US 5200956

I'll verify the patent's status and check for any litigation activity, then summarize.

US 5,200,956 — Verification Summary

Search scope note: I ran targeted searches for the literal string "5200956" and for the patent in combination with litigation/docket terms. I did not find any CAFC 2026 docket, district court case, or PTAB proceeding involving US 5,200,956. The only hits for the bare number "5200956" were unrelated identifiers (a 2025 French trademark filing "ESPERANCE BANLIEUES," a 1972 RAF officer service number, and unrelated CourtListener/registry record IDs). I have no authoritative evidence of litigation involving this patent, and I will not infer any. The substantive data below comes from the patent's full text (Google Patents, fetched 2026-09-21) plus corroborating sources; I did not execute a live USPTO Patent Center / CourtListener docket query, so treat the litigation finding as "no evidence found," not as a certified negative.

Bibliographic data

Abstract (as published)

A digital cordless telephone system comprises primary/base stations (PS) coupled to a system controller interfacing with the PSTN, and secondary stations (SS). TDMA is used for forward and reverse transmissions. Normally one duplex voice channel (one forward slot paired with one reverse slot per frame) is allocated per transaction. For fast data rates, additional duplex voice channels should be made available quickly and then relinquished. To enable rapid setup, a map store in each data secondary station lists the usage and quality of all duplex voice channels, so the station can immediately use additional listed channels instead of scanning all channels first.

Independent claims — plain language

Claim 1 (system). A communication system with at least one primary and one secondary station communicating over a radio channel divided in time into frames of 2n slots. The first n slots are for one direction; the remaining n for the opposite direction; slot x and slot (x+n) together form a duplex voice channel. Each station can determine duplex-channel usage, and a transmitting station picks an acceptable available channel and transmits in its slot. Key addition: at least one secondary station has (a) a map store holding indicia of duplex-channel usage, and (b) a controller that detects (i) it has a data message and (ii) the message's data rate exceeds that of a single duplex voice channel. On that determination, the controller selects at least one additional duplex voice channel stored as available, causes non-identical packets of the data message to be transmitted substantially in parallel in at least two duplex voice channels, and relinquishes the additional channel(s) after the message is transmitted.

Claim 10 (secondary station apparatus). The station-level counterpart of claim 1: a radio transmitter/receiver; means for determining duplex-channel usage; a map store of usage indicia; means for receiving an input data message for onward transmission; and a controller that detects the data message and that its rate exceeds a duplex voice channel's. In response it selects at least one additional channel stored as available, causes non-identical packets to be sent substantially in parallel in at least two duplex voice channels, and relinquishes the additional channel after transmission.

Practical characterization of the inventive core (common to both independents): the pairing of (1) a pre-populated channel-usage map enabling near-instant multi-channel access without a fresh scan, with (2) rate-triggered parallel transmission of different packets across multiple TDD duplex channels, with (3) prompt release of the extra channels. The "non-identical packets … in parallel" and "relinquishing" language reads as the likely point of novelty over the cited art (e.g., US4831373, US4785450), though I have not reviewed the prosecution history to confirm the exact basis for allowance.

Dependent claims (brief)

• 2: ≥2 frequency channels; map store lists lesser-used channels per frequency channel; the secondary station may transmit in a listed channel from a frequency channel other than the one already assigned to it.
• 3: Secondary station has signal-quality measuring means; quality indications stored in the map store with usage indicia.
• 4: Primary station measures its local radio environment and periodically transmits that information; secondary station stores it and stores a channel-quality indication alongside usage indicia.
• 5: As in 4, plus cross-correlation of measured signal quality with the primary station's local-environment info to derive overall channel quality, stored in the map store.
• 6 (dep. on 3): Controller determines which are the better duplex voice channels from the map store.
• 7 (dep. on 6): Controller updates the map-store list including quality information.
• 8: Control means coupled to the primary station controls assignment of lesser-used channels and signals their details to the secondary station via the base station.
• 9: Usability details for duplex voice channels are carried in the signalling portion of a message from primary to secondary station.
• 11 (dep. on 10): Means for measuring quality of radio transmissions from a primary station; indications stored in the map store with the usage indicia.
• 12 (dep. on 10): Means for storing the primary station's local radio environment details as transmitted.
• 13 (dep. on 11): Combines claims 11 and 12.
• 14 (dep. on 13): A cross-correlator correlating signal-quality measurements with local radio-environment details, storing overall quality indications in the map store with the usage indicia.

Notable technical details from the specification

• Example system: five frequency channels C1–C5 at 1.152 Mbit/s each, 1.728 MHz channel spacing, 10 ms frames, 24 slots (F1–F12 forward, R1–R12 reverse), forward/reverse slots "twinned" as duplex voice channels.
• Message structure: 2 bytes preamble, 2 bytes sync, 8 bytes signalling, 40 bytes data; 32 kbit/s digitization buffered to 1.024 Mbit/s bursts.
• Motivation: bursty data and 2B+D ISDN at 144 kbit/s exceed one slot's capacity; channels should be surrendered after the burst (ISDN link may persist).
• Frequency agility: two synthesisers SYN1/SYN2 with switching means 32, or per-channel continuously running local oscillators.

Uncertainty flags

• Litigation: No CAFC 2026 (or any) docket activity found for this patent. Given the listed expiration of 2010-05-22 and the 35 U.S.C. § 286 six-year damages lookback, active 2026 infringement litigation on this patent is highly improbable. I state this as an inference from the record, not verified docket fact.
• Assignee chain: The 2009 transfer to "IPG Electronics 503 Ltd" is consistent with public commentary describing that entity as a Philips-related patent divestiture vehicle, but I have not independently verified that characterization for this specific patent.
• Prosecution history: I did not retrieve the file wrapper, so my note on which claim language was added to overcome prior art is an inference, not a documented finding.

I'll research litigation involving US patent 5200956 across patent litigation databases and court records.

Let me search more specifically across litigation databases and the patent's post-2009 owner, IPG Electronics 503 Ltd.

Let me check litigation-specific databases directly.

Litigation Search Results: US Patent 5,200,956

Bottom line

I found no litigation involving US Patent 5,200,956. My searches of patent-based databases (Google Patents, FreePatentsOnline, SumoBrain, EPO/Espacenet family data) and general web/court-record sources returned no district court complaints, no CAFC appeals, and no ITC investigations asserting this patent. I also found no evidence that its post-2009 owner ever filed suit on it.

Confidence level: moderate. I could not run direct queries against PACER, Docket Navigator, Lex Machina, or the Unified Patents litigation portal within this session, so I cannot rule out an obscure or short-lived case. However, the absence of any "Cited By"/litigation cross-references in the patent records and the patent's expired status make litigation unlikely.

What I verified about the patent (grounding facts)

From the authoritative record at https://patents.google.com/patent/[US5200956](/patent/US5200956)/en:

• Title: "Communications system for data transmission over a time division duplex frequency channel"
• Patent number: US 5,200,956 A (interpreted literally, not auto-corrected)
• Application: US 07/528,016; filed 1990-05-22
• Priority: 1989-05-26 (GB 8912175)
• Granted/published: 1993-04-06
• Original assignee: U.S. Philips Corporation
• Reassignment (2009): To IPG Electronics 503 Limited, effective 2009-04-13 (recorded 2009-05-05) — a Philips-related patent holding/transfer entity
• Legal status: Expired – Lifetime; anticipated expiration 2010-05-22
• Inventors: Christopher D. Pudney; Frank C. G. Owen
• Family members: EP 0399611 B1, GB 2232326 A, JP 3040429 B2, DE 69025353 T2, FI 902549 A7

Why no litigation is the expected result

1. The patent expired May 22, 2010 — only about 13 months after the 2009 reassignment to IPG Electronics 503. Even if IPG or a downstream acquirer had wanted to assert it, the enforcement window was extremely short and it would have had to act almost immediately. Any suit filed near the end would have faced an inevitable mootness/expiry defense.

2. IPG Electronics 503 Limited functioned as a Philips patent transfer vehicle, not an operating litigant. As reported in coverage of the 2012 Yahoo v. Facebook dispute (Ars Technica, https://arstechnica.com/tech-policy/2012/04/yahoo-accuses-facebook-of-being-mean-and-dealing-with-dirty-patent-trolls/), IPG Electronics 503 sold patents to parties such as Facebook, Funai, and entities associated with Acacia (e.g., Mobile Enhancement Solutions). I found no record indicating US5200956 was among the patents transferred into any litigating NPE portfolio, nor any suit naming IPG Electronics 503 as plaintiff on this patent.

3. No litigation cross-references appear in the patent records. The Google Patents "Also Published As / Similar Documents / Cited By" data and the EPO family citation data list no litigation events for US5200956.

Cases found

None. No plaintiff, defendant, jurisdiction, case number, filing date, or outcome can be reported, because no case asserting US 5,200,956 was identified.

Caveats and recommended next steps

• Do not treat this as a definitive clearance. A negative from secondary web sources is not the same as a certified PACER/RECAP search. If you need a legally defensible litigation history, run:
  • PACER (Case Locator / nationwide docket search) for "5,200,956" and "5200956"
  • CourtListener / RECAP (free) — https://www.courtlistener.com
  • Unified Patents litigation portal — https://portal.unifiedpatents.com/litigation/caselist
  • Docket Navigator / Lex Machina (commercial) for NPE-portfolio tracking
  • USPTO Patent Assignment Search to trace any post-2009 chain of title out of IPG Electronics 503 Ltd, which would identify any potential litigating acquirer
• Separate the identity of the patent carefully. US 5,200,956 is a distinct patent from its sibling US 5,117,423 ("Data transmission over a time division duplex channel," also a Philips TDD patent). If your real interest is the DECT/TDD Philips patent family generally, that sibling and its successors should be searched independently. I found no litigation for US 5,200,956 and am not resolving that question for any other patent number.

I'll verify the canonical "no proceedings" finding against independent sources before drafting.

Let me run a few more targeted checks for tribunal-specific records and any pre-AIA reexamination.

AIA Trial Proceedings — US 5,200,956

Proceedings overview

Total AIA trial proceedings on file: 0 (zero). The USPTO Open Data Portal returns no IPR, PGR, or CBM proceeding naming US 5,200,956 as the challenged patent, and no institution denial, settlement, or FWD exists to report. Because the patent issued 1993-04-06 — long before the AIA's 2013 effective date — the only AIA vehicle that could ever have reached it is an IPR (PGR is limited to post-2013 filings; CBM requires a "covered business method" claim, which a TDMA cordless-telephony claim set is not). Defensive posture: there is no PTAB record to work with either way — no petitioner has ever knocked down any claim, and no panel has ever blessed one. A defendant cannot point to a canceled claim, but equally cannot cite a favorable FWD. Any invalidity theory for this patent must be built from scratch in district court or a fresh IPR, and the patent's 2010-05-22 expiry means the practical problem is § 286 damages lookback, not claim survival.

No proceedings to list

There is no most-impactful-first ordering to produce because the list is empty. Verbatim from the structured source: "The USPTO ODP API returns no AIA trial proceedings for this patent as of the most recent ingest."

I independently attempted to surface an un-indexed proceeding (older filings, CBM-era petitions, or recently-filed petitions the ODP may lag on) and found none:

That last point matters for accuracy: the appearance of the digits 5200956 inside a PTAB exhibit file or inside a patent's "Referenced Cited" table is not evidence of a proceeding. The only substantive occurrences of this patent's number in the record are (a) as a family member of EP 0 399 611 and (b) as a cited reference in later patents (e.g., US 5,768,264, US 6,367,566). Neither is a trial event.

Strategic summary

Claim status: every claim of US 5,200,956 is UNTESTED at the PTAB. Claims 1 and 10 (the independents) and dependent claims 2–9 and 11–14 have never been the subject of an AIA challenge, so none is canceled, none is confirmed, and none carries an FWD-based estoppel or preclusive effect. Contrast this with the situation the task template anticipates — "the patent has survived two IPRs and is hardened" — which does not describe this patent. The accurate description is the inverse: an untested patent.

Estoppel landscape: essentially empty, and that cuts both ways. Because no IPR was ever instituted, § 315(e)(2) estoppel never attached to anyone. No petitioner is barred from raising any ground, which means a fresh petitioner retains the full universe of § 102/§ 103 art — including the eight references already cited on the face of the patent (WO1984004435A1, US4556972, US4638479, US4656645, EP0261127A1, US4785450, US4831373, and the US4785450 continuation). The flip side is that no examiner or panel has ever narrowed the claims on the merits, so there is no prosecution-to-PTAB narrowing to exploit either. One notable lever: the patent's own EP family member, EP 0 399 611, was later cited as X-category prior art against EP 0 732 826 ("column 3, line 33 – column 4, line 25," etc., per the EP 0 732 826 search report). That confirms the disclosure — the map-store/near-instant-access teaching — has been treated as anticipating-level art for at least one later claim set, which is a useful starting point for a § 102 attack on 5,200,956's broader claims.

Pattern signals: none. There is no multi-petition petitioner, no serial IPR filer, no patent-owner appeal activity (nothing to appeal), and no evidence of a defensive aggregator such as Unified Patents ever targeting this patent. The absence is itself diagnostic. Well-asserted patents do attract IPRs — that is the normal, almost mechanical pattern. This patent was never asserted (consistent with the prior litigation section finding zero suits), so it never attracted a petition. A 1993-issued patent that expired 2010-05-22, with a maximum six-year damages lookback under § 286, is a poor candidate for either assertion or a defensive IPR in 2026.

Recommended next steps

• If you are a defendant: there is no FWD to quote and no canceled claim to point your adversary to. Do not represent to a court or opposing counsel that any claim of this patent has been invalidated at the PTAB — that assertion would have no support in the record. The accurate statement is: no AIA trial proceeding has ever been filed against US 5,200,956.
• The real defense is the calendar, not the PTAB. Confirm the expiration basis directly: Google Patents lists legal status "Expired – Lifetime" with anticipated expiration 2010-05-22 (https://patents.google.com/patent/US5200956/en). If a 2026 demand letter cites this patent, the § 286 six-year lookback (reaching back only to 2020) post-dates expiry entirely, so there is no actionable damages window. Verify the maintenance-fee chain in the file wrapper to be certain the patent was not revived in a way that alters the expiry date — I did not pull the fee record in this session.
• If you actually need an invalidity position, you must build it yourself. With no PTAB history, there is no estoppel and no panel findings to inherit. The productive path is a §§ 102/103 attack grounded in the patent's own cited references plus the EP 0 399 611 X-citation discussed above — but that is district-court or fresh-IPR work, not something the existing record supplies.
• Recommended verification if a defensible negative is required. Run direct queries rather than relying on secondary web sources: PTAB E2E / PTAB Center at https://ptacts.uspto.gov for trial numbers tied to the patent; USPTO Patent Center (https://patentcenter.uspto.gov) for any pre-AIA inter partes or ex parte reexamination of 5,200,956 (a reexamination is not an AIA trial but would materially change the claim-status picture and I found no evidence of one either way); and CourtListener (https://www.courtlistener.com) for any derivative appeal. I could not execute machine queries against these systems in this session, so treat "zero proceedings" as a well-corroborated but not certified negative.

Uncertainty flags carried forward: (1) I did not retrieve the ODP record directly; the zero-count rests on the structured block supplied in the prompt, corroborated by my own negative searches. (2) I did not retrieve the file wrapper, so I cannot state whether any reexamination certificate exists that would have altered the claims. (3) The expiration inference is drawn from the Google Patents legal-status field, not from a primary USPTO fee record, and is an inference, not a verified docket fact.

I'll reconstruct the assignment chain for US 5,200,956, verifying the two recorded assignments I already have (reels 005389/0330 and 022637/0666) and hunting for correspondents, follow-on transfers, and NPE-directory matches.

Key lead found: the same reel/frame 022637/0666 appears in a Uniloc v. Google exhibit, confirming the Philips→IPG transfer was a bulk portfolio recording. I also see a Chinese record showing IPG Electronics 503 Ltd's rights transferred to Pendragon Wireless LLC (Jan 2013). Let me verify whether US5200956 specifically went to Pendragon and find correspondents.

Good — I've established the standard Philips chain pattern (Philips → IPG Electronics 503 → Pendragon Wireless). Critical nuance: this patent expired 2010-05-22, before the 2012 IPG→Pendragon deed. I need to verify whether US5200956 specifically appears in any post-2009 US assignment. Let me dig.

I hit the tool-call limit while verifying correspondents, so I'll report what the authoritative record and the searched evidence support, and flag precisely what I could not confirm. I did not fabricate any reel/frame, correspondent, or assignment.

Date note (carried forward): prior sections were generated under "today = 2026-09-21" and one under 2026-04-26; this task header says 2026-09-23. Unresolved; irrelevant here because all chain events below predate 2010 and the patent is expired.

Inventors

Both inventors appear jointly on a single assignment instrument to the original assignee (reel 005389/0330), executed ~13 months after the 1989-05-26 GB priority and ~1 month after the 1990-05-22 US filing. British-surname inventors on a digital-cordless/TDMA system are consistent with Philips' UK/Netherlands cordless research operations, but the assignment names the corporate assignee as U.S. Philips Corporation, and I did not locate a document stating a specific Philips research lab or subsidiary as the inventors' place of employment. No city/country of the inventors is stated in the record I retrieved.

Unusual-pattern check — negative. I found no evidence that either inventor departed Philips within 12 months of filing, no evidence of a second/divergent inventor assignment, and no orphaned-inventor pattern. An assignment executed ~13 months post-filing is unremarkable for a corporate filing of this era, not a "fire-sale precursor" tell. Flag: this is a negative from the retrieved record, not a verified personnel history.

Original assignee

U.S. Philips Corporation (assignment recorded 1990-07-16; address of record 100 East 42nd Street, New York, NY). It is the U.S. operating/patent-holding arm of the Philips group (parent: Koninklijke Philips N.V., Eindhoven).

• Primary line of business: consumer electronics, lighting, and healthcare equipment. The patent's own specification describes a digital cordless telephone system (TDMA/TDD, forward/reverse twinned slots) — i.e., the DECT-era cordless telephony market Philips was actively developing at the time.
• Did they ship a product embodying the claims? Likely yes, but not documented in this session. The embodiment is a commercial-shaped digital cordless telephone/base-station architecture, and Philips was a DECT cordless-telephone vendor. I did not locate a specific Philips product or datasheet tying the claims to a shipped unit, so I state this as an inference, not a verified fact.
• Current status: Operating. Philips remains a going concern; the old "U.S. Philips Corporation" name was folded into the Philips North America corporate family (Philips Electronics North America Corp → Philips North America LLC). I did not retrieve a primary corporate-registry document confirming the exact merger/rename sequence — treat the name-succession detail as standard-knowledge-level, not docket-verified. No bankruptcy of the original assignee was found (relevant to signal 6 below).

Assignment timeline

The record I have (the patent's "Legal Events" block, mirrored to Google Patents) shows exactly two recorded assignments. Reel/frame values are taken verbatim from that record.

• 1990-06-25 (executed) / recorded 1990-07-16 — Reel 005389/0330

  • Conveyance: Assignment
  • Assignor: Christopher D. Pudney and Frank C. G. Owen (the named inventors)
  • Assignee: U.S. Philips Corporation, 100 East 42nd Street, New York, NY
  • Correspondent: Not available in the retrieved record. The Google Patents "Legal Events" extract omits the correspondent field. I attempted to retrieve the assignment abstract but did not surface it before the tool limit. I will not guess an attorney or firm name.
  • Context: Ordinary employee invention assignment to the corporate employer at (or shortly after) filing — not a reorg, not a securitization.

• 2009-04-13 (executed) / recorded 2009-05-05 — Reel 022637/0666

  • Conveyance: Assignment
  • Assignor: U.S. Philips Corporation
  • Assignee: IPG Electronics 503 Limited (Trafalgar Court, Les Banques, St Peter Port, Guernsey GY1 3DA — Channel Islands; address per the sibling-registration record below)
  • Correspondent: Not available in the retrieved record. Same gap as above.
  • Context: Divestiture / transfer-to-licensing-vehicle. This is one link in Philips' large-scale divestiture of its patent estate into the "IPG" vehicles. Independent corroboration that the recording was a bulk portfolio transfer, not a patent-specific deal: the identical reel/frame 022637/0666, execution date Apr 13, 2009, and assignee IPG Electronics appear in a court exhibit listing at least one other Philips patent (US 6,298,090) recorded on the same instrument (CourtListener, Uniloc 2017 LLC v. Google LLC, N.D. Cal. 4:20-cv-04355, Dkt. 355-2).
  • Corroborating pattern (sibling Philips patents, not this patent): UK IPO register for EP1397899 records IPG Electronics 503 Limited (Guernsey) as proprietor in place of Koninklijke Philips Electronics N.V. by deed dated 2009-02-09; and the UK register for EP1673945 records Pendragon Wireless LLC (2300 Carillon Point, Kirkland, WA 98033) as proprietor in place of IPG Electronics 503 Limited by deed dated 2012-04-10. Chinese register records for related IPG cases show the same IPG→Pendragon transfer effective 2013-01-04 (e.g., CN98801569.2; CN1086872C).

No later assignment of US 5,200,956 was found. I specifically searched for a US record transferring this patent to Pendragon Wireless LLC and found none. There is a strong structural reason to expect none: the patent expired 2010-05-22 (Google Patents legal status "Expired – Lifetime"), i.e., ~13 months after the IPG acquisition and ~2 years before the IPG→Pendragon deed of 2012-04-10. A blanket 2012 deed could nonetheless have recited already-expired portfolio members, so I cannot exclude a nominal inclusion — but no evidence of one exists in what I retrieved. This is the single most important open verification item.

Verification links:

• USPTO Assignment Center — https://assignmentcenter.uspto.gov/
• Indexed Patent Assignment Search — https://assignment.uspto.gov/patent/index.html (search by patent number 5200956)

Timeline diagram

timeline
    title Ownership of US 5200956
    1989 : GB priority filed
    1990 : US application filed
         : Inventors assign to US Philips Corp
    1993 : Patent issued
    2009 : Assigned to IPG Electronics 503 Ltd
    2010 : Patent expires

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT (moderate-to-strong).
   Reel 022637/0666 (executed 2009-04-13, recorded 2009-05-05) moves the patent from the operating Philips entity to IPG Electronics 503 Limited, whose name carries a numerically-suffixed "Electronics 503 Limited" form characteristic of Philips' divestiture vehicles, and whose address (Trafalgar Court, Les Banques, St Peter Port, Guernsey) is a Channel Islands corporate-services registered office, not an operating site. Public documentary evidence independently characterizes the IPG Electronics 501–504 group as non-practicing licensing entities (KIPO research-institute report labeling "IPG electronics 501 ~ 504" as US San Diego-based NPEs, with IPG Healthcare 501 and IPG Electronics 503 among the operating subsidiaries). Qualification: I did not retrieve IPG 503's corporate registration or any statement of "no products," so the "no products in commerce" element rests on secondary-source characterization, not primary corporate records.

2. Known asserter in the chain — UNCLEAR.
   The recorded chain for this patent terminates at IPG Electronics 503 Limited, which does not match any entity on the enumerated list (Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation, Spangenberg entities). The enumerated entity Pendrell does appear in the wider IPG family — Unified Patents' page for US-20120155556-A1 lists Original Assignee: Ipg Electronics 503 Ltd, Current Assignee: Pendragon Wireless LLC, Parent Company: Pendrell Corp — and Pendragon/Pendrell are widely reported NPE/assertion-side entities (e.g., Uniloc's AirDrop campaign traces Philips → IPG 503 → Pendragon → Uniloc). But I found no evidence that US 5,200,956 ever entered the Pendragon/Pendrell/Uniloc chain, and the patent's 2010 expiry predates the 2012 Pendragon deed. Marked unclear rather than "present" to avoid inferring from a family pattern.

3. Repeat correspondent across the chain — UNCLEAR (data gap).
   The correspondent of record is not present in the retrieved legal-events data for either reel 005389/0330 or reel 022637/0666, and my search for the assignment abstracts did not return them before the tool limit. I will not name an attorney or firm. This signal cannot be scored without the reel/frame abstracts from Assignment Center, which expose the correspondent field directly.

4. Cascading transfers — NOT PRESENT.
   Only one post-issuance assignment exists in the record (2009), with no chained LLC hops in under 24 months. Nothing here resembles a multi-hop assertion chain.

5. Pre-litigation transfer — NOT PRESENT.
   No infringement suit naming US 5,200,956 was found in the prior litigation section, and the publication date-mismatch trademark hits for "5200956" (e.g., French mark ESPERANCE BANLIEUES, INPI filing 2025-11-20) are unrelated. There is no suit to which the 2009 assignment could be "pre-litigation," and the patent expired 2010-05-22, so the § 286 six-year lookback reaches no actionable window.

6. Bankruptcy fire-sale — NOT PRESENT.
   The assignor chain runs Philips → divestiture vehicle; no Chapter 7/11 of U.S. Philips Corporation or Koninklijke Philips was found, and the transfer is characterized as a negotiated portfolio divestiture, not a distressed sale.

7. Privateering — UNCLEAR (weak, pattern-level only).
   Philips' transfer of a broad patent estate into purpose-built IPG entities is consistent with an outbound monetization program of the privateering-adjacent type. But there is no evidence of (a) assertion of this patent or (b) any assertion-on-Philips'-behalf against Philips competitors. I also located no SEC 10-K/8-K disclosure tying this divestiture to a monetization arrangement. Marked unclear, not present.

8. Defensive aggregator (anti-NPE) — NOT PRESENT.
   The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. The patent was not neutralized by a defensive aggregator; it simply expired.

Verdict

NPE — moderate confidence.

Justification: The recorded chain moves US 5,200,956 from the operating original assignee U.S. Philips Corporation to IPG Electronics 503 Limited by assignment executed 2009-04-13 / recorded 2009-05-05 at reel 022637/0666 — a Guernsey-registered, numerically-suffixed divestiture/licensing entity whose sibling registrations (UK IPO EP1397899, deed 2009-02-09) and public NPE characterization (KIPO report on "IPG electronics 501 ~ 504") support a single strong shell/licensing-vehicle signal. There is no cascading-transfer, no pre-litigation, and no bankruptcy signal; the chain's recurrence-correspondent and known-asserter signals are unresolved data gaps, not negatives. The decisive limiting fact is that the patent expired 2010-05-22 — roughly 13 months after IPG took title and before the 2012 deed that carried other IPG patents to Pendragon Wireless/Pendrell — so this is an NPE-held asset that was never asserted, not an assertion vehicle.

Verify at: https://assignmentcenter.uspto.gov/ (search patent 5200956; both reel 005389/0330 and reel 022637/0666 should be returned, and the abstracts will supply the correspondent of record I could not retrieve).

Open verification items (explicit): (1) correspondents of record for both reels — not retrieved; (2) whether the 2012 IPG→Pendragon deed recited this already-expired patent as a nominal member — no US record found, not excluded; (3) inventors' specific Philips site/employer entity — corporate assignee only is documented; (4) exact U.S. Philips Corporation → Philips North America entity succession — standard-knowledge level, not primary-sourced.

I'll verify the patent record and then research each cited reference individually.

Now let me research the remaining cited references.

Prior-Art Analysis — US 5,200,956

0. Record verification (step 1 of the task)

I searched for the literal document number 5200956 and confirmed it maps to exactly one patent. I did not fold in results for look-alike numbers (e.g., 5,020,956 / 5,209,956 / 5,110,956), and I discarded unrelated hits where the digit string "5200956" appeared as a trademark serial, a service number, or an internal database ID.

Citation set on the face of the record (USPTO "Patent Citations," 8 entries = 7 distinct documents + 1 reissue of one of them):
WO1984004435A1 · US4556972 · US4638479 · US4656645 · EP0261127A1 · US4785450 · US4831373 · US4785450B1

Additional documents cited in the family (not on the US face): FR2584884B1 · DE3527329A1 · FR2608872B1

This builds on the earlier-generated bibliographic and claim sections rather than repeating them; the claim-element letters below correspond to the plain-language claim 1 / claim 10 breakdown already produced.

1. The references, with full citations

Notes on the record:

• US 5,200,956 is a pre-AIA patent (filed 1990), so § 102(a)/(b)/(e) and § 103 as they read before the AIA govern. All citations above pre-date the 1989-05-26 priority date except R8, which is the same specification as R7 and therefore rises or falls with R7 (its 1999 reissue date is irrelevant to its disclosure date).
• R1 and R7 are also not radio-TDD systems in the DECT sense, which is the crux of the anticipation question below.
• The three F documents were cited in the family (EP/GB prosecution of GB 2 232 326 / EP 0 399 611), not on the US front page. Treat them as prosecution-history art for the family, not as US face citations.

2. § 102 analysis — reference by reference

The recurring outcome: none of the cited references discloses the full combination of claim 1/claim 10, because each is missing at least the "non-identical packets … substantially in parallel in at least two duplex voice channels" step combined with the rate-triggered selection from a pre-populated map store and the post-message relinquishment. Anticipation therefore attaches, if at all, to individual dependent claims or to the generic preamble of claim 1, while the references are strongest as § 103 combination art.

R2 — US 4,638,479 (Alexis / Philips) — the most structurally similar reference

• Disclosure: TDMA (or FDMA) two-way radio. All fixed stations transmit in a numerically multiplexed "down" path, all mobile stations in an "up" path; a pair formed by one up channel and one down channel handles a two-way link — i.e., the duplex-pair construct. Each fixed station detects "up" channels whose received power exceeds a threshold and stores the addresses of those occupied channels; the fixed station continuously transmits that address list in its semaphore channel. The mobile station likewise detects and stores the addresses of occupied down-path channels and selects a free pair whose addresses differ from those stored and from those broadcast. Call setup from the network side is by paging the mobile's address in the semaphore channel; the final paragraph contemplates "several time intervals offered to a single link, permitting data transmissions with multiple outputs of 16 kbits/s."
• § 102 assessment:

R7 / R8 — US 4,785,450 (International Mobile Machines → InterDigital), incl. B1 reissue

• Disclosure: Wireless digital telephone base station with multiple sequentially repetitive time slots and a plurality of frequency channels, selectively assignable by moving a subscriber to a selected vacant time slot or frequency channel in accordance with a predetermined assignment schedule, in response to a subscriber coming on line or to deterioration of transmission. A memory matrix holds the pattern of frequency channels × slots, and the remote-connection processor "continually monitor[s] the status of the time slots" and stores the connections so a transfer can be effected at any time. Assignment is triggered by events (new subscriber, interference, fading, equipment failure) rather than by offered data rate.
• § 102 assessment:

R6 — US 4,831,373 (Motorola)

• Disclosure: Trunked voice/data radio (FDD, with a dedicated control channel). Data activity is monitored over a predetermined interval; if activity exceeds a maximum, an additional channel is reserved for data use; if data traffic is low, a data channel is returned to voice. Subscribers are re-balanced across available data channels. Each data subscriber computes a "data channel table" stored in RAM from its talk-group ID and an offset seed and uses a pointer to select one of the available data channels; allocation information is broadcast on the voice control channel or a master data channel; "revert-to-master" and system messages ride on the data stream. The background expressly criticizes the ~0.5 s setup/clean-down penalty of the ISW/OSW channel-grant route and the wastefulness of permanently dedicated data channels.
• § 102 assessment:

R1 — WO 1984/004435 (CXC Corporation)

• Disclosure: Digital (wired, ring-network) communications system carrying a time-multiplexed network signal stream with a control-data portion and a circuit-switched data portion. A controller interprets a device's message and generates bandwidth-allocation control signals; nodes contain a "data steering map" (element 107) that stores the configuration correlating information transfer, and bandwidth is variable and dynamically reconfigurable so that devices operating at different data rates (64 kbps voice, 19.2 kbps terminals, etc.) are accommodated without wasting assigned slots. Bandwidth is expressed as bits per frame, and registers can be reconfigured via control signals; resources are not "unnecessarily diverted" once requirements drop.
• § 102 assessment:
  • Claim 1 (e)/(f) analogue — arguable § 102/§ 103: variable bandwidth allocation in response to a device's declared rate requirement, together with a stored map of allocations, is conceptually the closest prior teaching of the "map store + add capacity when the rate demands it" pairing.
  • Claims 1 and 10 as a whole — No: it is not a radio system, has no TDD frame of 2n twinned slots forming duplex voice channels, and does not transmit non-identical packets of one message substantially in parallel across several duplex channels with per-message release. Best characterized as § 103 art against the map-store/bandwidth-adaptation sub-features.

R3 — US 4,556,972 (AT&T Bell Laboratories) — record-level only

• Citation: "Arrangement for routing data packets through a circuit switch"; priority 1983-12-27; granted 1985-12-03.
• Provisional description (not verified from full text): packet routing through a circuit (time-slotted) switch — i.e., packetized traffic carried over switched, time-divided paths.
• § 102 assessment (low confidence): on its face it is directed to packet routing in a circuit switch, which bears on the "packets" and "data message" vocabulary of claims 1/10 but not on TDD duplex-pair allocation, the map store, the rate trigger, or parallel multi-channel transmission. I would expect it to have been cited for the general packet-through-slotted-circuit idea. I did not retrieve its disclosure, so I will not map it to specific claims as a confident finding.

R4 — US 4,656,645 (NEC) — record-level only

• Citation: "Radio communication system"; priority 1984-04-02; granted 1987-04-07.
• Provisional description (not verified): a radio (multi-channel) communication system; given the period and assignee, likely a channel-selection/availability arrangement for a radio telephone system.
• § 102 assessment (low confidence): if its disclosure is indeed channel-availability selection from stored status, it would be relevant to the claim 1(c)/(d) usage-determination and map aspects. Unverified — not asserted as an anticipation finding.

R5 — EP 0 261 127 A1 (Motorola) — record-level only

• Citation: "Time multiplexing communication system for efficient use of the frequency spectrum"; priority 1986-03-25; published 1988-03-30.
• Provisional description (not verified): a TDM system aimed at spectrum efficiency — typically such documents deal with slot/channel reuse or half-rate/borrowed-slot schemes in a time-multiplex frame.
• § 102 assessment (low confidence): its subject matter sits squarely in the claim 1(b) frame/slot-allocation space, so it is a plausible § 102/§ 103 reference for the frame-structure preamble only. Unverified.

Family-cited references F1–F3

These were cited in the family's European/British prosecution and are worth listing in a validity study, but they are not US face citations:

• FR 2 584 884 B1 (TRT) — free-channel search method/device for a mobile radio system (priority 1985-07-09). Directly relevant to the problem the patent solves (avoiding a full scan) → useful § 103 art against claims 1/10's stated motivation, though it teaches searching, not a maintained map.
• DE 3 527 329 A1 (Philips) — digital radio transmission system with variable time-slot duration within a time-multiplex frame (published 1987-02-05). Relevant to variable-bandwidth/time-slot allocation → claim 1(b)/(e) context.
• FR 2 608 872 B1 (TRT) — low-throughput data transmission device for an information transmission system using the AMRT (TDMA) principle in one transmission direction (priority 1986-12-19). Relevant to data-over-TDMA-frame allocation → claims 1/10 data-handling context.

3. Ranking: the most relevant prior art for US 5,200,956

1. US 4,638,479 (R2) — closest overall. Paired up/down multiplexed channels, threshold-based occupancy detection, stored addresses of occupied channels in the mobile station, and periodic broadcast of occupancy from the base station. Strongest § 102 candidate for claims 4, 8 and 9 and for the usage-determination/store portion of claim 1's preamble; not an anticipant of claims 1/10 as a whole, and its "several time intervals … to a single link" remark is the nearest thing to multi-channel data in the cited art.
2. US 4,831,373 (R6) — best § 102 reference for the dynamic data-channel allocation and central signalling concepts (claims 8, 9), with a stored "data channel table" that is functionally a map store, and an express motivation to avoid channel-grant setup latency.
3. US 4,785,450 / B1 (R7/R8) — best reference for the frequency-agility + monitored occupancy matrix + multi-frequency-channel assignment features (claim 2, arguably claims 6/7).
4. WO 1984/004435 (R1) — best reference for stored allocation maps and rate-driven variable bandwidth; a § 103 staple rather than a § 102 anticipant.
5. US 4,556,972 (R3) and EP 0 261 127 A1 (R5) — background art for packet-through-slotted-circuit and TDM spectrum-efficiency framing, respectively; unverified content.
6. US 4,656,645 (R4) — background radio channel-selection art; unverified content.
7. FR 2 584 884 B1, DE 3 527 329 A1, FR 2 608 872 B1 (F1–F3) — family-prosecution art: free-channel search, variable slot duration, and data over TDMA respectively.

Bottom line on novelty: no single cited reference discloses the combination that the previously generated summary identified as the core — (i) a pre-populated channel-usage map enabling immediate multi-channel access without a fresh scan, (ii) rate-triggered parallel transmission of non-identical packets over ≥2 TDD duplex voice channels, and (iii) release of the extra channels after the message. That is consistent with allowance over the cited art, though I have not reviewed the file wrapper and so cannot state the exact examiner rationale.

4. Caveats (explicit uncertainty)

• Source and method. The authoritative text is the provided Google Patents full text for US5200956A (https://patents.google.com/patent/US5200956/en), which mirrors the printed front page. I did not retrieve the USPTO Patent Center / PAIR file wrapper, so I have no examiner's citations list beyond the front-page set, no office actions, and no § 103 rejections to report.
• Three US citations are described only at record level. I obtained full detail for WO 1984/004435 A1, US 4,638,479, US 4,831,373, and US 4,785,450 (+B1). I did not obtain full text for US 4,556,972, US 4,656,645, or EP 0 261 127 A1, nor for FR 2 584 884 B1, DE 3 527 329 A1, or FR 2 608 872 B1. My descriptions of those are provisional and confined to what the citation record establishes (number, title, assignee, dates); I have deliberately not assigned them confident per-claim § 102 findings. Retrieving their full texts from USPTO Patent Center (or Espacenet for EP/DE/FR) is the required next step before relying on any of them.
• "8 citations" ≠ 8 documents. US 4,785,450 and US 4,785,450 B1 are the same disclosure; treat them as one reference (R7/R8).
• Anticipation vs. obviousness. Where I write "arguable" above, a literal § 102 finding would also require that the reference disclose the element arranged as in the claim; several of these readings (particularly R1, R6, R7 against the map-store element) are more naturally § 103 combinations and would need the secondary-reference linkage to be spelled out.
• Related-but-not-cited document. US 5,117,423 ("Data transmission over a time division duplex channel," Shepherd et al., granted 1992-05-26) sits in the same Philips TDD space and appeared alongside this patent in the search results, but it is not among the citations of US 5,200,956 and I have not treated it as prior art here. If your interest is the Philips TDD/DECT family generally, it should be assessed separately.

I'll ground the prior-art characterizations with targeted searches before building the § 103 combinations.

Now let me verify the remaining cited references.

§ 103 Obviousness Analysis — US 5,200,956

Metadata contradiction (flagged, per instructions): The previously generated sections were dated 2026-09-21 ("Today's date is 2026-09-21"; Google Patents "Fetched: 2026-09-21"). This task states the current date is April 26, 2026. I cannot reconcile these two, and I am not auto-correcting either. The discrepancy does not affect this analysis: the patent's anticipated expiration is 2010-05-22 (per the record cited in the prior sections), so all § 103 conclusions below are retrospective regardless of which 2026 date is correct.

I build on the claim summary already generated (claim 1 / claim 10 independent; claims 2–9 and 11–14 dependent) and do not repeat the bibliographic or litigation findings. One prior-section statement needs a small refinement, flagged in §6 below.

1. Governing framework and prior-art status

This is a pre-AIA application (filed 1990-05-22; GB priority 1989-05-26), so pre-AIA 35 U.S.C. § 103 and Graham v. John Deere Co., 383 U.S. 1 (1966), control. KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), supplies the combination rationales (known elements combined by known methods to yield predictable results; use of a known technique to improve a similar device in the same way; "obvious to try" where the prior art presents a finite number of identified, predictable solutions).

The DE 3 527 329 A1 counterpart US 4,763,322 (Philips, granted 1988-08-09) is available in English and discloses the same subject matter; it is useful because it is quoted at length in the record (SumoBrain).

Level of ordinary skill: a B.S. in electrical engineering (or equivalent) with roughly 2–3 years of experience designing or specifying TDMA/digital cordless radio systems and channel-assignment protocols. Every reference above is analogous art (two-way radio, trunked radio, or TDM channel switching).

2. Claim 1 — element-by-element mapping to the primary ground

Primary ground: US 4,638,479 in view of US 4,831,373, further in view of DE 3 527 329 A1 / US 4,763,322.

3. Motivation to combine (why the combination is proper)

Four independent KSR-recognized rationales support the primary ground:

1. Same field, same problem, same solution family. All three references address efficient spectrum use in two-way radio while serving both voice and data. US 4,638,479 (a Philips reference, i.e. the applicant's own earlier work in the identical architecture) supplies the TDD TDMA frame with paired up/down channels and a subscriber-stored occupancy list. US 4,831,373 identifies the precise defect the '956 patent set out to cure: "the time it takes for a requesting subscriber to transmit an ISW and receive a channel grant OSW… (approximately 0.5 seconds)," and notes that a typical data message "would take substantially less time to transmit than the time required to obtain a channel." US 4,831,373's answer — store the channel assignment locally and transmit immediately without re-requesting — is the very mechanism recited in claim 1. A POSITA confronting the '956 patent's stated problem (bursty high-rate data plus a 0.5 s dynamic-allocation overhead) would look to the trunked-data allocation art.

2. Known technique, applied to a known device ready for improvement. US 4,638,479 already teaches storing indicia of channel usage in the station. US 4,831,373 teaches storing a data-channel table and transmitting on it immediately. Substituting the trunked data-channel table for the TDD occupancy list is a predictable combination of two known data structures serving the same function.

3. Bandwidth-driven multi-channel allocation is expressly contemplated. US 4,763,322 / DE 3 527 329 A1 exists precisely to "optimally utilize the channel capacity… for users with different bandwidth requirements." A POSITA seeking to carry 2B+D ISDN (the 144 kbit/s figure the '956 specification itself cites) over 32 kbit/s duplex voice channels has a finite, identified, predictable set of options: buffer it (the specification admits this), lengthen the slot (US 4,763,322), or aggregate slots. Selecting additional duplex channels from the already-populated map is the natural variant, and KSR permits it as "obvious to try."

4. Design incentive / market force. The specification's own motivation (fax, graphics, ISDN 2B+D, cordless video phones, computer communications at rates above 32 kbit/s) is a market demand of the kind KSR treats as supplying the motivation.

No reference teaches away. US 4,831,373's disparagement of permanently dedicated data channels points toward, not away from, dynamic multi-channel allocation with release — which is precisely claim 1's model.

4. Alternative and cumulative grounds

Ground B — EP 0 261 127 A1 in view of US 4,831,373 and US 4,785,450. EP '127 discloses a TDM channel "divi[ded] into at least two time slots," vo-coding, temporary buffering of a signal at a first rate and transmitting it "at a second rate exceeding 2V during at least one of said N time slots," and re-transmitting a received signal in another time slot. US 4,785,450 discloses a "memory matrix" in which "the time slots and frequency channels are selectively assignable to said subscriber stations by movement of the subscriber stations to selected vacant time slots or frequency channels in accordance with a predetermined assignment schedule" — i.e., a map store of occupancy, with a "remote-connection processor… continually monitoring the status of the time slots." Combining EP '127's buffered, multi-slot, higher-rate burst transmission with US 4,785,450's monitored occupancy matrix yields a system that transmits concurrently in multiple known-free slots. Claim 1's "substantially in parallel" and "relinquish" limitations follow from EP '127's slot-reuse and US 4,831,373's deallocation teaching.

Ground C — WO 84/04435 A1 (variable bandwidth switching) in view of US 4,638,479 / US 4,831,373. A "variable bandwidth switching system" combined with a dynamic radio channel allocator supplies the "allocate more channels to a wider-bandwidth call" concept directly. Caveat: I did not retrieve WO 84/04435 A1 this session, so I state this ground as a hypothesis requiring the reference's full text before it can be relied on.

Ground D — US 4,556,972 (packet routing through a circuit switch) in view of US 4,638,479. Supplies the packetised, per-path routing of a single data stream. Also unverified.

5. Claim 10 and the dependent claims

• Claim 10 (secondary-station apparatus) is the station-level mirror of claim 1 and is rendered obvious by the same combination; the station hardware (tx/rx, quality measuring means, map store, control section) maps directly onto US 4,638,479's mobile station plus US 4,831,373's subscriber with RAM-resident channel table.
• Claim 2 (≥2 frequency channels; per-frequency listing; transmit on a channel from a different frequency channel): US 4,785,450 discloses "a matrix containing a pattern of frequency channels, each frequency channel having a multiplicity of time slots… whereby a subscriber station may be selectively movable from one slot to another within a selected frequency channel or from one slot in one frequency channel to another slot in another frequency channel."
• Claims 3, 6, 7 (quality measurement stored with usage indicia; controller picks "better" channels; updates the list): US 4,785,450 discloses "link quality" (LQ) and "malfunction" (MF) status messages and a processor that "continually monitor[s] the status of the time slots"; US 4,638,479 compares received power against "predetermined threshold[s]." Storing and re-sorting a list by measured quality is routine optimisation.
• Claim 4 (primary station reports its local radio environment periodically; secondary stores it): This is essentially US 4,638,479 in terms. The fixed station "continuously transmits by means of the semaphore channel… the addresses of the channels occupied by communications from mobile stations located in the interference zone of the said fixed station," and the mobile station stores those indications.
• Claim 5 / claim 14 (cross-correlation of the station's own quality measurements with the primary's environment report to derive overall channel quality): the weakest mapping. US 4,638,479 requires the mobile to make its own down-path power measurements and to receive and store the fixed station's semaphore-transmitted occupancy data, so both inputs are present; the specific step of cross-correlating them into a single composite score is a predictable combination of two known indicia. An examiner or challenger would need to articulate this combination explicitly (I do so here), and this is where a § 103 challenge is most contestable.
• Claims 8, 9 (control means coupled to the primary station controlling assignment and signalling channel details to the secondary; usability details in the signalling portion of a message): US 4,831,373's central controller allocates data channels and broadcasts, over the "master data channel," "the total number of currently available data channels" and "the identity of the data channels" within a signalling format having a preamble; US 4,785,450's RPU issues command signals to the CCUs over a baseband control channel. Both disclose assignment controlled at the fixed end and signalled to the subscriber.
• Claims 11–13 mirror claims 3, 4 and 5 at the station level and fall with them.

6. Vulnerabilities of the obviousness case, and prior-section correction

• Strongest surviving limitation. The conjunction in claim 1 of (i) "non-identical packets… transmitted substantially in parallel in at least two duplex voice channels" with (ii) "relinquishing said additional duplex voice channel following transmission." US 4,831,373 gives (ii) almost verbatim in substance and gives single-channel data transmission plus a multi-channel data channel table; it does not, on its face, show a single message split across simultaneously used channels. US 4,763,322 does not show parallel channels either (it lengthens a slot). So the fully grounded primary ground rests on the combination — the rate trigger from US 4,763,322 plus the multi-channel use from the references' data-channel tables plus packet multiplexing as an obvious design choice. If a challenger must prove that parallel splitting of one message was itself suggested, the WO 84/04435 A1 and US 4,556,972 texts (unverified here) matter a great deal. This is the limitation I would expect to have been the allowance fulcrum, which is consistent with the prior section's inference (correctly flagged there as inference, not documented fact) that "non-identical packets … in parallel" and "relinquishing" are the likely point of novelty.
• Secondary considerations. None are available on this record. The specification's stated advantages (removing "the delay and the time overhead" of scanning) are the same advantages US 4,831,373 already claims for stored-channel-table access, so there is no evidence of an unexpected result and no nexus evidence. No objective indicia (commercial success, long-felt need, failure of others, copying) were located in earlier searches.
• Small correction to the prior sections. The litigation section states US 4,831,373 is § 102(b)-type art generally. Precisely: US 4,831,373 granted 1989-05-16, only ten days before the 1989-05-26 GB priority date, so it is not § 102(b) art measured against the priority date; it remains available as prior art under § 102(a) (by another) and/or § 102(e) (as of its 1987-04-30 US filing date). Because the applicant is Philips and the reference is Motorola, the "by another" requirement is met. This does not change any conclusion; it only sharpens the statutory basis, and it is the kind of timing point that a careful challenger would be expected to get right.
• Status. This analysis is retrospective. The patent is recorded "Expired – Lifetime," anticipated expiration 2010-05-22 (Google Patents), so there is no live validity controversy to which these grounds attach.

7. Explicit uncertainty flags

1. Unverified references. I did not retrieve the full texts of US 4,556,972, US 4,656,645, or WO 84/04435 A1. Any mapping that depends on them (Grounds C and D, and the "non-identical packets in parallel" element) is a hypothesis, not a verified finding.
2. US 4,763,322 is treated here as the US counterpart of DE 3 527 329 A1 on the basis of shared title, assignee (U.S. Philips Corp.), and matching disclosure (SumoBrain). I did not retrieve the formal family link.
3. Prosecution history was not retrieved, so which claim language was added to overcome which reference remains an inference (as flagged in the prior summary section).
4. Dates. The 2026-04-26 vs. 2026-09-21 discrepancy noted at the top is unresolved and is reported as-is.