US 5982540

I'll search for this specific patent number across the requested sources.

I have the patent's bibliographic data and claims. Now let me check for any litigation or CAFC 2026 docket activity.

Let me confirm there's no 2026 CAFC activity specific to this patent, and get the full claim set.

US Patent 5,982,540 — Summary

Important scoping note: I searched for the patent number literally as 5982540. My searches surfaced no USPTO litigation docket and no CAFC 2026 docket involving this specific patent. I cannot confirm from the search results that any such 2026 proceeding exists, and I am not going to imply one does. Details and caveats on litigation are at the end.

---

Bibliographic data

Sources: Google Patents US5982540A, FreePatentsOnline 5982540, Justia Patents 5982540.

---

Abstract (as issued)

Light from a fluorescent lamp L enclosed by a silver foil sheet R is supplied into a light incident surface 2, producing a parallelized illuminating light flux out of a light exit surface 3. On the back surface 4 side of the light flux parallelizer 1, across or without a thin air layer 5, a polarization converter Ε is disposed, which reinforces the device's polarization function via polarization conversion accompanying reflection in the prism. A polarization separating plate 8 is disposed opposite the exit surface 3 across an air layer 7; an exit light direction modifier 10 is disposed outside it across an air layer 9. A bright polarized light flux is emitted from exit surface 11; when used as LCD backlight, a liquid crystal panel sits outside the modifier.

---

Plain-language overview of the independent claims

The patent's core idea: a wedge-shaped light-scattering light guide gives emitted light a known, oblique propagation direction. A polarization-separating plate placed at roughly the Brewster angle then (a) transmits most P-polarized light and (b) reflects most S-polarized light back into the guide. The guide's internal scattering "scrambles" that returned S light into mixed polarization, which is re-emitted and re-filtered. Repeating this recycle loop means little energy is thrown away — unlike an absorbing polarizer, which discards ~half.

Claims 1–14 and 16–17 each open with "A surface light source device with polarization function comprising…" — i.e., they are drafted as independent claims on different combinations of the same building blocks. Claim 15 is the one dependent claim (it depends on claims 1, 2, and 3–14). Note: several claim bodies are truncated in the publicly indexed HTML I retrieved, so where I say "appears," treat the precise wording as needing verification against the granted PDF.

• Claim 1 — Core "Brewster plate" device. A wedge-section scattering light guide + light feed means at the thick end + a light-transmissive polarization-separating plate along the exit surface, where the plate has refractive index 1.3–2.0 and a repetitive slope on its light-incident surface set to approximately satisfy the Brewster angle for the parallel flux's exit direction. (This is the corrugated-plate third embodiment.)
• Claim 2 — Adds a beam steerer. Same core, plus an exit light direction modifier outside the polarization separating means (prism sheet that bends the oblique output toward the viewer/front).
• Claim 3 — Adds a polarization scrambler. Same core, plus a polarization scrambler along the surface opposite the exit surface (extra strong-scattering layer to ensure the returned S light is fully depolarized).
• Claim 4 — Beam steerer + polarization converting means. Exit light direction modifier outside the separating means, with polarization conversion on the back side.
• Claim 5 — Retardation-plate converter. Polarization converting means on the non-exit side comprising a phase difference plate (retarder) plus light reflecting means outside it — the S→P conversion uses a half-wave plate and mirror ("HW → reflector → HW" round trip).
• Claim 6 — Converter on the exit-facing side. Polarization converting means along the surface confronting the exit surface.
• Claim 7 — Prism-block converter. Polarization converting means made of a parallel array of one-piece compound prism regions, each formed by uniting three right-angled-isosceles triangular-column elements with their axes in three mutually orthogonal directions (the "U-turn," 90° polarization-rotation prism unit).
• Claim 8 — Converter variant (body truncated in indexed text).
• Claim 9 — Arrays of unit volume regions. As claim 7, but the compound prism region is built by coupling multiple such unit volume regions in parallel.
• Claim 10 — Converter variant (body truncated).
• Claim 11 — Prism region on a corrugated substrate. Compound prism regions as in claim 9, plus a substrate region with repeated slopes on both faces, the prism region sitting on the slope and being optically and mechanically coupled to the substrate.
• Claim 12 — Converter variant (body truncated).
• Claim 13 — Prism region on a one-sided corrugated substrate; the compound prism region is on the repetitive-slope side and coupled to the substrate.
• Claim 14 — Converter variant (body truncated).
• Claim 15 — Dependent; the "correlation distance" limitation. Depends on claims 1, 2 and 3–14, requiring correlation distance a in the range 0.06 µm ≤ a ≤ 35 µm, where the refractive-index non-uniformity's correlation function is approximated by γ(r) = exp(−r/a), r being the distance between two points.
• Claim 16 — Core device defined by the scattering parameter. The three-part device (wedge scattering guide + light feed means + polarization separating means along the exit surface) without the Brewster-slope feature, but with the 0.06 µm ≤ a ≤ 35 µm correlation-distance limitation.
• Claim 17 — Core device plus beam steerer (body truncated).

---

Litigation / CAFC check — what I did and did not find

• No docket appears to exist for '540 itself. Searches for 5982540, "5,982,540", and combinations with litigation/appeal terms returned no infringement suit, PTAB proceeding, or Federal Circuit appeal naming this patent.
• Do not confuse it with the Enplas case that did go to the CAFC and Supreme Court. Enplas Display Device Corp. v. Seoul Semiconductor Co., Ltd., N.D. Cal. No. 3:13-cv-05038-NC, CAFC No. 16-2599 (Nov. 19, 2018, panel: Newman, Hughes, Stoll), cert. denied Nov. 4, 2019 (No. 18-1530), concerned US 6,473,554 and US 6,007,209 — different patents, not '540. That litigation concluded in 2019, well before 2026.
• Uncertainty flagged: I could not query the CAFC's own docket system or PACER directly, only web search. A 2026 filing that is very recent, sealed, or thinly indexed could exist without surfacing in search results. I am not asserting a blanket negative; I am reporting that no 2026 CAFC activity for patent 5982540 was found.
• The patent itself was cited by later patents roughly 103 times (heavily cited in display/backlight art), which reflects its technical influence rather than any litigation.

One additional caveat: I did not auto-correct anything. Note the source you supplied states the priority is claimed from JP 07274694A and JP 08371794A — I have reproduced those identifiers exactly as given, not normalized them.

I'll search litigation databases for US patent 5982540.

Let me search additional litigation databases with more targeted queries.

Let me run a couple more targeted checks before concluding.

Litigation Search Results for US Patent No. 5,982,540

Conclusion: No litigation involving US Patent No. 5,982,540 was found.

I searched patent-litigation-oriented sources (Google Patents/Patents, Justia Dockets, Docket Alarm, Unified Patents materials, Finnegan case notes, and general web indices for court records, CAFC, and PACER-derived dockets). None of the results identify any district court, ITC, or Federal Circuit proceeding in which US 5,982,540 ("Surface light source device with polarization function," Enplas Corp., inventors Yasuhiro Koike and Takayuki Arai, issued November 9, 1999) was asserted, challenged, or otherwise at issue.

Important clarification — do not conflate with the Enplas litigation that exists

There is well-documented Enplas litigation, but it involves different patents. The following case is frequently returned when searching "Enplas patent litigation," and I confirm it does not involve US 5,982,540:

Sources: https://dockets.justia.com/docket/california/candce/3:2013cv05038/[271409](/patent/271409) ; https://www.enplas.co.jp/en/english/wp-content/uploads/2016/04/20160420_en.pdf ; https://www.finnegan.com/print/v2/content/66728/Enplas-Display-Device-Corporation-v-Seoul-Semiconductor-Company-Ltd.pdf

Why this matters for your litigation question

• The '540 patent's listed status is "Expired – Lifetime," with anticipated expiration November 9, 2016 (see https://patents.google.com/patent/US5982540/en). A patent this old and expired is an unlikely candidate for an active assertion campaign.
• The Enplas corporate family has been active in patent enforcement, but the enforcement I found concerned LED diffusion lenses and the '554/'209 patents, not the '540 backlight/polarization patent.
• The '540 patent's claims were also continued in a related application, US 09/127,822, which issued as US 6,172,809 (per the Google Patents "priority to" entry). If you are evaluating freedom-to-operate or portfolio exposure, US 6,172,809 is a different patent number and should be searched separately — it is not the same as 5,982,540.

Caveats / confidence level

• Confidence that no reported litigation exists on '540: moderate-to-high. My conclusion is based on searches that returned no docket or case matching the number.
• Databases not fully exhausted. I was unable to directly query PACER, Lex Machina, RPX Insight, Unified Patents' litigation portal caselist, or Docket Alarm's patent-case search within the available steps. An exact-number query in those tools (e.g., Unified Patents Portal case list, Docket Alarm "Patent No. 5982540," or PACER's civil case search) would be the authoritative confirmation.
• I am not asserting that litigation definitely never occurred; I am stating that no litigation was found in the sources searched, and I am explicitly declining to attribute the Enplas v. SSC case (a different patent pair) to '540.

Recommended next-step verification

If you need a negative-clearance-grade answer, run the literal string "5982540" (and the format "5,982,540") in:

1. Unified Patents Portal litigation caselist — https://portal.unifiedpatents.com/litigation/caselist
2. Docket Alarm / Justia Dockets patent search
3. PACER Case Locator (civil, all districts)
4. CAFC docket search
5. The patent-in-suit field of Lex Machina or RPX Insight

If you'd like, I can attempt these one at a time on a follow-up.

I'll verify the structured block against live sources before concluding.

Let me run a couple more targeted checks to make sure nothing is hiding under alternate naming.

One more targeted pass on the PTAB docket specifically, then I'll write it up.

PTAB / AIA Trial Analysis — U.S. Patent No. 5,982,540

Patent: Surface light source device with polarization function — Enplas Corp (Kawaguchi, JP)
Inventors: Yasuhiro Koike; Takayuki Arai
Application: US 08/403,323 · Filed: 1995-03-13 · Priority: 1994-03-16 (JP 07274694A / JP 08371794A)
Granted: 1999-11-09 · Anticipated expiration: 2016-11-09 (status: Expired – Lifetime)

---

Proceedings overview

There are zero AIA trial proceedings on file against U.S. Patent No. 5,982,540 — the USPTO Open Data Portal returns no IPR, PGR, or CBM for this patent, and my independent web checks surfaced none; the defensive posture this gives a defendant is therefore not "the patent survived and is hardened" but rather "the patent is expired and has never been tested at the PTAB, so a demand letter citing it is directed at a dead right."

Breakdown by status: active 0 · claims invalidated 0 · claims sustained 0 · settled 0 · institution denied 0. There is no FWD, no institution decision, no termination, no appeal, and no estoppel to map.

Verification trail (so you can rely on the null result)

I found no recently-filed or un-indexed proceedings that the ODP ingest might have missed. Flagging explicitly per your instruction: nothing to flag.

Why the absence is structural, not accidental

Two independent reasons explain the empty docket, and both matter to the defense:

1. The patent expired on 2016-11-09 and the § 286 six-year damages look-back closed on 2022-11-09. As of today (2026-09-27), there is no injunctive relief and no recoverable past damages left for any infringing act — even if a claim were literally met. Filing an IPR to invalidate an unassertable, expired claim would be a pure fee-burn.
2. The patent-owner never ran an assertion campaign on '540. The litigation record involving Enplas in the U.S. is one in which Enplas was the defendant, and it concerned other companies' patents. This matters because a Google-search for "Enplas patent" will surface that case prominently and it is easy to mis-associate it with '540.

---

Adjacent records — NOT PTAB proceedings on this patent

I am recording these so they are not mistaken for '540 proceedings. None of the following is an AIA trial against U.S. 5,982,540, and I am not presenting any of them as such.

Enplas Display Device Corp. v. Seoul Semiconductor Co., Ltd. — N.D. Cal. Case No. 3:13-cv-05038-NC

• Type: District court declaratory judgment action with infringement counterclaims — not a PTAB proceeding
• Filed: 2013-10-29 (DJ complaint by Enplas entities; SSC counterclaimed 2014-04-21)
• Patents at issue: U.S. 6,473,554 and U.S. 6,007,209 — both owned by Seoul Semiconductor, neither related to U.S. 5,982,540
• Outcome: Jury verdict 2016-04-18 found that Enplas's overseas sales of certain LED diffusion lenses induced infringement of claims 1, 6, 33, 34 and 35 of the '554 patent and claim 20 of the '209 patent under § 271(b). December 2015 summary judgment had already cleared Enplas on all other asserted claims and on direct/contributory infringement.
• Appeal: CAFC No. 2016-2599, decided 2018-11-19 — affirmed in part, vacated in part, remanded (Newman, Hughes, Stoll; authored by Stoll; Newman concurring-in-part/dissenting-in-part). Affirmed the inducement and anticipation rulings; vacated the damages award and remanded. Certiorari denied 2019-11-04 (Supreme Court No. 18-1530).
• Sources: CAFC opinion via CourtListener · SSC's brief in opposition
• Why it's in this memo: Because "Enplas" + "backlight patent litigation" is a high-ranked search result, a defendant or a junior associate could easily pull this case and assume it involved '540. It did not. Any assertion of '540 today would be a standalone event with no litigation history, no invalidity adjudication, and no license/estoppel tail.

U.S. 5,982,540 as prior art against a third party

Separately, the '540 specification (cited as the "Koike" reference, Ex. 1008) appears to have been relied on in an IPR petition against U.S. 7,914,196 (light redirecting film system), where the petitioner charted '540's light flux parallelizer, fluorescent lamp L, PMMA/silicone-resin scattering medium, and exit light direction modifier against the '196 claims under § 103.

• Hedge: I identified this from a petition document hosted at ptacts.uspto.gov, whose claim chart text maps closely onto '540 (Fig. 4, prism surfaces 10a/10b, the silicone-resin-powder suspension-polymerization example). I did not independently verify the exhibit's bibliographic identity, so treat the mapping as probable rather than confirmed.
• Relevance: This cuts against a defendant. The '540 disclosure was persuasive enough as § 103 art to be charted element-by-element in a real petition. If you were contemplating an invalidity theory against '540 itself, the same disclosure-heavy specification is a two-edged sword — but it is also a well-documented, publicly charted teaching that supports obviousness combinations against downstream patents you may be accused of infringing.

---

Strategic summary

Claim status: all claims UNTESTED — none canceled, none sustained. Claims 1–15 of '540 have never been construed by the PTAB, never been the subject of an institution decision, and never been adjudicated invalid or valid by any tribunal that I can find. There is no narrowing claim amendment, no certificate of correction cancelling claims, and no reexamination certificate of cancellation in the record I retrieved. The practical implication is that there is no "claim 1 is dead" shortcut available to you, and equally no "the patent survived two IPRs, so it's hardened" warning. Neither narrative applies. The correct narrative is mootness: the claims may well be of questionable validity over the 1990s Japanese backlight art cited on the face of the patent (JP 51-89888, JP H2-221926, JP H4-145485, WO 93/06509), but you will never need to litigate that, because the patent cannot be asserted.

Estoppel landscape: empty, and it will stay empty. Section 315(e)(2) estoppel is petitioner-specific and arises only from an instituted IPR that reaches a final written decision. With no IPR ever instituted, no party anywhere is estopped as to U.S. 5,982,540. That means: (a) there are no unavailable prior-art grounds — every ground is nominally available to anyone; and (b) more importantly, there is no prior petitioner whose invalidity work product you could inherit or whose § 315(b) time bar has run. If you ever needed to challenge '540, you would be starting from zero with no estoppel baggage. Given the expiry, that scenario is not real.

Pattern signals: none of the usual ones. No petitioner has filed multiple IPRs on this patent (zero, in fact). The patent owner has never pursued a PTAB appeal because it has never been before the PTAB. There is no defensive aggregator in the chain — no Unified Patents, no RPX, no IPR-based defensive campaign. The patent's citation count (~103 forward citations per third-party ranking data) confirms it was technically influential and heavily built-upon, which is exactly the profile that normally attracts IPRs — the fact that it attracted none is best explained by the 2016 expiry closing the enforcement window before the modern IPR filing wave peaked. The sibling case US 6,172,809 (continuation, priority to the same 1994-03-16 JP filings, from US 09/127,822) is subject to the same expiry math.

---

Recommended next steps

If you are a defendant and '540 has been asserted against you, or appears in a demand letter:

1. Lead with expiry, not invalidity. The patent expired 2016-11-09 and the § 286 six-year damages recovery window closed 2022-11-09. Quote it plainly: no act occurring today can infringe an expired patent, and no act occurring after 2022-11-09 can support damages. There is no FWD to link because none exists — the dispositive document is the Google Patents legal-status entry ("Expired – Lifetime," anticipated expiration 2016-11-09) at patents.google.com/patent/US5982540/en.
2. Demand the asserted claims and the accused acts with dates. If the letter cites claims 1–15 but the conduct is post-2016, the demand is facially defective. If the conduct is pre-2011, it is outside the § 286 look-back as of today and also time-barred under § 286. Either way, ask them to state the dates.
3. Do not spend money on an IPR. There is nothing to invalidate that matters, and § 315(b)'s one-year bar and the estoppel consequences are irrelevant when the patent is unassertable. An IPR here would not even clear a real threat off your books — it would generate a public record of the patent's invalidity that the owner has no incentive to contest, at your expense.
4. If a customer is being pressured through you (the classic induced-infringement squeeze, which is precisely how the Enplas/SSC case ran), note that SSC's own inducement theory there required proof of a predicate direct infringement in the U.S. under § 271(a) or contributory infringement under § 271(c) — see CAFC No. 2016-2599. That doctrinal structure is unchanged, but it is irrelevant to '540 because no direct infringement of an expired claim is possible.

If no PTAB activity exists, say so plainly — and it does not. The absence here is not the soft signal ("well-asserted patents eventually attract IPRs") that it would be for a live patent. For '540 the absence has a hard explanation: the enforceable life of the patent ended in 2016 and the damages tail ended in 2022. The null docket tells you the patent was never worth attacking because it was never worth enforcing on its own.

Caveats I want on the record:

• I could not access a PTAB E2E docket page directly for '540; the null result rests on the ODP structured block plus the web checks itemized above. If you need belt-and-braces confirmation, search PTAB E2E by patent number 5,982,540 and application 08/403,323.
• I did not verify whether any foreign opposition or nullity action (e.g., in Japan against JP 3545803B2 or JP 3384867B2, or in Europe against the WO 95/03559 family) ever existed. Foreign proceedings would not create § 315(e) estoppel in any event, but if your exposure is global you should run that separately.
• I did not exhaustively review whether '540 appears as an instituted ground (as opposed to merely charted) in the '196 IPR. The article of manufacture — a well-documented, publicly charted prior-art teaching — is useful either way, but do not represent to a court that '540 was necessarily relied upon in an FWD without pulling that FWD first.

I'll research the assignment chain for US 5982540, starting with the USPTO records and cross-referencing sources.

Let me dig into inventor affiliations and the USPTO assignment records specifically.

Assignment Reconstruction — US Patent No. 5,982,540

Scope note / evidence limitation (stated up front): I was not able to open the USPTO Assignment Center UI directly in this session, so the reel/frame numbers below could not be read off the raw record. What I can confirm comes from the assignment data as mirrored in the Google Patents legal-events table for the patent (https://patents.google.com/patent/US5982540/en) plus the face-of-patent bibliographic data (law firm, inventors, assignee). Where a reel/frame or correspondent name is not in hand, I say so rather than invent it. The one recorded assignment is unambiguous in existence; only its reel/frame and recorded correspondent are unverified.

---

Inventors

Pattern notes:

• This is a university–industry co-inventorship, not a pure in-house filing. Koike is the inventor of record on a large family of light-scattering light-guide applications (the patent itself cites Koike's co-pending applications 08/064,175; 08/214,589; 08/403,880 and Arai's 08/402,483). The scattering-medium physics claimed in claims 15/16 (a≤0.06–35 µm, Debye correlation function) is Koike's academic specialty, so the co-inventorship split (Keio physicist + Enplas engineer) is technically coherent rather than a red flag.
• No "inventors departed within 12 months" signal. Koike remained at Keio (he is still there as Distinguished Professor / KPRI Director). There is no evidence either inventor exited to trigger a fire-sale.
• Oddity worth flagging: the sole indexed assignment names only Arai as assignor, even though Koike is a co-inventor. That is either (a) a second, separately indexed Koike→Enplas record I could not surface, or (b) an artifact of Google's abbreviated entry ("SEE DOCUMENT FOR DETAILS"). It is not evidence of a defect in title; it is an indexing gap I am flagging rather than papering over.

---

Original assignee

Enplas Corporation (Kawaguchi-shi, Saitama, Japan), listed on the issued patent as assignee. Japanese publicly listed operating company (Tokyo Stock Exchange). It is the original and, on all evidence, the sole owner — the patent has never left it.

• Did it ship a product embodying the claims? Yes, at least in the relevant product category. Enplas is a long-standing manufacturer of light guide plates (LGPs) for LCD backlights — its own IR materials describe a "小型LGP事業部門" (small-size LGP business division) that was reorganised internally (Enplas FY2008/3 presentation, https://enplas.co.jp/jp/wp-content/uploads/2012/06/presen_2007_07.pdf). The '540 claims are directed to a polarized backlight using a scattering LGP, squarely inside that product line.
• Primary line of business: precision optical and semiconductor parts — test sockets/contactors, LED diffusion lenses, optical connectors, light guide plates, micro-fluidic devices (Justia assignee listing: https://patents.justia.com/assignee/enplas-corporation).
• Current status: operating, not dissolved, not in bankruptcy. Enplas remains an active filer (fluid-handling devices, optical receptacles, sockets, planetary gears) as of the most recent records indexed. Its subsidiary Enplas Display Device Corporation was the party in the Enplas Display Device v. Seoul Semiconductor litigation — but note that case turned on SSC-owned US 6,473,554 and US 6,007,209 and is irrelevant to title in '540 (see prior litigation section, and https://www.enplas.co.jp/en/english/wp-content/uploads/2016/04/20160420_en.pdf).

---

Assignment timeline

Chronological list of recorded assignments:

• 1995-05-08 (executed) / recorded 1995-05-08 — Reel/frame: NOT VERIFIED (could not be read from an accessible source; confirm at Assignment Center)
  • Conveyance: Assignment (Google Patents descriptors: "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
  • Assignor: Takayuki Arai (only assignor named in the indexed entry — see inventor oddity note)
  • Assignee: ENPLAS CORPORATION
  • Correspondent: Not captured. The only firm of record I can document is the prosecution firm on the patent face — Staas & Halsey (Washington, D.C.) — which is a different field from the assignment-record correspondent. I decline to assert that Staas & Halsey filed the assignment record; that name appears on the patent, not necessarily on the assignment cover sheet.
  • Context: ordinary employment/inventor assignment to the operating company at filing-time (routine pre-issuance title cleanup, filed ~2 months after the 1995-03-13 application). Not an acquisition, not a fire-sale.

That is the entire recorded chain. There is no post-issuance assignment, no security interest, no merger, no change of name, and no transfer to any third party, licensing entity, or defensive aggregator in the sources I could reach.

Positive finding (not a gap): the absence of any post-1995 assignment is itself the key result — it means Enplas Corporation has owned US 5,982,540 continuously from filing through expiration. The only related prosecution event (not an assignment) is the continuation filed 1998-08-03 as US 09/127,822, which issued as US 6,172,809 — a sibling patent, not a transfer of '540.

To obtain the exact reel/frame, query patent number 5982540 at:

• https://assignmentcenter.uspto.gov/ (recorded-assignment search)
• https://assignment.uspto.gov/patent/index.html (legacy interface, same records)

---

Timeline diagram

timeline
    title Ownership of US 5982540
    1994 : JP priority filed
    1995 : US application filed
         : Assigned to Enplas Corporation
    1998 : Continuation filed
    1999 : Patent issued
    2016 : Patent expired

---

NPE / troll-pattern signals

Signals present: none. All eight are negative or inapplicable. The only weak positive is an unverified metadata field (assignment correspondent), which is an evidence gap, not a signal.

---

Verdict

Insufficient data — and specifically the sub-case the framework defines as "only the original assignment."

The chain consists of exactly one recorded link: the 1995-05-08 inventor→Enplas Corporation assignment, dated roughly two months after the 1995-03-13 filing of application 08/403,323. Nothing after that date appears in any source I could reach, and the patent expired 2016-11-09 while still held by its original, operating Japanese assignee. That absence of post-issuance transfers is the finding: unless the raw Assignment Center record discloses a link I could not retrieve, US 5,982,540 never entered an NPE chain, which is the opposite of a troll pattern. I cannot upgrade this to a clean "non-asserting operating company" verdict only because I could not read the reel/frame and correspondent fields directly.

Verification link: https://assignmentcenter.uspto.gov/ — search patent number 5982540 (also mirrored at https://assignment.uspto.gov/patent/index.html and in the legal-events table at https://patents.google.com/patent/US5982540/en).

---

Two explicit cross-reference flags for the record

1. Consistency with the prior litigation section: the earlier finding that no litigation names '540 is reinforced here — a patent that is never assigned away from its operating owner and that expires in that owner's hands is a poor fit for an assertion campaign. No contradiction between the two sections.
2. Do not conflate Enplas with an NPE. The Enplas entities that do appear in patent litigation (Enplas Display Device Corp., Enplas Tech Solutions, Enplas U.S.A.) were asserting/defending on US 6,473,554 and US 6,007,209 — LED diffusion-lens patents belonging to Seoul Semiconductor, with Enplas on the receiving end — in N.D. Cal. 3:13-cv-05038-NC. That is an operating-company product dispute, not an NPE chain, and it involves different patent numbers than 5,982,540.

I'll search the USPTO and reference databases for the prior-art citations on US 5,982,540, then verify each reference's identity and dates.

I have the face-of-patent citation list. Now let me identify each reference's subject matter and dates.

I hit the tool-step ceiling partway through reference-by-reference verification. Below I give the complete citation set exactly as it appears on the face of US 5,982,540, identify what I verified from live sources, and — critically — explain why the § 102 picture is much narrower than the citation list makes it look.

A note on method and honesty: the face-of-patent citation list was retrieved from two independent mirrors and cross-checked (Google Patents, Justia, and the printed front page via patentimages.storage.googleapis.com). Everything I was able to open is marked [verified]. References whose subject matter I could not open before running out of steps are marked [unverified] — I have not invented descriptions for them. Per your operating rules, I have not normalized any identifier (including the anomalous "7253511" and the "Treka"/"Trcka" spelling split).

---

1. The citation set on the face of US 5,982,540

1A. U.S. Patent Documents

1B. Foreign Patent Documents

1C. Other cited material (co-pending applications)

• US App. 08/064,175 — Koike, filed May 25, 1993
• US App. 08/214,589 — Koike, filed Sep. 22, 1993
• US App. 08/403,880 — Koike, filed Mar. 21, 1995
• US App. 08/402,483 — Arai, filed Mar. 10, 1995

These are the applicants' own co-pending applications (the "light flux parallelizer" family), cited for continuity, not third-party art.

---

2. The § 102 frame (pre-AIA, and it matters a lot here)

Application 08/403,323 was filed 1995-03-13; priority runs to 1994-03-16 (JP 07274694A → JP3545803B2) and 1994-03-31 (JP 08371794A → JP3384867B2). So:

• § 102(b) art = published/patented more than one year before the filing (measured from the foreign priority date under §§ 119/102(b)), i.e., before ~13/16 March 1994.
• § 102(a)/(e) art = everything else that pre-dates invention/filing but is not a >1-year bar; the four 1994-issued U.S. patents (5,295,048 / 5,303,322 / 5,359,691, and 5,548,670) fall here, not under § 102(b).
• Anything published after the filing (the JP "7253511" as printed, and arguably WO 95/03559) is not prior art to '540.

---

3. § 102 anticipation analysis, reference by reference

The controlling point first. Under § 102, a claim is anticipated only if one single reference discloses every element as arranged in the claim. Every independent claim of '540 (claims 1–14, 16–17) requires the combination of (i) a wedge-shaped light-scattering light guide ("light flux parallelizer"), (ii) light feed means at the thick end, and (iii) light-permeable polarization-separating means arranged along the exit surface. Claim 1 adds the Brewster-angle repetitive slope on the separating plate's incident face. No reference on the face — verified or not — discloses a reflection/transmission polarization-separating element (still less one at roughly the Brewster angle) mated to the output of a scattering wedge guide. The prior art cited is generic backlight / light-guide / LC-illumination art, and it is § 103 material, not § 102 material.

Accordingly, my good-faith mapping is:

Bottom line on § 102: none of the cited references anticipates any claim of US 5,982,540. If you need to attack or defend these claims, the fight is entirely under § 103 (obviousness), and even there the citations are individually weak because none of them supplies the polarization-separating/ Brewster element.

---

4. Assessment of "most relevant prior art"

Ranked by relevance to the actual point of novelty:

1. Koike's own earlier work — US 5,548,670 (Aug. 20, 1996) and the co-pending applications 08/064,175 / 08/214,589 / 08/403,880, with JP H5-349479 (Dec. 28, 1993) named in the specification — these establish the "light flux parallelizer" / wedge light-scattering guide with defined scattering parameters. They are the closest art to the guide element, but they are the inventor's own and contain no polarization separation.
2. US 5,359,691 (Tai et al.) and US 5,303,322 (Winston et al.) — the two most likely to disclose directional/wedge light-guide output among the U.S. citations. (Unverified — worth opening first.)
3. US 4,059,916 and US 4,618,216 — verified wedge/tapered edge-lit guides with rear reflectors.
4. The JP background documents (51-89888, H2-221926, H4-145485) — admitted conventional art only.

Notable gap: the face citations contain no reflective-polarizer / polarizing-beam-splitter / Brewster-angle art. The element that carries the invention — the polarization-separating plate — was not met by anything the examiner cited. The true closest art on that element (e.g., the mid-1990s reflective-polarizer and beam-splitter patents) is not a face citation here; I did not have steps left to verify it and I am not asserting any specific number.

---

5. Verification gaps and caveats you should carry forward

• Twelve cited U.S. patents [unverified]: I retrieved number/date/inventor from the face for all of them, but ran out of steps before opening their specifications. Their descriptions above are drawn from citation metadata and their class-of-art, not from reading the documents. Do not rely on my one-line characterizations of 4,453,200 / 4,706,173 / 4,794,492 / 5,050,946 / 5,123,077 / 5,157,526 / 5,202,950 / 5,211,463 / 5,295,048 / 5,303,322 / 5,359,691 / 5,548,670 without opening them.
• Identifier anomalies, preserved literally: (a) "7253511" prints with a 10/1995 date that cannot be prior art — possible OCR/transcription artifact, but I did not auto-correct it; (b) the inventor is "Treka" on Justia and "TRCKA" in a later ISR — I've shown both rather than normalizing; (c) Google's OCR gives US 5,123,077 a missing month ("0/1992") where Justia gives June 16, 1992.
• § 102(e) subtleties I did not fully resolve: the 1994-issued patents (5,295,048, 5,303,322, 5,359,691) and 5,548,670 are § 102(e) candidates only; 5,548,670 raises a same-inventor ("by another") question because co-inventor Koike is its sole named inventor — that requires a claim-by-claim § 102(e)/§ 103(c) analysis I did not complete.
• The prior-litigation/PTAB sections of this analysis stand: no litigation and no AIA trial involves '540, so this citation analysis is being done for portfolio/FTO understanding, not in response to an active assertion.

If you want, the single highest-value next step is to open US 5,359,691, US 5,303,322, and US 5,548,670 and chart them element-by-element against claims 1 and 16 — those three are where a § 103 combination would most plausibly be built from this citation list, and none of them can carry a § 102 case on its own.

§ 103 Obviousness Analysis — US 5,982,540

Building on the Prior Art section. That section established the controlling § 102 result — no face-of-patent reference anticipates any claim. This section takes the same reference set and asks the harder question: would the claimed subject matter have been obvious to a person of ordinary skill in the art (PHOSITA) at the 1994 priority date? I am not repeating the § 102 charts; I am extending them.

Let me ground a few of the key references before writing.

Good — the searches confirm the key subject matter. Let me ground the polarization-element side of the analysis.

§ 103 Obviousness Analysis — US 5,982,540

Building on the Prior Art section. That section established the controlling § 102 result — no face-of-patent reference anticipates any claim — and closed with a recommendation to chart the three most likely § 103 building blocks. This section takes that recommendation up and answers the § 103 question directly. I am not repeating the § 102 element charts; I am extending them.

Scope of the analysis, stated honestly up front. The prior-art section flagged twelve cited U.S. references as [unverified] and noted the face citations contain "no reflective-polarizer / polarizing-beam-splitter / Brewster-angle art." In this session I confirmed the subject matter of four key references (Tachihara, Suzawa, Tai, Winston) and — importantly — surfaced a body of outside-the-face polarization art that changes the § 103 calculus considerably. I flag the verification status of every reference I use.

---

1. The legal frame (pre-AIA § 103)

Application 08/403,323 was filed 1995-03-13, so pre-AIA § 103 governs. The analysis proceeds under Graham v. John Deere (scope/content of prior art; differences from claims; PHOSITA level; secondary considerations) as modulated by KSR Int'l v. Teleflex:

• TSM is flexible. A teaching, suggestion, or motivation may come from the prior art, the problem itself, or the PHOSITA's own knowledge. KSR: "the combination of familiar elements according to known methods is likely to be obvious when it does no more than yield predictable results."
• "Obvious to try." Where there is a design need and "a finite number of identified, predictable solutions," pursuing a known option within the PHOSITA's technical grasp is obviousness, not invention.
• § 102(b) critical date ≈ 1994-03-13 (one year before the U.S. filing). Anything publicly printed before that date is a statutory bar with no swearing-behind. This matters: several of the polarization references below are cleanly pre-1993, hence unimpeachable § 102(b) art.
• Every reference in a § 103 combination must independently qualify under § 102.

Caveat on dates: the publication/filing dates for the EP and WO references I introduce in Combinations II–III are taken from background text in other documents and must be confirmed against the 1994-03-16 / 1994-03-31 priority dates before they are relied upon adversarially. I mark those entries "[date to verify]".

---

2. PHOSITA

A person with a B.S. in optical engineering or applied physics plus 2–4 years designing LCD backlight optics, or an M.S./Ph.D.-level worker in polymer photonics (the scattering-medium physics in claims 15/16 — effective scatter irradiation parameter E, Debye correlation distance a — is graduate-level light-scattering theory). This is the profile the '540 specification itself assumes: it discusses Fresnel interface transmissivity, critical angle, Brewster's angle, and Rayleigh–Gans/Debye scattering in the same document. A PHOSITA would therefore be comfortable with both guided-light optics and polarization optics.

---

3. Claim decomposition — what actually carries the claims

The prior-art section correctly identified that every independent claim is built from the same three blocks. For § 103, the useful decomposition is:

The decisive insight for § 103 is this: the '540 specification itself concedes A and admits the problem that motivates C. The Background recites the inventors' own prior "light flux parallelizing apparatus … having characteristic parameters for defining the scattering characteristic (effective scatter irradiation parameter E and correlation distance a) determined in a specific range" (JP App. H5-349479, filed Dec. 28, 1993), and then states the unmet need in the very next breath. That admission does most of the obviousness work.

---

4. The combinations

4.1 Combination I — The core claim-1/16 attack: admitted parallelizer + Tachihara/Suzawa + a known polarizing element

Why a PHOSITA would combine them. The motivation is in the specification itself and in the admitted state of the art:

1. Same field. Tachihara and Suzawa are edge-lit LCD illuminators. Tachihara's own claim 15 already mated the wedge plate to an LC cell with polarizing plates. A PHOSITA optimizing that device immediately confronts the absorptive polarizer.
2. The stated problem is the motivation. The '540 Background says the prior polarizing plates "eliminate the polarized components in a direction orthogonal to a specific direction," so "about half of the light energy is lost," and that a device generating polarized light "having a suppressed light energy loss … has been waited for." Under KSR, the problem in the art is itself a sufficient motivation.
3. Complementary geometry, not a lucky accident. The parallelizer's output is oblique (~65° per the spec); Brewster's angle is ~56°. Placing a Brewster-angle plate in an oblique output path is a predictable, deterministic match. That the guide produces obliquity makes C feasible — this is the classic "arrange old elements with each performing its known function" case.
4. Claim 16's guide is admitted art. Claim 16 is the three-block combination without the slope, with the 0.06–35 µm range. With element A conceded, claim 16's only remaining element is C — and C (a polarizing element in the exit path of a guide) is exactly what US 4,212,048 disclosed in 1980 and what a Brewster plate supplies trivially.

Reasonable expectation of success: high. The optical physics is deterministic; the spec's own tolerance analysis confirms the device works "about 10°" off Brewster.

---

4.2 Combination II — The polarization-recycling attack (the disclosed point of the invention)

The '540 specification asserts the recycle loop is "intrinsic of the invention" and claims amplification of the P component "to over 100%." This is where the patent hopes to rest. The record shows the concept was already a 20-year-old, well-documented technique:

Motivation. Once a PHOSITA has (a) a reflective polarization-separating element and (b) a light cavity with a rear reflector (Tachihara, Suzawa, and the '540 guide all have one), recycling is the recognized, known way to exceed the ~50% ceiling. The KSR "finite number of identified, predictable solutions" framing fits precisely: at the 1994 priority date, the menu was {absorptive polarizer (baseline loss), reflective polarizer + recycle, Brewster plate + recycle, retarder conversion + recycle}. Each was known; each was predictable; and converting rejected S to P via retarder + mirror was explicitly the Zehnder 1973 / projector art answer.

The one genuinely arguable step is the recognition that the scattering guide acts as the depolarizer, so that returned S light re-emerges as mixed polarization and can be re-filtered. But: (i) a light-scattering medium is the textbook description of a depolarizer; (ii) the '540 specification's own forward-scattering analysis (correlation distance a controls forward-scatter) is admitted to describe the guide's behavior, not a discovery; and (iii) claim 3 exists precisely because the inventors knew the scramble can be insufficient — which concedes that scrambling is an expected property of the medium, merely needing augmentation.

---

4.3 Combination III — The architecture attack on claims 1, 5, and 16: edge-lit guide + polarization separator on one face + reflector on the other + retarder

This is the most damaging teaching in the material I retrieved, and it comes from the background section of EP 0 778 818 B1:

"EP 0 597 261 discloses an illumination device for a liquid crystal display. The device comprises an edge-lit light guide having a polarisation separator on one surface and a reflector on an opposite surface. A phase shift layer or retarder may be provided, for example on the reflector."

Map that onto the claims:

Motivation: identical to Combination I (the LCD half-loss problem), and EP 597 261 is expressly directed to an LCD illuminator — same field, same problem.

Date caveat — this is the load-bearing reference, so the date must be nailed down. Based on the citing-source context, EP 0 597 261 appears to have a 1992 UK priority and to have published in May 1994 — i.e., after the 1994-03-16/31 priority dates but before the 1995-03-13 U.S. filing. If that is correct, EP 597 261 is § 102(a)-only art and cannot be used unless the applicant's invention date is shown to post-date its publication (which is difficult given the constructive priority date). I flag this rather than assume it away. If, however, the published specification's filing predates and its disclosure is complete, or if it qualifies as a § 102(b) bar by a different route, Combination III becomes the single strongest attack on claims 1, 5 and 16.

Also flagged: WO 94/11776 (multilayer reflective polarizer between an optical cavity and an LCD, 3M) and WO 95/17692 / WO 95/17699 are PCT publications. The WO 95/17692 and WO 95/17699 documents published June 1995 — after the '540 filing and are therefore not prior art at all as printed publications. The corresponding U.S. grants (e.g., the 3M multilayer reflective-polarizer family asserting a December 1993 priority) could be § 102(e) art, but that requires pulling the U.S. specifications and confirming both the priority date and support. Do not assume these WO documents are citable.

---

4.4 Combinations for the dependent claims

On the compound prism (claims 7–14) specifically: the retrieved record contains no art on a monolithic triple-reflection polarization-rotating prism. A PHOSITA could plausibly arrive there from US 5,359,455 (which already uses "prism formed plates" to redirect converted beams) via routine engineering of prism geometry, and the polarization-preservation-on-TIR physics is elementary. But that is an argument I cannot presently ground, and I will not assert it as established. Claims 7–14, if anything survives, survive here.

---

5. Claim-by-claim disposition (weighted)

---

6. Countervailing (nonobviousness) considerations — the honest case for the patentee

I owe the patentee an accurate statement of the other side:

1. No single reference or combination on the face of the patent supplies element C. The examiner cited only generic backlight/light-guide art; the Brewster/reflective-polarizer element was met by nothing cited. On the face-of-patent record alone, claim 1 was allowable, and there is a real argument that the § 103 combination requires impermissible hindsight reassembly of the applicant's own parallelizer with separately-sourced optics art.
2. The claimed synergy — that a scattering wedge produces oblique output which happens to land near Brewster's angle, enabling a flat, simple plate to polarize efficiently while the same medium depolarizes the return — is a two-way interaction. Under KSR, a patentee may argue the combination yields more than the sum of its parts. The counter is that each part performs exactly its known function, and the spec itself frames the result as "intrinsic."
3. The >100% P-amplification result is the strongest "unexpected result" candidate — but it is arguably a straightforward consequence of energy recycling, not an unpredictable discovery.
4. Claims 7–14 are genuinely unimpeached on this record.
5. No teaching away was found, but also no secondary-considerations record: there is no evidence of commercial success of the claimed embodiment, no nexus, no industry praise, and no copying. Per the Assignment section, Enplas is an LGP manufacturer, but no nexus between the '540 claims and any revenue has been shown. Weak objective indicia cut against patentability.

---

7. Bottom line

• Claims 1, 2, 3, 4, 5, 6, 15, 16, 17 are, in my assessment, vulnerable to § 103. The strongest attacks are (i) the applicant's own admission of the wedge-scattering parallelizer + any known polarizing element, and (ii) the polarization-recycling technique that was documented from Zehnder (1973) forward and industrialized in the LCD-projector polarization-converter art (Imai 1990, Shinsuke 1991, US 4,913,529, US 4,969,730, US 5,042,921, US 5,124,841, US 5,359,455). The § 102(b) status of those pre-1993 references is their most important attribute — they need no date gymnastics.
• EP 0 597 261 is the single highest-value reference I found (edge-lit guide + polarization separator on one face + reflector on the other + optional retarder), but its date must be verified before it can be relied upon; as characterized it may be § 102(a)-only and thus possibly unusable.
• Claims 7–14 (compound prism converter) are the patent's survival position. Nothing in the face citations or in what I retrieved discloses the monolithic three-orthogonal-axis prism.
• Claim 15's range and claim 16's broadest formulation are the easiest to take down on numerical-range/routine-optimization grounds.

Caveats I want on the record:

• I could not verify the specifications of twelve face-cited U.S. references (they remain [unverified] per the prior-art section), nor the publication dates of EP 0 597 261, EP 615,148, EP 573,905, WO 94/11776, or the 3M multilayer-polarizer U.S. grants. Every date-dependent assertion above is contingent on that verification.
• The EP/WO/3M references were surfaced from the background discussion of another document; I have not opened their full texts. Treat their characterizations as probable, not confirmed.
• This is a technical/analyst assessment, not legal advice, and it is academic as to enforcement — per the PTAB and Litigation sections, US 5,982,540 expired 2016-11-09 with no assertion history, so there is no live controversy to which this § 103 analysis attaches.

Highest-value next step: pull EP 0 597 261 (date + full specification) and the 3M multilayer reflective-polarizer U.S. grant with a December 1993 priority, and chart each against claims 1, 5 and 16. Those two documents, if they date correctly, convert the "moderate" § 103 case on the core claims into a strong one.