the NPE analysis is not "negative on the merits" but empty of evidence in either direction, which is precisely what "Insufficient data" is reserved for. On the available record the practical reading is that of a non-asserted patent owned and maintained by an operating manufacturer (Qisda, TWSE 2352, an active projector ODM that paid the 4th-year maintenance fee), with no litigation, no NPE chain, and no defensive-aggregator transfer — but the affirmative NPE signals cannot be ruled in or out beyond what the single link shows.

Two honest limitations, flagged rather than papered over:

1. I could not retrieve the correspondent of record for reel 054569/0878 — the one field most diagnostic for NPE-adjacent activity. The reel/frame itself is taken from the patent's indexed legal events ("REEL/FRAME: 054569/0878"), which is reliable, but the correspondent field was not exposed in anything I could reach.
2. Verify directly at the USPTO Assignment Center before relying on this for a filing or opinion: https://assignmentcenter.uspto.gov/ (also indexed at https://assignment.uspto.gov/patent/index.html) — search by patent number 11422382 or application 17/114,434, and open the 054569/0878 record to capture the correspondent. This write-up should be treated as a null result with one verified link, not as a clean bill of health or as an NPE finding.

Cross-reference note: this section is consistent with the earlier litigation summary — no suit, ITC action, PTAB proceeding, or CAFC appeal involves US 11,422,382, which independently supports the "no assertion" conclusion and means signals 5 and 7 have no trigger event to point to.

I'll verify the citation metadata and reference details against live sources before analyzing.

Prior-Art Analysis — US 11,422,382 B2

Patent under analysis: US 11,422,382 B2 (Aligement module with preferred illumination uniformity applied for a projector — title reproduced literally, including the "Aligement" typo in the granted record). App. No. 17/114,434; filed 2020-12-07; priority CN 202010910530.5 (2020-09-02); granted 2022-08-23; assignee Qisda Corporation; inventors Jia-Ming Zhang, Ching-Tze Huang, Tsung-Hsun Wu. Status: Active (4th-year maintenance fee paid 2026-02-11; anticipated expiration 2040-12-07).

Scope note / how "each patent citation" was read. The record for US 11,422,382 carries two distinct lists, and I treat them separately because they are legally different kinds of art:

1. "Citations (9)" — the references of record in the US case (the nine listed under Patent Citations). Google Patents displays these with the * marker whose legend is "* Cited by examiner, † Cited by third party"; the fetched HTML flattens the marker into its own line, so I treat these as references that were before the examiner but I flag that I cannot, from this record alone, confirm for every row whether it was examiner-cited or applicant-submitted.
2. "Family Cites Families (9)" — nine CN/JP documents cited in the Chinese counterpart record (CN114200752B family). These are not US-case citations, but several are substantively closer art, so I analyze them in a secondary section.

Prosecution consequence worth stating up front: the record shows a single Non-Final Action (2022-01-28), a response (2022-04-20), and a Notice of Allowance nine days later (2022-04-29). Whatever these nine references were, the issued claims were allowed over them. That does not immunize the claims (§ 282(b) allows any prior art to be reasserted; the nine cited references are only a floor), but it does mean no reference below was treated as anticipatory of the issued independent claims.

Method caveat (transparency). I verified US 4,995,718; US 2011/0235175 A1; WO 2013/016163 A2; and US 2015/0168817 A1 against live full-text/PDF sources this session. My tool budget was exhausted before I could pull full text for US 5,903,388; US 2002/0176054 A1; US 2006/0164607 A1; US 2006/0244871 A1; and KR 20080004990 A. Those five are described from the patent's own citation record plus general knowledge, and are labeled [unverified full text] below. I have not invented any content for them.

The § 102 legal standard I applied

A reference anticipates a claim only if each and every element of the claim, arranged as recited, is disclosed in that single reference — expressly or inherently. Verdegaal Bros. v. Union Oil Co. of California, 814 F.2d 628, 631 (Fed. Cir. 1987); MPEP § 2131. Secondary considerations are irrelevant to § 102; obviousness (§ 103) is a separate inquiry from different prior art, combinations, and articulated reasoning. KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398, 418 (2007); In re Kahn, 441 F.3d 977, 988 (Fed. Cir. 2006).

Element checklist for the four independent claims

The crux is a single element (the "polarized dichroic mirror") that is simultaneously wavelength-selective and polarization-selective — reflecting the first beam and the first-polarization-state second beam while transmitting the second-polarization-state second beam and the third beam — with the second beam doing a double pass through a quarter-wave retarder off a behind-it non-polarizing dichroic mirror. That double-pass polarization flip is what lets the returning beam pass back through the very same element that reflected it.

Reference-by-reference analysis (the nine US case citations)

1. US 4,995,718 A — Honeywell Inc.

• Full citation: US 4,995,718, Full color three-dimensional projection display, Honeywell Inc. (granted as US 4,995,718 A).
• Dates: priority/filing 1989-11-15 (US 437,488); granted/published 1991-02-26. EP family member EP 0 428 971 (search report 1991-12-27). [Verified]
• Description: A bi-prism/quad-prism combiner for a 3D projection display. A broadband polarizing beam-combining coating (20) sits at a prism interface; dichroic coatings (32, 48) transmit red/green and reflect blue; quarter-wave retarders (30, 46) sit between a prism face and each dichroic coating. In the bi-prism embodiment, p-polarized blue 22 exits the prism, enters QWP 30, reflects off dichroic coating 32, and, on the return trip through QWP 30, emerges as s-polarized blue 26 that is then reflected by polarizing coating 20. Red/green beams pass through the dichroic coatings and QWPs and combine with blue. An electronically controllable (half-wave) retarder 16 time-multiplexes the blue channel for left/right eye views.
• Potentially anticipatory claim(s): None of claims 1–20 as issued. This is the closest of the nine to the polarization-flip sub-combination of claim 1 (polarization-selective coating + QWP + blue-reflecting dichroic, with the QWP double-passed), but it lacks: (a) a single element that is both wavelength- and polarization-selective in the claim-1 sense; (b) discrete light sources — its inputs are image sources (CRT/light valve), not "light sources" outputting beams as claimed; (c) a third light source positioned so that the retarder lies between it and the polarized dichroic mirror with the third beam traversing dichroic-then-retarder (in the '718 device the red/green path transits coating 32 and retarder 30, but the beam originates on the far side of the dichroic, not through both in the claim-1 sequence); and (d) any optical equalizer/light pipe (claim 2). § 103 use: strong as general evidence that polarization-conversion loops using QWPs between a polarizer and a dichroic were long known.

2. US 5,903,388 A — Sedlmayr, Steven R.

• Full citation: US 5,903,388, High efficiency electromagnetic beam projector and systems and method for implementation thereof, Sedlmayr, Steven R.
• Dates: 1992-06-11 → 1999-05-11. [unverified full text — described from the citation record]
• Description (per record/general knowledge): A reflective-LCD projection architecture using polarizing beamsplitters and dichroic mirrors to split and recombine color channels while managing polarization state, with retarders used in the recombination path for light-efficiency ("high efficiency") reasons.
• Potentially anticipatory claim(s): None. Even on its face this is a § 103 background reference for the generic proposition that projectors combine a polarizing beam splitter, dichroic color separation/combination, and retarders. It does not disclose the claim-1 three-source layout, the third-source/retarder spatial requirement, or the claim-5/15/18 second polarization-selective dichroic mirror. I flag this as the lowest-value of the nine and note I could not verify its disclosure this session.

3. US 2002/0176054 A1 — Mihalakis, George M.

• Full citation: US 2002/0176054 A1, Reflective liquid-crystal-on-silicon projection engine architecture.
• Dates: 1999-12-30 → 2002-11-28. [unverified full text]
• Description (per record): An LCOS projection engine architecture — typically PBS-cube based, with dichroic color splitting/combining and waveplates used to rotate polarization between passes.
• Potentially anticipatory claim(s): None. Same posture as Sedlmayr: background art for PBS + dichroic + waveplate engine configurations, not a disclosure of the claimed combiner topology or source placements. Relevant under § 103 only.

4. US 2006/0164607 A1 — Morejon, Israel J.

• Full citation: US 2006/0164607 A1, Light-emitting diode (LED) illumination system for a digital micro-mirror device (DMD) and method of providing same.
• Dates: 2005-01-25 → 2006-07-27. [unverified full text]
• Description (per record): An LED illumination system that combines multiple color channels with dichroics and delivers the mixed light to a DMD via homogenizing/integrating optics.
• Potentially anticipatory claim(s): None. Its relevance is narrow and two-fold: the dichroic combining of multiple solid-state color sources (claim 1's mirrors 12/14, claim 5's 26/28) and the light-pipe/integrator "optical equalizer" element of claim 2 (and claim 6/16/19). It says nothing about polarization-selective dichroics or the QWP double-pass loop, so it cannot anticipate even claim 1 read alone. § 103 candidate for the "receive and equalize" limitation only.

5. US 2006/0244871 A1 — Tzong, Ruey-Yau

• Full citation: US 2006/0244871 A1, Liquid crystal projection system.
• Dates: 2005-04-27 → 2006-11-02. [unverified full text]
• Description (per record): An LC projection system; the family/record context places it among polarization-managed, PBS-and-dichroic projection architectures.
• Potentially anticipatory claim(s): None. I flag this reference as the one where my confidence is lowest — I have the citation record but not the disclosure, and I will not characterize its teachings further than the record supports. Treat as § 103 background only.

6. KR 20080004990 A — Samsung Electronics Co., Ltd.

• Full citation: KR 20080004990 A, Laser display device with speckle reduction optical unit (삼성전자주식회사).
• Dates: 2006-07-07 → 2008-01-10. [unverified full text; Korean-language]
• Description (per record): A laser display device incorporating optics for speckle reduction in the laser illumination path.
• Potentially anticipatory claim(s): None. Its evidentiary value is at most the use of laser (RGB) light sources in a display illumination path — relevant background for the specification's statement that "the RGB laser light sources can have pure color index and wider color gamut," and marginally to the claim-1/5/15/18 "light source" elements. It discloses nothing of the polarization-selective combiner geometry, the retarder placement, or the source-adjacency relations. § 103, low value.

7. US 2011/0235175 A1 — Poon, Yarn Chee (3M Innovative Properties Co.)

• Full citation: US 2011/0235175 A1, Polarization converting color combiner; filed as US 13/129,893 (PCT/US2009/064931; WO 2010/059684 A2); priority to US provisionals 61/116,108 and 61/116,072. Sibling application US 2011/0242653 A1 (Ouderkirk et al.); the family includes granted US 8,654,444.
• Dates: priority 2008-11-19; PCT filed 2009-11-18; published 2011-09-29; application abandoned. [Verified]
• Description: A color combiner in which each color input passes through a color-selective dichroic filter, then a quarter-wave retarder orientated at 45°, then intercepts a reflective polarizer at approximately 45°. One polarization state passes through; the other is reflected, exits the prism, is circularly polarized by the QWP, reflects off a second color-selective dichroic filter (which reverses the circular handedness), and is converted back to the desired linear state by a second QWP pass — i.e., recycled and polarization-converted rather than lost. The combined output is a single polarized multicolor beam; an optional light tunnel or integrator may be placed at the output for uniformity. Light sources "can be lasers, laser diodes, organic LEDs, or non-solid-state lamps."
• Potentially anticipatory claim(s): This is the most on-point of the nine, and it is still not a § 102 reference to any independent claim. The disclosed dichroic-filter + reflective-polarizer + QWP-double-pass stack is functionally the claim-1 polarization-conversion loop, but the reference (a) splits the two functions across separate elements (a color-selective dichroic filter and a reflective polarizer) rather than the claimed single "polarized dichroic mirror"; (b) uses unpolarized LED inputs, so it does not disclose a "second light source … adapted to output the second beam so that the second beam with the first polarization state arrives the polarized dichroic mirror"; (c) has no third source arranged so the retarder sits between the third light source and the polarized dichroic mirror with the third beam passing dichroic-then-retarder (claim 1, and the parallel hook in claim 18); (d) contains no second polarization-selective dichroic mirror (claims 5, 15) and no second dichroic mirror/second retarder pair (claim 15); and (e) discloses the integrator only as optional, so the claim 2 "optical equalizer" limitation is not disclosed as part of the claimed arrangement. § 103 value: very high — this is the reference I would lead with in an obviousness theory against claims 1 and 18.

8. WO 2013/016163 A2 — 3M Innovative Properties Company

• Full citation: WO 2013/016163 A2, Illumination module, 3M Innovative Properties Company.
• Dates: priority 2011-07-22; published 2013-01-31. [Verified]
• Description: An illumination module with first/second/third light collimators for three light sources; first and second dichroic mirrors (explicitly "disposed on glass plates" or "on prism diagonals"); reflective polarizers perpendicular to the collimated beams and aligned to pass a first polarization direction; and, in claim 12, "a quarter-wave retarder disposed between the first dichroic mirror and the reflective polarizer, the quarter-wave retarder aligned at 45 degrees to the first polarization direction." The reflective polarizers recycle the orthogonal polarization direction of the first and third (and second) beams back into the useful path. Claim 15 specifies red, green, and blue beams; claim 23 recites an image projector with the combined output directed to an imager.
• Potentially anticipatory claim(s): None of claims 1–20. Substantively adjacent to claim 1 (three color sources + dichroic mirrors + reflective polarizer + 45°-oriented QWP + recycling), but: the polarizer is perpendicular to the beam (normal incidence), not a 45°-inclined "polarized dichroic mirror"; the dichroic mirror in the module is a separate element from the polarizer, not a combined polarized-dichroic element; no third-source/retarder-between geometry; no second polarization-selective dichroic mirror (claim 5) or second dichroic/retarder pair (claim 15). Notably, the express recitation of a perpendicular reflective polarizer is mild evidence of a contrary design trend relative to claim 4's requirement that the polarizer's planar normal be at 45° to the dichroic mirror's and the retarder's — useful defensive rhetoric, not a dispositive teaching-away. § 103 value: high.

9. US 2015/0168817 A1 — Corning Incorporated

• Full citation: US 2015/0168817 A1, Light multiplexer with color combining element, Corning Incorporated; granted as US 9,638,988 B2 (grant date not verified this session); family WO 2015/089157 A1, CN 105980913 A.
• Dates: priority US provisional 61/915,126 (2013-12-12); application filed 2014-12-09; published 2015-06-18. [Verified for the publication; grant date flagged]
• Description: A light multiplexing system with three laser color channels (red, blue, green) whose beams are made parallel and incident on a color-combining element formed of two pieces of solid transparent material having three dichroic coatings on external surfaces: a first coating reflects the first band and transmits the second and third; a second coating reflects the second band and transmits the third; a third coating reflects the third band. The three reflected beams combine coaxially as an output beam, which may then go to a light integrator (fly's-eye integrator or integrator bar) or other light homogenizer and to a spatial light modulator. The reference expressly discusses low angles of incidence as the design goal and states that "one or more phase retarders can be added to the different color channels, but only at the expense of added cost and complexity."
• Potentially anticipatory claim(s): None. It is a non-polarization-selective combiner — it teaches away from adding retarders on cost/complexity grounds — so it cannot disclose the claim-1 "polarized dichroic mirror," the claim-5 "first/second polarized dichroic mirror" pair, nor the claim-18 mirror-image geometry. Its independent value is two-fold: (a) it is good § 103 art for the stacked-dichroic-plate combiner sub-combination (three sources, dichroic(s) reflecting one band and transmitting others, coaxial output); and (b) it discloses an integrator downstream of the combiner, i.e., material for the claim 2/6/16/19 "optical equalizer" limitation. Its "cost and complexity" statement about retarders is also teach-away ammunition against a hypothetical § 103 combination that grafts a retarder onto a Corning-style stacked-dichroic combiner.

Anticipation scorecard

Bottom line on the nine: I found no § 102 anticipation. No single one of the nine references discloses all elements of claims 1, 5, 15, or 18 as issued, and the gap in each case is the same one — no reference discloses a single element that is both wavelength-selective and polarization-selective in the claim-1 sense, combined with the claimed source/retarder spatial relationships and (for claims 5 and 15) the second polarized dichroic mirror.

Secondary: the nine foreign references in the family record ("Family Cites Families")

These were cited in the CN114200752B (Chinese counterpart) record, not in the US case, and are listed for completeness. I did not verify their disclosures this session — treat all as [record only].

Two of these — CN 109188709 B ("polarized light conversion") and CN 107193177 B (Appotronics light-source system) — are, on their titles alone, plausibly closer to the claimed subject matter than several of the nine US-case citations. I could not verify their disclosures and am not asserting they anticipate anything; I flag them as the items a full validity study should retrieve first, along with the Qisda-affiliate CN 106647126 A.

Cross-check against the earlier-generated sections

• No contradiction with the earlier litigation and PTAB sections: nothing here changes the finding of zero assertions and zero AIA trials, and the § 315(e)(2) estoppel analysis is unaffected. The prosecution history is consistent with the earlier summary: single Non-Final Action (2022-01-28), response (2022-04-20), allowance (2022-04-29), issue fee (2022-07-11), grant (2022-08-23) — no appeal, no PTAB.
• Consistent with the earlier observation that issued claim 1 is narrower than the Summary-of-Invention ancestor text (the "phase retarding component being disposed between the third light source and the polarized dichroic mirror" hook). That narrowing is materially helpful here: it is precisely the limitation the closest two references — US 2011/0235175 A1 and WO 2013/016163 A2 — do not disclose.
• One nuance to add to the earlier strategic summary: the earlier note that the family cite list includes citation traffic between Qisda's and BenQ's projector portfolios is correct, and I would extend it — the citation record here also includes Qisda's own affiliate (CN 106647126 A, Qisda Optronics Suzhou), which is a § 102(a)(1)-eligible document rather than a grace-period-exempt inventor disclosure. That is a self-inflicted-art risk worth checking, not a finding of invalidity.

Ranked answer to "the most relevant prior art"

1. US 2011/0235175 A1 (Poon, 3M; family includes WO 2010/059684 A2 and US 8,654,444) — closest disclosure of the dichroic-filter/reflective-polarizer/quarter-wave polarization-conversion loop that claim 1 builds on. Best § 103 primary reference.
2. WO 2013/016163 A2 (3M) — three R/G/B sources, dichroic mirrors, reflective polarizers, a 45°-oriented quarter-wave retarder, and polarization recycling. Second-best § 103 primary reference; also a mild teach-away on the 45° polarizer orientation of claim 4.
3. US 4,995,718 A (Honeywell) — earliest clean teaching of a QWP double-pass polarization flip between a polarization-selective surface and a dichroic coating (blue channel), i.e., the mechanism claim 1 relies on.
4. US 2015/0168817 A1 (Corning; granted as US 9,638,988 B2) — stacked-dichroic three-source combiner plus downstream integrator; useful for the combining and "optical equalizer" elements, and teaches away from retarders on cost/complexity grounds.
5. US 2006/0164607 A1 (Morejon) — dichroic LED color combining into integrating optics toward a DMD.
6. US 5,903,388 A (Sedlmayr), US 2002/0176054 A1 (Mihalakis) — § 103 background on PBS/dichroic/waveplate projector architectures.
7. KR 20080004990 A (Samsung) — laser sources with speckle-reduction optics; source-technology background only.
8. US 2006/0244871 A1 (Tzong) — record only; lowest confidence; I will not characterize its disclosure.

Firm conclusions: (i) No § 102 reference to any of claims 1–20 exists among the nine US case citations; the closest two references fail the same two limitations (single combined polarized-dichroic element; the third-source/retarder spatial requirement). (ii) Any invalidity case on this patent must be built under § 103, most plausibly from US 2011/0235175 A1 or WO 2013/016163 A2 as primary, with US 4,995,718 A and/or US 2015/0168817 A1 as secondary. (iii) The most under-examined art is outside the US citation list — specifically CN 109188709 B and CN 107193177 B from the family record, which should be retrieved before any opinion is finalized. (iv) Where I lack the full text, I have said so and have not filled the gap with invented disclosure.

I'll ground this obviousness analysis in the actual cited prior art. Let me pull details on the references of record.

Let me pull the family-cited references that look most structurally on-point.

Obviousness Analysis — U.S. Patent No. 11,422,382 B2 under 35 U.S.C. § 103

Analysis date: April 26, 2026
Claim set analyzed: All 20 claims (independent claims 1, 5, 15, 18; dependents 2–4, 6–14, 16–17, 19–20)
Posture: No IPR/PGR has ever been filed (see prior sections), so there is no Board claim construction, no FWD, and no § 315(e)(2) estoppel. The obviousness analysis below is therefore a first-impression validity assessment built on the intrinsic record and the references of record — not a review of any adjudicated outcome.

1. Scope, sources, and explicit limitations

What I relied on. This analysis uses (a) the full specification and claim text of US 11,422,382 B2 as provided, (b) the 9 references in the "Citations (9)" block of the file wrapper, (c) the 9 references in the "Family Cites Families (9)" block, and (d) full-text/abstract retrievals for the four most structurally relevant references — US 2011/0235175 A1 (Poon, 3M), WO 2013/016163 A2 (3M), US 2015/0168817 A1 / US 9,638,988 B2 (Corning), and US 5,903,388 (Sedlmayr).

What I did not verify. I did not retrieve full claim or paragraph text for US 4,995,718; US 2002/0176054; US 2006/0064607; US 2006/0244871; KR 20080004990; or for the CN/JP family citations (CN 101051176, CN 102162982, CN 102402016, CN 107193177, CN 106647126, CN 107357125, CN 207181934, JP 6841269, CN 109188709). Where I characterize those, I do so only at the level supported by the record's titles and the snippets retrieved, and I flag it. I will not assert an element-by-element disclosure for a reference I have not read.

One reference I found that is not of record. During search I identified WO 2013/062930 A1 (3M) — a color combiner having first and second PBS, a quarter-wave retarder, a first tilted dichroic plate (first/second/third dichroic reflectors) and a second tilted dichroic plate, producing "a combined color light beam having the first polarization direction." This is directly germane to independent claim 15's two-PDM/two-DM architecture. I retrieved only partial text (claim excerpts), so I treat it as a lead for a full-text pull, not as an established disclosure. It was not cited by the examiner.

2. Claim construction anchors that drive the analysis

Two constructions control the outcome:

(a) "Polarized dichroic mirror" (PDM). The specification defines it functionally: it "has a specific polarized dichroic design, and can reflect the first beam B1 and the second beam B2_1 with a first polarization state, and allow passing of the second beam B2_2 with a second polarization state and the third beam B3." In plain terms it is a single optical interface carrying both a wavelength-selective (dichroic) coating and a polarization-selective (reflective-polarizer) function — reflecting red and blue-S, transmitting blue-P and green. This is the single most important limitation, because the art of record discloses the two functions separately and adjacently (Poon) rather than fused (the patent).

(b) The "double-pass phase-retarder loop." Claim 1 requires the phase retarding component between the PDM and the dichroic mirror, with the second beam reflected by the PDM → through the retarder → reflected by the dichroic mirror → back through the retarder → returned to the PDM in the second polarization state. This is a textbook polarization-conversion-by-recycling loop; the specification itself calls the retarder "a quarter wave plate," and the physics (two passes through a λ/4 plate with an intervening reflection = 90° net rotation) is elementary.

A third, smaller anchor: claim 4's "equal to or similar to forty-five degrees" with "ten percentage of tolerance" is broad enough that any reference stating "approximately 45 degrees" reads on it.

3. Level of ordinary skill in the art (PHOSITA)

A POSITA here is a person with a B.S. in optics, physics, or electrical engineering and 2–5 years of experience designing projector illumination/light-combining optical engines, or an M.S. with less experience — someone who would know (i) dichroic filter design and its polarization and angle dependence, (ii) reflective polarizers / multilayer optical film polarizers and their 45° use as beam splitters, (iii) quarter-wave and half-wave retarder placement for polarization rotation, and (iv) light-pipe/fly's-eye integrators for homogenization. That level of skill is fully consistent with the references of record, all of which are projector-illumination patents by 3M, Corning, Honeywell, Samsung, and Sedlmayr.

4. The prior art of record and its disclosure of the claim-1 elements

The table below maps claim 1's elements to the references of record. Poon and 3M '163 together cover every element except the specific source-side geometry.

Reading of the table: Poon's light path — unpolarized color light → dichroic filter → λ/4 retarder → reflective polarizer → (S reflects back) → λ/4 retarder → dichroic filter → (reflects, handedness flips) → λ/4 retarder → reflective polarizer (now passes as P) → integrator — is the same physical loop the patent claims, executed by the same three component types in the same order, for the same purpose (recycle the rejected polarization to raise efficiency). 3M '163 supplies the identical sandwich as an express claim element (claim 12), plus the three-source color-combining context and the red/green/blue assignment (claims 15, 20). Corning supplies the three-laser, dichroic-plate, coaxial-output, integrator-terminated architecture with express recognition of the laser-polarization-matching problem.

5. The gap: what no single reference shows

Intellectual honesty requires stating that no one reference of record discloses claim 1's exact source-side geometry: a first light source on the far side of the PDM, a second light source on the opposite side (co-planar with the dichroic stack), and a third light source beyond the dichroic mirror whose beam is transmitted through the dichroic mirror before reaching the retarder. The references disclose:

• the components (Poon, '163, Corning: dichroic + reflective polarizer + retarder + three sources + integrator);
• the loop (Poon, '163: retarder-mediated polarization recycling between a reflective polarizer and a dichroic);
• the sandwich order ('163 claim 12: λ/4 retarder between the dichroic mirror and the reflective polarizer, at 45°);
• the three-color combining with a coaxial output into an integrator (Corning, Morejon, '163 claim 15/20, X-cube art in Corning's background).

What remains is arrangement — which side of the PDM each source sits on. That is a § 103 question, not a § 102 question, and it is where the patent (if ever asserted) would have to fight.

6. Grounds of rejection

Ground 1 — Claims 1–4 obvious over 3M WO 2013/016163 A2 in view of Poon US 2011/0235175 A1 (and, for the PDM construction, Corning US 9,638,988)

• Statutory basis: § 103; all references are § 102(a)(1) printed publications (published 2013-01-31 and 2011-09-29 respectively).
• '163 discloses: a three-source illumination module for a projector; first and second dichroic mirrors each reflecting a color beam toward a combined output region; first and second reflective polarizers aligned to pass one polarization direction and recycle the orthogonal direction; a quarter-wave retarder disposed between the dichroic mirror and the reflective polarizer, aligned at 45° (claim 12); red/green/blue source assignment (claims 15, 20); output directed to an imager (claim 23). This is the whole claim-1 architecture minus the fused PDM and minus the rear-mounted first source.
• Poon supplies: (i) the fused/near-fused PDM concept — a reflective polarizer with wavelength-specific "blue layers" disposed proximate to color-selective dichroic filters within the same prism; (ii) the exact recycling loop with a λ/4 retarder interposed between the reflective polarizer and the dichroic filter, with the dichroic filter acting as the return mirror; (iii) the output integrator/light tunnel; (iv) source types expressly including lasers and laser diodes.
• Corning supplies: the three-laser, low-incidence-angle, dichroic-plate-combiner context with an express statement of the very problem the patent addresses (matching polarization among laser diodes, optional phase retarders), and the integrator output.

Motivations (KSR-rationale-tagged):

1. Same field, same problem, same result — and same assignee. '163 and Poon are both 3M projector color-combiner/illumination disclosures. "Ordinary creativity" and the "known technique" rationales apply most forcefully where the references arise from the same technical team addressing light-combining efficiency and uniformity. (KSR, combination of known elements according to known methods.)
2. The '163 and Poon disclosures are complementary halves of one design. '163 teaches the sandwich order and the three-source combination but uses a separate reflective polarizer and dichroic mirror; Poon teaches co-locating the wavelength-selective layer with the reflective polarizer and demonstrates the recycling loop against the dichroic filter. Substituting a single dual-function "polarized dichroic mirror" for the '163 pair is a simple substitution of one known element for another, with a predictable result (fewer interfaces, less loss, tighter packaging) — the very benefit the challenged patent claims ("tight structural design").
3. Recognized, calculable benefit. '163 states the purpose outright: reflective polarizers "recycle light having an undesired polarization state," and its background notes this "at least doubl[es] the effective brightness." Poon states the goal is "improving the light utilization efficiency." A POSITA would adopt the loop to capture that benefit; there is no teaching away.
4. The patent's own background supplies the motivation. The specification admits that in the conventional module "the illuminated beam and the activated beam ... are difficult to uniformly mix, and a projected image ... cannot have expected color uniformity." A problem the applicant itself concedes, addressed by known polarization-recycling hardware, is the paradigm KSR fact pattern (identified problem → known solution → predictable result).

Claim 1's narrowing does not save it. As noted in the earlier strategic section, issued claim 1 was narrowed relative to the Summary's "by means of" language to require the retarder "between the third light source and the polarized dichroic mirror" and the third beam to reach the PDM "by passing through the dichroic mirror and the phase retarding component." That is exactly the '163 claim-12 spatial relationship (retarder between the dichroic mirror and the polarizer) applied to the third beam — so the narrowing that apparently persuaded the examiner is met by '163's own claim language.

Claims 2, 3, 4. Claim 2's optical equalizer = Poon's optional integrator/light tunnel or Corning's fly's-eye/integrator bar — a known output homogenizer, expressly described as improving uniformity. Claim 3's λ/4 plate = Poon's retarder 220 and '163 claim 12, both λ/4, both at 45°. Claim 4's "equal to or similar to forty-five degrees" = Poon's "approximately 45 degrees" and '163's "45 degrees" (and Corning's stated preference for low incidence angles). All three are obvious, and none adds a structural difference.

Ground 2 — Claim 18 obvious over Poon alone or Poon in view of Corning

Claim 18 is the mirror-image geometry: the PDM passes the first beam and the second beam in the second polarization state and reflects the second beam in the first polarization state and the third beam; the dichroic mirror passes the first and third beams and reflects the second. Poon's FIG. 4b embodiment discloses a reflecting prism 460 and a mirror 468 handling the transmitted p-polarized output and the reflected s-polarized path, with the reflective polarizer both passing and reflecting different polarization states of different color beams. Whether Poon's FIG. 4a/4b pair literally meets claim 18 requires the full figure set and claim text, which I have not parsed — so I characterize claim 18 as strongly obvious over Poon (optionally with Corning or '163 for the third-source path), and flag a possible anticipation that a full-text pull should test. Given that claim 18 differs from claim 1 only in which polarization state the PDM passes versus reflects, and that a reflective polarizer's pass/reflect axes are freely selectable ("the polarizer can instead be aligned to pass the s-polarization state ... if desired," Poon), swapping which state is transmitted is a design choice with no new structural consequence — a classic "obvious to try" / "finite number of identified solutions" rejection.

Ground 3 — Claims 5–14 obvious over '163 (claims 1, 11, 16) in view of Poon and Sedlmayr US 5,903,388

Claim 5 adds a second PDM downstream of the dichroic mirror that reflects beam 2 and beam 3 in the first polarization state and passes them in the second. This is a second instance of the same recycling stage, which '163 claims 11 and 16 disclose in substance — two dichroic mirrors with a reflective polarizer between them, cooperating to recycle the orthogonal polarization of the beams — and claim 12 supplies the intervening λ/4 retarder. Duplicating a known recycling stage in series to handle an additional input is a predictable, mechanical extension, supported by the "design incentive" and "finite number of solutions" rationales.

• Claim 8 (two first light sources on opposite sides): routine duplication of a known element to increase brightness, with the admitted motivation of symmetry for color uniformity; Corning expressly contemplates "arrays with multiple light emitters."
• Claims 9–10 and 12–13 (second retarder before the second/third source; half-wave plate or two quarter-wave plates): Sedlmayr discloses a half-wave retarder 38 converting one orthogonal polarization to the other in a projector illumination path, and expressly teaches that "means for converting ... other than the half-wave retarder 38 can be provided, such as reflection from coated mirrors." Corning teaches "one or more phase retarders ... added to the different color channels." A λ/2 plate for 90° rotation and two λ/4 plates for the equivalent net rotation are interchangeable textbook equivalents. Predictable, equivalent structures.
• Claims 11 and 14 (two second/third sources at different polarization states): disclosed in substance by Poon, where the PBS is fed unpolarized beams that are split into s- and p-components and recombined in one output polarization; the patent merely moves the polarization diversity upstream to the sources.

Ground 4 — Claims 15–17 obvious over '163 claims 11/12/16 in view of Poon, Corning, and (subject to full-text verification) WO 2013/062930 A1 (3M)

Claim 15 requires the full two-PDM / two-dichroic-mirror / two-retarder cascade. Two of record give the building blocks ('163 claims 11/12: two dichroic mirrors + reflective polarizer + λ/4 retarder; Poon: reflective polarizer + λ/4 + color-selective dichroic in the recycling loop; Sedlmayr: half-wave conversion). The unverified but highly relevant WO 2013/062930 A1 discloses a color combiner with first and second PBS, a quarter-wave retarder adjacent them, and first and second tilted dichroic plates with three/four dichroic reflectors, producing "a combined color light beam having the first polarization direction" — which is the two-stage cascade in a single document. If WO 2013/062930's full text bears out the retrieved claim excerpts, claim 15 becomes a straightforward § 103 (arguably § 102) candidate, and I would recommend a full-text and figure pull as the single highest-value next step for this claim group. Claims 16 and 17 follow for the reasons given in Ground 1 (integrator; λ/4 plates).

7. Dependent-claim roll-up

8. Secondary considerations

None is supported in the record. There is no evidence in the file wrapper of unexpected results, no comparative data against the admitted conventional module, no commercial-success evidence tied to the claimed arrangement, and no competitor praise. Two points matter here:

1. No nexus, because no showing. With no litigation or IPR, no secondary-consideration record has ever been developed. A patent owner relying on "preferred illumination uniformity" would have to prove the uniformity is attributable to the claimed arrangement rather than to the known λ/4 recycling loop present in Poon and '163.
2. "Preferred illumination uniformity" is a result, not a difference. The title and the specification frame the invention by the outcome achieved, not by a new physical mechanism. Uniform mixing and increased brightness are the recognized purposes of the very recycling hardware in Poon and '163. A desired result, achieved by known means, does not rebut obviousness.

9. Where the obviousness case is weakest (counterarguments the patent owner will press)

I want to be explicit about the soft spots, because an honest petition would need to brief them:

1. The "polarized dichroic mirror" as a single element. The patent can argue no reference discloses a single mirror that is simultaneously wavelength-selective and polarization-selective with the claimed pass/reflect combination. The counter is Poon's reflective polarizer "including blue layers disposed proximate [the] color-selective dichroic filters" plus the 3M PBS-with-dichroic-layer art — but that counter depends on proving the co-located structure was known, and I have not verified the 3M PBS patent number or its claim text. This is a real evidentiary gap and should be closed before filing.
2. The source-side geometry. No reference of record places the first source on the far side of the PDM with the second source on the opposite side and the third source injecting through the dichroic mirror. The petition must make the folding-for-compactness motivation explicit and tie it to a concrete benefit the prior art already sought (path-length equalization, reduced volume/height — the patent itself cites "volume, height and cost of the projector").
3. Fast allowance. The file wrapper shows a single non-final action (2022-01-28) and allowance four months later (2022-04-29) — suggesting the art of record was overcome on a narrow point. That cuts both ways: it hints the examiner may not have had Poon and '163 combined in the manner of Ground 1 (a strong IPR lead), but it also means the examiner's search may have already confronted and disposed of part of this theory. The prosecution history should be pulled and read before any petition is drafted.

Validity items I noticed but that are outside § 103 (flagged only): (i) the specification asserts the third (green) beam passes the quarter-wave retarder "without transformation of the polarization state" — that is only physically true if the green linear polarization is aligned with a principal axis of the λ/4 plate, which is not explained; this may raise § 112 enablement/written-description questions; and (ii) claim 4's "equal to or similar to forty-five degrees" is indefinite-adjacent under § 112(b) — though as noted in § 2 above, broader is worse for the patent owner on § 103.

10. Bottom line and recommended next steps

• All 20 claims are, on the present record, vulnerable under § 103, but to differing degrees. The strongest grounds are claims 1–4 and 18–20, where Poon and 3M '163 disclose the claimed components, the claimed sandwich order (λ/4 retarder between the dichroic mirror and the reflective polarizer, at 45°), and the claimed recycling loop, leaving only source placement as the difference. Claim 15 is potentially the most vulnerable of all if WO 2013/062930 A1's two-PBS/two-dichroic-plate cascade is as retrieved — but that reference is unverified and not of record.
• Nothing here is adjudicated. Per the earlier PTAB section, no claim has been canceled, confirmed, or construed. Any petition would be a first impression on this specification. The PGR window closed 2023-05-23, so § 112-based challenges are unavailable at the PTAB; an IPR may rest only on patents and printed publications under § 102/§ 103.
• Highest-value verification steps: (1) full-text + figures for WO 2013/062930 A1; (2) full text of Poon's figures 4a–4b to test anticipation of claim 18; (3) pin down the 3M PBS patent disclosing "a polarizing layer and a dichroic optical layer" (number not resolved from the retrieved assignee listing); (4) pull the CN 202010910530.5 / CN 114200752 prosecution and any Chinese invalidation record, since the CN family shares this specification; and (5) obtain full text for the family-cited CN/JP references, particularly CN 107193177 (Appotronics) and CN 106647126 (Suzhou Qisda Optical) — note that a foreign application publication is § 102(a)(1) art only if publicly accessible, and the common-ownership exception of AIA § 102(b)(2)(C) does not reach § 102(a)(1) art, so Qisda's own Suzhou publication cannot be automatically disqualified as prior art on that basis.

Confidence: High on the § 103 theory as to claims 1–4 and 18–20 against Poon + '163. Moderate on claims 5–14, which require stacking two recycling stages. Lower — pending the WO 2013/062930 pull — on claim 15.