US 7404660

I'll run targeted searches for US patent 7404660 and any 2026 CAFC activity.

Let me verify whether any 2026 CAFC docket actually names the '660 patent, and confirm the claim-text discrepancy I'm seeing between sources.

Both restricted searches returned zero hits. One more check on the patent's post-grant and litigation record.

US 7,404,660 — search results summary

Bottom line up front: I found no 2026 CAFC docket activity naming 7,404,660. The only 2026 Federal Circuit decision my searches surfaced involving a "740" number is Manufacturing Resources International, Inc. v. Squires, Nos. 2024‑2224, 2024‑2231 (Fed. Cir. Apr. 7, 2026), which concerns U.S. Patent Nos. 10,506,740 and 11,013,142 (display cooling systems). That is a different patent number and a different technology — it is not 7,404,660, and I am not treating it as related. I also found no 2026 PTAB or district court filing for the '660 patent.

Bibliographic data (per Google Patents, Docket Alarm/PTAB records)

Continuity: Divisional of 10/784,527 (now US 7,160,015), itself a divisional of 09/256,275 (now US 6,712,481), a CIP of 08/778,089 (now US 6,079,838), a division of 08/495,176 (now US 5,613,751). This is a very large continuation family (34 listed family applications), with three later continuations off this patent: 12/164,315 (US 7,513,672), 12/164,508 (US 7,524,101), 12/164,702 (US 7,563,012).

Assignment chain (per Google Patents reassignment records): Solid State Opto Limited → Rambus International Ltd. (2010‑02‑18) → Rambus Inc. → Rambus Delaware LLC (2013‑03‑04) → Acacia Research Group LLC (2013‑06‑26) → Innovative Display Technologies LLC (2013‑06‑27).

Abstract (as printed)

"Light emitting assemblies include a generally planar optical conductor having at least one input edge with a greater cross-sectional width than thickness and at least one light source having a light output distribution with a greater width component than height component positioned adjacent to the input edge for directing light into the optical conductor and emission of the light from at least one output region of the optical conductor. A transition region is disposed between the light source and output region that is configured to spread and transmit the light by the light source to the output region. A plurality of faceted surfaces in close proximity to the light source maximize or otherwise change the light emitted from the light source."

Plain-language overview of the independent claims

Claim 1 (assembly): A light-emitting panel assembly with four required elements:

1. a generally planar optical conductor (a light-guide panel) whose input edge is wider than it is thick;
2. a plurality of light sources (note: plural, not "at least one") that produce light with an output distribution wider than it is tall, placed next to that input edge to inject light sideways into the conductor;
3. the conductor has an output region plus a predetermined pattern of deformities that causes light to escape from that region; and
4. the conductor has a transition region between the light source and the output region.

Claim 33 (assembly, LED-specific variant): Same four-element structure as claim 1, but the light sources are positively recited as a plurality of LED light sources each having a greater width than height, positioned adjacent the input edge and each generating light with a wider-than-tall output distribution, again with an output region/deformity pattern and an intervening transition region.

All remaining claims are dependent. The most consequential dependents:

• Spreading/mixing: transition region spreads the light (claim 2); thicker than the output region (8) or tapering down toward the output region (9); constant thickness (10); conductor tapers along its length (11).
• Color mixing: LEDs (5); an LED with at least two different color chips (6); differently colored LEDs (19) whose light mixes and spreads in the transition region to yield a desired output color (7, 20).
• Faceting: faceted surfaces near the light sources to maximize/change the emitted light (4, 15, 22, 24, 32, 36, 39, 40); reflective surfaces on the transition region producing internal reflection back into the conductor (30–32).
• Slots: slots bored in from the proximal edge to receive the light sources, where each slot's inner edge acts as the input edge (14, 23, 38) — see the FIG. 15 sliding-LED embodiment.
• Reflectors/mechanics: distal-edge reflector (12), side-edge reflector (13), a tray housing the conductor and acting as side/distal/bottom reflector (25–29); focused light sources (16, 17) directing light perpendicular to the input edge (18).

Post-grant and enforcement history (all pre-2026)

• IPRs: IPR2014‑01094 (LG Display — not instituted on the merits), IPR2015‑00487 (LG Display), IPR2015‑01113 (Mercedes‑Benz USA), plus related settlements/non-institution: IPR2015‑00363, ‑00495, ‑00745, ‑00855, ‑00897, ‑01113. Challenged claims included 1, 3, 5, 10, 13, 16, 17, 25–29, 33–35; claim construction focused on "deformities" and "a plurality of light sources configured to generate light having an output distribution defined by a greater width component than height component." Prior art relied on included Pristash (US 5,005,108 — a patent the '660 specification itself incorporates by reference), Ciupke (US 5,461,547), Tai (US 5,359,691), Funamoto (US 5,619,351), Tsuchiyama (US 5,548,271), Oe (US 5,711,589), and JP publications H5‑210014 / 07‑064078.
• District court campaign (2013–2015): Delaware Display Group / IDT v. LG, Sony, Vizio, Lenovo, Pantech (D. Del. 1:13‑cv‑02106–02112); IDT v. Hyundai, Toyota, Mazda, Mercedes‑Benz, Nissan, BMW, GM, Ford, VW, Sprint, etc. (E.D. Tex. and D. Del. 2014–2015), with accused products including navigation/display units.
• CAFC: Docket 17‑1413 is listed for this family on Unified Patents (a 2017 appeal), not 2026.

Uncertainty and caveats — please read

1. No 2026 CAFC docket hit for 7,404,660. My searches for the specific number plus 2026/federal-circuit terms returned zero results, and a broad CAFC-2026 search returned only the unrelated 10,506,740 case. I cannot positively state the '660 patent has no 2026 docket activity; I can only report that I found none. Because the patent's statutory term ended June 27, 2015, and its PTO status is "Expired – Fee Related," new appellate activity in 2026 would be unusual.
2. Claim 1 wording differs between sources — this matters. The granted patent (Google Patents full text, and confirmed by the LG petition's quote of the granted claim) recites "a plurality of light sources." However, the pre-grant publication US 2006/0232965 A1 renders claim 1 as "at least one light source" (see Justia). Source: https://patents.justia.com/patent/20060232965. This is consistent with a prosecution amendment from "at least one" to "a plurality." If you are analyzing scope, use the granted language (plural), not the publication.
3. A clerical inconsistency appears in the PTAB record. One exhibit list in an LG Display petition lists "U.S. Patent No. 7,440,660" (Ex. 1001) while the petition body repeatedly refers to "Patent 7,404,660." I am reporting that literally rather than correcting it; it is almost certainly a typographical error in the petition, but the operative patent is 7,404,660.
4. Abstract vs. claims mismatch is real, not my error. The printed abstract says "at least one light source," while granted claim 1 requires a plurality. This is a known consequence of amending claims without conforming the abstract.
5. Date discrepancy in the request. The prompt states the current date is April 26, 2026, while my operating context sets today at September 10, 2026. This affects how I characterize "2026 activity," and I flag it rather than silently choosing one.
6. Assignee: Google Patents lists the current assignee as Innovative Display Technologies LLC and the original as Solid State Opto Ltd. I did not independently verify current ownership beyond those records.

Sources: https://patents.google.com/patent/US7404660/en · https://uspto.report/patent/grant/7,404,660 · https://patents.justia.com/patent/20060232965 · PTAB IPR2015‑00487 and IPR2015‑01113 petitions (ptacts.uspto.gov / docketalarm.com) · https://www.cafc.uscourts.gov/opinions-orders/24-2224.OPINION.4-7-2026_2672479.pdf

I'll research litigation involving US patent 7404660, cross-referencing the litigation data already embedded in the patent record with live searches.

Let me gather additional details on the specific cases, outcomes, and PTAB proceedings.

Litigation Involving US Patent 7,404,660 ("Light emitting panel assemblies")

Bottom line: Yes — US 7,404,660 was heavily litigated. It was asserted in dozens of district-court actions (E.D. Tex. and D. Del.) by Innovative Display Technologies LLC and Delaware Display Group LLC (both Acacia/related NPEs), and it was challenged in nine inter partes review (IPR) proceedings at the PTAB, plus one Federal Circuit appeal. The patent itself is now Expired – Fee Related (anticipated expiration listed as 2015‑06‑27).

Important scope caveats I want to be explicit about:

• The litigation list on the patent page is the "family has litigation" block. Google Patents' family-level docket list can include cases in which this patent was one of several asserted; I have cross-checked against the Stanford NPE Litigation Database entry for 7404660 to confirm specific assertion.
• I could not verify a case-by-case disposition for every 2013–2014 district-court action from live sources in this session. Where the outcome is not confirmed below, I say so rather than guess.

1. PTAB – Inter Partes Review (the most fully documented proceedings)

All nine IPRs are listed on the patent record, and several are confirmed by RPX, Patexia, GreyB/IPVerse, and Docket Alarm:

Sources: https://insight.rpxcorp.com/ptab/2548; https://insight.rpxcorp.com/ptab/3758-lg-display-v-innovative-display-technologies-llc-ipr-of-660; https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2015-00487; https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2015-01113; https://services.patexia.com/lawsuits/LG-Display-Co-Ltd-v-Innovative-Display-Technologies-LLC-id-60515; https://casetext-staging-alt.com/admin-law/lg-display-co-ltd-v-innovative-display-technologies-llc-2.

Net PTAB result: The '660 patent survived every instituted challenge — no claims were held unpatentable.

2. Court of Appeals for the Federal Circuit

• Case No. 17‑1413 — appeal arising from the IPR proceedings (IPR2015‑00487 / IPR2015‑01717). Listed on the patent record and on RPX/Patexia. Appeal disposition not confirmed in this session.

3. District Court Litigation

A. E.D. Tex. — Initial 2013 wave (Innovative Display Technologies LLC)

Filed on or about 2013‑06‑28 (per Mandour Law and the STPI/Chinese case summary), asserting the '660 patent together with U.S. 6,755,547; 7,300,194; 7,384,177; 7,434,974; 7,537,370; and 8,215,816:

(Defendants also named in the wave included Research In Motion/BlackBerry.) Sources: https://www.mandourlaw.com/technology-company-accuses-dell-hp-of-patent-infringement/; https://iknow.stpi.niar.org.tw/Post/Read.aspx?PostID=8222.

B. D. Del. — Delaware Display Group LLC / Innovative Display Technologies LLC (2013)

Filed late 2013 (dockets 1:13‑cv‑02106 through 1:13‑cv‑02112), including LG Display's U.S. entity (per the Korean report referencing a December 2013 D. Del. suit against LG Display America):

C. E.D. Tex. — 2014 wave (Innovative Display Technologies LLC)

D. D. Del. — 2014–2015 wave (Delaware Display Group LLC et al.)

Source for the Dec. 30, 2015 LG filings: http://www.thebell.co.kr/free/content/ArticleView.asp?key=201601150100026580001613&lcode=00.

Claim construction: In Innovative Display Techs. LLC v. Acer Inc., the E.D. Tex. court issued a claim-construction order covering the '660 patent (among the seven patents-in-suit); the court struck Plaintiff's expert declaration. Source: https://case-law.vlex.com/vid/innovative-display-techs-llc-890271239.

4. What I could NOT confirm

• Exact filing dates for most individual district-court dockets (beyond the June 28, 2013 E.D. Tex. wave and the Dec. 30, 2015 D. Del. LG filings) — these dockets' filing dates are shown as blank/truncated in the NPE database.
• Outcomes/dispositions for the individual district-court cases (settlement, dismissal, judgment). At this scale of parallel NPE litigation the cases were overwhelmingly resolved by settlement or dismissal, but I do not have a verified per-case record and will not state one.
• The Federal Circuit (17‑1413) disposition.

Authoritative starting points for the dockets/PTAB papers:

• Stanford NPE Litigation Database for 7404660: https://npe.law.stanford.edu/patent/7404660
• Google Patents litigation block (source of the case list): https://patents.google.com/patent/US7404660/en
• Unified Patents litigation portal (caselist): https://portal.unifiedpatents.com/litigation/caselist

If you want, I can drill into a specific docket (e.g., the E.D. Tex. 2:13‑cv‑00526 Dell action or the Dec. 2015 LG Display D. Del. action) to try to pull the filing date and disposition.

I'll start by verifying the structured data against public sources, since the structured block reports no AIA trials while the patent page itself lists nine PTAB case numbers.

Significant PTAB activity is surfacing. Let me pin down which proceedings target the '660 patent and what happened on appeal.

⚠️ Threshold reconciliation — the structured block is wrong

The "PTAB proceedings on file" block states the ODP ingest returns no AIA trial proceedings for US 7,404,660. That is contradicted by both the Google Patents family-litigation data embedded in this same patent record (nine PTAB case numbers, six of them labeled "Settlement" or "Not Instituted") and by the public PTAB record I retrieved. The '660 patent was the target of a multi-petitioner IPR campaign in 2014–2016. I flag the ODP gap as an ingestion error, not a fact about the patent. Everything below is sourced from PTAB papers/decisions (docketalarm/PTACTS mirrors), RPX Insight, and GreyB ipverse.

Proceedings overview

Five AIA trials on the '660 patent proper: 0 claims invalidated · 11 challenged claims sustained on the merits · 2 settled (no estoppel) · 1 institution denied · 0 active. Two proceedings reached Final Written Decision, and the patent won both — every challenged claim survived. Bottom line for a defendant: the troll's patent is hardened, not dead — but it is also expired (2015-06-27, "Expired – Fee Related"), so any assertion today is a damages-only, backward-looking case, and LG's § 103 theory over Oe + Etsuo is now twice-rejected precedent you cannot simply re-run.

IPR2015-00487 — LG Display Co., Ltd. (w/ LG Electronics, Inc.) v. Innovative Display Technologies LLC

• Type: Inter Partes Review
• Filed: 2014-12-23
• Status: Final Written Decision — "No Claims Unpatentable" (claims 1, 3, 5, 10, 13, 16, 17, 25–29, 33–35 all sustained)
• Judge panel: Lora M. Green, Thomas L. Giannetti, Beverly M. Bunting (Bunting authored the FWD). Oral hearing 2016-03-16.
• Petition grounds: § 103(a) over U.S. Patent No. 5,711,589 ("Oe") in view of JP 07-064078 ("Etsuo"), against claims 1, 3, 5, 10, 13, 16, 17, 25–29, and 33–35. No § 102 or § 112 grounds were instituted.
• Institution decision: Instituted 2015-07-16 on the single Oe + Etsuo obviousness ground — the Board found a reasonable likelihood of prevailing. (Note: a prior petition on this patent, IPR2014-01094, had already been denied on the merits, so this was LG's second, narrower bite — see below.)
• Final Written Decision (2016-07-15): Petitioner failed to carry its burden by a preponderance. The dispute turned on the "plurality of light sources … output distribution defined by a greater width component than height component" limitation: Patent Owner argued (and the Board accepted) that Etsuo's lamp units are combined into one light source, so the petition could not rely on them as a plurality while simultaneously aggregating them to manufacture the greater-width-than-height distribution. PO also attacked Petitioner's expert (Escuti) as a word-for-word copy of another expert's declaration, and disputed the Oe/Etsuo combination as teaching away. Independent claims 1 and 33 both survived, along with dependents 3, 5, 10, 13, 16, 17, 25–29, 33–35. Claim construction was governed by Phillips (expired patent).
  • FWD docket/papers: https://www.docketalarm.com/cases/PTAB/IPR2015-00487/Inter_Partes_Review_of_U.S._Pat._7404660/
• Settlement / termination: none — decided on the merits.
• Appeal: Appealed by Petitioner to the Federal Circuit, CAFC No. 17-1413 (Notice of Appeal dated 2016-12-23). Appeal issues: (1) whether the Board erred in finding claims 1, 3, 5, 10, 13, 16, 17, 25–29, 33–35 not unpatentable over Oe + Etsuo; (2) wrong legal standard for obviousness; (3) claim construction/prior-art interpretation. I could not verify the disposition of No. 17-1413 from my sources — do not assume affirmance or vacatur without pulling the CAFC docket. (CourtListener docket: https://www.courtlistener.com/?q=%2217-1413%22&type=r)
• Defensive value: The two strongest independent claims, 1 and 33, are confirmed patentable against the best art LG could muster — so an IPR-based knock-out of this patent has already been tried and failed. Your realistic defenses are § 102 on art LG never raised, or invalidity/§ 101/§ 112 in district court (where the expired-patent Phillips construction applies and IPR estoppel limits you to grounds you could not have raised).

IPR2015-01717 — LG Electronics, Inc. et al. v. Innovative Display Technologies LLC

• Type: Inter Partes Review (joinder — joined IPR2015-00487)
• Filed: 2015-08-13
• Status: Final Written Decision (2016-07-15) — "No Claims Unpatentable" via the joined -00487 FWD
• Judge panel: Lora M. Green, Miriam L. Quinn, Thomas L. Giannetti, Beverly M. Bunting (+ further panel members per RPX)
• Petition grounds: same Oe + Etsuo § 103 challenge, filed by the LG entity that was not the named petitioner in -00487 (LG Electronics, Inc.); the Board granted the motion for joinder in its 2016-02-05 institution decision.
• Institution decision: Joinder granted 2016-02-05.
• FWD: decided with -00487 (see above). No separate claim-level verdict.
• Settlement: none.
• Appeal: carried with CAFC No. 17-1413.
• Defensive value: Purely additive estoppel exposure for the LG entities. Because it joined rather than ran its own trial, the effect on the patent is nil — but § 315(e)(2) estoppel will bind both LG Display and LG Electronics for anything raised or reasonably raisable from Oe/Etsuo.
• Source: https://insight.rpxcorp.com/ptab/3758-lg-display-v-innovative-display-technologies-llc-ipr-of-660

IPR2015-00363 — Mercedes-Benz USA, LLC & Mercedes-Benz U.S. International, Inc. v. Innovative Display Technologies LLC

• Type: Inter Partes Review
• Filed: 2015 (pre-institution)
• Status: Terminated — Settlement (Judgment entered 2015-07-02; joint motion filed 2015-06-30)
• Judge panel: Lora M. Green, Thomas L. Giannetti, Miriam L. Quinn (shared caption with 11 companion IPRs)
• Petition grounds: challenged the '660 patent (the petition was filed but never substantively briefed; PO filed no response and no declaration, no depositions taken).
• Institution decision: none reached — terminated in the preliminary stage.
• FWD: none.
• Settlement: Yes. Mercedes-Benz settled a 12-IPR block (IPR2015-00359, -00360, -00361, -00363, -00366, -00368, -00994, -01044, -01067, -01113, -01114, -01115) covering multiple IDT patents; settlement agreement filed as Exhibit 1025/1010 and treated as confidential business information under § 37 C.F.R. § 42.74(c). Each side bore its own fees; the underlying E.D. Tex. action was dismissed. The joint motion expressly states no § 315(e) estoppel attaches to Mercedes and no § 42.73(d)(3) estoppel to IDT.
• Appeal: none.
• Defensive value: Zero precedent value, zero estoppel value. Mercedes bought peace cheaply and its settlement gives you no cover — and is not citable as an admission.

IPR2015-01113 — Mercedes-Benz USA LLC et al. v. Innovative Display Technologies LLC

• Type: Inter Partes Review
• Filed: 2015-04-24
• Status: Termination — Settlement (2015-07-02) — note the Google Patents listing labels this "(Settlement)"; one aggregator (GreyB) mislabels the status as "Institution Denied," which is an error — the paper of record is a joint-motion termination.
• Judge panel: same combined panel (Green, Giannetti, Quinn).
• Petition grounds: '660 patent. Petitioner's exhibits bundled the '660 petition, the '660 file history, Pristash, Ciupke, Kisoo, Seraku, an infringement contention, and the "1094 Institution Denial" — i.e., the Mercedes petition was built on prior-art combinations different from LG's Oe/Etsuo, and expressly used an earlier denial as its roadmap.
• Institution decision: none — terminated 2015-07-02 before institution.
• FWD: none.
• Settlement: same Mercedes-Benz consolidated settlement; confidential; no estoppel.
• Appeal: none.
• Defensive value: Same as -00363 — no citable outcome. But the exhibit list is a useful free roadmap: Pristash, Ciupke, Kisoo and Seraku are art Mercedes thought strong enough to assert on the '660, and they were never tested. For a defendant today, those references are the most obvious starting point for a fresh § 102 challenge that LG's estoppel does not touch.
• Source: https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2015-01113

IPR2015-00495 — Petitioner not identified in my sources v. Innovative Display Technologies LLC

• Type: Inter Partes Review (instituted as a motion for joinder vehicle)
• Filed: ~2014-12/2015-01 (PO Power of Attorney dated 2015-01-19)
• Status: Institution DENIED; Motion for Joinder Dismissed (Decision 2015-07-20)
• Judge panel: not retrieved
• Petition grounds: § 103 against claims 1, 3, 10, 16, 17, 25, 28, 33, and 34 of the '660 patent — a narrower subset of the LG -00487 claim set (compare: -00487 challenged 1, 3, 5, 10, 13, 16, 17, 25–29, 33–35).
• Institution decision: Denied 2015-07-20 under 37 C.F.R. § 42.108, and the accompanying joinder motion dismissed under § 42.122. The Board's denial came after it had already instituted -00487 (2015-07-16), consistent with a redundant/improper joinder posture.
• FWD: none. Settlement: none. Appeal: none.
• ⚠️ Caveat: I could not confirm the petitioner's identity from the retrieved decision text. Do not attribute this petition to any named party without pulling Paper 10.
• Defensive value: Confirms that a third party tried to piggyback on the LG trial and was turned away at the door — the Board does not run parallel trials on this patent for free.

Flagged from the family-litigation data — '660 attribution unverified

The Google Patents family block lists five further PTAB numbers. I verified three of them (-00363, -00495, -01113 are '660). The remaining two I could not confirm target the '660 itself rather than a sibling in this 34-application family (e.g., the '177/-'973 patents hit by LG's IPR2015-00489 and -00506):

There was also a Unified Patents-linked IPR stream in the family (the structured block's PTAB links include UNIFIED-flagged cases), and the E.D. Tex./D. Del. campaigns (IDT and Delaware Display Group, 2013–2015) that drove these filings — but I have no verified Unified Patents IPR specifically against the '660.

Strategic summary

Claim status. Independent claims 1 and 33 are SUSTAINED — the Board, in the only merits decision ever rendered on this patent (IPR2015-00487 FWD, 2016-07-15), held all of claims 1, 3, 5, 10, 13, 16, 17, 25, 26, 27, 28, 29, 33, 34, 35 not unpatentable over Oe + Etsuo. No claim of the '660 has ever been canceled. Untested claims — i.e., claims no petitioner has ever put at risk — are 2, 4, 6, 7, 8, 9, 11, 12, 14, 15, 18, 19, 20, 21, 22, 23, 24, 30, 31, 32, 36, 37, 38, 39, 40. Several of these are substantively interesting: claims 6–7 and 37 (multi-color-chip LEDs mixing in the transition region), claims 14–15 and 38–39 (slots extending inward from the proximal edge), and claims 30–32 (reflective surfaces in close proximity to the light sources) were never reached by LG or Mercedes. If a plaintiff is asserting only claims 1 and 33, it is asserting the two claims its opponents already failed to kill.

Estoppel landscape. § 315(e)(2) estoppel attaches only on FWD. That means: LG Display / LG Electronics are estopped as to any ground raised or that reasonably could have been raised from Oe + Etsuo (subject to whatever happened in CAFC No. 17-1413). Mercedes-Benz is NOT estopped — its joint motion in -00363/-01113 recorded that termination came before any FWD, so § 315(e) never attached. And because -00495 was denied institution without a FWD, its unnamed petitioner is also free. For you, as a non-petitioner, no statutory IPR estoppel applies at all — you may file a fresh IPR, and the Oe/Etsuo failure is not binding precedent against the Board. What is foreclosed as a practical matter is re-running Oe + Etsuo, or any obviousness theory a POSITA would have constructed from those two references. The open ground is § 102: neither LG's nor Mercedes' art was ever tested for anticipation, and the Mercedes exhibit trail (Pristash, Ciupke, Kisoo, Seraku) is untouched.

Pattern signals. This is a textbook Acacia-family monetization cascade: Solid State Opto → Rambus International (2010) → Rambus Delaware/Rambus Inc. (2013-03-04) → Acacia Research Group (2013-06-26) → Innovative Display Technologies LLC (2013-06-27), with IDT and its sister Delaware Display Group filing dozens of E.D. Tex. and D. Del. suits in 2013–2015 (the family block lists 40+ district court cases). The IPR response was defensive and coordinated but fragmented: LG ran one full merits trial (plus a joined twin), Mercedes bought its way out with a 12-IPR block settlement, and at least one further petitioner was denied joinder. The patent owner (IDT) litigated hard and won on the merits, but did not itself appeal. LG appealed and pursued it to the Federal Circuit.

Recommended next steps

1. If your demand letter or complaint cites claims 1, 3, 5, 10, 13, 16, 17, 25–29, 33, 34 or 35 — those claims were adjudicated patentable on the merits. Do not build an IPR on Oe + Etsuo. Pull the IPR2015-00487 FWD (2016-07-15) and read Section on the "plurality of light sources" limitation before you draft anything: https://www.docketalarm.com/cases/PTAB/IPR2015-00487/Inter_Partes_Review_of_U.S._Pat._7404660/
2. Resolve CAFC No. 17-1413 first — this is the single highest-value open question. LG's appeal issues were limited to the obviousness standard and claim construction on Oe/Etsuo. An affirmance locks in the FWD as the last word on those claims; a vacatur/remand (or the appeal never having been perfected) changes the estoppel math for the LG entities. I could not verify the disposition and flag it as unverified rather than guess. Check the CAFC docket and CourtListener before relying on the FWD as final.
3. Exploit the expiration. The patent's term ran out 2015-06-27 and the record status is "Expired – Fee Related." No injunction, no ongoing royalties; only pre-expiration damages, and claim construction is Phillips, not BRI. That substantially devalues a 2026 assertion and should be your lead negotiating point.
4. Attack the untested claims, not the tested ones. If IDT (or a successor) asserts claims 2, 4, 6–9, 11, 12, 14, 15, 18–24, 30–32, 36–40, there is no PTAB record on them at all — those are the claims where a fresh petition has the most room, and where the Board's institution calculus is unencumbered by a prior FWD.
5. Mine the Mercedes exhibit trail. Pristash, Ciupke, Kisoo and Seraku were assembled by a sophisticated petitioner for the '660 and were never substantively adjudicated. Pair them as § 102 references before reaching for Oe/Etsuo.
6. Watch for Unified Patents. The structured block carries UNIFIED-flagged PTAB links in this family, but I found no verified Unified-filed IPR on the '660. If one exists post-ingest, it would be the third independent merits challenge to claims 1/33 — check PTAB E2E/PTACTS for any 2016+ filing against 7,404,660 before finalizing your position.
7. Correct the record internally. The ODP-derived "no PTAB activity" statement is flatly wrong for this patent. Do not let it drive a freedom-to-operate or clearance conclusion.

I'll research the assignment record and litigation posture for US 7,404,660. Let me run several searches in parallel.

Strong signal set emerging. Let me try to pin down reel/frame and correspondent details.

Ownership reconstruction — US 7,404,660 B2 ("Light emitting panel assemblies")

Data-quality caveat up front: The authoritative full text you supplied (Google Patents legal-events table) exposes the sequence of reassignments, assignors and assignees, but not the reel/frame numbers, execution dates, or correspondents of record. Those fields live only in USPTO Assignment Center. I was unable to retrieve them from the sources reachable in this session (searches for the reel/frame and correspondent of this specific patent returned unrelated Acacia-family recordings). I therefore report reel/frame as "not retrieved" rather than fabricating numbers. Every date below is a recording/publication date from the legal-events table — treat execution dates as potentially earlier.

Inventors

Unusual-pattern check: None observed. The '660 patent names a single inventor, and his assignment to the company was recorded ~3 weeks after the 2006-06-16 filing — a routine confirmatory inventor-to-assignee assignment, not an inventor-departure / fire-sale precursor. I found no evidence of all inventors leaving within 12 months of filing (or of any co-inventor dispute).

Original assignee

Solid State Opto Limited — the "Original Assignee" on the issued face of US 7,404,660.

• Line of business: Optical light-guide / backlight components. Its recorded corporate address and downstream reporting tie it to the Global Lighting Technologies (GLT) group (Lee Mang-Shiang / 茂林光電). GLT is a real component manufacturer, so the original assignee is plausibly an operating company — but I could not confirm that Solid State Opto itself shipped a product embodying the '660 claims. Status: unclear / effectively dormant as a patent holder after the 2010 sale.
• Current status: The rights were transferred to Rambus in early 2010 (Rambus's Q2-2010 10-Q describes a December 2009 acquisition of lighting technology from GLT). Taiwan trade press reported the optical-patent package sold to Rambus for ~US$26 million. No bankruptcy, dissolution, or Chapter 7/11 proceeding involving Solid State Opto was found.

Assignment timeline

All entries derive from the Google Patents legal-events table (the same records surfaced by USPTO Assignment Center). Reel/frame and correspondent = not retrieved for any link.

2006-07-05 (recorded) — Reel not retrieved

• Conveyance: Assignment of assignors' interest (inventor → company)
• Assignor: Parker, Jeffery R.
• Assignee: Solid State Opto Limited
• Correspondent: not retrieved — cannot assess recurrence.
• Context: Confirmatory inventor assignment executed just after the 2006-06-16 filing; routine, not a fire-sale indicator.

2010-02-18 (recorded) / executed ~Dec 2009 per Rambus 10-Q — Reel not retrieved

• Conveyance: Assignment
• Assignor: Solid State Opto Limited
• Assignee: Rambus International Ltd.
• Correspondent: not retrieved.
• Context: Portfolio acquisition / sale to a licensing-oriented buyer (Rambus "lighting technology" acquisition; ~$26M reported). This is the point where the patents leave a manufacturing entity.

2013-03-04 (recorded) — Reel not retrieved

• Conveyance: Assignment
• Assignor: Rambus International Ltd.
• Assignee: Rambus Inc.
• Correspondent: not retrieved.
• Context: Internal corporate reorganization (up-streaming into the parent).

2013-03-04 (recorded, same day) — Reel not retrieved

• Conveyance: Assignment
• Assignor: Rambus Inc.
• Assignee: Rambus Delaware LLC
• Correspondent: not retrieved.
• Context: Internal reorganization into a dedicated Delaware IP-holding LLC — the classic pre-sale packaging step. Note the same-day, two-hop structure.

2013-06-26 (recorded) — Reel not retrieved

• Conveyance: Assignment
• Assignor: Rambus Delaware LLC
• Assignee: Acacia Research Group LLC
• Correspondent: not retrieved.
• Context: Transfer-to-asserter. Acacia Research Group LLC is the centralized Acacia (now "Acacia Research Corp") acquisition/assertion entity. A PTAB exhibit in an unrelated Acacia matter shows Acacia Research Group LLC recorded at 6136 Frisco Sq., Frisco, TX, but I cannot tie a correspondent attorney to this recording.

2013-06-27 (recorded) — Reel not retrieved

• Conveyance: Assignment
• Assignor: Acacia Research Group LLC
• Assignee: Innovative Display Technologies LLC
• Correspondent: not retrieved.
• Context: Transfer to a single-purpose assertion LLC, one day before the first infringement suits. IDT LLC was reportedly formed 2013-03-13 and operates from the shared Acacia address at 1701 Directors Blvd, Ste 300, Austin, TX (which also houses Database Sync Solutions, Brilliant Optical Solutions, Cellular Communications Equipment, etc.). Textbook clean-standing + venue setup.

2015-06-27 — Statutory (anticipated) expiration; subsequent status "Expired – Fee Related." No further recorded assignments found after 2013.

Timeline diagram

timeline
    title Ownership of US 7404660
    1995 : Priority application filed
    1997 : US 5613751 issued
    2006 : Application filed
         : Assigned to Solid State Opto Ltd
    2008 : US 7404660 issued
    2010 : Assigned to Rambus International Ltd
    2013 : Assigned to Rambus Inc
         : Assigned to Rambus Delaware LLC
         : Assigned to Acacia Research Group LLC
         : Transfer to Innovative Display LLC
         : First infringement suits filed
    2015 : Patent expires
         : LG institutes IPR

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT.
The patent travels from Solid State Opto Limited (manufacturing-side) → Rambus → Acacia Research Group LLC → Innovative Display Technologies LLC (recorded 2013-06-26 and 2013-06-27, respectively). IDT LLC carries an "LLC" IP-assertion suffix, no product line, no website, and a shared registered-agent address (1701 Directors Blvd Ste 300, Austin TX) shared with ~10 other Acacia shells per Techdirt; it was formed 2013-03-13, months before the transfer. The corporate address pattern is the concrete evidence — not the name alone.

2. Known asserter in the chain — PRESENT.
Both Acacia Research Corp (via Acacia Research Group LLC) and Innovative Display Technologies LLC are on the public PAE lists. Unified Patents' PTAB case list labels the owner "Acacia Research Corp / Innovative Display Technologies LLC … NPE (Patent Assertion Entity)." RPX tracked IDT's campaign from June 2013 onward (see RPX Insight articles cited below). Techdirt's Acacia shell inventory lists "INNOVATIVE DISPLAY TECHNOLOGIES LLC – 37" lawsuits.

3. Repeat correspondent across the chain — UNCLEAR (not verifiable).
Correspondent-of-record data was not retrievable for any of the six recordings. This is the single most important gap — I am not asserting a repeat correspondent, and I decline to guess a name. Action item: pull reel/frame + correspondent for all six recordings at Assignment Center; the 2013-03-04 and 2013-06-2x recordings are the ones to compare first.

4. Cascading transfers — PRESENT.
Five recorded hops in roughly 40 months (2010-02-18 → 2013-06-27), of which four occur inside ~4 months in 2013 (two same-day Rambus hops on 2013-03-04, then Acacia 2013-06-26, then IDT 2013-06-27). The two December 2013-recipient entities (IDT and Delaware Display Group) share the same Acacia principal/address ecosystem.

5. Pre-litigation transfer — PRESENT (strong).
The assignment to Innovative Display Technologies LLC is recorded 2013-06-27; the first IDT infringement suits asserting the '660 patent were filed ~2013-06-28 (Texas E.D., later Delaware). Transfer within ~24 hours of the first complaint is the clearest indicator that the chain was arranged to establish standing and venue.

6. Bankruptcy fire-sale — NOT PRESENT.
No Chapter 7/11, receivership, or 363 sale involving Solid State Opto, Lumitex, GLT, or Rambus was found. The 2010 exit appears to have been a negotiated private sale (~$26M reported), not a distressed sale.

7. Privateering — UNCLEAR.
There is a documented operating/licensing-side transfer to a pure asserter (Rambus → Acacia, 2013), and Rambus retained its own licensing program — but I found no SEC filing or press confirmation that Rambus retained a revenue share / back-end interest in IDT's recoveries. Without that, this is a transfer-to-asserter, not a proven privateering arrangement. Marked unclear by design.

8. Defensive aggregator — NOT PRESENT.
The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. It terminates at an asserting LLC and then expires by law.

Litigation / validity overlay (context for the verdict):

• IPR2015-00487 (LG Electronics v. Innovative Display Technologies) — instituted, '660 patent; RPX reports IDT/DDG were forced to drop '660 from the Delaware cases and refile separately, agreeing to stay pending IPR.
• IPR2015-01717 (LG) — reached Final Written Decision.
• IPR2014-01094 — Not instituted (merits).
• Delaware cases: 1:13-cv-02106 through -02112; 1:14-cv-00849, -00850; 1:15-cv-01220, -01221, -01222. Texas E.D.: 2:13-cv-00524…-00527, -00783, -00784; 2:14-cv-00030, -00106, -00142…-00146, -00200…-00202, -00300…-00302, -00532, -00535, -00624, -00720…-00723. Federal Circuit appeal 17-1413.

Verdict

NPE — high confidence.

Four independent strong signals converge on a single 2013 window: (i) shell-entity transfer — Rambus Delaware LLC → Acacia Research Group LLC (recorded 2013-06-26) → Innovative Display Technologies LLC (recorded 2013-06-27), an entity formed 2013-03-13 with no products and a shared Acacia registered-agent address; (ii) known asserter — Unified Patents classifies the owner as "NPE (Patent Assertion Entity)" for Acacia/IDT; (iii) cascading transfers — five hops in ~40 months, four of them within four months in 2013; and (iv) pre-litigation transfer — assignment to IDT recorded one day before the first infringement suits (~2013-06-28) asserting the '660 patent. The original assignee (Solid State Opto Limited) was a component maker, so the assertion is not an operating-company-vs-competitor dispute. Corroborating secondary sources: RPX Insight (https://insight.rpxcorp.com/news/3720 and https://insight.rpxcorp.com/news/4389), Unified Patents PTAB owner list (https://portal.unifiedpatents.com/ptab/caselist?patent_owners=Acacia+Research+Group+LLC), and Techdirt (https://www.techdirt.com/2014/08/12/know-your-troll-innovative-display-technologies-targeting-any-company-that-creates-product-with-lcd-screen/).

Verification link: USPTO Assignment Center, searched by patent number 7,404,660 — https://assignmentcenter.uspto.gov/ (mirror: https://assignment.uspto.gov/patent/index.html). Retrieve the six recordings' reel/frame and correspondent there; that is the one field that could harden the repeat-correspondent signal I was unable to confirm.

I'll research US 7,404,660 and its cited prior art.

Prior-Art Analysis — US 7,404,660 B2

1. Verification of the patent (searched, not auto-corrected)

Continuity chain (critical for prior-art date): 11/454,822 is a division of 10/784,527 (now US 7,160,015), which is a division of 09/256,275 (now US 6,712,481, filed 1999‑02‑23), which is a continuation‑in‑part of 08/778,089 (now US 6,079,838), which is a division of 08/495,176 (now US 5,613,751, filed 1995‑06‑27).

Legal framework: Filed June 16, 2006 → pre‑AIA 35 U.S.C. § 102 governs. Because 09/256,275 was a CIP, any claim element not supported by the 1995 disclosure is only entitled to the 1999‑02‑23 date. References published more than one year before the controlling date qualify under § 102(b); others fall under § 102(a)/(e). Everything below predates both candidate dates, so each qualifies as prior art regardless of which priority date applies.

2. Sourcing limitation (read this before relying on the list)

I could not pull the official USPTO IDS/“References Cited” enumeration for 7,404,660 directly through my available tools. The citations below are reconstructed from the machine‑readable citation records of the same patent family on Google Patents and Justia:

• https://patents.google.com/patent/US20060274554 (family record listing “Patent Citations (63)”)
• https://patents.justia.com/patent/[7963687](/patent/7963687) and https://patents.justia.com/patent/[8142063](/patent/8142063) (family “Referenced Cited” lists)
• https://www.freepatentsonline.com/7963687.html and /7513672.html (family “US Patent References”)

Because all members share one specification, the citation sets overlap heavily, but I cannot guarantee the list is element‑for‑element identical to 7,404,660's own IDS, and I could not verify the foreign patent documents or non‑patent literature citations (unknown / not retrieved — I will not invent them). Treat the “potentially anticipates” column as an issue-spotting screen, not a legal conclusion.

3. Patent citations — citation, date, subject matter, and § 102 exposure

Group A — Earliest US patents (1949–1973)

Group B — 1974–1986 (the most structurally pertinent block)

Group C — Later references (1991–2003, from the continuation IDS records)

The following appear in the family “US Patent References” lists (FreePatentsOnline, above) and are largely secondary/general lighting-art references rather than edge‑lit‑panel anticipatory art; they are most useful for § 103 combinations and for the “deformity/extraction pattern” and “reflector” limitations:

US 4,974,124; US 4,991,064; US 5,005,108; US 5,050,946; US 5,096,765; US 5,134,545; US 5,136,480; US 5,150,965; US 5,184,882; US 5,190,370; US 5,225,818; US 5,236,215; US 5,289,351; US 5,307,241; US 5,315,491; US 5,349,502; US 5,353,291; US 5,363,470; US 5,377,084; US 5,390,088; US 5,396,350; US 5,398,179; US 5,402,327; US 5,410,454; US 5,432,876; US 5,438,484; US 5,448,468; US 5,467,417; US 5,475,539; US 5,528,720; US 5,530,626; US 5,550,672; US 5,575,549; US 5,588,737; US 5,611,613; US 5,618,096; US 5,671,994; US 5,711,592; US 5,719,649; US 5,749,646; US 5,771,328; US 5,775,791; US 5,779,337; US 5,779,338; US 5,808,784; US 5,828,488; US 5,844,720; US 5,851,062; US 5,890,791; US 5,917,664; US 5,919,551; US 5,931,555; US 5,947,578; US 5,961,198; US 5,971,559; US 5,999,685; US 6,011,602; US 6,036,329; US 6,091,547; US 6,120,280; US 6,130,730; US 6,151,169; US 6,172,809; US 6,402,334; US 6,406,158; US 6,565,225; US 6,582,103; US 6,676,284.

⚠️ Caveat: these later numbers were returned as family-record entries; I could not individually confirm that each was cited against 7,404,660 rather than against a later continuation in the same family.

4. Intrinsic references (not “prior art” citations, but important)

The specification itself incorporates by reference:

“any of the types disclosed in U.S. Pat. Nos. 4,897,771 and 5,005,108, assigned to the same assignee as the present application, the entire disclosures of which are incorporated herein by reference.”

These are not § 102 prior art to the extent they are commonly owned and incorporated (they are part of the intrinsic record and are highly relevant to claim construction of “light source,” claims 1, 5, 16, 33).

Also not prior art but decisive for the priority date: the parent/priority patents US 5,613,751; US 6,079,838; US 6,712,481; US 7,160,015, and the four sibling continuations in the family (US 7,513,672; US 7,524,101; US 7,563,012; and the cohort US 7,404,661, US 7,465,043, etc.).

5. Claim-by-claim anticipation screen (§ 102) — strongest candidates

Because claim 1 and claim 33 are the only independent claims, anticipation must be assessed there first.

Claim 1 requires, in combination: (a) planar optical conductor; (b) input edge with width > thickness; (c) a plurality of light sources each with an output distribution wider than tall; (d) an output region with a predetermined pattern of deformities; (e) a transition region between the source and the output region.

• US 4,573,766 (Bournay, Cordis) — best § 102 candidate: staggered plural LEDs along the input edge of a planar LCD backlight, with light extraction. Anticipatory exposure primarily to claims 1, 5, 33 (and claim 6/19/20 if multi‑color LEDs are disclosed — I could not confirm that from the retrieved data).
• US 3,752,974 (Baker, Coastal Dynamics) and US 4,043,636 (Eberhardt, Intel) — planar, edge‑illuminated conductor with a controlled (non‑uniform) extraction pattern; each reads on (a), (b), (d) and arguably (e). Exposure mainly to claims 1, 2, 10, 11.
• US 4,323,951 (Pasco) — the “laminar light guide” limitation maps to claim 1(b); claims 1, 10, 13.
• US 4,277,817 (Hehr, Blaupunkt) — maps to the enlarged transition/thickness limitations: claims 1, 8, 9.
• US 4,128,332 (Rowe, Xerox) — the “illuminator illuminating a guide” maps to the transition region limitation of claims 1, 2, 3.

Claim 33 requires LED sources wider than tall with a wider‑than‑tall output distribution, extraction pattern, and transition region. The only retrieved reference squarely disclosing plural LEDs at the edge is US 4,573,766; the wider‑than‑tall LED package geometry appears more likely to be addressed by the later package‑art references (e.g., the US 5,xxx/6,xxx block in Group C). I would expect claim 33 to be the harder claim to anticipate but easier to attack under § 103 (LED + edge‑lit panel + extraction pattern).

Dependent-claim exposure by limitation:

• Reflective/refractive or faceted surfaces near the source (claims 4, 15, 22, 24, 30–32, 36, 39, 40): US 4,277,817; US 4,128,332; US 4,791,540.
• Tray / edge / bottom reflectors (claims 25–29): US 4,791,540 (Dreyer); US 4,791,807‑class reflector art in Group C; the Cordis and Intel references.
• Slots from the proximal edge defining the effective input edge (claims 14, 23, 38) and multiple input edges (claim 21): no retrieved reference discloses this geometry — these look like the most defensible dependent claims.
• Multi‑color chips mixing in the transition region (claims 6, 7, 19, 20, 37): the multi‑chip LED art in Group C is the relevant field; no retrieved reference appears squarely anticipatory.

6. Contested validity — what actually happened to these claims

The record shows this patent was heavily litigated, which is where the authoritative validity determinations live:

• IPRs (all filed 2015, per the Google Patents litigation block): IPR2015‑00487 and IPR2015‑01717 reached Final Written Decision; IPR2015‑00363, ‑00745, ‑00855, ‑00897, ‑01113 reached Settlement; IPR2015‑00495 was Not Instituted (procedural); IPR2014‑01094 was Not Instituted (merits).
• District court: numerous D. Del. and E.D. Tex. actions (e.g., 1:13‑cv‑02106 through ‑02112; 2:13‑cv‑00524 through ‑00527).
• Appeal: Fed. Cir. No. 17‑1413.

I cannot state the substantive outcomes (which claims were cancelled or upheld) — the retrieved record identifies the proceedings only, not their holdings. If you need the operative anticipation ruling, the IPR2015‑00487 and IPR2015‑01717 Final Written Decisions are the controlling documents and should be pulled from the PTAB record directly.

7. Bottom line

1. US 7,404,660 B2 is confirmed: Parker, Solid State Opto, app. 11/454,822, filed 2006‑06‑16, granted 2008‑07‑29, priority 1995‑06‑27, expired.
2. Its cited prior art is a 1949–1995 edge‑lit‑panel corpus, dominated by US 4,573,766 (Bournay/Cordis), US 3,752,974 (Baker), US 4,043,636 (Eberhardt/Intel), US 4,323,951 (Pasco) and US 4,277,817 (Hehr) — my strongest § 102 candidates for claims 1 and 33 and their thickness/transition dependents.
3. The slot‑defined input edge claims (14, 23, 38) and the faceting/transition‑region reflector claims (30–32) have the least apparent anticipatory art in the retrieved set.
4. I could not verify the exact IDS enumeration for 7,404,660, nor any foreign‑patent or non‑patent‑literature citations, nor the PTAB/Fed. Cir. holdings. Those should be confirmed against the USPTO PatentCenter file wrapper and the PTAB FWDs before any validity conclusion is relied upon.

I'll research the prior art cited on this patent and the PTAB challenges to it before building the § 103 analysis.

Strong leads. Let me pull the details of the key references and the Board's reasoning.

Scope and Caveat on Sources

The page you supplied for US7404660 ("Light emitting panel assemblies," Jeffery R. Parker, filed 2006-06-16, granted 2008-07-29, priority 1995-06-27, current assignee Innovative Display Technologies LLC) rendered only the prior-art metadata — "Prior art keywords: light / assembly / optical conductor / light sources / edge," prior art date 1995-06-27 — with the reference listing itself not populated in the fetched text. I therefore supplemented from the "US Patent References" listing on the FreePatentsOnline mirror of the same patent (freepatentsonline.com/7404660.html) and from the institution/final-written-decision record of the PTAB proceedings against this exact patent (IPR2015-00487; IPR2015-01717). Where I rely on the IPR record I say so explicitly, because that record is the best available proxy for "the prior art of record" against these claims. Anything I could not verify is flagged.

I. Effective date and legal framework

• Priority chain (from the face of the patent): US 11/454,822 (2006-06-16) ← division of 10/784,527 (2004-02-23, now US 7,160,015) ← division of 09/256,275 (1999-02-23, now US 6,712,481) ← CIP of 08/778,089 (1997-01-02, now US 6,079,838) ← division of 08/495,176 (1995-06-27, now US 5,613,751).
• Governing statute: Because every application in the chain was filed before 2013-03-16, the AIA's § 103 does not apply. The analysis is under pre-AIA 35 U.S.C. § 103(a), with § 102(b)/§ 102(e) defining what is prior art. The PTAB analyzed it this way (the FWD cites "35 U.S.C. § 103(a)"), and applied Oe, whose US filing traces to a 1993-06-22 parent application.
• Critical practical point: The presumption is that the claims are entitled to 1995-06-27, which makes Etsuo (JP 07-064078, published 1995-03-10) and Tsunoda (JP H5-210014, published 1993-08-20) § 102(b) art, and makes Oe, Tai, Tsuchiyama and Funamoto § 102(e) art via their earlier US filing dates. However, several references cited on the face of US7404660 issue in 1999–2004 (e.g., US 6,582,103 Popovich; US 6,676,284 Wynne Willson; US 6,176,809 Koike; US 6,402,334 Yu-San; US 6,131,730 Jannson). Those can only be prior art if some claims are not entitled to 1995-06-27 because the supporting disclosure was added in the 1997 or 1999 CIP. That is a written-description/priority question I cannot fully resolve from the materials supplied; I treat it as a separate vulnerability in § IV.D below.

II. What the claims actually require

Two independent claims, and they are not coextensive:

Claim 1 requires: (a) a generally planar optical conductor; (b) at least one input edge with greater cross-sectional width than thickness; (c) a plurality of light sources configured to generate light whose output distribution has a greater width component than height component; (d) the sources positioned adjacent the input edge; (e) at least one output region with a predetermined pattern of deformities causing emission; and (f) a transition region disposed between the light source and the output region.

Claim 33 is narrower in one respect and parallel in another: it requires a plurality of LED light sources, "each having a greater width than height" (i.e., a physical package dimension of the emitter), positioned adjacent the input edge, each configured to generate light having an output distribution defined by a greater width component than height component.

Dependent claims cluster into five families:

One of skill in the art (POSITA) here is a backlight/light-guide design engineer with a mechanical-optics background and working familiarity with LCD edge-lit backlights, LED packaging, and total-internal-reflection extraction structures.

III. Prior art of record

IV. Obviousness combinations

A. Combination 1 (primary, and the only one the Board instituted): Oe + Etsuo

This was the ground actually tried: IPR2015-00487, LG Display Co., Ltd. v. Innovative Display Technologies LLC, instituted 2015-07-16 on claims 1, 3, 5, 10, 13, 16, 17, 25–29, 33–35 under § 103(a) over Oe and Etsuo; final written decision 2016-07-15, with the Board finding no claims unpatentable; appeal docketed at the Federal Circuit as 17-1413. (FWD/Notice of Appeal)

Element mapping (Petitioner's theory):

• Optical conductor, input edge wider than thick → Oe's acrylic light guide (150 mm × 250 mm × 3 mm in the worked example), with a side end surface 7 as the light incident surface.
• Output region + predetermined pattern of deformities → Oe's roughened surface / printed white-ink pattern on the surface opposite the emitting surface, with flat-area ratio increasing nearer the incident surface — i.e., a graded, non-uniform extraction pattern for uniformity. This is a strikingly close antecedent to the '660's own "denser concentration of light extracting deformities with increased distance from the light source."
• Transition region between source and output region → the portion of Oe's light guide adjacent to the lamp but before the extraction region (Petitioner characterized it as coextensive with the guide, which is how the '660 describes FIGS. 1 and 3).
• Light sources adjacent the edge → Oe's fluorescent/linear/point source 4 in a lamp case with reflector 5 (Oe's case is a cylindrical reflector with a 3 mm aperture in the worked example).
• Plurality of light sources; width > height output distribution → supplied from Etsuo: multiple LED lamp units L at the edge, each with reflector plates, and the rectangular LED packages.

Motivation argued: Both references are in LCD edge-lit backlighting; LEDs were a known substitute for fluorescent lamps at a light-guide edge, offering thinner form factor, lower power, and longer life; Etsuo expressly uses LEDs in an edge-lit configuration with reflectors to direct light toward the conductor edge.

Why the Board rejected it — and why it matters to your analysis:

1. Insufficient articulated motivation. The Board held that "the mere fact that both Oe and Etsuo are in the same field of endeavor falls short of an adequate rationale," and that a rationale directed only to "both use light sources at the edge of a light guide" does not explain why a POSITA would replace Oe's single source with Etsuo's plurality. This is a KSR "articulated reasoning with rational underpinning" failure, not a technical impossibility.
2. The width-vs-height output distribution was not shown. Patent Owner argued, and the Board apparently credited, that light exiting Etsuo's LEDs "has a negligible distribution in its width direction as compared to its height direction," and that light spreads in the width direction only after entering the light conductor (expert declaration excerpts). This limitation is the crux of both independent claims.
3. "Plurality of light sources adjacent to the input edge" was not shown. Etsuo's Fig. 6 configuration of six lamp units L on each side was read by Patent Owner either as one light source per side or as sets placed on opposite edges — either reading fails a claim requiring a plurality of sources adjacent the input edge.

My assessment of the combination's strength, independent of the Board's ruling: On the merits, Oe is a remarkably close primary reference — it teaches the graded-deformity extraction pattern, the edge-lit planar conductor, the reflector layer, and even mentions thickness variation along the light direction (relevant to claims 8–11). The combination's vulnerability is entirely in (i) the output-distribution limitation, which is a light-source characterization limitation, and (ii) the motivation. Notably, the Vince A. Thomas declaration (Ex. 1030) on commercial success in IPR2015-00487 shows Patent Owner affirmatively defended the claims on objective indicia, and the Board's decision not to invalidate means Oe+Etsuo should be treated as adjudicated insufficient for claims 1, 3, 5, 10, 13, 16, 17, 25–29, and 33–35. Any new § 103 theory would need either (a) a better motivation (e.g., a reference teaching that a wide-but-short emitter footprint improves in-coupling into a thin wide guide edge), or (b) a different secondary reference that expressly discloses wide-angle or width-dominant emission from an edge-mounted LED.

B. Combination 2: Tai + Funamoto (the § 103 ground directed at claim 28)

This ground was asserted (Petitioner argued Tai meets all elements of claim 1, with Funamoto supplying only the tray of claim 25, from which claim 28 depends) but was not the instituted ground; the Board instituted only Oe+Etsuo. (exparte analysis of the -01717 record)

• Tai (US 5,359,691) supplies: light pipe (optical conductor) with pyramid microprisms (the "predetermined pattern of deformities" causing extraction), a cross-section with greater width than thickness, a plurality of light sources at an input edge, and a collimating/multi-reflection injection assembly that functions as the claimed transition region.
• Funamoto (US 5,619,351) supplies only the tray in which the assembly is housed (claims 25 and 28: tray including a reflector associated with the bottom surface of the conductor).
• Motivation: a tray/case is a conventional housing for protecting and integrating an edge-lit backlight into an LCD module; both references share the objective of a more efficient edge-lit panel; the combination is a simple substitution of a known element (a housing tray) for a known purpose, yielding predictable results — the classic KSR rationale. Expectation of success is high because no functional interaction between the tray and the optical extraction mechanism is required.

Note the structural dependency: claim 28 → claim 25 → claim 1. A Tai+Funamoto theory therefore stands or falls with the claim-1 showing on the width > height output distribution and plurality of light sources adjacent to the input edge limitations — precisely the two elements the Board found lacking in Etsuo. Whether Tai's multi-source injection assembly cures that gap (Tai expressly uses multiple sources feeding a single wide edge, which is a stronger "plurality adjacent one input edge" showing than Etsuo's Fig. 6) is the key question, and I cannot verify from the available record whether the Board reached that ground on the merits or declined institution.

C. Combination 3: Tsuchiyama + Tsunoda

Asserted in the IPR papers against claims 1, 3, 5, 10, and 25–29 (Tsuchiyama US 5,548,271 for the edge-lit light guide/surface light source architecture; Tsunoda JP H5-210014 for the surface-light-source features). This was also not the instituted ground. The rationale would again rest on the shared field of edge-lit surface light sources and the conventionality of arranging point/linear sources along a light-guide side surface — the same rationale the Board found too thin when offered for Oe+Etsuo, which is a caution for this pairing as well.

D. Combinations targeting the dependent-claim families (a la carte)

Because these limitations are narrower than the independent claims, they are where § 103 is most likely to succeed — but each requires its own primary reference pair:

• Claims 8, 9, 11 (thickness gradient): Oe's own background expressly describes prior art in which "the thickness of the light guide is varied along the direction of light," and describes light guides "having a thickness which varies according to the distance from a lamp." Combined with any edge-lit LED guide (e.g., Etsuo's lamp units), the motivation is direct: a wedge guide is a known technique for improving uniformity/efficiency in edge-lit panels. (Oe background text)
• Claims 12, 13, 25–29 (reflectors and tray): Oe itself recites a light reflecting layer on the surface opposite the light emitting surface (claims 12/28-adjacent), and US 6,131,730 (Jannson, "Backlight assembly for a display") and the Etsuo reflector plates 11–15 supply edge reflectors. A tray-with-reflector is conventional housing art (Funamoto).
• Claims 4, 15, 22, 24, 32, 36, 39, 40 (faceted surfaces in close proximity to the light sources): This is the one family where the effective-date vulnerability matters most. The face-of-patent art US 6,582,103 (Popovich, "Lighting apparatus"), US 6,676,284 (Wynne Willson, "Apparatus and method for providing a linear effect"), and the prism-sheet art (US 5,771,328 Wortman; US 5,919,551 Cobb; US 5,917,664 O'Neill) all post-date 1995. If the "faceted surfaces in close proximity to the light source" language is supported only by the 1997/1999 CIP disclosures, then these 1999–2004 references become § 102(e) art against the claims whose effective date moves forward. I cannot confirm the CIP-additions question from the materials provided, and I flag this as an inference, not a verified finding. If those references are available, the combination of an edge-lit guide (Oe/Tai) with faceted/prismatic surfaces at the source region is straightforwardly obvious — faceting a transition region and a prism sheet to redirect off-axis rays is the standard, predictable technique in this art and is exactly what Oe and Tai already do downstream of the source.
• Claims 6, 7, 19, 20 (multi-color chips / LEDs, color mixing in the transition region): These are the hardest. The '660's own continuation US 11/013,833 was titled "Light emitting panel assemblies having LEDs of multiple colors," which tells you the family treated multi-color mixing as a distinct contribution. I did not verify a specific reference in the supplied prior-art section that expressly discloses a red/green/blue LED cluster adjacent a light-guide edge for mixing into white. Stating this plainly: I do not have a grounded reference for claims 6–7 and 19–20 from the record available to me, and I will not fabricate one. If such art exists (RGB edge-lit backlight art is dense in the mid-1990s), the motivation is still the classic one — LED intensity control is a known way to synthesize arbitrary colors, and the "transition region" mixing requirement is satisfied by any sufficiently long pre-extraction guide segment, which Oe already has.
• Claims 14, 21, 23, 38 (slots extending inward from the proximal edge, whose inner edges act as the input edge): This is a receptacle/insert architecture (the '660 describes FIGS. 9 and 15 slot-and-insert embodiments). The face-of-patent art US 6,036,329 (Iimura, "Flat light source and passive display utilizing bypass light guide") and US 6,011,602 (Miyashita, light guiding body with trapezoid projections) are candidates, but I did not verify either discloses edge slots. Flagged as unverified.
• Claims 16, 17, 18 (focused light sources; light focused substantially perpendicular to the input edge): Notably, the Board's institution decision considered "focused light sources" — the ptacts record includes a construction discussion around claim 1 and the limitation set, and the printed/roughened-surface regulation function of Oe. A lens-ended or collimating source feeding a guide edge is conventional (Tai's microprism injection is a collimating structure); motivation is efficiency of in-coupling.

V. Motivation-to-combine rationales a POSITA could articulate

To survive scrutiny after the Board's criticism of the Oe+Etsuo rationale, a theory must state a technical reason, not a field-of-endeavor reason. The defensible ones here:

1. Substitution of a known element to obtain predictable results (KSR factor (4)): Replacing a fluorescent/linear lamp at a light-guide edge with LED lamp units is a known substitution in edge-lit backlights, with the predictable benefits of reduced thickness, lower power, longer life, and elimination of an inverter.
2. Known technique to improve a similar device (KSR factor (1)): Graded extraction patterns (Oe's varying flat-area ratio) are the known technique for luminance uniformity; the '660's own specification concedes this is a well-known approach ("[v]arying the percentages and/or size of deformities in different areas of the panel is necessary in order to provide a uniform light output distribution").
3. Design incentive / predictable use of a wedge (KSR factor (5)): Greater thickness near the source increases in-coupling aperture and the number of TIR bounces available; reducing thickness downstream matches extraction demand. This is the express rationale for claims 8, 9, 11.
4. Housing/integration expedient (KSR factor (4)): A tray with reflective surfaces is a standard module housing that also performs back/side reflection — the Funamoto rationale for claims 25–29.
5. Obvious to try (KSR factor (7)): For the width-vs-height output distribution, one could argue the art recognized that a wide, short emitting aperture is preferable for coupling into a wide, thin guide edge; but the Board's finding shows this must be evidenced, not assumed.

VI. Countervailing evidence that must be weighed

• The adjudicated result. For claims 1, 3, 5, 10, 13, 16, 17, 25–29, 33–35, the PTAB's final written decision was "No Claims Unpatentable" on the only instituted ground. That is strong, specific, on-the-record evidence against the strongest asserted § 103 theory. Petitioner's appeal to the Federal Circuit (17-1413) challenged (a) the non-obviousness finding and (b) the Board's application of the legal standard for obviousness, and (c) the claim construction — i.e., Petitioner attacked the legal framework, which signals the factual record on the output-distribution limitation was the losing issue.
• Objective indicia. Patent Owner submitted a commercial-success declaration (Ex. 1030, Vincent A. Thomas). Any new obviousness theory must contend with the nexus question (does the commercial success of an LCD backlight derive from the claimed combination or from unrelated market factors?), which is where these showings usually fail or succeed.
• Expiry / status. The patent's anticipated expiration is 2015-06-27 and the status is "Expired - Fee Related." Practically, the stakes are historical damages, not prospective injunctive relief.

VII. Bottom line

1. The single best-grounded § 103 combination is Oe (US 5,711,589) as the primary reference + Etsuo (JP 07-064078) for claims 1, 3, 5, 10, 13, 16, 17, 25–29 and 33–35. Every element of claim 1 except the "plurality of light sources" / "greater width component than height component" output-distribution limitations is met by Oe with unusual closeness (graded deformities, edge-lit planar guide, transition region, reflector layer). But this exact combination was tried and rejected, on motivation and on the output-distribution limitation.
2. Tai (US 5,359,691) + Funamoto (US 5,619,351) is the most promising re-packaging for the claim 1 / claim 25 / claim 28 path, because Tai expressly teaches a plurality of light sources feeding one wide input edge with microprism extraction — which addresses the very "plurality adjacent to the input edge" defect the Board identified in Etsuo. Its weakness is the motivation rationale, not the disclosure.
3. Dependent-claim families split by difficulty: thickness-gradient claims (8, 9, 11) are the most clearly obvious (Oe's own background describes varying guide thickness); reflector/tray claims (12, 13, 25–29) are close behind; faceted-surface claims (4, 15, 22, 24, 32, 36, 39, 40) depend on a priority/effective-date determination I could not complete — if the "faceted surfaces in close proximity to the light sources" language was added in the 1997 or 1999 CIP, the 1999–2004 cited patents (Popovich, Wynne Willson, Wortman, Cobb, O'Neill) become available and these claims are readily obvious; and the multi-color mixing claims (6, 7, 19, 20) are the weakest link in any § 103 theory I can construct from the available record, because I could not verify a reference in the cited art expressly disclosing multi-chip / multi-color LED mixing in an edge transition region.
4. Unverified items I am explicitly not asserting: the specific disclosures of Tsuchiyama (US 5,548,271) and Tsunoda (JP H5-210014); whether the Tai+Funamoto and Tsuchiyama+Tsunoda grounds were denied institution or simply not instituted in -00487; whether US 6,036,329 or US 6,011,602 disclose edge slots; and the CIP-additions/priority analysis underpinning § IV.D. Each of these requires the reference documents and the parent applications themselves, which were not in the supplied page text.