US 10706412

I'll run several targeted searches for this specific patent number and check for any appellate docket activity.

Let me pull the actual claim language and confirm the litigation/appellate posture.

US 10,706,412 B2 — Analyst Summary

Note on dates: Your prompt states today is April 26, 2026, but my session date is 2026-09-19, and the search results below include items dated into August 2026. I have not filtered out post-April-2026 material; flagging this so you can reconcile it against whatever "as-of" cutoff you intended.

1. Bibliographic record (as reported by the authoritative source text)

Conflicting secondary-source values (do not reconcile silently): The Unified Patents portal page for US-10706412-B2 reports priority date 2002-07-08, application date 2015-03-03, and grant date 2020-07-06 — i.e., each value is one day earlier than Google Patents. These are almost certainly convention/time-zone artifacts, not different events, but I am reporting the discrepancy rather than picking a winner. It also reports "Lifetime Renewal Fees: $0."

Point of real consequence: both Google Patents and Unified Patents place the anticipated expiration in early July 2023 and show the patent as lapsed. That is internally consistent (a patent granted 2020 with a 2002-family term would owe no post-grant maintenance fee if it lapsed in 2023). If accurate, the '412 would have been expired before the 2026 suits discussed below were filed, which would confine any recovery to past damages under the § 286 lookback rather than injunctive relief. I am not making a legal conclusion here — Google expressly labels its status as "an assumption … not a legal conclusion," and I have not verified the PTA/terminal-disclaimer history that produces the 2023-07-09 date.

2. Abstract (verbatim)

"The present invention provides a method and system for verifying and tracking identification information. In an embodiment of the invention, a system for delivering security solutions is provided that includes at least one of the following: a radio frequency (RF) identification device, an identification mechanism (e.g., a card, sticker), and an RF reader."

3. Independent claims — I do not have authoritative claim text for this patent

I want to be direct about this rather than paraphrase something I cannot verify. The full patent text you supplied contains no claims section. It runs from the title through the end of the "II. RF Registered Item and Method of Use" description and stops at the "Secure Data Base" paragraph. Google Patents' rendering of this record likewise only surfaces a fragmentary claim-reference snippet (…response Effects 0.000 claims description 3). My targeted searches for the claim text of the '412 returned unrelated patents and unrelated subject matter, not the '412's claim set.

Accordingly, I cannot give you the number, count, or verbatim language of the independent claims of US 10,706,412, and I will not reconstruct them from the specification's embodiments, because doing so would be fabrication. This is a hard gap, not a hedging statement.

Two indirect, clearly-labeled inferences (evidence-backed but not claim text):

1. Claim-drafted subject matter (supportable): The assigned CPC codes are claim-derived and cluster tightly on mobile wireless payment: G06Q20/3278 ("RFID or NFC payments by means of M-devices"), G06Q20/327, G06Q20/34, G06Q20/3674 ("…involving authentication"), H04L63/0861 ("using biometrical features, e.g. fingerprint"), G07C9/257, G06K19/0723, H04W4/80, and H04W12/06. That profile strongly indicates the claims recite an NFC/RF mobile-device payment architecture with biometric authentication — which is also exactly how the '412 is being asserted in the 2026 Samsung case (see §5).
2. Claim vocabulary (weaker): Google Patents' auto-generated "Prior art keywords" for this record are: information, memory, biometric, payment information, security. The term "payment information" appears nowhere in the description text you supplied, which suggests it is drawn from the claims rather than the specification. Treat this as a hint about claim terminology, not as claim language.

4. Plain-language overview of the disclosure (specification-based, not claim-based)

The '412 shares a common specification with the rest of the Neology family (it cross-references U.S. App. Ser. No. 10/118,092 filed 2002-04-09 and PCT/IB02/01439 filed 2002-04-30). It is a kitchen-sink disclosure — consistent with its ~20-year pendency — organized as six embodiments:

1. General RF identification system — an RF device (chip + antenna) embedded in an identification mechanism, read contactlessly by an RF reader/writer (FIGS. 1–2, 4–5).
2. RF registered item / license plate — VIN, owner biometric data, and photo written to an embedded chip so an officer in the field can verify vehicle and driver identity without a central-database round trip; includes selective demetalization and conductive-ink antenna fabrication (FIGS. 5–8).
3. RF passport — biometric template written to a chip in a passport inlay, with authentication keys and personalization applied post-manufacture so an uninitialized booklet is useless if stolen (FIGS. 9–11).
4. RF communications device (the transaction embodiment) — the closest thing to "secure transactional solutions." An RF cellular telephone/PDA holds identification and transaction information (credit card, airline/hotel loyalty), paired with a POS having an RF reader/writer and a fingerprint scanner; a one-to-one local biometric verification confirms the holder is the enrolled owner before payment data is released. The chip is described as passive and secure, with the EEPROM memory "totally separated from the rest of the communication" so that memory is inaccessible if security requirements are not met (FIGS. 12–14).
5. Border crossing control — RF decals on cab/trailer/containers plus an RF laser card for the driver, pre-registered at origin, read in sequence at checkpoints; closed (all data on-card) or open (networked) architectures (FIGS. 15–18).
6. Software-download licensing and airport security — an RF smart label locking a software license to a CPU serial number, and RF boarding passes/baggage tags cross-referenced to a biometrically verified passenger (FIGS. 3A–3B, 19–23).

Recurring technical spine across all embodiments: a chip with modulator, AC/DC converter, encoder/decoder, power control, instruction sequencer, security management unit, hardware cryptographic block with exchange keys, and EEPROM memory — plus a thick layer of physical/optical anti-counterfeiting (holograms, microprint, taggants, UV fluorescence, up-conversion IR materials).

5. Enforcement posture

District court (active, 2026 filings):

The Samsung suit asserts a seven-patent family that includes 10,706,412 alongside U.S. 8,151,345; 8,548,924; 8,766,772; 8,933,807; 10,235,513; and 10,762,187. Reported characterization: the family covers secure-element, tokenization, NFC, and biometric-authentication techniques in mobile payment.

CAFC / appellate — I found nothing, and none would be expected yet. My searches returned no Federal Circuit docket, opinion, or Rule 36 judgment involving US 10,706,412 in 2026 (or any year). That is consistent with the procedural posture: the two district court cases were filed in Feb/Mar 2026, Samsung's answer was not filed until June 2026, and neither case has reached judgment. I cannot confirm a CAFC appeal exists, and I would treat any claim that one does as unsupported. State this as "not found," not "does not exist." The docket listings I retrieved were also flagged "docket is incomplete" in at least one aggregator, so a negative finding here is weak evidence.

6. Uncertainty register

Recommendation if you need the claim analysis: pull the granted claims directly from USPTO Patent Center or the USPTO Patent Public Search full-text database for application 14/639,034, or from the Google Patents PDF of US10706412B2. That is the only way to satisfy your "plain-language overview of each independent claim" requirement authoritatively, and I would rather hand you an exact gap than a plausible-sounding reconstruction.

I'll run targeted searches for litigation on this specific patent number.

Let me search more specifically on the litigation portals and the plaintiff entity.

Let me verify the Apple complaint's asserted patents and check for any Federal Circuit or additional proceedings.

Litigation involving US 10,706,412

I searched the litigation aggregators (Unified Patents patent/litigation portal, Docket Alarm, Justia dockets, PACER Monitor, the Ex Parte docket mirror), plus secondary press. Below is what I can ground in retrieved sources, with confidence levels.

Confirmed asserters of the '412 (the patent is in suit)

Sources: Korean trade press item explicitly listing 10,706,412 among the seven asserted patents (seoulwire.com); E.D. Tex. docket mirror (docketupdate mirror); ai-lab.exparte.com case page.

Reported, but I could not independently confirm the '412 is among the asserted claims

Sources: Docket Alarm, Ex Parte case page, Patexia, Justia.

Important caveat. The Apple complaint analysis I retrieved reproduces claim-element charts for U.S. 8,151,345 and U.S. 8,548,924 only; it does not list the full set of asserted patents, and I found no source stating that 10,706,412 specifically is asserted against Apple. Google Patents' own record links this patent to W.D. Tex. case 7:26-cv-00075, which is corroborating but not claim-level proof. Treat Apple as "probably in suit on the '412" — unconfirmed — not as established.

Do NOT list as '412 litigation (near-miss identifiers)

• Mobility IP Holding, LLC v. Para Mobility PTY LTD, N.D.N.Y. 3:26-cv-01456 (filed July 2026). Different plaintiff entity (LLC, not Inc.) and different patents — U.S. 9,863,577 and 10,125,920. It does not involve the '412. (PACER Monitor)
• Blaze Mobile / Fisher v. Samsung (N.D. Cal. 5:21-cv-02989) involves a patent numbered 10,621,612 and 10,699,259 — not the '412. Do not conflate.

Appellate (CAFC) / PTAB

• No Federal Circuit docket, opinion, or Rule 36 judgment involving the '412 was found. This is expected given the posture: both district cases were filed in Feb./Mar. 2026 and the Samsung answer was not filed until June 2026. State this as "not found," not "does not exist" — my search was not exhaustive.
• No IPR, PGR, or other PTAB proceeding on the '412 was found in my searches. Again: "not found," not a negative finding.

Two contradictions with the earlier-generated section — flagging explicitly

1. Legal status string changed. The previously generated section recorded the status as "Expired – Fee Related." The Google Patents page as retrieved in this session now returns "Expired - Lifetime." These are different status labels from the same source. One of the two is stale; I cannot tell which. The Unified Patents portal value ("Lifetime Renewal Fees: $0") is consistent with no maintenance fee having been paid, which sits better with "Fee Related," but I am not reconciling this silently.
2. Enforcement window. Both aggregators still place anticipated expiration at ~2023-07-08/09 and show the patent lapsed. If accurate, the '412 was expired before the 2026 suits were filed, which would confine any recovery to past damages within the 35 U.S.C. § 286 six-year lookback (reaching back to roughly mid-2020) rather than injunctive relief. I am not making a legal conclusion — Google expressly disclaims legal effect and I have not verified the PTA/terminal-disclaimer history driving the 2023 date. This is material to both cases and worth verifying against the face of the patent.

Also note: your prompt states today is April 26, 2026, but the retrieved dockets and press include events dated June–September 2026 (Samsung's 2026-06-23 answer; a docket "last updated 09/18/2026"). I have reported the post-April-2026 material rather than truncating it, consistent with the prior section's flag.

Bottom line: Confirmed litigation = one case squarely asserting the '412 (Samsung, E.D. Tex. 2:26-cv-00157, filed 2026-02-27, currently open). A second case (Apple, W.D. Tex. 7:26-cv-00075, filed 2026-03-04) is linked to this patent by Google Patents' family-litigation record and by plaintiff identity, but the '412's inclusion in the Apple asserted-claim set is unverified. No appellate or PTAB matter found.

I'll research this proceeding before writing anything. Let me run targeted searches.

Let me dig for the petition itself and any parallel-proceeding stipulations.

PTAB / AIA Trial Analysis — U.S. Patent No. 10,706,412 B2

Scope note, stated up front and unhedged: the canonical structured feed for this patent contains exactly one AIA proceeding, and it was filed on the same date as this data pull (2026-09-19). That has one dominant consequence for your deliverables: there is no claim-level outcome of any kind to report. No institution decision, no final written decision, no panel, no settlement, no appeal, no canceled claim, no sustained claim. Everything in the requested template under "Final Written Decision," "Settlement," and "Appeal" is N/A by reason of ripeness, not by reason of my failing to find it. I flag this hard because the alternative — writing a plausible-sounding FWD narrative — would be fabrication, and because a defendant is entitled to know whether a "PTAB win" exists to cite. As of today, it does not.

Proceedings overview

Total AIA proceedings on US 10,706,412: 1 (IPR2026-00501). Status breakdown: 1 active/pending; 0 claims invalidated; 0 claims sustained; 0 settled; 0 institution denials. The petition was filed by [Apple Inc.](/litigations/by-plaintiff/Apple%20Inc.) et al. on 2026-09-19 — zero days of elapsed trial time as of this analysis.

Bottom-line defensive posture: This is neither "the patent survived two IPRs and is hardened" nor "claims 1–5 are canceled." It is the third posture, and the most under-appreciated one for a defendant: the patent is PTAB-naive — it has never been adjudicated at the Board, in any AIA vehicle, and no claim of it is dead or blessed. Any demand letter built on U.S. 10,706,412 is relying on claims that no tribunal has yet construed for validity. The only thing that has changed as of 2026-09-19 is that Apple has committed real money to a § 102/§ 103 challenge, which (a) will produce a public, claim-by-claim merits record within ~18 months if instituted, and (b) may generate § 315(e)(2) estoppel that constrains Apple's litigation defenses but not yours.

IPR2026-00501 — Apple Inc. et al. v. Mobility IP Holdings, Inc.

• Type: Inter Partes Review (35 U.S.C. §§ 311–319). Not a PGR — unavailable here: the '412 carries a 2002 priority date and issued 2020-07-07, so the § 321 nine-month PGR window closed years ago and the patent is not a first-inventor-to-file patent. Not a CBM — the AIA § 18 transitional program's authority for new petitions expired in September 2020.
• Filed: 2026-09-19 (verbatim from the structured feed; feed "last modified" date is also 2026-09-19).
• Status: Pending — verbatim from the structured data. Gloss: petition on file, trial not yet instituted; the proceeding is in the pre-institution window in which (i) the patent owner's preliminary response is due ~2026-12-19 under 37 C.F.R. § 42.107(b), and (ii) the institution determination is statutorily due ~2027-03-19 under § 314(b) (three months from the POPR deadline, which is the standard six-months-from-petition cadence in practice).
• Judge panel: Not yet public. The Board does not assign the merits panel in a way that is reliably reported until/unless institution is granted. Do not accept any panel list you see for this proceeding as of today.
• Petition grounds: Not publicly available as of 2026-09-19. I could not retrieve the petition, its grounds table, or its challenged-claim list; the P-TACTS/PATENTS END-TO-END record for a same-day filing is not yet indexed in any source I could reach. What can be said from the statute rather than from the petition: an IPR can only assert § 102 and/or § 103 grounds based on patents and printed publications (§ 311(b)). No § 101 and no § 112 theory is available to Apple in this forum, even though the '412 (a 2002-priority continuation with a kitchen-sink specification, as noted in § 3–4 of the earlier summary) is the kind of patent that often draws § 112 written-description attacks in district court. Do not assume Apple's IPR grounds are coextensive with its district court invalidity contentions — they cannot be.
• Institution decision: Not issued / not due yet. Statutory deadline ~2027-03-19 (§ 314(b)). Standard-of-review: whether there is a reasonable likelihood that Apple prevails on at least one challenged claim (§ 314(a)). Note the 2026 practice environment: the Director is exercising discretionary denial aggressively, and the newly precedential Magnolia Medical Technologies, Inc. v. Kurin, Inc., IPR2026-00097 (PTAB May 14, 2026) restated that institution is never compelled and that parallel-litigation leverage (overlapping art, use of system art in court, no meaningful narrowing, poor prospects of a stay) weighs against institution. Apple filed this petition roughly six months after Mobility IP Holdings sued it in W.D. Tex. (see below), so a discretionary denial under Magnolia/§ 314(a)/§ 325(d) is a live outcome, not a formality.
• Final Written Decision: None. Not due until ~2028-03-19 at the earliest — § 316(a)(11) requires the FWD within one year of institution, extendable by up to six months for good cause. There is no claim of the '412 that has been canceled, and none that has been held patentable. Any representation to the contrary is false.
• Settlement / termination: None. No termination, no adverse judgment, no request for adverse judgment on the record.
• Appeal: None, and none is possible yet. No FWD exists to appeal to the Federal Circuit. This is consistent with the earlier-generated enforcement section, which found no CAFC docket activity for the '412 — that negative finding remains accurate and is now explainable by ripeness rather than by search incompleteness.
• Real parties in interest / privity: The feed lists the petitioner as "Apple Inc. et al." The "et al." matters more than it looks. Whoever appears on Apple's RPI list determines the reach of § 315(e)(1)–(2) estoppel and of § 315(b) timeliness. Verify the mandatory notices once the petition is posted. In particular, if any Samsung entity is named as an RPI or privy, the E.D. Tex. defendant (Mobility IP Holdings, Inc. v. Samsung, 2:26-cv-00157) inherits estoppel exposure it does not currently have; if it is not, Samsung retains the full § 282 prior-art defense unencumbered.
• Timeliness: Apple's own one-year clock under § 315(b) runs from service of Mobility's complaint. The W.D. Tex. complaint (7:26-cv-00075, Mobility IP Holdings, Inc. v. Apple Inc.) was filed 2026-03-04; a 2026-09-19 petition is comfortably timely against a March-2026 service date. I have not verified Apple's actual service date, and I flag that as a check item rather than a conclusion. Note also that Apple's documented 2026 practice is to file Sotera / Sotera-plus stipulations in its IPRs (e.g., IPR2026-00240, IPR2026-00241, IPR2026-00253); expect one here, which would bar Apple — but not Mobility's other defendants — from re-running the same § 102/§ 103 art in the district court.
• Defensive value: Low today; potentially high by 2028. There is nothing here to cite in a motion or a damages negotiation yet. What it buys an unrelated defendant is a free option: Apple is funding the validity attack, the petition and any institution decision become public prior art/claim-construction data, and a claim canceled in an FWD would be canceled as to the world (not just Apple). What it does not buy is estoppel protection — § 315(e)(2) runs against the petitioner and its privies, not against the rest of the industry.

Strategic summary

Claim status: 100% UNTESTED. The requested three-way split resolves to a single category. Canceled: none. Sustained: none. Untested: all claims of the '412. I cannot even identify the independent claim numbers (a gap already recorded in the earlier summary — the supplied full text contains no claims section, and the granted claims must be pulled from Patent Center or the Google Patents PDF). The practical consequence is straightforward: a defendant today has no PTAB-derived argument that any claim of the '412 is invalid, and equally no argument that any claim has been validated. Both directions cut. A demand letter citing the '412 is citing unadjudicated claims; a defendant who assumes the patent is weak because it is old and from a 2002 family should note that it survived prosecution and has never been tested at the Board.

Estoppel landscape. Only Apple and its RPIs/privies are exposed to § 315(e) estoppel, and only if the Board institutes and reaches a final written decision. If, and only if, that happens, Apple will be barred in the W.D. Tex. action from asserting any ground it raised or reasonably could have raised under §§ 102/103 on patents and printed publications — a bar that sweeps in art Apple knew of but chose not to put in the petition. For any other defendant (including Samsung in 2:26-cv-00157, absent a privity/RPI finding), the full § 282 universe remains available: §§ 101, 102, 103, 112, on-sale/prior-use system art, and public-use art. That is a genuine asymmetry worth planning around — system-art and § 112 theories that Apple cannot assert at the PTAB are precisely the theories available in the district court. Conversely, if you are coordinating with Apple, note that a broad Sotera stipulation plus a § 315(e)(2) bar can strip Apple of its strongest district court defenses for the benefit of a co-defendant who never filed.

Pattern signals. (1) Single petition, not a barrage — Apple has filed one IPR, not the multi-petition parallel attacks it has used elsewhere in 2026 (e.g., the Headwater, WeCrevention, and IngenioSpec matters). One petition against a seven-patent family being asserted in two districts suggests either a prioritization decision or a targeted lead patent. (2) No defensive aggregator in the chain. There is no Unified Patents or RPX filing here; the petitioner is the accused infringer itself, which means the usual aggregator benefits (broad, privately-funded, non-infringement-motivated challenge) are absent. (3) Patent owner has not yet engaged — Mobility IP Holdings acquired the '412 from Neology on 2026-02-11 and asserted it within weeks, in two districts, before Judge Albright and Judge Gilstrap. Expect an aggressive POPR, and expect it to raise § 314(a)/Magnolia discretionary arguments and possibly a § 325(d) argument that the art is cumulative of what the examiner considered over the patent's long prosecution.

Two open questions that change the strategic picture and that I could not resolve. First, term/expiration. The record shows an anticipated expiration of 2023-07-09 with a lapsed status, which is internally consistent with 2002-priority-plus-PTA arithmetic — but it is in tension with a 2026 assertion campaign and a February-2026 acquisitions of the asset, since a rational acquirer does not pay for an expired patent. Either the Google/Unified expiration data is an automated artifact (most plausibly a mis-attributed earliest-priority date, in which case the term may run from the 2015-03-04 filing to ~2035), or the patent genuinely lapsed and Mobility is litigating a closed damages window under the § 286 six-year lookback. This is the single highest-value fact to verify, and it is verifiable in minutes from the Patent Center maintenance-fee and term-adjustment record. Note that an expired patent can still be the subject of an IPR, but expiry removes any amendment path for the patent owner and collapses the remedy to past damages — which changes both the stay calculus and Apple's incentive to settle. Second, no institution decision exists to reason from, so any Fintiv/Magnolia outcome here is speculation. I am not offering one.

Recommended next steps

• Do not cite a PTAB disposition for the '412 — there is none. If you are drafting a defense and need a dispositive PTAB document to point at, you do not have one. Cite the pending petition instead: IPR2026-00501, Apple Inc. et al., filed 2026-09-19, status Pending, docketed at the Board's P-TACTS/PTAB Center (https://ptacts.uspto.gov/ptacts/), with the institution decision statutorily due ~2027-03-19 (§ 314(b)) and any FWD due ~2028-03-19 (§ 316(a)(11)).
• Calendar the milestones now: POPR due ~2026-12-19 (37 C.F.R. § 42.107(b)); institution decision ~2027-03-19; if instituted, oral hearing typically set in the ~2027-Q4/2028-Q1 window on the Board's scheduling order, FWD ~2028-03-19. Whether a W.D. Tex. trial date (Judge Albright) or an E.D. Tex. trial date (Judge Gilstrap, 2:26-cv-00157) lands first is the Fintiv/Magnolia variable that will decide institution — get both scheduling orders on the same page.
• Pull the petition the day it is indexed, and read three things only: the challenged claims, the grounds table, and the real party in interest list. The third determines who is estopped and who is not.
• Verify the term. Retrieve the maintenance-fee history, any terminal disclaimer, and PTA from USPTO Patent Center for application 14/639,034, and confirm the actual expiration against the Google Patents entry (https://patents.google.com/patent/[US10706412B2](/patent/US10706412B2)/en) and the Unified Patents docket (https://portal.unifiedpatents.com/litigation/Texas%20Western%20District%20Court/case/7%3A26-cv-00075). If the patent is in fact expired, the litigation is a past-damages-only case and the entire IPR calculus shifts toward early settlement rather than trial.
• Get the claims. The earlier-generated summary correctly flagged that the granted claims are absent from the supplied text; both the independent-claim count and the claim text must come from Patent Center or the Google Patents PDF before any infringement or validity opinion is written. This remains an unresolved hard gap.
• No Federal Circuit step is available. With no FWD, there is nothing appealable; corroborate by checking the Federal Circuit docket and CourtListener (https://www.courtlistener.com) rather than accepting my search results as dispositive.

Explicit non-findings. No panel names, no grounds, no challenged-claim list, no institution decision, no FWD, no settlement, no termination, no CAFC appeal — none exists for IPR2026-00501 as of 2026-09-19, and I have not inferred any of them from the patent's litigation posture. The only proceeding number I will report is the one in the structured feed. One date reconciliation to note: the earlier-generated summary flagged that its originating prompt specified an "as-of" of 2026-04-26 while session material post-dates that; this analysis is written as of 2026-09-19, which is also the petition's filing date and the feed's last-modified date.

I'll search for the assignment records and ownership chain for this patent.

Let me dig deeper into the specific ownership records.

Analysis note before the record

I was able to reconstruct the ownership chain logic for US 10,706,412 from Google Patents legal events (supplied as authoritative in the prior section) plus corroborating primary documents, but I was not able to retrieve the reel/frame numbers or the correspondent-of-record names for the two post-issuance events. My searches against the USPTO Assignment Center index and the PTAB document corpus returned sibling-patent records (e.g., the '436 patent's 2003 assignment history) but not the '412's own record. I am flagging this as a hard gap rather than filling it in, per the constraints. Where I have a reel/frame for a different patent in the family, I say so explicitly and do not attribute it to the '412.

Inventors

Unusual-pattern check: Not present. Neither inventor shows a departure around filing. Martinez de Velasco is still publicly identified as Neology's Founder as late as a 2022-02-22 Neology press release (GlobeNewswire). There is no 12-month post-filing inventor-exodus signal that would precede a fire-sale, and Neology/Smartrac retained the portfolio for a decade after the 2015 filing.

Original assignee

Neology, Inc., a Delaware corporation, HQ San Diego, CA (with Mexico City offices; also given as Poway, CA in some filings).

• Line of business / product: Yes — a genuine operating company. Neology is a supplier of secure RFID transponders and integrated solutions for electronic vehicle registration (EVR), electronic toll collection (tolling), and public safety. It was the exclusive provider of the Mexico REPUVE program (national vehicle-tagging). It shipped product and held 100+ granted patents (Neology "100th granted patent" milestone). The '412's transaction/NFC subject matter is adjacent to, not squarely within, Neology's tolling/EVR core — relevant later.
• Status: Operating, and acquired. Smartrac N.V. signed to acquire Neology on 2011-12-08 for $30.0M cash/debt-free plus up to $9.0M earn-out; founding shareholder Tenedora de Empresas S.A. de C.V. retained 15% and rolled earn-outs forward. By the 2015 PTAB record, Smartrac owned 100% of Neology, Inc. (Exhibit 2005, Uhl Declaration, filed May 1, 2015: "Smartrac owns one hundred percent (100%) of the shares of Neology, Inc."). Neology continued operating under Smartrac, and still operates today (www.neology.net, 2022 press activity).
• Bankruptcy: No Neology/Smartrac bankruptcy identified in my searches.

Assignment timeline

Chain anchors from Google Patents legal events (authoritative supplied text), with the data gaps called out:

• Executed date not shown / recorded 2015-07-30 — Reel NNNNNN/NNNN — NOT RETRIEVED

  • Conveyance: ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS) — i.e., employment/inventor-to-company assignment
  • Assignor: Manfred Rietzler; Francisco Martinez de Velasco Cortina
  • Assignee: Neology, Inc.
  • Correspondent: not retrieved
  • Context: Original inventor assignment — the founders/inventors conveying rights to the operating company Neology, consistent with the family's early practice (the sibling '436 patent shows the same two inventors assigning to Neology with an assignment logged 09/25/2003 on the 2002 parent family).

• Executed date not shown / recorded 2026-02-11 — Reel NNNNNN/NNNN — NOT RETRIEVED

  • Conveyance: ASSIGNMENT OF ASSIGNOR'S INTEREST
  • Assignor: Neology, Inc.
  • Assignee: Mobility IP Holdings, Inc.
  • Correspondent: not retrieved
  • Context: Transfer-to-asserter / divestiture — the operating company Neology (Smartrac-side) conveying the patent to a non-operating holding entity that filed suit against Apple (W.D. Tex. 7:26-cv-00075, filed 2026-03-04) and Samsung (E.D. Tex. 2:26-cv-00157, filed 2026-02-27) within ~3 weeks of the recording.

If the Assignment Center returns no additional records beyond these two, that is itself informative: only the original inventor-to-company link and the single transfer-to-asserter link exist — no security agreements, no re-recorded chain-of-title.

⚠️ Contradiction to flag explicitly: The prior section records Google's anticipated expiration 2023-07-09 and legal status "Expired – Fee Related." If correct, the 2026-02-11 assignment and the 2026 suits occurred after the patent had already lapsed, which is difficult to reconcile with a live assertion campaign. Either (a) the 2023-07-09 date is an aggregator artifact (e.g., unadjusted 20-year-from-2002 math ignoring PTA/terminal disclaimer), or (b) the '412 was asserted for past damages only under the 35 U.S.C. § 286 six-year lookback. I did not verify the PTA/terminal-disclaimer record, so I am not resolving this. See the Uncertainty register in the prior section.

Timeline diagram

timeline
    title Ownership of US 10706412
    2002 : Priority application filed
    2011 : Smartrac acquires Neology
    2015 : 412 application filed
         : Inventors assign to Neology Inc
    2020 : Patent issued
    2026 : Neology assigns to Mobility IP Holdings
         : Suits filed vs Apple and Samsung

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT (moderate strength).
Patent moved from a genuine operating assignee (Neology, Inc., a Smartrac subsidiary shipping RFID/EVR product) to Mobility IP Holdings, Inc., a licensing/holding-style entity, via a single ASSIGNMENT OF ASSIGNOR'S INTEREST recorded 2026-02-11. The "Holdings" suffix is a naming tell only; the concrete evidence is (a) the operating company is the assignor, and (b) the assignee's first observable act is filing two infringement suits within ~3 weeks (2026-02-27 Samsung; 2026-03-04 Apple) with no product offering identified. I could not retrieve the assignee's address or LLC/incorporation details, so the "registered-agent address" and "single-purpose LLC" sub-tells are unclear.

2. Known asserter in the chain — UNCLEAR.
Mobility IP Holdings, Inc. does not appear on the named public NPE lists supplied in the brief (Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation, Spangenberg entities), and RPX/Unified directories were not successfully queried for it. Conversely, it is also absent from the operating-company universe. The high-frequency-plaintiff signal is not established from my searches — do not treat the mere filing of two suits as a directory match.

3. Repeat correspondent across the chain — UNCLEAR (data gap).
This is the single most diagnostic signal and I could not capture it. No correspondent-of-record name was retrievable for either the 2015-07-30 reel or the 2026-02-11 reel. What I can say from adjacent records: the '412 litigation is being run by the McManis Faulk firm (Jason McManis, Warren McCarty III, Thomas DelRosario, Andrea Fair, Zachary Grinovich, plus Howard Wisnia, Justin Kenney, Weining Bai — E.D. Tex. appearance notices 2026-03-02 to 2026-04-02). That is litigation counsel, not the assignment-recording correspondent, and I am explicitly not equating them. A recurring same-attorney-on-every-reel pattern would move this from "unclear" to "present," and you need the Assignment Center correspondent field to make that call.

4. Cascading transfers — NOT PRESENT.
Only two recorded links over ~11 years (2015 inventor→Neology; 2026 Neology→Mobility IP Holdings). No chained LLC hops in <24 months, no shared-correspondent clustering. This is a clean two-hop chain, which cuts against a classic troll rotation.

5. Pre-litigation transfer — PRESENT (strong).
The assignment to Mobility IP Holdings was recorded 2026-02-11, and the first infringement suits naming this patent followed within ~16 days (Samsung, 2026-02-27) and ~21 days (Apple, 2026-03-04). This is squarely within the 6-month pre-suit window and indicates the transfer was arranged to create a clean standing/ownership record for assertion and to place the plaintiff as the named owner in E.D. Tex. / W.D. Tex.

6. Bankruptcy fire-sale — NOT PRESENT.
No Chapter 7/11 for Neology, Smartrac, or Tenedora was identified. The 2011 Smartrac deal was an M&A acquisition, not a distressed sale.

7. Privateering — UNCLEAR.
Structurally this looks like possible privateering: an operating RFID parent (Neology/Smartrac lineage) transferring an assert-worthy patent to a holding entity that sues Apple and Samsung — i.e., competitors/adjacent giants in the mobile-payment space, not Neology's tolling rivals. But privateering is a finding only when SEC filings or journalist coverage ties the transferor to the assertion. I found no SEC 10-K/8-K or Patent Progress/EFF coverage tying Smartrac or Neology to Mobility IP Holdings' suits. Do not call it privateering on structure alone.

8. Defensive aggregator — NOT PRESENT.
The chain terminates at Mobility IP Holdings, Inc., an asserting entity, not at RPX, AST, LOT, or OIN. Inverse signal confirmed: the patent is not neutralized.

Verdict

NPE — moderate confidence (1 strong signal + 1 moderate).

The chain is short and clean — one inventor-to-Neology assignment (recorded 2015-07-30) and one transfer from operating company Neology, Inc. to Mobility IP Holdings, Inc. (recorded 2026-02-11) — so the classic multi-LLC-cascade, bankruptcy, and defensive-aggregator signals are all absent. What drives the call is the pre-litigation transfer (strong, present): a licensing-style "Holdings" entity acquired the patent and filed against Apple (2026-03-04) and Samsung (2026-02-27) within roughly three weeks, with no product in commerce identified, plus the shell-entity transfer pattern (operating assignor → holding assignee) at moderate strength. I stop short of high confidence because the two most diagnostic fields — reel/frame numbers and the correspondent-of-record — were not retrievable, and Mobility IP Holdings was not confirmed against any RPX/Unified high-frequency-plaintiff directory. Verify the reel/frame and correspondent at the USPTO Assignment Center search page: https://assignmentcenter.uspto.gov/ (indexed mirror: https://assignment.uspto.gov/patent/index.html), searching by US 10,706,412. If two consecutive reels share the same correspondent attorney or firm, upgrade to high confidence; if a single 2026 reel names one attorney also appearing across the sibling patents (10,726,414; 10,762,187; 10,235,513; 8,933,807, etc.), that corroborates a coordinated assertion program.

I'll search for the prior-art citations on US 10,706,412 and verify the specific patent record.

Let me try to retrieve the actual "References Cited" section from the patent documents.

I've hit the tool-step limit, so let me give you a direct, honest answer based on what I actually retrieved — including a significant gap I can't paper over.

Prior art analysis — US 10,706,412 B2

⚠️ Bottom line up front: I could not retrieve the patent's own "References Cited" list

You asked me to "look at each patent citation for 10706412" and map each to the claim(s) it potentially anticipates under § 102. I was not able to retrieve the front-face "References Cited" (Box 56) list for US 10,706,412 B2. My searches for the (56) References Cited section of this specific patent returned:

• the Google Patents landing page itself (which renders only the abstract, description, classification, and OCR-fragmentary claim-reference snippets — no visible Box 56 list);
• the (56) sections of unrelated patents (US 11,389,291; US 11,008,518; US 11,847,378; US 11,445,261; US 8,764,326; USD510,866); and
• the search reports of different applications (PCT/US2015/027997 → WO2015168126A1; PCT/US2014/058599; PCT/KR2013/009354; PCT/US2018/059278; PCT/US2016/039715).

Do not treat any of those lists as the '412's references. The WO2015168126A1 list (Richards US 2009/0151925; Badalamenti US 2011/0094742; Halley US 2003/0111224; Patel US 2009/0008078; Bode US 6,371,210) is from a different PCT application and — note the Bode reference is a Halliburton downhole-tool patent — belongs to an oilfield family, not to Neology's RFID family. Attributing it to the '412 would be a fabrication.

This is a hard gap, not a hedge, and it compounds the gap flagged in the earlier section: I have neither the claim text nor the reference list. A faithful per-reference § 102 mapping is therefore impossible for me to produce right now without inventing either side of the equation.

1. What I did verify about the record's citation structure

The two incorporated-by-reference documents are the strongest, most defensible citations I can hand you, and they are exactly what a § 102/§ 103 challenge would start from:

Both predate the claimed 2002-07-09 priority date by ~3 months. That timing is what makes them useful, and it is also why the priority date is the pivot for this entire analysis.

2. The issue that dominates every § 102 question here: which date is the critical date?

This is more important than any individual reference, and I want to flag it explicitly because it changes the answer to your question by roughly a decade.

The '412 is a late-filed continuation — application 14/639,034, filed 2015-03-04, claiming priority to 2002-07-09 (Google Patents labels that priority "an assumption … not a legal conclusion"). If the claims are fully supported by the 2002 disclosure, § 102 art must predate 2002-07-09 — a very small universe of RFID/NFC/biometric-payment art. That would make most of the modern NFC-mobile-payment prior art (2005–2015) categorically ineligible for § 102 against these claims.

But there is a real, documented basis to question full priority entitlement, and it is the single most consequential fact in this record:

• Google Patents' own auto-extracted "prior art keywords" for the '412 include "payment information."
• The earlier-generated section established — and I independently confirm — that "payment information" does not appear anywhere in the 2002-era specification text supplied, while the specification speaks instead of "identification information," "authentication information," and "transaction information (e.g., credit card information …)."
• If the granted claims recite "payment information" (and per the earlier section's §3 inference they likely do, consistent with the CPC cluster G06Q20/3278 — "RFID or NFC payments by means of M-devices" — and G06Q20/3674 "involving authentication"), a challenger can argue those limitations are new matter unsupported by the 2002 disclosure. The claims would then be entitled only to the 2015-03-04 filing date.

Consequence: under the 2015 date, the prior-art universe explodes to include the entire 2002–2015 NFC/secure-element/tokenization/biometric-payment corpus — which is precisely the art the 2026 defendants (Apple Pay, Samsung Pay) would marshal. This is not a hypothetical: it is the obvious first battleground in both suits.

I am not making a legal conclusion that the priority claim fails — that requires the actual 2002 specification and the granted claims side by side, which I do not have (the supplied full text stops at "Secure Data Base" and contains no claims).

3. What I therefore can and cannot give you

Identity confirmation (so you know I searched the right number):

4. Contradictions / flags against the earlier-generated sections

Per your instruction to flag contradictions rather than silently reconcile them:

1. Legal status string changed again. The authoritative text in your prompt says "Expired - Fee Related." The Google Patents page as retrieved this session returns "Expired - Lifetime." The earlier-generated litigation section said the prior section recorded "Expired – Fee Related" while the current session showed "Expired - Lifetime." So across three reads we now have two different status labels from the same source. I cannot determine which is current. Materially, "Expired – Lifetime" would indicate the patent ran its full term (consistent with the ~2023-07-09 anticipated expiration), while "Expired – Fee Related" indicates a lapse for non-payment. Do not rely on either without checking the USPTO Patent Center fee history.
2. My § 102 analysis is date-bounded, not reference-bounded. Because I lack the Box 56 list, I have answered the structure of your question (critical date, candidate references, analytical framework) rather than a per-reference chart. If your workflow requires a reference-by-reference chart, this section does not satisfy it and should not be represented as doing so.
3. Date note (carried forward). Your prompt states today is April 26, 2026; my session date is 2026-09-19 and retrieved material post-dates April 2026. I have not truncated.

5. Confidence register

6. What you should pull to complete this task properly

1. USPTO Patent Center / Patent Public Search, application 14/639,034 → the "References Cited" / "Patent References Cited" tab. This is the authoritative Box 56 list and is the only correct source for your per-reference requirement.
2. The granted claims (also from Patent Center, or the front matter of the Google Patents PDF). Without claim numbers and text, no § 102 mapping is possible.
3. The file wrapper → to see whether the examiner made any § 102/§ 103 rejections and which references drove allowance; this tells you which references the USPTO itself considered material.
4. The 2002 priority application (Ser. No. 10/118,092 lineage) → to test whether "payment information" limitations are supported, which determines whether the critical date is 2002-07-09 or 2015-03-04.

If you can supply the Box 56 list and the granted claims, I can produce the full citation-by-citation § 102 chart you asked for. Until then, the honest answer is: two attributable references, one dominant critical-date question, and a missing reference list.

Obviousness (35 U.S.C. § 103) — US 10,706,412 B2

Note on scope of this memo. Interpreting "the Prior Art section of this page" as the Google Patents prior-art metadata block for this record — prior-art date 2002-07-09 and prior-art keywords: information, memory, biometric, payment information, security — plus the prior-art documents the specification itself references (U.S. App. Ser. No. 10/118,092, filed 2002-04-09, and PCT/IB02/01439, filed 2002-04-30). The supplied page text contains no "References Cited" table and no "Cited By" table, and, as the two previously generated sections correctly flagged, no claims. If your platform holds a separate prior-art panel with those tables, that content did not reach me; say so and I will fold it in.

Because § 103 is applied claim by claim under Graham v. John Deere Co., 383 U.S. 1 (1966), this is a framework-plus-candidate-combinations memo, not an invalidity opinion. I flag every inference as an inference.

1. Threshold issues that dominate everything else

1.1 There are no claims to analyze — restated, not softened

I have no verbatim claim text for the '412, no independent-claim count, and no claim differentiation data. The prior two sections reached the same conclusion. Any element-by-element mapping below is to an inferred claim profile, not to claim language, and the legal conclusion ("would the claims have been obvious") cannot be closed out without the claims.

1.2 Two critical dates, and the one provision that decides which

The application was filed 2015-03-04 and claims a 2002-07-09 priority date. That gap is the single most consequential § 103 fact about this patent, because it produces two entirely different prior-art universes:

Which applies: the AIA's § 102/§ 103 apply to any application that contains or contained at any time a claim with an effective filing date on or after 2013-03-16 (AIA § 3(n)(1)). A 2015-filed continuation therefore applies AIA § 102/§ 103 unless every claim, at every point in prosecution, was entitled to the 2002 date. Practically, the difference is the common-ownership carve-outs (pre-AIA § 103(c) vs. AIA § 102(b)(2)(C)) and the Hilmer doctrine — both of which matter here because the most dangerous art against this patent is Neology's own earlier family publications (see § 5.4).

Google's own metadata encodes the assumption: its "Prior art date" is 2002-07-09, i.e., it is treating the 2002 priority as valid. Google's priority date is expressly "an assumption … not a legal conclusion." So the entire prior-art taxonomy Google presents is contingent, and the assumption is exactly what a defendant would attack.

1.3 A structural inference about the priority chain (labeled as inference)

Google reports anticipated expiration 2023-07-09 — precisely 21 years after the 2002-07-09 date, and 20 years after 2003-07-09. The cleanest explanation is a provisional filed 2002-07-09 and a non-provisional filed one year later, 2003-07-09, with no material PTA. The supplied text does not itemize the chain, so treat this as an inference with real consequences: if the chain includes any continuation-in-part, or if any current claim is not supported by the 2002/2003 disclosure, that claim drops to 2015 and Scenario B opens.

1.4 The "payment information" anomaly — the best available claim-scope proxy

Google's keyword for this record includes "payment information." That exact phrase does not appear anywhere in the description text supplied. The description says "credit card information," "transaction information," and "hotel mileage card information." A term that appears only in the keyword list and not in the specification is most plausibly drawn from the claims. Treat this as the strongest available signal that the claims recite payment information as a stored/gated data element — not as a specification term.

1.5 Date contradiction carried forward

The prompt says today is April 26, 2026; my session date is 2026-09-19, and the source page was fetched 2026-09-19. Both prior sections flagged this and I do not resolve it. Also, the authoritative page text now reads "Expired - Fee Related" — matching the first prior section, not the second (which recorded "Expired - Lifetime"). The supplied authoritative text controls: "Expired - Fee Related."

Remedy consequence (material to any § 103 fight): if expiration really is 2023-07-09, the '412 was already expired when the 2026 suits were filed. Any recovery is past damages only, and the § 286 six-year lookback running from the 2026-02-27/03-04 filings reaches back only to ~2020-02 — inside the patent's remaining life. So the practical damages window is roughly February 2020 → July 2023, and no injunction is available. That is precisely the window in which Apple Pay and Samsung Pay were already mature, which is why a priority-date challenge is worth so much more to a defendant here than a typical § 103 defense.

2. Level of ordinary skill in the art (Graham factor 3)

Scenario A (critical date ~2001-07-09). PHOSITA = B.S. in electrical engineering or computer science, or equivalent, plus 2–4 years in contactless smart-card/RFID design (or an M.S. with 1–2 years). Presumed knowledge: ISO/IEC 14443 and 15693 signaling and anti-collision; ISO/IEC 7816-4 APDU command structure; the EMV ICC Specifications (static and dynamic data authentication, application cryptograms); hardware cryptographic coprocessors and key storage in EEPROM; challenge-response authentication; capacitive/optical fingerprint sensor ICs and on-card template matching; and the fact that battery-less chips are powered by the reader field and gate access to memory on authentication. The '412's own specification reads as a PHOSITA-level recitation of exactly this skill set ("A modulator is a device that receives baseband signals from an antenna… A cryptographic block is a device that stores the security keys. These keys are checked and validated to grant or deny access to the memory chip").

Scenario B (critical date 2014-03-04). Add: NFC (ISO/IEC 18092, NFC Forum), GlobalPlatform secure-element management and Java Card applet firewalls, ARM TrustZone/TEE, host card emulation, EMVCo payment tokenization, mobile OS biometric APIs (Touch ID, Android fingerprint), and secure-element-to-POS contactless transaction flows.

A well-drafted petition would offer both definitions, because the narrower Scenario-A PHOSITA is the one a patentee will argue for.

3. Inferred claim profile (proxy elements) — explicitly not claim language

Derived from the assigned CPC codes (G06Q20/3278 RFID or NFC payments by means of M-devices; G06Q20/327; G06Q20/34; G06Q20/3674 …involving authentication; H04L63/0861 …using biometrical features; G07C9/25/G07C9/257; G06K19/0723; H04W4/80; H04W12/06) and the keyword set.

Caveat that a court would care about: E3's "memory totally separated" and E4's "one-to-one" are specification phrases. Whether they are claim limitations — and whether E6 imports "NFC" specifically or any short-range RF — cannot be established from what I have.

4. Prior art actually available from this record, organized by element

Tiered by how firmly I can ground each item in this session.

Tier 1 — dated standards and deployed commercial systems (high confidence as to existence and period):

• ISO/IEC 14443 parts 1–4 (proximity cards, 1999–2001): RF power and data transfer, anti-collision, and the SELECT/APDU exchange — supplies E1, E5, E6 signal layer and the battery-less, field-powered model.
• ISO/IEC 15693 (vicinity cards, 2000) and ISO/IEC 10536 (close-coupled, mid-1990s).
• ISO/IEC 7816-4 (interindustry APDU commands): supplies the reader→card command/response structure the 2026 Samsung complaint itself invokes.
• EMV '96 and EMV 2000 Integrated Circuit Card Specifications: on-card keys, application cryptograms, dynamic authentication — supplies E3.
• GSM/ETSI SIM and SIM Application Toolkit (GSM 11.11 / 11.14, mid-to-late 1990s): a secure element already resident in every cellular telephone, with the A3/A8 authentication algorithm and Ki key, gating access to applets and data. The '412 specification admits this explicitly: "the GSM telephones identify the telephones to a network carrier to avoid, for example, telephone cloning." That admission supplies E1 + E3 in the phone context and is applicant-admitted prior art.
• WAP Forum "WAP Identity Module" (WIM) specifications (1999–2000): a tamper-resistant hardware module in the handset that holds keys/certificates and requires a PIN to release an identity for a transaction — a direct E3 + one-factor-authentication reference.
• Stored-value / e-purse systems: Mondex and Proton (1990s); Sony FeliCa with Octopus (Hong Kong, 1997) and Suica (Japan, 2001) — deployed contactless payment in public hands well before the critical date, supplying E2 + E5 + E6.

Tier 2 — well-populated reference categories I can identify as classes but whose specific patent numbers I will not assert without verification: self-authenticating smart cards carrying an on-card fingerprint template and on-card matcher; biometric-access payment terminals; "mobile wallet" / wireless-payment architectures predating 2002 (several filed by card networks, handset makers, and carriers); and on-card template-storage-with-external-sensor arrangements. I flag these as categories rather than invent numbers. If you need specific documentary art, the productive CPC neighborhoods are G06K19/07x × G07C9/25x × G06Q20/34x filtered to pre-2002 priority — that is where the Tier-2 class lives.

Tier 3 — the record's own prior-art documents and the specification's admissions:

• U.S. App. Ser. No. 10/118,092 (2002-04-09) and PCT/IB02/01439 (2002-04-30), both incorporated by reference. Note the § 103(c)/(AIA § 102(b)(2)(C)) common-ownership complication: these are Neology's own filings, so if they qualify only as § 102(e)/(a)(2) art, the common-ownership exception may disqualify them for § 103 purposes. They are not a clean § 103 base.
• Applicant-admitted prior art throughout the specification (the component glossary repeated in every figure; "There are a variety of ways to make identification cards that are resistant to fraud and counterfeiting"; the security-feature catalog; the taggant/hologram discussions). KSR permits this art to be used, and it does a lot of work against E3.

What this record does not contain: no third-party patent citations, no examiner-cited art, no "Cited By" listing. That absence materially limits how concrete I can make the § 103 mapping.

5. Candidate combinations and the motivation to combine

5.1 Combination I — Contactless payment card + separate biometric verification at the terminal (Scenario A)

References: ISO/IEC 14443 + EMV 2000 (card side: E2, E3, E6) + a deployed contactless payment system such as FeliCa/Octopus or Mondex (E5, E6) + a biometric smart-card or biometric-terminal reference from the Tier-2 class (E4), with the terminal's RF reader/writer (E5).

Mapping: E2 (payment data in card memory), E3 (on-card keys and cryptograms), E5/E6 (proximity RF to a networked POS) all present. The gap is E4 — 1:1 fingerprint verification as a condition to releasing payment data — and E1 (the handset form factor).

Motivation (why a PHOSITA would combine): the specification itself supplies the market rationale. It states that ordinary photo IDs are trivially reproduced "by the ready accessibility of dye sublimation card printers, color scanners, color printers, heat sources (e.g., hair dryers), and color photo copiers," and it frames the entire invention as responding to "credit card fraud, through donation, substitution of identity, or counterfeit documents." Once the problem is framed as binding a portable payment credential to the person carrying it, adding a biometric comparison to a contactless payment card is arguably the most predictable move available: the payment credential is the thing being lost, stolen, or cloned, and the biometric is the one factor that travels with the holder.

KSR/MPEP 2143 rationales: (a) combining known elements each performing its known function (RF payment + biometric verification) to yield a predictable result; (d) applying the known technique of biometric access control to a known device (a payment card/terminal) that is ready for improvement; (f) known work in physical access control combined with payment, prompted by the identity-fraud market forces the specification itself describes.

Reasonable expectation of success: high, in that both halves were separately deployed and the combination is a structural addition, not a predictive-science leap. But note the counter: on-card biometric matching with a template stored in the same EEPROM from which the payment keys are read was, in 2001, not obviously trivial — power budget on a field-powered card, sensor integration, and template-integrity are real engineering problems, and a patentee could argue a PHOSITA would not have had a reasonable expectation of success. This is the strongest available nonobviousness argument if the priority date holds.

5.2 Combination II — The telephone as the payment instrument (Scenario A)

References: GSM SIM and SIM Toolkit (secure element + authentication already in the phone: E1, E3) + WAP WIM (hardware key storage gating a transaction identity, PIN-gated: E3) + a short-range wireless link (Bluetooth, IR, or ISO/IEC 14443: E5, E6) + an e-purse or payment-network backend (E2).

Motivation: eliminating the physical card is the stated advantage of the whole "RF communications device" embodiment — "reduces or eliminates the need to carry credit cards, documents (e.g., a boarding pass) or other information; reduces or eliminates the need for an entity to issue or update credit cards… with associated fees (e.g., mailing costs)." Those are textbook design incentives and market forces under MPEP 2143(f). Further, the phone already had the two hardest pieces — a tamper-resistant secure element and a user-authenticated key release — so putting payment credentials there is a simple substitution of one known element for another (2143(b)).

Weak point: E4 remains unaddressed in a phone-based combination unless a Tier-2 biometric reference is added; and pre-2002 short-range payment from a handset was fragmentary (Bluetooth payment pilots existed but were not deployed at scale), which a patentee would use to argue the combination was not "obvious to try" but speculative. Counter-counter: 2143(e) (finite number of identified, predictable solutions) applies once the handset is chosen, and the 2002 specification's own assertions about GSM cloning show a PHOSITA recognized the phone's secure element as a payment-capable platform.

5.3 Combination III — Biometric smart card + contactless payment card + handset (Scenario A, full element coverage)

References: Combination I + Combination II + any Tier-2 self-authenticating-card reference teaching the 1:1 on-card match (E4), optionally with the '412's admitted 1:1-verification rationale.

Mapping: covers E1–E6 on paper.

Motivation: three independent, mutually reinforcing reasons, all of which a PHOSITA would have had: (i) the payment credential must be bound to its holder; (ii) the credential must not be usable if separated from the holder; (iii) templates should not leave the device (the specification's own "e-prompt… totally separated from the rest of the communication, so if the security requirements are not met, a user cannot access the memory of the chip"). That third point is privacy/security engineering, and it is also the whole point of the on-card matcher — so the motivation is overdetermined.

This is the combination I would expect a defendant to lead with under Scenario A, because it needs no post-2002 art at all. Its vulnerability is the same as Combination I's: the "reasonable expectation of success" story for an integrated, field-powered, fingerprint-gated payment card in 2001 is contestable, and the '412's ~13-year prosecution history suggests the examiner never had the right combination in front of them.

5.4 Combination IV — the 2011–2015 NFC payment stack (Scenario B; the high-leverage attack)

References (all post-2002, pre-2014-03-04): an NFC Forum / ISO 18092 contactless mobile-device reference (E1, E6); a secure-element wallet reference — GlobalPlatform secure-element management, or an Android 4.4 HCE (2013) reference — teaching a secure element storing a tokenized PAN and gating it behind device authentication (E2, E3); an EMVCo Payment Tokenisation Specification (March 2014) reference (E2, E6); and an on-device biometric release reference — Touch ID / Secure Enclave (2013), or the earlier Motorola Atrix fingerprint handset (2011), or a TrustZone/TEE-based fingerprint-matching disclosure (E4); all combined with a standard contactless POS accepting NFC (E5).

Why this combination is close to unanswerable if the priority date fails: every element is disclosed, the field of endeavor is identical to the claimed field (mobile payments), and the motivation is not merely inferable but documented — the NFC Forum and EMVCo existed to create precisely this interoperability, so the standards themselves are the suggestion to combine (MPEP 2143(g): "the nature of the problem to be solved"; and 2143(f): design incentives/market forces). The result is the mechanically predictable assembly of a standard-conformant payment flow.

Also note the 2026 pleadings are their own § 103 exhibit. The Samsung complaint is reported to accuse an embedded secure element, tokenized PAN, biometric authentication, NFC transmission to POS, and ISO 7816-4 reader interrogation. If the accused features are all things the patent alleges infringe, those features were, by 2013–2015, the industry-standard, jointly-specified way to build an NFC wallet. Standardization is powerful evidence that the combination was obvious to try with a reasonable expectation of success — the specifications themselves supply the expectation of success.

Explicit chain of reasoning (Samsung/Apple petition form):

1. Teach, show, or suggest E1 + E5 + E6 (NFC handset ↔ contactless POS).
2. Teach, show, or suggest E2 + E3 (secure element storing payment credentials, access gated by keys).
3. Teach, show, or suggest E4 (biometric match on the device releases the credential).
4. Claim the combination as a "secure transactional solution"; argue the elements are known, each performing its known function, in the known field, per published interoperable standards, with a documented market incentive (Apple Pay/Google Wallet era).
5. Rebut a nexus argument by showing the specification adds nothing beyond the assembly.

5.5 Combination V — Secondary reference for the memory-isolation limitation (both scenarios)

Reference: GlobalPlatform / Java Card (1999+) applet-firewall and secure-element API art, or any secure-processor reference teaching that a protected memory region is unreadable unless a key check succeeds.

Motivation: security best practice; 2143(c) (use of a known technique to improve a similar device in the same way). This is the least contestable element in the portfolio and should be treated as a secondary reference rather than the lead.

5.6 Combination VI — the self-collision attack (both scenarios)

If priority fails, the '412's own 2003–2007 family publications become § 102(b) art against the 2015 claims. The common-ownership exception does not rescue a patentee where the reference is the patentee's own earlier printed publication more than one year before the 2015 filing — § 103(c) removes only § 102(e)/(g) art from the § 103 reach, and a published parent is § 102(b) art against later claims that lack the parent's priority. This may be the single most efficient invalidity theory available, because the parent disclosure is by definition the patentee's own enabling description of the same subject matter.

6. § 102 cross-check (obviousness is not the only tool)

• If any single NFC-wallet reference discloses handset + secure element + on-device biometric release + NFC to a POS, that is § 102 anticipation, not § 103 — one reference, no motivation analysis, no KSR. A defendant should search for such a reference before building a combination.
• Same for a pre-2002 reference disclosing a field-powered card with on-card fingerprint template and on-card matcher that releases a stored-value balance.
• Also run § 102 against the incorporated 2002 applications and the family publications under the priority-failure theory.

7. Secondary considerations (Graham factor 4)

• Nexus is the kill switch. Any commercial-success evidence will be Apple's and Samsung's success with Apple Pay and Samsung Pay, not the patentee's. Without a licensing program tying the '412 to revenue, there is no presumption of nexus, and the accused products launched 12–13 years after the 2002 specification — with technology (HCE, tokenization, secure enclave, NFC Forum compliance) that the '412 does not describe. Expect an argument that the claims cover the idea of mobile biometric payment rather than the implementation that made it commercially successful, which defeats nexus.
• Long-felt but unmet need is genuinely available if the priority date holds — a security company described mobile biometric payments in 2002, and the market did not deliver them at scale until 2014. But the 13-year pendency between the 2002 disclosure and the 2015 filing cuts the other way: if the need were so clearly felt, why was the application not converted into allowed claims for a decade? A defendant will also argue the delay is the patentee's, not the industry's.
• Copying is unavailable as a positive factor and affirmatively harmful: the accused features postdate the 2002 disclosure but are standard-driven, and standardization supports obviousness.
• Expired patent — no injunction, no ongoing royalty; § 286 lookback only, into a window ending 2023-07 (see § 1.5).

8. Prosecution-history / robustness angles worth checking (not § 103, but decisive to it)

Given a ~13-year pendency (2002 priority → 2015 filing → 2020 grant), expect to check for prosecution laches, prosecution history estoppel, obviousness-type double patenting across the sibling family (the record links US 14/724,593 → US 10,726,414 B2), any terminal disclaimer (which would explain why the expiration tracks 2023-07-09 and would also create a per se ODP vulnerability if any claim is later held invalid), and the PTA calculation. A terminal disclaimer filed to overcome ODP over a sibling is itself a lever: a successful attack on the sibling's claims can cascade.

Fora: the Samsung and Apple district cases were served in early March 2026, which sets a roughly March 2027 § 315(b) IPR bar for the served defendants. Ex parte reexamination has no time bar and is available even on an expired patent; IPR on an expired patent is permitted (the Board can cancel claims but issues no amendment-based relief). Given that the patent is expired and the damages window has already closed at ~2023-07, a reexam/IPR posture is plausibly more valuable than litigation to the defendants — which is itself a reason to expect the § 103 attack to be run hard and early.

9. Uncertainty register

10. Bottom line

Without claim text, I can rank the theories but cannot conclude on any of them. Subject to that:

1. If the 2002 priority date holds, the best § 103 case is Combination III — pre-2002 contactless payment + SIM/WIM secure element in a handset + a biometric smart-card reference for the 1:1 match. All elements were known; the specification's own admissions about card fraud and GSM authentication supply the motivation. The patentee's strongest rebuttal is the 2001-era reasonable-expectation-of-success problem for an integrated field-powered fingerprint-gated payment card.
2. If the priority date fails (a claim lacking § 112 support in the 2002/2003 disclosure, any CIP, or the AIA transition provision), Combination IV — the 2013–2014 NFC/HCE/tokenization/secure-element/on-device-biometric stack — renders the claims obvious essentially as a matter of standard conformance, with the NFC Forum and EMVCo specifications supplying the motivation and the expectation of success. On top of that, Combination VI turns the patentee's own family publications into § 102(b) art.
3. Either way, the expired status confines the remedy to a closed, ~February 2020–July 2023 damages window under § 286, and no injunction is available.

To convert this from a framework into an opinion, I need four things, in this order of importance: (i) the granted claim text of US 10,706,412 (USPTO Patent Center or Patent Public Search, application 14/639,034, or the Google Patents PDF); (ii) the References Cited table from the face of the patent; (iii) the full priority chain, with each link's filing date and whether any link added new matter; and (iv) the PTA/terminal-disclaimer record explaining the 2023-07-09 date. I would rather hand you that list than a plausible-sounding element chart built on inferred claim language.