US 7639244

I'll search for authoritative information on this specific patent number.

Let me run targeted searches for any 2026 litigation or appeal involving this specific patent number.

Summary of US 7,639,244 (from the authoritative patent text)

Important caveat up front: I found no CAFC 2026 docket entry, PTAB proceeding, or district-court case specifically involving US 7,639,244. I cannot confirm any such activity from the available search results, and I will not fabricate one. Multiple search hits containing the string "7639244" or a "244 patent" are unrelated identifiers (see "False matches" below).

Bibliographic data

Source: Google Patents US7639244B2; Justia.

Abstract (verbatim)

"A display includes an array of pixel circuits and data drivers to drive the pixel circuits. The data drivers include a first data driver to receive pixel data according to a first clock frequency and to forward some of the pixel data to a second data driver according to a second clock frequency, the second clock frequency being different from the first clock frequency."

Plain-language overview of the independent claims

Claim 1 — "Designated-driver" display with unequal signal-line widths (display apparatus).
A display has an array of pixel circuits. A first data driver receives pixel data from a timing controller and uses it to drive a first portion of the pixels; it also receives additional pixel data that it does not use. A second data driver receives that additional data from the first driver (not directly from the controller) and drives a second portion of pixels. The distinguishing limitation: the first driver receives its own data from the timing controller over a first number of signal lines, and receives the additional (pass-along) data over a second, different number of signal lines.

Claim 12 — Cascade with unequal line counts (display apparatus).
A display with an array of pixel circuits and data drivers, where a first data driver receives pixel data through a first number of signal lines and forwards some of it to a second data driver through a second, different number of signal lines; the second driver uses the received data to drive its own pixels. (Dependent claim 14 adds that the second number is less than the first; claim 13 covers sending portions to two drivers simultaneously.)

Claim 17 — Multi-clock, staged buffering across five drivers (display apparatus).
A substrate carrying pixel circuits; a timing controller outputs pixel data plus first, second and third clock signals, with the second and third clocks running slower than the first. Three data drivers: during a first time period the first driver receives data per the first clock and buffers it; during a second time period the first driver receives data per the first clock and forwards portions to the second and third drivers per the second and third clocks respectively, each buffering what it receives. The claim as issued also positively recites fourth and fifth data drivers that, during a third time period, receive forwarded data from the second/third drivers and buffer it. (Claim 18 adds that all five then drive their pixels from their buffers.)

Claim 19 — Method of operating a display (method claim).
Transmitting pixel data from a timing controller to a first data driver through a first number of signal lines, and transmitting that pixel data from the first data driver to a second data driver through a second number of signal lines, the first number being different from the second. (Claim 20 adds different transmit clock frequencies; claim 21/22 add driving pixels with the second driver; claim 23 adds sending first/second/third pixel data to the first driver; claim 24 specifies glass-substrate signal lines; claim 25 recites the data representing chroma values for different portions of a pixel row.)

Technical gist (why it exists)

The patent addresses electromagnetic interference (EMI) and bezel width in flat-panel (LCD) displays. Rather than the timing controller driving every data driver in parallel over the FPC, all pixel data goes to one "designated" data driver, which keeps its own share and forwards the rest over wire-on-array (WOA) transmission lines formed directly on the glass to neighboring drivers. Two main levers reduce EMI and line count: (1) lowering the inter-driver clock frequency (e.g., half the controller clock, using odd/even clock phases); and (2) reducing the number of signal lines between drivers (e.g., 10 lines in vs. 5 lines out), shrinking the bezel. It also describes TTL signaling (vs. mini-LVDS/whisper-bus) for noise tolerance and simpler, lower-power drivers, and a post-passivation gold-bump structure to lower on-glass line impedance. Embodiments cover a 17‑inch SXGA (1280×1024) panel with 384-channel drivers.

Litigation / prosecution check — no confirmed 2026 activity

• Family: two continuation applications (US 12/618,176 → US 2010/0060617 A1, and US 14/200,052 → US 2014/0184576 A1), both abandoned; the '244 patent is the only granted member in the US family. Foreign counterparts include JP 5071701 B2, KR 101189922 B1, and TW I345214 B.
• Assignee context: the patent passed from Chi Mei Optoelectronics → Chimei Innolux (2010 merger) → Innolux Corporation. Innolux has appeared in unrelated proceedings (e.g., Phenix Longhorn v. Innolux, E.D. Tex.; IPR2025‑00043/‑00044 against Phenix Longhorn's '305 and '788 patents — institution denied), but none of those involves US 7,639,244.
• No CAFC appeal, IPR, or district-court assertion of US 7,639,244 was found. Given its "Expired – Fee Related" status with adjusted expiration 2028‑02‑04, active 2026 appellate activity on this specific patent is unlikely, but I cannot rule it out definitively from the searches performed — treat "no known 2026 CAFC docket" as a search result, not a certified clearance.

False matches explicitly ruled out (do not conflate with US 7,639,244)

• Japanese Patent No. 7639244 — 特許第7639244号 (2025) — an unrelated pipe-type folding-door lock device (E06B classification), assignee 株式会社TOKO. A different national patent with a coincidentally identical number.
• "the '244 patent" in InterDigital v. ZTE — refers to an InterDigital patent in PTAB remand, not this patent.
• "the '924 Patent" in the Deere seed-delivery complaint — a seed delivery apparatus patent, unrelated.
• Various "US7639244B2 *" entries on other Google Patents pages (JP3779687B2, JP3895897B2, etc.) are merely this patent appearing in citation lists, confirming it has been cited as prior art but not that it was litigated.

Uncertainty statement: The bibliographic data and claim text above come from the authoritative source provided and are high-confidence. The absence of any 2026 CAFC/PTAB activity is based on my searches and is not a guarantee; a query of the CAFC's CM/ECF docket and USPTO PatentCenter/PTAB API under the exact number "7,639,244" (and application 11/445,943) would be needed for a definitive negative.

I'll search for litigation involving US patent 7,639,244 across patent litigation databases.

Let me run more targeted searches on litigation databases and Innolux/Chi Mei enforcement activity.

Let me check a few more specialized litigation sources and the Justia patent page directly.

Litigation search results for US Patent No. 7,639,244

Bottom line: I found no known litigation — district court, ITC, PTAB, or CAFC — involving US 7,639,244. Searching litigation-oriented sources (Unified Patents litigation portal, Justia, CourtListener/RECAP, PACER-adjacent aggregators, SEC filings, and general web) produced zero cases in which US 7,639,244 ("Flat panel display using data drivers with low electromagnetic interference") was asserted, challenged, or otherwise at issue.

Cases: none to report

Critical warning — every "244" hit returned is a different patent

The search hits containing "244" or "7639" are false matches. Do not conflate any of them with US 7,639,244:

Context supporting a "no litigation" finding

1. Legal status: US 7,639,244 is "Expired – Fee Related" with an adjusted expiration of 2028‑02‑04. A patent in expired/fee-lapsed status is a poor candidate for assertion, which is consistent with the absence of any identified suit.
2. Family history: The two US continuations in the family — US 12/618,176 (US 2010/0060617 A1) and US 14/200,052 (US 2014/0184576 A1) — were both abandoned. The '244 patent is the sole granted US member.
3. Assignee activity is unrelated: Innolux/Chi Mei has appeared in proceedings such as the In re TFT‑LCD (Flat Panel) Antitrust Litigation (N.D. Cal., MDL 3:07‑md‑1827‑SI), Eidos Display v. Chi Mei Innolux (E.D. Tex. 6:11‑cv‑00201), ITC Inv. No. 337‑TA‑737, and IPR2025‑00043/‑00044 (Phenix Longhorn, institution denied). None of these involves US 7,639,244 — the asserted patents were different (e.g., '092, '019, '241, '958, '480, '102, '413, '311, '978, '204).

Consistency with the prior section

This confirms — and is consistent with — the earlier-generated section of this analysis, which likewise found no CAFC 2026 docket entry, no PTAB proceeding, and no district-court case specifically involving US 7,639,244. No contradiction to flag.

Uncertainty statement (required)

• The absence of litigation above is a search result, not a certified negative. I did not obtain a direct query confirmation from PACER/CM‑ECF, the USPTO PatentCenter/PTAB API, or Docket Navigator under the exact identifiers "7,639,244" and application "11/445,943." A definitive negative would require querying those systems directly, since patent-number searches on general web engines are heavily polluted by the coincidental matches listed above.
• Per the operating rules, I have not auto-corrected any identifier. Every number is reported exactly as encountered.
• I cannot rule out a small, recently filed, or sealed action that has not been indexed by the sources searched. But on the evidence available, the correct answer is: no litigation is known involving US 7,639,244.

I'll verify the canonical empty list against public sources before writing up.

Let me run a few more targeted checks before finalizing.

Proceedings overview

Total AIA trial proceedings on US 7,639,244: zero. The canonical structured feed — the USPTO Open Data Portal "PTAB proceedings on file" block supplied in this prompt — returns no IPR, no PGR, and no CBM for this patent or its application (11/445,943). There is therefore no proceeding to break down by active / invalidated / sustained / settled / institution-denied: there is no institution decision, no Final Written Decision, no termination, and no Federal Circuit appeal to report. The bottom-line defensive posture for a defendant is not "this patent has been hardened by surviving IPRs," and it is not "the claims are canceled." It is a third, less-common posture: an unexercised patent. The claims are fully intact and completely untested in any adversarial validity forum. That cuts both ways — no claim has been canceled, so a demand letter citing claims 1–12 is not sanction-bait on its face; but the patent has also never been through a Board validity scrub, so an IPR petitioner would be writing on a blank slate with no prior art, expert, or claim-construction record to inherit or attack. The searches corroborate the structured feed — see the false-positive disambiguation below, which is where the "244" string keeps surfacing.

Sourcing note: the count of zero comes from the structured ODP block (authoritative for this task) and is corroborated by (a) web searches for IPR/petition activity tied to "7,639,244," and (b) the absence of any PTAB/IPR section on the Google Patents US7639244B2 page, whose only litigation-adjacent content is a Cited By list (i.e., the patent has been cited as prior art by later filings, which is not the same as being challenged). Absence of a proceeding from the ODP feed is strong but is a search result, not a certified negative — I cannot rule out a very recently filed petition not yet ingested.

Proceeding-by-proceeding

None. No IPR####-#####, PGR####-#####, or CBM####-##### number exists for this patent to analyze, so the prescribed per-proceeding template (type / filed / status / panel / grounds / institution / FWD / settlement / appeal / defensive value) has no entries. I will not invent one to fill the section.

Number-collision false positives — explicitly ruled out (do not brief these as this patent)

These came up repeatedly in search and are different patents that happen to share the digits "244." Conflating any of them with US 7,639,244 would produce a materially wrong invalidity opinion:

The existence of the Phenix Longhorn IPRs is worth flagging in one respect only: it shows Innolux is an active, sophisticated PTAB litigant that files IPRs and litigates claim construction aggressively. That is a signal about the assignee, not about any proceeding on this patent.

Strategic summary

Claim status: all 25 claims are UNTESTED — none canceled, none sustained, none narrowed. Independent claims 1, 12, 17, and 19 and all twenty-one dependent claims (2–11, 13–16, 18, 20–25) stand exactly as issued on 2009-12-29, including the limitations the prior section flagged as the likely attack surface — the "second number of signal lines being different from the first number" requirement of claims 1, 12, and 19, and the fourth-and-fifth-driver / third-time-period recitation that claim 17 carries as issued (note: claim 17's five-driver staging is a claim limitation, not merely a described embodiment — a petitioner attacking claim 17 must address the fourth and fifth drivers expressly). The two US continuations (12/618,176 → US 2010/0060617 A1; 14/200,052 → US 2014/0184576 A1) were abandoned, so the only live US claim set is the '244's. Because the patent has never been before the Board, there is no narrowing amendment, no certificate of correction record, and no Board claim construction to constrain the district court's reading of these terms.

Estoppel landscape: essentially clean, which favors a defendant. With zero IPRs, there is no § 315(e)(2) estoppel attaching to any party, and no IPR-based § 315(a)(1) or § 325(e)(2) estoppel either. A defendant asserted today faces no estoppel bar on any prior-art ground — patents, printed publications, and (in district court, unlike the PTAB) product/on-sale/public-use art all remain available. Two timing constraints do bite, though: (1) § 315(b) — if the defendant was served with a complaint alleging infringement more than one year ago, it is already time-barred from filing its own IPR, so the "blank slate" advantage may be unusable at the Board; and (2) forum availability is narrow — this is a pre-AIA patent (effective filing 2005-06-15), so PGR is unavailable by statute, and the CBM transitional program sunset on 2018-09-16, so IPR is the only AIA vehicle left. An expired/lapsed patent can still be the subject of an IPR in principle, but the practical incentive collapses once enforceability ends.

Pattern signals: none inbound. There is no repeat petitioner (because there is no petitioner), no patent-owner appeal history at the Federal Circuit, and no defensive aggregator (e.g., Unified Patents) in the chain for this patent. The only notable cross-proceeding pattern is on the assignee side — Innolux's involvement in Phenix Longhorn IPR2025-00043/-00044 (as petitioner, institution denied 2025-05-15) and the 2011-era ITC LCD investigations, none of which touch the '244.

Flagged inconsistency for resolution (not a contradiction I can settle): the prior section records the legal status as "Expired – Fee Related" and an "Adjusted expiration" of 2028-02-04. Those are in tension — a fee-related lapse means the patent ceased earlier than its nominal adjusted term. Google Patents' "adjusted expiration" field reflects the maximum term (20 years from 2005-06-15 priority, plus PTA), not the date enforcement actually stopped. Verify the maintenance-fee payment record in PatentCenter before relying on either date — it determines whether any damages window remains open at all.

Recommended next steps

• If you are a defendant: there is no FWD to cite and no canceled claim to point at. The strongest validity posture is a fresh one. Because no IPR record exists, you are free to (i) file your own IPR under § 102/§ 103 on the two-interface/TTL and unequal-signal-line limitations if you are still within the § 315(b) one-year window from service, and/or (ii) run the full prior-art case in district court with no § 315(e)(2) estoppel shadow. Note there is no institution decision, no panel, and no Board reasoning to borrow — anything you file is first-impression.
• Confirm the expiration/preclusion facts before spending money. Pull the maintenance-fee history for US 7,639,244 in USPTO PatentCenter to confirm whether the patent actually lapsed for non-payment and, if so, on what date. If it expired, the practical assertion risk is limited to past damages within the pre-lapse window, and an IPR is likely uneconomic.
• Do not brief the number collisions. If an adversary's demand letter or an internal docket entry cites "the '244 patent," verify the number digit-for-digit against the front page before importing any analysis from Sotera v. Masimo (RE47,244), Toyota v. Emerging Automotive (11,396,244), the Bowman/Vahalia storage-system IPR, or JP 特許第7639244号. All four appeared as search hits for "7639244" and all four are unrelated to US 7,639,244.
• If you are the patent owner: the absence of any IPR in ~17 years is mildly favorable (no successful invalidity challenge on record), but it is not evidence of "hardening" — it more likely reflects the patent being shorter-lived, in a crowded LCD-driver art, or simply not commercially worth attacking. Any 2026 assertion should be evaluated against the expiration flag above first.
• Definitive negative confirmation (if you need it in writing): query the PTAB E2E / PTAB API and USPTO ODP directly under both "7,639,244" and application 11/445,943, and check the CAFC CM/ECF docket for any appeal originating from that application number. The ODP feed in this prompt is the canonical list and it is empty; treat that as the operative fact unless an E2E query surfaces a newly filed, un-indexed petition.

Uncertainty statement: The zero-count finding is grounded in the structured ODP "PTAB proceedings on file" block plus corroborating searches of PTAB petition documents, PTAB decision summaries, Google Patents, and CAFC-adjacent sources. I found no proceeding, panel, FWD, settlement, or appeal for US 7,639,244, and I have not fabricated any. A very recently filed (post-ingest) petition could exist outside these sources.

I'll research the USPTO assignment records for US 7,639,244 and the assignee chain.

Let me search for the specific assignment records and the corporate history.

Let me look for the specific reel/frame records and the correspondents.

I hit the search-step limit, so I'll compile this from the authoritative Google Patents legal-events record (provided in the patent text) plus the assignment records I was able to cross-reference. I'll be explicit about what is confirmed versus what needs a direct Assignment Center query.

Inventors

Pattern notes:

• Both inventors assigned to their employer (Chi Mei Optoelectronics), which is the ordinary employee-invention assignment pattern for a Taiwanese panel maker — not an inventor-held-then-sold pattern.
• I found no evidence of either inventor departing within 12 months of filing, and no evidence of a portfolio fire-sale involving them. I could not retrieve inventor-employment histories or subsequent patent filings from the available sources; treat the absence of a departure finding as no data, not as a confirmed negative.
• Only two inventors for a 25-claim driver/interface family is unremarkable.

Original assignee

Chi Mei Optoelectronics Corporation (CMO) — the entity named in the 2006-08-28 assignment and the original assignee shown on the face of the patent.

• Primary line of business: TFT-LCD panel manufacturing (a major Taiwanese flat-panel maker; one of the parties in In re TFT-LCD (Flat Panel) Antitrust Litigation, N.D. Cal. MDL 3:07-md-1827-SI).
• Did they ship a product embodying the claims? CMO shipped LCD panels/modules throughout the relevant period, and the specification itself targets commercial product configurations (a 17-inch SXGA 1280×1024 panel with 384-channel data drivers). The claims cover the driver architecture and signaling, not a standalone product; CMO's panels are the natural embodiment. This is high confidence on "operating company", moderate confidence on any specific panel reading the claims.
• Current status: No longer an independent entity — absorbed by merger. CMO merged with Innolux Display Corp. and TPO Displays Corp. effective 2010-03-18 (Innolux Display was the surviving entity and was renamed Chimei Innolux Corporation). Chimei Innolux was then renamed Innolux Corporation (company-side name change dated ~2012-12-19 / effective 2013-01-01). Innolux remains a going concern (now part of the Hon Hai/Foxconn group). No bankruptcy, no wind-down, no fire-sale.

Source: Google Patents US7639244B2; Himax 2010 Annual Report (merger description).

Assignment timeline

Important qualification on reel/frame data. The Google Patents legal-events listing for US 7,639,244 records three assignment events but prints only "SEE DOCUMENT FOR DETAILS" — it does not surface the reel/frame numbers, execution dates, or correspondents. I could not open the Assignment Center record directly (search-step limit reached). Below I give the three confirmed events (assignor/assignee/type/date as recorded by Google Patents) and I mark reel/frame as not retrieved rather than invent one. Where I can supply a verified reel/frame, I do — but only from a sibling patent in the same CMO portfolio, clearly flagged as a cross-reference, not as the '244 record.

• Executed 2006 (date not retrieved) / recorded 2006-08-28 — Reel not retrieved

  • Conveyance: Assignment ("Assignment of Assignors' Interest")
  • Assignor: Tzong-Yau Ku; Yung-Yu Tsai (inventors)
  • Assignee: Chi Mei Optoelectronics Corporation (Taiwan)
  • Correspondent: not retrieved (Google Patents shows only "SEE DOCUMENT FOR DETAILS"). No recurrence finding can be made without the recorded correspondent.
  • Context: Initial employee-invention assignment — inventors convey rights to their employer; this is the ordinary original-assignment link, not an acquisition.

• Executed 2010-03-18 (company merger date) / recorded 2010-05-10 — Reel not retrieved for '244; see cross-reference below

  • Conveyance: Merger
  • Assignor: Chi Mei Optoelectronics Corp.
  • Assignee: Chimei Innolux Corporation (160 Kesyue Road, Jhunan Science Park, Miaoli County 350, Taiwan)
  • Correspondent: not retrieved.
  • Context: Internal corporate reorg — the three-way merger (CMO + Innolux Display + TPO Displays) with Innolux Display as surviving entity renamed Chimei Innolux. Not an asset sale.
  • Cross-reference (verified, different patent): For sibling CMO patent US 6,937,297 B2, the identical CMO→Chimei Innolux merger is recorded at Reel 024329 / Frame 0752, paperwork executed 2010-03-18, recorded 2010-05-05 (assignor "CHI MEI OPTOELECTRONICS CORP."; assignee "CHIMEI INNOLUX CORPORATION"). The '244 merger shares the execution date (2010-03-18) but Google Patents shows a recording date of 2010-05-10, i.e. a different (later) recording batch. Do not assume Reel 024329/0752 is the '244 record — verify directly.

• Executed 2012/2013 (name-change date) / recorded 2014-04-07 — Reel not retrieved

  • Conveyance: Change of Name
  • Assignor: Chimei Innolux Corporation
  • Assignee: Innolux Corporation
  • Correspondent: not retrieved.
  • Context: Change of name only — Chimei Innolux was renamed Innolux Corporation (Chinese-language registry shows a name-change announcement dated 2012-12-20, effective 2012-12-19; Innolux's own CSR report says the name took effect 2013-01-01). The USPTO recording lagged to 2014-04-07. No new party, no consideration, no asset transfer.

No other recorded assignments. The chain contains zero third-party transfers. The only two continuation applications in the family — US 12/618,176 (filed 2009-11-13, pub. US 2010/0060617 A1) and US 14/200,052 (filed 2014-03-07, pub. US 2014/0184576 A1) — are abandoned, and their filing dates are application events, not assignments. US 7,639,244 remains the family's sole granted US member.

Verified vs. unverified summary: assignors, assignees, conveyance types and event dates are confirmed from the authoritative patent record. Reel/frame numbers and correspondents for the three '244 events were not retrieved and must be pulled from the Assignment Center (assignmentcenter.uspto.gov, search "7639244", or application 11/445,943).

Timeline diagram

timeline
    title Ownership of US 7639244
    2005 : Taiwan priority filed 94119899
    2006 : US application 11 445 943 filed
         : Ku and Tsai assign to Chi Mei Optoelectronics
    2009 : Patent granted 29 Dec
    2010 : Chi Mei Optoelectronics merges into Chimei Innolux
         : Merger assignment recorded at USPTO
    2013 : Chimei Innolux renamed Innolux Corporation
    2014 : USPTO records the change of name

(Slash-free application number "11 445 943" used to keep the parser happy.)

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. Every assignee in the chain is an operating panel maker: Chi Mei Optoelectronics → Chimei Innolux → Innolux Corporation. No "IP/Holdings/Ventures/Licensing" suffix, no Delaware/Texas single-member LLC, no registered-agent address. All three links are the same corporate lineage crossing a merger and a rename (recorded 2010-05-10 and 2014-04-07).

2. Known asserter in the chain — NOT PRESENT. Neither the current assignee (Innolux Corporation) nor any predecessor matches the listed NPE rosters (Acacia, Marathon, IV, IPNav, Wi-LAN, Conversant/Mosaid, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, DGC, Spangenberg entities). Innolux is an operating company; while it has asserted its own patents and, separately, been a defendant and an IPR petitioner (e.g., IPR2013-00066/-00068; IPR2025-00043/-00044), it does so as a manufacturer, not a licensing-only NPE. None of those matters names US 7,639,244.

3. Repeat correspondent across the chain — UNCLEAR / no data. The Google Patents record suppresses the correspondent ("SEE DOCUMENT FOR DETAILS") and I could not retrieve it. I will not name a correspondent I cannot cite. This is the single most important item to recover on a direct Assignment Center pull — the merger/name-change filings for the CMO portfolio were typically handled by a consistent outside prosecution firm, and if a single firm recurs across the 2010 and 2014 recordings, that is worth noting (though ordinary corporate housekeeping for an operating company would not, by itself, be an NPE tell).

4. Cascading transfers — NOT PRESENT. Only two post-issuance transfers, four years apart (2010 and 2014), both corporate structural events within one company. No chained LLCs, no shared correspondent addresses, no <24-month cascade.

5. Pre-litigation transfer — NOT PRESENT. No infringement suit naming US 7,639,244 was found (consistent with the prior litigation section). There is therefore no assignment dated within 6 months of any suit on this patent.

6. Bankruptcy fire-sale — NOT PRESENT. No Chapter 7/11 for Chi Mei Optoelectronics or Innolux; the 2010 transaction was a merger, not a distressed sale, and Innolux remains solvent and operating.

7. Privateering — NOT PRESENT. There is no transfer of '244 to any NPE. Innolux's patent enforcement has been self-directed (operating company suing operating companies, e.g., the 2010 actions against Sony and the Thomson Licensing dispute), not delegation to a troll. No SEC-filing or EFF/Patent Progress coverage ties this patent to a privateering arrangement.

8. Defensive aggregator — NOT PRESENT. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN; it terminates at Innolux Corporation.

Verdict

Defensive / non-asserting.

Justification (with an explicit caveat). The ownership chain recorded at the USPTO is: inventors → Chi Mei Optoelectronics (assignment, recorded 2006-08-28) → Chimei Innolux (merger, recorded 2010-05-10) → Innolux Corporation (change of name, recorded 2014-04-07). There is not a single third-party or NPE transfer in the record, no litigation naming this patent, and the patent is now Expired – Fee Related (adjusted expiration 2028-02-04). Every NPE signal above is "not present" on the evidence available.

Caveat you should not skip: the "Defensive / non-asserting" label is the closest fit, but the framework defines that category as a chain terminating at a defensive aggregator, and here the chain terminates at an operating company. So read this verdict as "non-asserting operating-company chain," not as "neutralized by RPX/AST." If you need a technically precise label, this patent falls just outside the framework's categories — there is no assertion and no aggregator.

Uncertainty statement (required).

• Confirmed: assignor/assignee identities, conveyance types, and event dates, per Google Patents US7639244B2 legal events.
• Not retrieved: reel/frame numbers and correspondents for the three '244 recordings. I did not fabricate these. The only verified reel/frame I hold (024329/0752, executed 2010-03-18, recorded 2010-05-05) belongs to sibling patent US 6,937,297 B2, not to '244 — cross-reference only.
• No direct query of the USPTO Assignment Center or PatentCenter/PTAB API was completed under "7,639,244" / application "11/445,943." A definitive reel/frame-and-correspondent reconstruction — and a certified "no further assignments" finding — requires that pull. Verify at https://assignmentcenter.uspto.gov/.
• Consistent with the two earlier-generated sections: no litigation and no NPE activity involving US 7,639,244. No contradiction to flag, though note the same reel/frame blind spot affects all three sections.

I'll research the prior art citations for US 7,639,244. Let me start with targeted searches.

Let me retrieve the complete cited-references list for this patent.

Prior-Art Analysis — US 7,639,244 B2

Method note / limitation up front: This deliverable is built on the authoritative patent text and on web-search retrieval of the patent's own "Patent Citations" list. The full face-of-patent list is 25 references (per the Google Patents "Patent Citations (25)" table). My searches recovered a substantial subset but not a verified 25/25 enumeration — the aggregator snippets truncate the table and mix in the citation lists of the two abandoned continuations (US 12/618,176 / US 2009…? and US 14/200,052). I flag every item below with how confident I am that it is a genuine §102 citation of the '244 patent. I have not auto-corrected any number.

1. Identified references cited on / of record for US 7,639,244

Non-patent literature referenced in the specification (background, not a §102 citation): the '244 specification distinguishes its TTL signaling from "mini-CVDS or whisper-bus signals." The WhisperBus reference traces to McCarthy et al., "WhisperBus™: An Advanced Interconnect Link For TFT Column Driver Data," SID '01, pp. 106-109 (2001) — which appears in the prosecution-file prior-art list I retrieved under a PTAB exhibit. It is contextual, not a claim-by-claim §102 reference against the '244 claims.

2. Claim-by-claim §102 relevance (independent claims 1, 12, 17, 19)

Key point: the distinguishing limitations of the '244 claims are (a) a designated first driver that receives MORE pixel data than it uses and forwards the surplus to downstream drivers, and (b) unequal numbers of signal lines in vs. out (claim 1: first number ≠ second number; claim 12: forwarding-line count ≠ receiving-line count; claim 19 method analog) and (c) the alternate odd/even clocking / staged buffering of claim 17. None of the identified references (1)–(13) discloses all of it. They are best characterized as:

• Anticipating/obviousness-relevant to dependent features only (glass-substrate lines, clock generation, TTL, LCD preamble): refs 1, 3, 4, 6, 8, 11, 13.
• Closest to the core concept — US 2002/0008682 A1 (ref 7): "Flat panel display with an enhanced data transmission." This is the single reference I would flag as the primary §102 candidate for the data-forwarding-between-drivers gist underlying claims 1, 12, 19. A §102 rejection would require that its disclosed line-counts or forwarding path map onto the "first number different from second number" limitation — I cannot confirm that mapping from the retrieved abstract alone.
• Clock-phase support — US 3,172,042 (ref 1) is the natural §102 anchor for the odd/even pulse clocking language of claims 7–8 and part of claim 17.
• Glass-substrate support — US 6,379,785 (ref 4) is the natural §102 anchor for the "transmission lines disposed on a glass substrate" limitations of claims 5, 9, 15.

No identified reference anticipates claim 1 as a whole (array + first driver receiving both used and unused data + second driver receiving the unused data + unequal signal-line counts), and none anticipates claim 12 or claim 19 as a whole on the retrieved text.

3. Caveats and required uncertainty statement

1. Coverage gap. I retrieved the citation table only partially; the aggregator source truncated the list and interleaved the citation lists of the two abandoned continuations (US 12/618,176 → US 2010/0060617 A1; US 14/200,052 → US 2014/0184576 A1). A definitive, exhaustively enumerated §102 table for US 7,639,244 requires reading the printed "(56) References Cited" block on the patent front page directly (USPTO PatentCenter or the Google Patents PDF), which I could not open as an image here.
2. Date/§102 framework. The '244 patent claims TW priority 2005-06-15 (TW Ser. No. 94119899) and was US-filed 2006-06-02, i.e., pre-AIA. Publications dated before 2005-06-15 are §102(a)/(b) art; references published after the filing date (e.g., JP 4390451 B2) are not §102 prior art on their face unless an earlier effective date is established.
3. No auto-correction. Every alphanumeric identifier above is reproduced as encountered. Where the retrieved snippet truncated an entry (e.g., ref 10), I say so rather than guessing.
4. The reference date/inventor fields for refs 10–13 in particular may belong to the continuation's citation list rather than the '244 patent's own list. Treat those four rows as lower-confidence.
5. "Potentially anticipates" is used cautiously — the retrieved abstracts are short; a true §102 finding requires element-by-element comparison against the full reference disclosure, which is beyond what these search results support.

Bottom line: The most relevant identified prior art for US 7,639,244 is US 2002/0008682 A1 (enhanced flat-panel data transmission — primary candidate for claims 1, 12, 19), supported by US 6,388,651 B1 and JP 2001-324967 A for the timing-controller/driver architecture, US 3,172,042 A for the odd/even clock limitations (claims 7–8, 17), and US 6,379,785 B1 for the glass-substrate transmission-line limitations (claims 5, 9, 15). I could not confirm that any single reference anticipates claims 1, 12, 17, or 19 as issued.

I'll build the §103 analysis on the prior-art set already identified, first verifying the two most load-bearing references so the combination theory rests on confirmed disclosures rather than bare titles.

The two key references are confirmed with substantive disclosure. Let me verify whether a promising third reference (Yi et al., glass channels between adjacent source drivers) predates the '244 priority, and check the general state of cascade/daisy-chain data distribution art.

This is a materially richer prior-art landscape than the earlier section's partial list captured — including direct cascade/daisy-chain driver references. Let me pin down the two most important dates before relying on them.

Obviousness Analysis — US 7,639,244 B2 under 35 U.S.C. § 103

Builds on: the Bibliographic, Litigation, PTAB, Assignment, and Prior-Art sections already generated. This section does not repeat that background; it starts from the reference set already identified and extends it where the prior-art section flagged its enumeration as incomplete.

0. Threshold correction / update to the Prior-Art section (flagged, not hidden)

The Prior-Art section concluded: "The most relevant identified prior art … is US 2002/0008682 A1 … I could not confirm that any single reference anticipates claims 1, 12, 17, or 19." That conclusion was explicitly limited to a partially-retrieved citation list (13 of the ~25 cited references) and to bare titles/abstracts.

Supplemental searching for this §103 task located a materially closer class of art that the earlier list did not surface: the cascade-connected / daisy-chained data-driver art, in which a timing controller feeds one driver and each driver forwards the surplus data to the next. That art maps directly onto the '244's core "designated driver forwards the rest" limitation. This is not a contradiction of the earlier section (it disclaimed completeness), but it does supersede the earlier "closest reference = US 2002/0008682 A1" characterization for the forwarding limitation. I flag it because the earlier section is authoritative per the task instructions.

Also confirmed/corrected against search results (search results control per operating rules):

• Ref 7 is confirmed as the pre-grant publication of US 6,657,622 B2 (Park Jin-Ho, Samsung; filed 2001-06-22; KR priority 2000-07-18; US 2002/0008682 A1 published 2002-01-24). — freepatentsonline 6657622
• US 7,215,312 B2 = Fujitsu Limited, Kumagai & Udo, filed 2003-04-30, published as US 2003/0218588 A1 on 2003-11-27, granted 2007-05-08 — i.e. §102(b) art relative to the '244 priority. — Justia 7215312; Google Patents US7215312

1. Governing law and framework

• The '244 is a pre-AIA patent (TW priority 2005-06-15; US filing 2006-06-02 — before the 2013-03-16 AIA change). Pre-AIA §103(a) applies, but the controlling obviousness standard is KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) (decided while this patent was pending).
• Graham v. John Deere factors: (1) scope/content of the prior art; (2) differences between the prior art and the claims; (3) PHOSITA level; (4) secondary considerations.
• KSR / MPEP 2143 rationales applicable here: (A) combining prior-art elements to yield predictable results; (B) simple substitution of one known element for another; (C) use of a known technique to improve a similar device in the same way; (D) applying a known technique to a known device ready for improvement; (F) "obvious to try" from a finite number of identified, predictable solutions; and the "design incentive / market pressure" and "optimization of a result-effective variable" rationales.
• All identified references are in the same field of endeavor (flat-panel display driver architecture and TCON-to-driver data transmission), so the analogous-arts hurdle is not a barrier.

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

A person with a B.S. in electrical engineering (or equivalent) and 2–4 years of experience designing/or integrating display timing controllers and source (data) driver ICs, or an M.S. with 1–2 years, and familiarity with: TTL / LVDS / RSDS / mini-LVDS / WhisperBus-style interfaces; clock generation and division; COG / WOA / TCP / COF packaging; and the tradeoffs of EMI, bezel width, line count, and power on a glass substrate. This level is well-supported by the reference set itself (WhisperBus, the Samsung and Fujitsu disclosures, and the LCD-panel architecture references).

3. Claim construction of the dispositive terms

Key observation: because claims 1/12/19 require only that the two line counts be different — not narrower downstream — the limitation is a result-effective design variable (bus width allocation) rather than a structural advance. That weakens it under KSR's optimization rationale. Only claim 14 pins the direction ("second number is less than the first"), which is the commercially meaningful case.

4. The §103 arsenal (prior-art section refs. + supplemental)

Already in the Prior-Art section (refs 1–13, abbreviated): US 3,172,042 (odd/even pulse generator); US 5,394,165; US 6,388,651 B1 (Toshiba picture control); US 6,379,785 B1 (glass-coated substrates, high-frequency); US 2001/0003447; US 2002/0008682 A1 / US 6,657,622 B2 (Samsung); US 2002/0027540; US 2002/0030653; JP 2001-324967 A; JP 2004-354567 A; JP 4390451 B2 (post-dates — not §102(b)).

Supplemental references located for this task:

Post-dating reference flagged out: US 2010/0315406 A1 ("Image data transfer to cascade-connected display panel drivers," Nose & Furihata) describes almost verbatim the '244 claim-1 architecture (driver "transfers received data to (i+1)th driver when not associated with i-th driver") and even claims a second bus with more lines than the first. It post-dates the '244 priority and is NOT prior art — I cite it only as evidence the concept was in the art stream. It cannot be used in a §103 attack on the '244.

5. Motivation to combine (KSR / MPEP 2143)

The record supplies an unusually strong, documented motivation set — this is not a hindsight reconstruction:

1. Reduce EMI — the express object of the '244 and of WhisperBus [G] ("reduction in EMI"), US 6,657,622 [H] ("eliminating problems of EMI"), and US 2003/0117350 [A]. Rationale (C)/(D).
2. Reduce signal-line count / bezel / FPC cost — [A] ("remarkably reducing a number of signal lines"), [G] ("reduced … signal trace count"), [B] (cascade "reduces the number of mounted circuit board layers … manufactured at low cost"). Rationale (D).
3. Lower the inter-device data rate — [G] expressly: the TCON "buffers the incoming data at the dot clock rate but re-transmits it … at a rate that uses the whole line time … to further minimize the data rate to [the] column driver." That is a direct, explicit teaching of the '244's claims 2/4 (forward at a lower clock). Rationale (A)/(D).
4. Cascade is a recognized, cost-driven design choice — [B], [C], [D], [E], [F] all present cascade connection as the standard way to avoid bus/wire crossings and reduce board layers. Rationale (A)/(C).
5. Avoid routing every line from the FPC to every driver — WhisperBus [G] and US 6,657,622 [H] both attack the multi-drop/many-line problem, so a PHOSITA had reason to localize distribution (TC → designated driver → neighbors). Rationale (F) — a finite set of known options (broadcast vs. cascade).
6. Interface selection (TTL vs. differential) — [H] teaches conversion among TTL/LVDS/RSDS/TMDS, and the '244 specification itself admits TTL, mini-CVDS and whisper-bus are known inter-driver signaling schemes. A specification admission of known art is strong §103 evidence. Rationale (B).

Contemporaneous-problem unity: every reference is aimed at the same triad (EMI, line count/bezel, power), so the "combine" step has both a reason and a reasonable expectation of success (the aggregate results — fewer lines and lower clock → lower EMI and thinner bezel — are predictable, not synergistic/unexpected).

6. Combination grounds and claim mapping

GROUND 1 — The "designated driver forwards the surplus" architecture

Claims 1, 9, 12, 19 (and dependents 13, 14, 15, 21, 22, 23, 24).
Combination: [C] JP 2001-174843 (or [A] US 2003/0117350 / [B] US 2003/0218588) as primary + [B] + [G] WhisperBus + [H] US 6,657,622.

Why obvious: every element existed; the only "difference" is allocating different line counts to two input paths — a result-effective variable a PHOSITA would set to match bandwidth/layout constraints (KSR; MPEP 2143 optimization rationale). Claim 12's "in ≠ out" asymmetry is even more directly the WhisperBus/[A] teaching (fewer lines downstream). Claim 19 is the method analog; claim 14 (second < first) is literally what [A]/[G] teach (fewer lines to the downstream device).

GROUND 2 — Clock-frequency and phase differences on the forwarding path

Claims 2, 3, 4, 7, 8, 20 (and 13).
Combination: Ground 1 + [G] WhisperBus + ref 1 (US 3,172,042) + [B].

• Claims 2/4/20 (forward at a different/lower clock): [G] expressly teaches re-transmitting at a reduced rate "to further minimize the data rate to [the] column driver." Combined with standard clock division, this is a predictable, expressly-motivated modification.
• Claims 7/8 (second clock = odd pulses, third clock = even pulses): ref 1 (US 3,172,042, "Precision phased pulse generator") discloses exactly odd/even phased clock generation; [B] teaches clock phase manipulation (180° inversion / half-cycle delay) between cascade stages. The odd/even split is a known technique applied to a known pipeline — Rationale (C).
• Claim 3 (alternate cycles to second and third drivers): obvious from odd/even phases + two output ports.
• Claim 13 (simultaneous to second and third): [G] point-to-point — all drivers receive simultaneously.

GROUND 3 — Glass-substrate transmission lines

Claims 5, 9, 15, 24.
Combination: Ground 1 + ref 4 (US 6,379,785) + the COG/WOA art.

• Glass-substrate high-frequency transmission is the express subject of US 6,379,785 (ref 4), and WOA/COG routing between adjacent drivers is standard. Motivation: eliminating the FPC run and shortening the path ( [H] "shorten transmission path," [B] fewer crossings). Rationale (D).
• Caveat: US 7,667,677 (Yi et al.) describes glass channels coupling two adjacent source drivers with TTL/P2P — highly on point, but its filing date could not be confirmed as pre-2005 in this session; treat as date-unverified, usable only if its effective date precedes 2005-06-15.

GROUND 4 — TTL and differential-signal interfaces

Claims 6, 10, 11, 16.
Combination: Ground 1 + [H] US 6,657,622 + [G] + the '244's own admissions.

• [H] expressly teaches TTL and LVDS/RSDS/TMDS data transmission and conversion. The '244 specification admits TTL, mini-CVDS and whisper-bus are known. Claim 6 (second driver has a first TTL interface to receive and a second TTL interface to forward to a third driver) maps directly onto [B]'s cascade driver IC with input buffers 20–23 and output buffers 28–31 and onto the general cascade-node architecture in [C]/[D]/[F]. Claim 11/16 differential signaling is [H]/[G] (LVDS/RSDS/current-mode link). Rationale (B) substitution of known signaling.
• The '244's asserted advantages of TTL over whisper-bus are inherent properties of a known standard, not unexpected results.

GROUND 5 — Staged, multi-clock, five-driver buffering

Claims 17 and 18 (the hardest claims).
Combination: [B] US 2003/0218588 + [F] JP 2004-264720 + [A] US 2003/0117350 + [G] + ref 1.

• Claim 17's elements: substrate + array + TC outputting three clocks with clocks 2, 3 slower than clock 1; first driver buffers during T1; forwards to second/third during T2 per the slower clocks; fourth/fifth drivers receive forwarded data during T3 and buffer.
• [B] supplies cascade forwarding with per-driver latches/buffers; [F] supplies staged, master/slave enable-out distribution to downstream drivers; [A] supplies the store-or-forward decision (by-pass register); [G] supplies lower-rate re-transmission and TCON line buffering; ref 1 supplies odd/even clocks.
• Assessment: the specific pipelined, branched (fan-out-to-2, then to 4/5) schedule is a predictable design of a cascade/pipeline, but claim 17 recites more structure than any single reference (the fourth/fifth-driver + third-time-period staging). This is the patentee's best defensive claim; a §103 attack on claim 17 should be built on a multi-reference ground with the pipeline/pipelining rationale made explicit, and is more exposed to a "no reasonable expectation / hindsight" rebuttal than claims 1/12/19.

GROUND 6 — Row-portion / chroma data

Claims 23, 25.

• Transmitting first/second/third pixel data to the first driver and having each refer to chroma values for different portions of one row is conventional — every multi-driver LCD splits a horizontal row among drivers, and RGB (including chroma) sub-pixel data per column is fundamental. Claim 25 adds nothing structurally; obvious over the LCD references (refs 6/8/11/13) alone. Rationale (A)/(C).

7. Claim-by-claim §103 conclusion

Net: on the assembled record, no independent claim presents a genuinely non-obvious combination; the strongest §103 case is against claims 1, 12, 19, 2, 4, 6, 7, 10, 11, 20; the weakest (most defensible) are claims 17–18, and the narrow line-count direction (claim 14) is a positive for the attacker (it is exactly what the art teaches).

8. Rebuttal considerations and counterarguments

Patentee's best arguments (and why they likely fail):

1. "Teaching away" — glass WOA lines attenuate high-speed data. The '244 spec states glass transmission lines have higher impedance and attenuate signals. Counter: the art expressly solves this — [H] shortens the path, ref 4 (US 6,379,785) addresses glass-substrate high-frequency transmission, and [G] teaches lowering the rate (the '244's own fix). At most this is a known design constraint with a known remedy, not a teaching away (KSR: a known problem with a finite set of known solutions is obvious to try).
2. "Cascade was for board-layer reduction, not EMI." Counter: [G] and [H] tie line/rate reduction directly to EMI; the '244's own stated advantage (reduced EMI) is therefore an expected consequence.
3. "Unexpected result / synergy." Counter: the aggregate result (fewer lines → thinner bezel; lower clock → less EMI) is the sum of predictable effects — no evidence of an unexpected property or of solving a problem others failed to solve.
4. Secondary considerations. No evidence of record: no commercial-success evidence with nexus to the claims, no copying, no industry praise, no long-felt-but-unsolved need (WhisperBus and the cascade art show the problem was being solved contemporaneously, undercutting "long-felt need"). The PTAB/Litigation sections already confirm no proceeding, no assertion, no licensing — which also means no objective indicia have been developed.

Weaknesses in the attacker's case (be candid):

• I have snippets, not full reference texts for [A], [B], [C], [D], [E], [F]. Element-by-element mapping (especially for claim 1's input-to-input line-count asymmetry, which is a slightly odd limitation) needs full-text and figure review.
• Claim 1's input-side asymmetry is the least clearly disclosed element; a strong ground must show either (i) an express disclosure, or (ii) that setting two input bus widths differently is a mere optimization. The latter is available under KSR but is the argument most vulnerable to a "that's not what the claim says" response.
• Claim 17 recites more structure (specific T1/T2/T3 staging, fourth/fifth drivers) than any single reference; it requires a genuine multi-reference combination with an articulated pipelining rationale.

9. Uncertainty statement (required)

• Identifiers reproduced literally; nothing auto-corrected. Where the prior-art section's numbering is referenced (refs 1–13) I rely on its entries; supplemental references are labeled [A]–[H] and are newly located for this task.
• Dates cited for [A]–[H] come from the search results (which control per the operating rules): [A] US 2003/0117350 A1 pub. 2003-06-26; [B] US 2003/0218588 A1 pub. 2003-11-27 / US 7,215,312 grant 2007-05-08 (Fujitsu, filed 2003-04-30); [D] JP 2002-351425 pub. 2002-12-06; [E] EP 0 497 378 A2 pub. 1992-08-05; [F] JP 2004-264720 pub. 2004-09-24; [G] SID '01 (2001). [C] JP 2001-174843's exact publication date was not independently confirmed (search-step limit reached); its content is known only from a later patent's citation of it — verify before relying on it as the primary reference.
• Full-text verification needed. The claim charts above are built on abstracts/partial passages and are provisional. A defensible invalidity contention requires pulling the complete [A], [B], [C], [D], [E], [F] documents (PatentCenter / Google Patents PDFs) and confirming figure-level support, particularly for claim 1's input-side asymmetry and claim 17's staged scheme.
• Post-dating references excluded from the §103 grounds: US 2010/0315406 A1 and US 7,667,677 B2 (Yi et al., glass channels) — [date-unverified], not to be used unless an earlier effective date is established; JP 4390451 B2 from the prior-art section likewise post-dates on its face.
• No secondary-considerations evidence was located; the absence is a search result, not a certified negative.
• No litigation/PTAB exists (per the earlier sections), so there is no Board claim construction or FWD to anchor these constructions — the district court would construe "different number of signal lines" and "additional pixel data not used" fresh.

10. Bottom line and next steps

1. The '244 claims are vulnerable under §103, most acutely claims 1, 12, 19 and the clock/interface dependents (2, 4, 6, 7, 10, 11, 20). The forwarding architecture was squarely in the cascade art ([A]–[F]); the EMI/line-count/rate-reduction motivation was expressly documented by WhisperBus [G] and US 6,657,622 [H] before the '244 priority.
2. The crown-jewel claims are 17–18; attack them with a multi-reference pipelining ground ([B]+[F]+[A]+[G]+ref 1), not a single reference.
3. Prioritize full-text retrieval of [A] US 2003/0117350, [B] US 2003/0218588/US 7,215,312, [C] JP 2001-174843, [D] JP 2002-351425, and [F] JP 2004-264720; confirm [C]'s date and pull English translations for the JP items.
4. Timing note (from the PTAB section): any IPR is subject to §315(b) (one year from service) and, being pre-AIA, no PGR is available and CBM has sunset — so a validity challenge is an IPR or district-court matter; the earlier section's expiration/maintenance-fee flag should be resolved first, since it drives whether a challenge is even economic.