US 9542542

I'll search for this specific patent number in relevant sources.

US 9,542,542 B2 — Summary

Important scoping note first (read this): I was asked to search the USPTO database and CAFC 2026 dockets for patent number 9542542 only. I could not directly query USPTO PatentCenter, the PTAB docket, or the Federal Circuit docket from here; the results below come from Google Patents, RPX/Unified Patents, patentimages (the issued-PDF scan), and legal-press reporting. I found no CAFC 2026 docket involving US 9,542,542. The high-profile Proxense Federal Circuit appeal briefed in April 2026 (Proxense v. Apple/Microsoft) concerns US 8,886,954 — a different patent, confusingly also called "the '954 patent." Do not conflate them.

Bibliographic data (US 9,542,542 B2)

Abstract (verbatim)

"A system and method provide efficient, secure and highly reliable authentication for transaction processing and/or access control applications in which only biometric input is required from the user. A Personal Digital Key stores a biometric profile that comprises a representation of physical or behavioral characteristics that are uniquely associated with an individual that owns and carries the PDK. The PDK wirelessly transmits the biometric profile over a secure wireless transaction to a Reader for use in a biometric authentication process. The Reader compares the received biometric profile to a biometric input acquired at the point of transaction in order to determine if the transaction should be authorized."

Plain-language overview of the independent claims

⚠️ Caveat on claim text: the authoritative full text provided to me was truncated before the claims, and I was unable to pull the issued claim set directly from USPTO. The following claim language is as reported by RPX/Unified Patents for US 9,542,542 B2. The wording is very close to claims in the later continuations (US 11,551,222 / US 12,626,262), so treat the exact scope as unverified and confirm against the issued patent before relying on it.

Independent claim 1 — method for electronic transaction processing. A reader/processor:

1. Automatically and wirelessly receives a first biometric sample from a first Personal Digital Key (PDK) once that PDK enters a proximity zone. The PDK stores both the sample and a full biometric profile of its valid user; the sample is a subset of the profile.
2. Does the same for a second PDK in the zone (so multiple PDKs' samples are on hand).
3. Receives a biometric input from a user (e.g., a fingerprint).
4. Compares the biometric input against the collection of samples from all PDKs in the zone.
5. Determines the input matches the first PDK's sample, and on that basis associates that PDK with the transaction.
6. Only after that association, checks whether the input also matches the full biometric profile of that PDK's valid user.
7. Authenticates the PDK without user intervention after it entered the coverage area, and authorizes the transaction if both the authentication and the full-profile match succeed.

In plain terms: a quick "screening" compare against short biometric samples from every nearby key is used to pick which key belongs to the person standing there, and only then is the slower/stricter full-profile verification done — so the user's only action is supplying the biometric.

Independent claim 15 — system. One or more processors plus memory holding instructions that, when executed, carry out the same automatic proximity-based sample reception, sample-matching, PDK association, full-profile verification, no-user-intervention PDK authentication, and transaction authorization.

I was not able to confirm whether the patent contains additional independent claims (e.g., further method, device, or CRM claims) — the sources I reached listed only claims 1 and 15 as independents.

Supporting disclosure worth noting

• The "PDK" is a compact, uniquely identifiable wireless device storing content in tamper-proof memory; profiles are provisioned through a trusted "Notary" process (see FIGS. 1–3 and the related applications incorporated by reference).
• The spec draws a distinction between biometric profile samples (subsets, transmittable over lower-security/public channels) and full biometric profiles (secure channel only) — the mechanism the claim-1 screening step relies on.
• FIGS. 8–10 describe three "single step" embodiments: biometric-only (FIG. 8), biometric + PDK authentication (FIG. 9), and a four-actor flow among user, PDK, Reader, and third-party system (FIG. 10).
• Readers may use registries (validation database, Central Registry, private registries) for additional, non-biometric authentication; the registries are described as not storing biometric data.

Enforcement / PTAB / appellate landscape (as retrieved)

• Google Patents flags that this family has litigation (Darts-ip family 38694426), but I found no specific 2026 CAFC docket for 9542542.
• The 2024–2026 Proxense proceedings I did find are directed at other patents: IPR2024-00233 and IPR2024-00846 (US 8,886,954, Apple/Google/Microsoft), IPR2024-00232 (US 8,352,730), IPR2024-00234 (US 9,298,905), plus ex parte reexaminations 90/015,052 and 90/015,053. Proxense's April 3, 2026 Federal Circuit opening brief challenges PTAB stays of the parallel reexamination. None of these identify 9542542 as the challenged patent.
• Proxense's district-court campaigns (W.D. Tex.) asserting related patents include Samsung, Google, Microsoft, Apple, LG, and Intel (Google and Intel reportedly settled in 2025 and March 2026 respectively).

Explicit uncertainties

1. No CAFC 2026 docket for 9542542 was found. Absence of evidence here is not proof of absence — I could not query the Federal Circuit's docket system directly.
2. Exact issued claim text is unverified; the claim 1/15 wording above is from a third-party aggregator and may reflect a later continuation.
3. Small internal discrepancy in the sources: Google Patents lists the priority date as 2005‑11‑30, which is the filing date of parent application 11/292,330 (a CIP chain ultimately tracing to Dec. 27, 2000); the earliest provisional benefit claims in the specification are 2006. Priority characterization should be treated as an assumption, not a legal conclusion.
4. Date note: my operating date was given as April 26, 2026, but some retrieved items (e.g., a Sept. 21, 2026 PTAB report) post-date that; I have reported them as found rather than adjusting.

I'll search for litigation involving US Patent 9542542 across patent litigation databases.

Let me search more specifically for Proxense litigation and Unified Patents case lists.

The searches surface a critical distinction: the "954 patent" asserted in Proxense's cases appears to be U.S. 8,886,954, not 9,542,542. Let me verify precisely.

Bottom line

I could not identify any litigation asserting U.S. Patent No. 9,542,542 itself. Every "954 patent" hit in Proxense's litigation campaigns refers to a different patent — U.S. Patent No. 8,886,954 — which is easy to confuse with 9,542,542 but is not the same patent. Per your strict rule I am not treating them as the same. So: no known litigation involving US 9,542,542 was found in the sources I could reach.

1. Critical identification warning (why "954" hits are false positives)

Search results and litigation documents repeatedly say "the '954 Patent," but the text makes clear it is 8,886,954, not 9,542,542:

• Proxense's Preliminary Infringement Contentions in Proxense v. Apple state: "U.S. Patent No. 8,886,954 (the '954 Patent)" — asserted claims 1, 2, 5, 6, 7.
• Google's Opening Claim Construction Brief in Proxense v. Google, No. 6:23-cv-00320-ADA: "U.S. Patent No. 8,886,954 ('954 patent')."
• Microsoft's Opening Claim Construction Brief in Proxense v. Microsoft, No. 6:23-cv-00319-ADA, likewise pairs "'730 Patent … ; '954 Patent, claim 1" — the device-ID/biometric family (8,352,730 / 8,886,954 / 9,298,905 / 10,698,989).

U.S. 9,542,542 ("Single step transaction authentication using proximity and biometric input," App. No. 11/745,249, filed 2007-05-07, priority from 11/292,330) is a different patent in a different continuation line from 8,886,954. I found no complaint, infringement contention, judgment, or ITC filing identifying 9,542,542 by number.

Caveat / verification limits: I was unable to load the Unified Patents litigation case list or PACER directly during this session; the searches surfaced patent pages and secondary aggregators rather than the litigation-portal record for this specific number. If you need a belt-and-suspenders confirmation, the authoritative checks are: (a) Unified Patents litigation portal case list, (b) Docket Navigator / Lex Machina filtered on "9,542,542," (c) PACER full-text search, and (d) the Darts-ip family link on the Google Patents page for 9,542,542 (family=38694426), which flags that the family has litigation — not that this patent was asserted.

2. What the Google Patents family flag actually means for 9,542,542

The 9,542,542 Google Patents record carries a "Family has litigation" flag pointing to Darts-ip family 38694426. That is a family-level flag. Based on the Google Patents priority data, 9,542,542 is a parent of later members, including US 11,551,222 and US 12,626,262, and is a family member alongside US 9,990,628. I could not confirm that the flag attaches specifically to 9,542,542 rather than to other family members.

3. Proxense's actual campaigns (context — these assert OTHER patents)

Because the likely source of any confusion is the Proxense portfolio, here are the campaigns I did verify, with the correct patent numbers noted:

Note: the Intel complaint's Exhibit E is U.S. 11,551,222, a patent that (per Google Patents priority data) claims priority from the 9,542,542 application chain. That is the closest connection I found — but 11,551,222 ≠ 9,542,542, so it is not litigation of 9,542,542.

Related PTAB proceedings (not litigation, and none on 9,542,542): Samsung IPR2021-01439 (9,235,700) and IPR2021-01447 (9,298,905) — institution denied/terminated Feb. 2022; Microsoft and Google IPRs (2024 series) on 8,646,042, 9,679,289, 10,073,960, 8,352,730-family patents.

Recall of the Samsung 6:21-cv-00210 five-patent set: sources describe "five biometric authentication patents" but I did not independently verify the full list, and I found no document naming 9,542,542 in that complaint. Treat any listing of 9,542,542 there as unverified.

4. Recommendation

If you need a definitive negative or positive for 9,542,542 specifically, run a full-text PACER search and a Docket Navigator patent-number query on "9,542,542" (and "9542542"), and check the Darts-ip family 38694426 record directly, since (a) all public "954" hits I located resolve to 8,886,954, and (b) aggregator taxonomy sometimes collapses family members. Based on the evidence available to me, no litigation is known involving US 9,542,542.

I'll start by verifying whether any AIA trial proceedings exist against this patent, since the structured block defaults to none.

The search surfaced heavy Proxense IPR activity — but much of it names U.S. 8,886,954 ("the '954 patent," a different Proxense patent). I need to pin down whether 9,542,542 itself was ever challenged.

Important threshold finding — read this before the template output

Web search contradicts the naive reading of the search results, not the ODP data. There is a large Proxense IPR campaign — but every proceeding I could tie to a patent number is against a sibling, not against US 9,542,542:

This is the single biggest trap in this assignment: "the '954 patent" in IPR2024-00233 / -00846 is US 8,886,954, not US 9,542,542. The two share an assignee (Proxense LLC), an inventor group (Giobbi et al.) and a 2005-11-30 priority lineage, and the PTAB even describes 8,886,954 as "the related 954 Patent." A defendant handed a demand letter citing "the '954 patent" must confirm which patent number is actually asserted.

I found no AIA trial proceeding naming US 9,542,542, and the structured ODP block is consistent: no PTAB activity on file. I flag below proceedings whose target patent I could not verify, so you can close that gap.

Proceedings overview

Total AIA trial proceedings on US 9,542,542: zero (0). Per the structured "PTAB proceedings on file" block (USPTO Open Data Portal, most recent ingest), there are no IPRs, PGRs, or CBMs against this patent, and no web result I could corroborate identifies one. Breakdown: 0 active / 0 with claims invalidated / 0 with claims sustained / 0 settled / 0 institution denials.

Defensive posture: the patent is untested, not hardened. It has never been through an AIA trial, so there is no FWD, no claim-level cancellation, and no § 315(e)(2) estoppel to exploit. Its defensibility must be assessed from its family's IPR record, its claim scope, and — critically — its adjusted expiration date of 2027-05-25, which is roughly eight months from today (2026-09-29). An IPR filed today would hit the roadblock of the March 26, 2025 "Interim Process for PTAB Workload Management" memo and the "settled expectations" discretionary-deny line of decisions; and the claim term would likely expire before any FWD issued.

No proceedings to enumerate for US 9,542,542

Because the canonical list is empty, there is no {PROCEEDING_NUMBER} entry to populate for this patent. Do not let anyone populate it with IPR2024-00233 or IPR2024-00846 — those are on 8,886,954.

Adjacent proceedings (sibling patents — context only, NOT estoppel on 9,542,542)

Full entries for the two most relevant, because they involve the identically nicknamed "'954 patent":

IPR2024-00233 — Google LLC v. Proxense, LLC (joined by Apple Inc. as IPR2024-01334)

• Type: Inter Partes Review
• Filed: 2024 (Google petition); Apple joined 2024-10-08 as a duplicate petition
• Status: Terminated as to Google (Paper 21), leaving Apple as sole petitioner; Final Written Decision issued
• Patent challenged: U.S. 8,886,954 B1 — not 9,542,542
• Judge panel: APJs Thu A. Dang, Kevin F. Turner, David C. McKone (opinion by McKone)
• Petition grounds: claims 1–7, 10, 12–19, 22–27 challenged; primary art Ludtke (US 7,188,110) and Kon (US 2002/0046336)
• Institution decision: instituted, filed under 35 U.S.C. § 314(a)
• Final Written Decision: all challenged claims — 1–7, 10, 12–19, and 22–27 — held unpatentable (35 U.S.C. § 318(a)). Apple's competing petition IPR2024-01485 (all 29 claims, Burger / Robinson / Orsini) was run in parallel and drew multiple-petition discretionary pushback. (FWD text, Banner Witcoff mirror; PTAB E2E)
• Settlement / termination: Google settled its district court dispute with Proxense (reported 2025-01-14/24) and was terminated from the IPR; Apple remained.
• Appeal: Yes. Consolidated Federal Circuit appeal from IPR2024-00233, IPR2024-01334, and IPR2024-00846; Proxense filed its opening brief ~2026-04-03, arguing (i) the PTAB violated the APA by staying the favorable ex parte reexam and disregarding the CRU's contrary findings, (ii) error in construing "a third party that operates a trusted authority," and (iii) an Appointments Clause defect in Director-review delegation. Pending as of 2026-09-29. (ip fray report)
• Defensive value (for 8,886,954, not 9,542,542): claims 1–7, 10, 12–19, 22–27 are canceled subject to appeal. If a demand letter cites 8,886,954 claims in that set, the § 315(e)(2)-adjacent posture is very strong.

IPR2024-00846 — [Microsoft Corp.](/litigations/by-plaintiff/Microsoft%20Corp.) v. Proxense, LLC

• Type: Inter Partes Review
• Filed: 2024 (petition on file 2024-04-26)
• Status: Final Written Decision issued — challenged claims unpatentable
• Patent challenged: U.S. 8,886,954 B1 — not 9,542,542
• Judge panel: APJs Thu A. Dang, Kevin F. Turner, David C. McKone
• Petition grounds: all 29 claims challenged; art includes Ludtke and Burger (US 2005/0050367); a § 325(d) dispute over the Examiner's prior consideration of Robinson and Greim is documented in the Patent Owner's preliminary response.
• Institution decision: instituted 2024-11-18 — "For the reasons explained below, we institute an inter partes review of the '954 patent." (Institution Decision)
• Final Written Decision: claims 1, 2, 4, 5, 7–13, 15, 16, 18–24, and 26–29 determined unpatentable under pre-AIA 35 U.S.C. § 103. Note that claims 3, 6, 14, 17, and 25 were not in the disposition — do not represent them as canceled.
• Settlement / termination: no settlement on the record for this IPR.
• Appeal: Yes — consolidated into the same Federal Circuit appeal as IPR2024-00233 (opening brief ~2026-04-03).
• Defensive value: another large claim block of the adjacent '954 patent is gone, subject to appeal and to Proxense's APA argument that the CRU had reached the opposite conclusion on the same art.

Other Proxense IPRs surfaced (sibling patents; verify each target before relying)

Ex parte reexaminations (not AIA trials): 90/015,052 (the '730), 90/015,053 (the '905), 90/015,054 (the '989). None reexamines 9,542,542. The '052 EPR produced a favorable Examiner amendment for Proxense and was stayed by the PTAB on 2025-03-27 — the basis of Proxense's current APA appeal.

Strategic summary

Claim status on 9,542,542: entirely UNTESTED. No claim of this patent has been canceled, narrowed, or confirmed by the PTAB, because no petition was ever filed against it. Every "claims 1–X canceled / unpatentable" headline you may have seen belongs to 8,886,954, 8,352,730, or 9,298,905. The continuation US 11,551,222 B2 (filed 2016-11-29, granted 2023-01-10, same title and inventors) is likewise untested. Whether '542 or its continuation '222 is the asserted instrument must be confirmed from the claim chart itself — the family's naming conventions are a live source of error.

Estoppel landscape: essentially empty for this patent. § 315(e)(2) estoppel is patent- and petitioner-specific. Because no IPR, PGR, or CBM ever reached 9,542,542, no petitioner is estopped as to any ground against '542, and no IPR ground for this patent has been "raised or reasonably could have been raised." A defendant retains the full prior-art field — including Ludtke (US 7,188,110), Burger (US 2005/0050367), Kon (US 2002/0046336), Scott (WO 1999/056429), Lapsley, Robinson, Baird, Berardi, and the "Pocket Vault" system — none of which has been tested against these claims. Conversely, the fact that the same references gutted the sibling '954 patent's claims is a strong signal about how a '542 validity fight could go, especially given that the '052 Examiner found the family's claims "not patentably distinct" across these patents.

Pattern signals. Google, Microsoft, and Apple each filed multiple petitions across the Proxense family, and Apple and Microsoft are the holdouts still litigating (stayed) while Samsung, Google, and Intel settled (2023, 2025-01, and 2026-03 respectively). There is no defensive aggregator (e.g., Unified Patents) proceeding in the record I found — Unified's portal merely hosts a patent page for 9,542,542; that is not a filed challenge. Proxense is appealing aggressively via Hecht Partners, and the conflict between the PTAB's FWDs and the CRU's reexamination findings is now squarely before the Federal Circuit — meaning the sibling patent's invalidation is not final and could be vacated.

A note on the timeline. Today is 2026-09-29 and 9,542,542 expires 2027-05-25 (20 years from the 2007-05-07 filing date plus 18 days of PTA). That is the dominant commercial fact: an IPR filed now would face the Director's "settled expectations" discretionary-denial posture and would likely outlive the patent. Any pre-2027 licensing or settlement math should be run against that date.

Recommended next steps

1. Confirm the asserted patent number first. Before responding to any demand citing "the '954 patent," pull the patent number. If it is 8,886,954, the demand is built on claims that the PTAB held unpatentable — quote the FWD dispositions: IPR2024-00233 ("we determine that Petitioner has demonstrated, by a preponderance of the evidence, that claims 1‑7, 10, 12‑19, and 22‑27 are unpatentable") and IPR2024-00846 (claims 1, 2, 4, 5, 7‑13, 15, 16, 18‑24, and 26‑29 unpatentable). FWD, IPR2024-00233 · PTAB E2E · CourtListener.
2. If it is genuinely 9,542,542, treat this as a fresh validity fight: no estoppel exists, so commission a full prior-art search, with Ludtke/Burger/Scott as the starting corpus given how those references performed against sibling claims. Check whether the family's double-patenting findings (claims "not patentably distinct") create a terminal-disclaimer or OTDP vulnerability.
3. Also check US 11,551,222 — if the continuation was asserted rather than '542, you may be facing a separate three-year-old patent with a later nominal term (the '222 carries a terminal disclaimer).
4. Watch the Federal Circuit. The consolidated appeal from IPR2024-00233 / IPR2024-01334 / IPR2024-00846 (briefed ~2026-04-03) could vacate the invalidation of the sibling patent on APA/claim-construction grounds. If Proxense wins there, the leverage calculus across the whole family changes. Docket via CourtListener.
5. Do not file a new IPR on '542 reflexively. With expiry on 2027-05-25 and the March 26, 2025 Director memo on workload management, expect discretionary denial — pursue invalidity in the district court or via ex parte reexamination instead.

Unverified gaps (flag for follow-up): the target patents of IPR2024-00784, IPR2025-00327, IPR2025-00562, and Intel's three denied petitions were not confirmed in the sources I retrieved; none showed any linkage to 9,542,542, but verify each against PTAB E2E before representing that this patent has "never been challenged." I also could not confirm the current status of IPR2024-01327.

I'll research this patent's assignment record and litigation history before writing. Let me run several searches in parallel.

Let me dig into the specific assignment records and Proxense's corporate status.

Ownership Reconstruction — US Patent 9,542,542

Patent: US 9,542,542 B2 — "Single step transaction authentication using proximity and biometric input"
Application: 11/745,249 · filed 2007-05-07 · granted 2017-01-10 · earliest priority 2005-11-30
Assignee of record: Proxense, LLC (Bend, OR)

Sourcing caveat (stated up front, per the no-fabrication constraint): I could not execute a live query against the USPTO Assignment Center in this session. Every statement below is grounded in the Google Patents legal-events record for US 9,542,542, the patent face, and litigation/file-history documents. I therefore do not report reel/frame numbers, execution dates, or a recording correspondent — those were not retrievable and I will not invent them. Verify reel/frame and correspondent directly at https://assignmentcenter.uspto.gov/ (search "9542542") or https://assignment.uspto.gov/patent/index.html.

Inventors

Unusual-pattern check — no "departure" red flag. The classic pre-fire-sale tell (all inventors leaving the assignee within 12 months of filing) is not present. Giobbi is a serial Proxense inventor who remains named on later Proxense patents in the same family (e.g., US 11,120,449; US 11,157,909) and is publicly identified as Proxense's CEO in 2021–2023 press and pleadings. This is a founder-controlled portfolio, not a team that walked away.

One minor inconsistency worth noting: local press dates Proxense's founding to 2001 in one article (The Bulletin, 2008) and 2005 in another (The Bulletin, 2011). This is a company-history discrepancy, not an ownership issue.

Original assignee

Proxense, LLC, Bend, Oregon — the assignee printed on the face of the patent (field 73 of the granted document).

• Primary line of business: secure proximity-based wireless authentication — the "Personal Digital Key" (PDK) carried by a user, read by a "Reader," with local biometric verification, applied to mobile payments, healthcare workstation access control, and item/employee tracking.
• Did it ship a product embodying the claims? Yes, historically. The Bulletin (2008) describes the PDK/key-fob system and a signed deployment deal with casino-systems provider Bally Technologies; the same paper (2011) reports Proxense installing its biometric proximity system for medical-records access at real customers (Central Oregon ENT; Bend Memorial Clinic), with expansion planned, and "more than 35 employees." Proxense's own 2021 litigation press release states that from 2004–2012 it "began to develop mobile payment technologies and commercial products, employing over thirty engineers." The past-tense framing suggests product development wound down around 2012.
• Current status: Operating as a patent licensing/enforcement company rather than a product vendor. Proxense self-describes an "IP portfolio with over 65 patents" (2021 press release) and has run a serial assertion campaign in the W.D. Tex. Waco Division from 2021 through 2025 (Samsung, Google, Microsoft, Apple). Third-party litigation-intelligence labels it "Biometric authentication IP licensor" (PatSnap). No source reviewed shows Proxense in bankruptcy, dissolved, or acquired — no successor entity appears anywhere in the chain.

Assignment timeline

Only one recorded assignment is evidenced in the sources reviewed. No post-issuance transfer of any kind appears in the Google Patents legal events for US 9,542,542.

• Executed: not shown in sources retrieved / recorded 2007-06-29 — Reel not retrieved (Assignment Center not directly queried; do not treat any reel/frame as verified here)
  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
  • Assignor: Giobbi, John J.; Brown, David L.; Hirt, Fred S.
  • Assignee: Proxense, LLC (Bend, OR)
  • Correspondent: Not determinable. For transparency: the prosecution correspondence of record across this family is "Patent Law Works / Proxense, 165 South Main St, Suite 2, Salt Lake City, UT 84111" (per the file histories in the IPR exhibits). That is prosecution counsel, not a confirmed assignment-recording correspondent, and I am not treating it as one. No repeat-player assignment attorney is observable because there is only one recorded lien on title.
  • Context: Formation-stage inventor-to-company assignment (founders/employees assigning their rights to the operating entity). It is not a fire-sale, reorg, securitization, or transfer-to-asserter.

Related priority-chain transfers: The parent application 11/292,330 (filed 2005-11-30, issued as US 9,613,483) and the later continuations (e.g., US 11,551,222; US 12,626,262) sit in the same Proxense, LLC ownership, per the Google Patents "priority to / priority from" events. Nothing in that related chain reassigns US 9,542,542 away from Proxense.

Timeline diagram

timeline
    title Ownership of US 9542542
    2005 : Priority date Nov 30
    2007 : Application 11 745 249 filed
         : Inventors assign to Proxense LLC
    2017 : US 9542542 granted Jan 10
    2021 : Proxense enforcement campaign begins
    2023 : Proxense sues Microsoft and Google

NPE / troll-pattern signals

Verdict

NPE — moderate confidence.

The recorded chain is unusually clean: there is exactly one assignment (recorded 2007-06-29, inventors Giobbi/Brown/Hirt → Proxense, LLC), and the current assignee on Google Patents is the same Proxense, LLC — so none of the transfer-based signals (shell transfer, cascading LLCs, pre-litigation transfer, bankruptcy sale, privateering) are present, and the strongest NPE indicators in this framework are absent. The NPE characterization rests instead on the owner's post-2012 business model: Proxense describes itself as holding an "IP portfolio with over 65 patents," is labeled a "biometric authentication IP licensor" by third-party intelligence, and has run a serial assertion campaign against Samsung, Google, Microsoft, and Apple in W.D. Tex. from 2021 onward while no longer shipping a competing product — i.e., a former operating company that has become a licensing/enforcement entity. Because that conclusion is supported by business-model and litigation evidence rather than by any assignment churn, it is a moderate, not high, confidence call, and it rests on a single 2007 lien on title whose reel/frame I could not independently verify.

Verify at: https://assignmentcenter.uspto.gov/ (search patent number 9542542) and the Google Patents legal-events tab https://patents.google.com/patent/US9542542/en.

I'll research the prior art citations for US 9,542,542 B2.

Prior-Art Analysis — US 9,542,542 B2

Scope and method (read first)

I searched for prior art citations directed to US 9,542,542 B2 ("Single step transaction authentication using proximity and biometric input," Proxense LLC, App. 11/745,249, filed May 7, 2007, issued Jan. 10, 2017). Two constraints materially limit this analysis, and I am flagging them up front rather than papering over them:

1. I could not retrieve the complete "References Cited" list from the face of the '542 patent. Google Patents renders a "Patent Citations (151)" block for the family, and the issued-PDF snippets expose only the first several U.S. references and the foreign/other-publication entries. I have therefore reconstructed the citation set from (a) the issued-PDF "References Cited" fragments for the '542 patent and its direct continuation US 11,551,222 B2, (b) the family citation block, and (c) the 151-entry list surfaced on the sibling US 8,412,949 B2. The references below are genuine family citations, but I cannot guarantee the exact per-reference membership in the '542 patent's own front page without the issued PDF.
2. §102 vs. §103 is important here. The overwhelming majority of the 151 cited references are classified as category "A" background art (they define the general state of the art, not "particular relevance"). In my assessment, none of the single cited references anticipates claim 1 or claim 15 under §102; the realistic exposure is §103 obviousness in combination. I say this explicitly because a §102 framing of every citation would overstate the art.

A. The reference set (cited U.S. patent documents)

The following are drawn from the '542/continuation front pages and the family citation block. Where the issued-PDF OCR is ambiguous (e.g., "5187352" vs. "5,187,352"), I note it.

Other publications of record (family):

• Anonymous, "Applying Biometrics to Door Access," Security Magazine, Sep. 26, 2002 — general biometrics-for-access background.
• "Alliance Activities: Publications: Identity," Smart Card Alliance, 1997–2007 — smart-card identity background.
• Zhang, "A User-Centric M-Payment Solution" (2005) — cited on the US 11,551,222 continuation front page as an "Other Publication."

B. The most relevant prior art (focused)

Filtering the 151-entry set down to what actually bears on claim 1 / claim 15 — i.e., automatic proximity reception of a biometric SAMPLE (a subset of a full profile) from each of two or more PDKs, screening the live biometric input against the collected samples, associating the matching PDK, then verifying against that PDK's FULL profile, and authenticating the PDK without user intervention — the citation set is notably thin. The genuinely relevant references are:

1. US 4,993,068 A — Piosenka et al. (Motorola), issued Feb 12, 1991

• Relevance: Discloses a portable carrier holding a digitized biometric template that is downloaded to a reader and compared with a live scan to authenticate a person. This is the closest cited art to the core "biometric data on a token, verified at the point of use" concept.
• §102 assessment: Does not anticipate claim 1 or claim 15. It lacks (i) wireless/proximity automatic reception, (ii) the multi-PDK sample screening, and critically (iii) the two-tier sample-vs-full-profile structure (subset screening first, full profile second). A §103 attack would need a second reference supplying the two-tier screening.

2. US 7,188,110 B1 — Ludtke (filed Dec 11, 2000; issued Mar 6, 2007)

• Relevance: Biometric verification of a user with a device that releases identifying/security data only upon a matching biometric scan. This is the reference the PTAB found persuasive as a §103(a) primary reference against claim 1 of the sibling US 8,886,954 (IPR2024-00233, in combination with Kon, US 2002/0046336 A1).
• §102 assessment: On its own it does not anticipate the '542 claims; the Board's holding against the sibling patent was a §103 combination, not §102 anticipation. Its importance here is as the leading obviousness reference if the '542 claims are ever challenged.

3. US 2002/0046336 A1 — Kon (Apr 18, 2002)

• Relevance: Secondary reference in the IPR2024-00233 combination. Contributes verification/authentication mechanics.
• §102 assessment: No independent §102 anticipation of '542 claim 1/15; §103 secondary reference only.

4. US 5,886,634 A (Muhme) and US 7,049,963 B2 (Waterhouse)

• Relevance: Establish the proximity/RFID-token detection element of claim 1 (step "automatically and wirelessly receives … once the PDK enters a proximity zone").
• §102 assessment: Address proximity/ID only; silent on biometrics entirely. Cannot anticipate.

C. Bottom-line §102 / §103 conclusions

The realistic prior-art threat is §103, not §102. The strongest combination a challenger would assemble is a token-based biometric-verification reference (Ludtke / Piosenka) as primary, plus an RFID/proximity reference (Waterhouse / Muhme / 5,426,632) for the automatic-proximity element. The two-tier "biometric sample is a subset of the full profile, screen against samples from all PDKs in the zone, then verify against the full profile" limitation is the feature I could not map cleanly onto any single cited reference — it is the strongest point of novelty on the face of the citation set.

D. Explicit contradictions / corrections relative to the earlier section

1. "The '954 patent" naming trap — real and material. The earlier section correctly flagged this, and my search confirms it again: US 9,542,542 (subject here) and US 8,886,954 ("Biometric Personal Data Key (PDK) Authentication") are different patents that are both routinely abbreviated "the '954 patent" in the W.D. Tex. and PTAB papers. All of the substantive IPR art I found — Ludtke (US 7,188,110) and Kon (US 2002/0046336) — is recited against US 8,886,954 in IPR2024-00233, not against US 9,542,542. Do not transplant that art to 9,542,542 without verifying it appears on the '542 face.
2. IPR status of the two patents differs. Consistent with the prior section, IPR2024-00846 (Microsoft v. Proxense) is directed at US 8,886,954; the reexaminations 90/015,052–054, and IPR2024-00232/-00234, target the '730, '905, and 10,698,989 patents. I still find no PTAB or CAFC proceeding whose challenged patent is US 9,542,542.
3. Claim-text caveat stands. The claim 1/15 characterizations in the prior section (subset-sample screening → full-profile verification) are from third-party aggregators; the authoritative claim text was truncated in my source. My §102 mapping above inherits that uncertainty.

E. What I could not verify (stated explicitly)

• I could not retrieve the full, exact 151-reference list for US 9,542,542 B2 itself, nor confirm each reference's per-document "consideration" category (§102 vs. mere background). The table in Section A is a reconstructed family list.
• The file-wrapper cites (USPTO form PTO-892/PTO-1449) for the '542 patent were not reachable from my sources; the examiner's "relevant to claim" annotations, if any, would be the authoritative §102 mapping and I do not have them.
• I therefore do not claim that any specific reference above is the examiner's ground of allowance or rejection for '542 — that requires the issued front page and the file history, which I could not pull.

I'll gather the prior-art references associated with this patent's Google Patents page before running the § 103 analysis.

Let me pull specifics on the key cited references and check whether any proceeding has actually challenged the '542 claims.

Obviousness Analysis — US 9,542,542 B2 under 35 U.S.C. § 103

Analyst note on the evidence base (read first). You asked me to use "the results from the Prior Art section of this page." The authoritative page text supplied to me is truncated before the claims and before the prior-art tables — the only prior-art content that survived the fetch is the line Prior art keywords: biometric, personal digital, digital key, key device, transaction. To reconstruct the prior-art section I therefore relied on (a) the printed patent's own "References Cited" list as reported by patentimages and Justia, (b) the "Patent Art (235)" similar-art list on the Unified Patents portal, and (c) citations to this family that appear in the sibling PCT search report and in USPTO/PTAB papers. I could not retrieve the Google Patents "Citations / Cited By / Similar Documents" tables themselves, so the reference set below is substantially but not certainly complete, and I flag each reference's provenance. Where I could not read a reference's disclosure, I say so rather than assuming.

1. Corrections and confirmations relative to the earlier sections

Useful new datum for § 103: the pre-grant publication claim 1 (US 2007/0245158 A1) was far broader — "wirelessly receiving data from a personal digital key (PDK); receiving a biometric input; authenticating the personal digital key; authenticating the biometric input; and performing a transaction responsive to the personal digital key and the biometric input being authenticated" (insight.rpxcorp.com/patent/US20070245158A1). That broad version is squarely the combination analyzed in Ground 1 below. The issued claim 1 adds the sample-subset/two-PDK screening architecture — i.e., that architecture is what carried the case over the art, and it is therefore where the § 103 fight concentrates.

2. Level of ordinary skill (POSITA)

Proxense's own proposed definition in the WDTX litigation is a useful anchor: "a bachelor's degree in computer or electrical engineering (or an equivalent degree) with at least three years of experience in the field of encryption and security (or an equivalent)" (Docket Alarm, Ex. 1018). A POSITA here is also presumed conversant with RF-ID/contactless-token design, fingerprint minutiae matching, and stored-value/credit-card transaction protocols.

3. The prior-art references available (with provenance)

4. Element-by-element mapping of issued claim 1

Claim 1 requires: (A) auto/wireless receipt of a first biometric sample from a first PDK entering a proximity zone, where the PDK holds both a sample and a full biometric profile of its valid user and the sample is a subset of the profile; (B) the same for a second PDK; (C) receipt of a biometric input; (D) comparison of the input to the collection of samples from PDKs in the zone; (E) determining the input matches the first PDK's sample; (F) on that basis, associating the first PDK with the transaction; (G) then checking the input against the first full profile; (H) authenticating the PDK without user intervention; (I) authorizing on authentication + full-profile match.

5. Grounds of rejection

Ground 1 — Piosenka + proximity/RF-token art (the "single-step, PDK-based transaction" concept)

Scope: meets the published claim 1 (US 2007/0245158 A1) almost directly, and meets the issued claim 1 except for the sample-subset/collection architecture.

• Piosenka discloses storing the biometric reference on a portable device carried by the user (not in a merchant/central database), reading it at the point of transaction, acquiring a live biometric input, comparing, and granting or denying the transaction based on privilege data also stored on the credential. That is elements (C), (G), (I) and the "PDK-vs-Reader" split.
• Piosenka's own background states the reason to make the token wireless/automatic: centralized biometric verification "must maintain an on-line data base," is "very expensive," and "poses intolerable access delays during periods of peak transactions." Substituting a wireless, proximity-triggered token interface for Piosenka's card-reader interface is the substitution of one known data-transfer mechanism (contactless RF-ID) for another (contact card reader), producing the same result — an art-recognized, predictable substitution under KSR.
• Motivation, articulated: (i) Piosenka itself wants the credential verified locally with no central database; (ii) the contactless-token references (WO 00/62505, WO 01/22724, Blair '352) want the token to work without physical presentation/contact; (iii) the Smart Card Alliance NPL shows the industry was already moving to contactless identity tokens. A POSITA seeking a "no-PIN, no-card-swipe" transaction has every reason to swap the contact interface for the proximity interface.
• The Office has already run a version of this combination in this family. In reexamination regarding the sibling '905 patent, the examiner rejected "proximity authorization unit" claims under § 103(a) over Grimes et al. in view of Piosenka et al., reasoning that "it would have been obvious … for the computer as taught by Grimes et al. to be replaced by … biometric data input and validation … as taught by Piosenka et al." (USPTO PTAB filing). A similar rationale applies here against the original '542 claim 1 and, with the secondary art below, against issued claim 1.

Ground 2 — Piosenka + Miller '177 (network-side biometric authentication of a client)

• US 7,249,177 (Miller/Sprint) was cited by the ISA as an "X,E" reference against claims 1–42 of the sibling "Two-Level Authentication" application, at col. 3:28–42; col. 4:19–25; col. 5:1–17 (WO2007130687A3.pdf). "E" status means it is prior art published on or after the international filing date but with an earlier effective date — here a Nov. 26, 2002 filing, available as § 102(e) art against the '542 regardless of the CIP chain.
• Combined with Piosenka, Miller supplies the network/client biometric-authentication architecture and the "the authenticating party holds a list/authority" aspect that the '542's Reader-registry embodiments rely on. Combined with a proximity token, it also supplies element (H) (device authenticated on entry, no user action).

Caveat: I could not retrieve Miller's specification to verify its disclosure columns. Its role in this ground rests on the ISA's citation, not on my own reading. Anyone relying on this ground should pull US 7,249,177 and confirm cols. 3–5.

Ground 3 — Adding the two-stage "coarse screen, then full verify" (elements A, D, G)

This is the ground that matters, because it targets the only limitation that plausibly distinguishes the issued claims from Grounds 1–2.

• Burger, US 6,219,439 B1 expressly claims a first comparison (scanned biometric vs. data on the card, performed in the portable reader) and, "when the first comparison signal indicates positive authentication," a second comparison (card's non-biometric identifier vs. data stored remotely). That is the same two-stage, escalating-verification architecture the '542 claims, in the same field, with the same rationale (screen locally and cheaply, then confirm definitively).
• Sony US 2002/0191816 A1 teaches using a biometric identifier to select which of multiple stored consumer profiles/accounts is the operative one — i.e., using the biometric as a differentiator among a plurality of candidate identities/profiles, a direct antecedent of steps (D)–(F) (compare the input against a collection associated with multiple nearby keys and pick the matching one).
• Motivation: the '542's own specification states the problem — a Reader in a dense client environment must "automatically determine[] which PDKs are valid and buffer the received information from each valid PDK." Once multiple PDKs are within the zone (the claim's own premise), the POSITA faces an obvious engineering choice: transmit full biometric profiles for every nearby key over the secure channel (slow, high-bandwidth, privacy-exposing), or transmit a compact subset and only pull the full profile for the one candidate that matches (fast, low-bandwidth, more private). KSR treats this as a predictable design tradeoff with an identifiable reason to prefer the cheaper screen. Byte-reduction of biometric templates for transmission (lossy/template representations, coarse-to-fine matching) was itself ordinary skill — see e.g. the revocable/triangle-encoded fingerprint template art in the broader field (US 2006/0104484 A1, for the general proposition that fingerprint representations were deliberately transformed/compressed for transport).

Why this ground is contestable: I found no verified reference in the '542's own prior-art section that expressly discloses transmitting a subset sample in advance, screening against a collection of samples from multiple keys, and only then fetching and comparing the full profile. The only document I located that describes that exact sequence is Proxense's own US 7,904,718, filed the same day (May 7, 2007) as the '542 application and therefore not prior art under § 102(a)/(b), and not § 102(e) art either, since the '718's filing date is not "before" the '542's effective date. That is the single most important finding of this analysis.

Ground 4 — Claim 15 (system)

Claim 15 recites "one or more processors … and a memory including instructions" performing the same steps. It rises and falls with claim 1: it adds no structural element that is not a generic processor/memory. If claim 1 is obvious, claim 15 is obvious; if claim 1 survives on the sample-subset architecture, claim 15 survives with it. (I could not obtain the text of claims 16–20; dependent-claim analysis is therefore unavailable — do not infer their content.)

6. Where the obviousness case is genuinely weak, and why

1. The "subset sample / screening-then-full-verify" architecture (elements A, D, G) is the last stand. On the record I can verify, its only unambiguous antecedent description is same-day, same-family US 7,904,718, which is not prior art. A petitioner would need an independent pre-2007 reference teaching coarse-to-fine template matching over a wireless link — I have not verified one exists in this patent's cited set.
2. Claims-construction risk cuts both ways. Element (D) is qualified — "the collection of biometric samples including the first … and the second … when the first … and the second … are within the proximity zone." If that condition is read as making the second-PDK elements conditional, then a single-PDK deployment practices claim 1, and the issued claim collapses back into the Ground 1 space that Piosenka + contactless-token art largely occupies. If instead the second-PDK receipt is read as mandatory, the claim is narrower but the screening rationale is stronger (you only need the screen because there are multiple keys). Either reading helps a challenger in different ways. Note that Proxense's own claim-construction positions in the WDTX cases construed "device ID code," "access message," and "tamper proof format" for the '730/'954/'905/'989 patents (Apple's PCC disclosure) — nothing there binds 9,542,542.
3. Secondary considerations Proxense can invoke: the '542 specification frames a long-felt need (PIN/password fatigue; the intrusiveness and repeat-enrollment burden of conventional biometric databases; the liability of merchants storing biometrics) and asserts a privacy-preserving result (biometrics stay on the user's own token; registries "do not store biometric information"). Proxense also has a substantial licensing/enforcement record across the family. These are real KSR counterweights, though they attach to the "single-step, no-external-biometric-database" concept — which Grounds 1–2 already reach — more than to the sample-subset limitation.
4. No PTAB challenge to 9,542,542 surfaced. The 2024–2026 proceedings I identified target other patents (US 8,886,954 → IPR2024-00233/00846; US 8,352,730 → IPR2024-00232 (Ludtke + Kon); US 9,298,905 → IPR2024-00234; reexams 90/015,052 and 90/015,053). I found no IPR, PGR, or reexamination directed at 9,542,542, so there is no petitioner record or Board construction to borrow from.

7. Bottom line

• Elements (C), (G), (H), (I) — acquiring the live biometric at the point of transaction, comparing it to a biometric reference carried on the user's own portable token rather than held by the merchant, automatically authenticating the token on entry, and authorizing — are plainly obvious over US 4,993,068 (Piosenka) in view of the contactless/proximity token art (WO 00/62505; WO 01/22724; Blair '352), with motivation supplied by Piosenka's own express criticism of centralized biometric databases. This is essentially the published claim 1, and it is why the application had to be narrowed.
• Elements (D)–(F) (screen the input against a collection of samples from several nearby keys to pick one) are obvious over Piosenka + US 2002/0191816 (Sony, biometric selection among multiple profiles/accounts), once the multi-PDK premise of the claim is taken as given.
• Elements (A) and (G)'s ordering (a subset sample transmitted in advance and screened before the full profile is fetched and compared) are the principal non-obviousness redoubt. The closest description of that architecture is same-day, same-family US 7,904,718, which is not prior art. Absent an independent pre-2007 reference teaching coarse-sample screening followed by full-template verification over a proximity link, a § 103 challenge to issued claim 1 is strong on everything except this, and this is where it would be fought.

Explicit uncertainties I am not papering over: (i) I could not read the Google Patents prior-art tables, so an additional on-point reference may exist that I did not see; (ii) I could not read Miller US 7,249,177, Ivi US 7,278,025, Privaris US 7,588,761, Sony US 2002/0191816, or AmEx US 7,314,164 directly — their content above is inferred from title/assignee/date metadata and the ISA's citation; (iii) I could not obtain the text of claims 2–14 and 16–20; (iv) claim 1's second-PDK limitation may be conditional, which materially changes the analysis.