US 8433919

I'll search for authoritative information on this specific patent, then summarize.

Let me check the specific litigation docket and any Federal Circuit proceedings involving this exact patent.

Let me verify whether this specific patent has been subject to PTAB challenges or Federal Circuit appeals.

US Patent 8,433,919 — Analyst Summary

1. Bibliographic record (as verified)

Sources: Google Patents US8433919B2, FreePatentsOnline 8433919, Justia 8433919, WO2007130687A3.

2. Abstract (verbatim)

"A system and method provide efficient, secure, and highly reliable authentication for transaction processing and/or access control applications. A Personal Digital Key stores one or more profiles (e.g., a biometric profile) in a tamper-proof memory that is acquired in a secure trusted process. Biometric profiles comprise 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."

3. Independent claims — plain-language overview

The patent has 37 claims, with five independent claims: 1, 13, 19, 25, and 34.

• Claim 1 (method; Reader-centric "two-level" method). Pull a biometric profile sample — transformed biometric data based on less than the entirety of the full biometric profile — from the memory of a Personal Digital Key (PDK); acquire a biometric input; compare the sample against the input; and authorize the transaction if they match. Separately, open a secure channel to a remote registry (a database administered by a trusted third-party organization with which the PDK is registered), send PDK-identifying information, and receive back a validation decision on whether the PDK is valid. Authorization requires both the sample match and the favorable registry decision.

• Claim 13 (apparatus; the Reader). A hardware claim covering: a biometric reader; a receiver/decoder circuit (RDC) wirelessly receiving the profile sample from the PDK; a processor that compares sample to input and indicates authorization on a match; and a network interface that establishes the secure channel to the registered remote registry, transmits PDK information, and receives the validity decision — with the processor deciding using both the registry decision and the sample comparison.

• Claim 19 (device; the PDK itself). A Personal Digital Key with: tamper-proof memory storing (i) a transformed biometric profile uniquely associated with an individual, (ii) a biometric profile sample that is a subset of that profile (less than its entirety), and (iii) a unique PDK ID; a transceiver that wirelessly receives, over a secure channel, data derived from a biometric input captured by an external device and wirelessly transmits the PDK ID; and control logic coordinating comparison of the biometric-input-derived data to the sample.

• Claim 25 (method; PDK-side). Store the full biometric profile and a subset sample in tamper-proof PDK memory; wirelessly receive data based on a biometric input acquired from the individual; then the external device authorizes the transaction based on two things: (a) the PDK-performed comparison of the sample against the biometric-input data, and (b) a validation decision received from a remote registry over a secure channel, where that decision authenticates the PDK at least in part on whether the PDK is registered with the registry.

• Claim 34 (method; Reader-side, most complete "two-level" recitation). Wirelessly receive uniquely identifying PDK information; transmit it to the remote registry; receive the validity decision; receive profile-type information from the PDK indicating the types of profiles it stores; determine whether those types are compatible with allowable authentication types; perform an authentication test in which a biometric profile sample is wirelessly received, a biometric input is acquired, and the profile is deemed valid when the input matches the sample; and authorize only when the PDK is valid, the profile types are compatible, and the profile is valid.

Notable dependents (context only): claim 4/5 device authentication (challenge-response) on entry into a proximity zone; claim 6 profile-type sufficiency check; claim 8/9 purchasing information plus remote validation database; claim 10 picture-profile/administrator confirmation; claim 11 PIN profile; claim 12 hashing of the biometric input; claim 14 credit-card-terminal I/O; claim 17 encrypted sample transmission; claim 20 programmer/notary/site IDs from a trusted initialization; claims 23–24 form factor (cell phone, PDA, badge, clothing, jewelry); claim 37 registry-profile authentication.

4. Enforcement, PTAB, and Federal Circuit posture (as searched)

• District court litigation. The Google Patents record flags "Family has litigation" and links to Proxense, LLC v. Intel Corporation, No. 7:25-cv-00531 (W.D. Tex., Midland Div., Judge Alan D. Albright), filed Nov. 14, 2025. Per the complaint analysis, US 8,433,919 is asserted, at least as to independent claim 25, against Intel's Platform Trust Technology (PTT) / Converged Security and Management Engine (CSME) implementations in Evo/vPro platforms (Windows Hello biometric flows, secure enclave storage, remote registry validation). The asserted family also includes US 9,405,898, US 10,469,456, US 11,182,792, US 11,551,222, and US 9,990,628. Counsel for Proxense: Hecht Partners LLP. Sources: RPX Insight 7:25-cv-00531, Ex Parte complaint analysis, Unified Patents litigation link.

• Broader Proxense campaign (context, different patents). Proxense has litigated against Samsung (settled), Google (settled Jan. 2025), Microsoft and Apple (stayed), LG (6:24-cv-00302, stayed Feb. 2026), and Intel (settled March 2026 per reporting). PTAB proceedings in the campaign involve the '730, '954, '905, '042, '289, '188 and other patents — not 8,433,919 itself as far as my searches show.

• Notable use of this patent as prior art. The published application of this patent, US 2007/0245157 A1 ("Giobbi '157"), is being asserted as prior art against other Proxense patents (e.g., Apple v. Proxense, IPR2025-00562, on US 9,049,188; Google v. Proxense, IPR2024-00783, on US 9,679,289).

• Federal Circuit, 2026. I found no CAFC docket in 2026 that names U.S. 8,433,919. The active Proxense Federal Circuit appeal reported in 2026 concerns U.S. 8,886,954 (Apple/Microsoft IPRs, PTAB stay of a parallel ex parte reexamination, and a "trusted authority" claim-construction dispute). Sources: ip fray on Proxense's CAFC opening brief. I also found no IPR/PTAB proceeding naming 8,433,919.

5. Stated uncertainties and literal-ID anomalies

I am flagging these rather than resolving them, per instruction:

1. Priority date conflict. Google Patents lists the priority date as 2005-11-30, "priority claimed from US 11/292,330," while Unified Patents' dataset lists the '919 priority date as 2005-11-29. The patent's own specification, however, claims benefit only under 35 U.S.C. §119(e) of five provisionals: 60/798,172 (May 5, 2006), 60/798,843 (May 8, 2006), 60/838,788 (Aug. 17, 2006), 60/824,758 (Sep. 6, 2006), and 60/894,608 (Mar. 13, 2007). The relationship asserted to US 11/292,330 is not explained on the face of the specification, and I cannot reconcile it authoritatively from the sources retrieved.
2. Inventor string on family members. US 8,433,919 names three inventors (Giobbi, Brown, Hirt). Later continuations in the same family (e.g., US 9,990,628; US 11,157,909) list only John J. Giobbi. The PCT publication renders the third inventor as "HRT, Fred, S." — an apparent OCR/typographical rendering of "HIRT" that I have preserved literally rather than corrected.
3. Assignee-history note. Google Patents' assignment entries show a June 26, 2007 record assigning to Proxense, LLC from Brown, Hirt, and Giobbi — consistent with Proxense and Gillette's filing.
4. Claim text provenance. The full independent-claim language above is drawn from the Google Patents full text (authoritative per instructions) and cross-checked against the Justia claim listing for patent 8433919. The Justia listing is the full issued claim set (claims 1–37). If exact claim wording is needed for a filing or opinion, the granted patent PDF should be the citation of record.
5. No CAFC 2026 activity for this number was located. If a 2026 appeal exists, it did not surface in my searches; treat the absence as unverified rather than as confirmation that none exists.

I'll research litigation involving US patent 8433919. Let me run several targeted searches.

Good initial results. Let me dig deeper into the specific cases asserting the '919 patent.

Let me verify whether the '919 patent has been asserted in any other case, and check for PTAB challenges.

Litigation Involving U.S. Patent No. 8,433,919

Bottom line

My searches surfaced one (1) district court case in which U.S. Patent No. 8,433,919 ("Two-level authentication for secure transactions") is affirmatively asserted: Proxense, LLC v. Intel Corporation, No. 7:25-cv-00531 (W.D. Tex.), filed November 14, 2025. This is also the case flagged in the "Family has litigation" field on the patent's Google Patents page.

I did not find any other district court case, ITC action, or appeal in which the '919 patent itself was asserted. Importantly, the '919 patent does not appear among the patents asserted in Proxense's numerous earlier suits against Samsung, Apple, Google, Microsoft, or LG (those asserted patents are the '730, '905, '954, '989, '332, '456, '979, '152, '042, '188, '289, and '960 patents, among others).

⚠️ Caveat on completeness: This is based on web-search-accessible litigation databases (PACER Monitor, Docket Alarm, CourtListener, RPX Insight, Unified Patents/Google Patents litigation links). I was not able to run a full PACER/Docket Navigator or Darts-ip family audit, so treat this as "known from searches," not a certified exhaustive docket history.

The one known case

Sources: PACER Monitor docket (complaint and exhibit list) — https://cdn.pacermonitor.com/public/case/61206235/Proxense,_LLC_v_Intel_Corporation_ ; RPX Insight — https://insight.rpxcorp.com/litigation/txwdce-1172875230-proxense-v-intel ; Ex Parte complaint analysis — https://ai-lab.exparte.com/case/dct/txwd/7:25-cv-00531/doc/analysis/1 ; Google Patents "Family has litigation" link — https://portal.unifiedpatents.com/litigation/Texas%20Western%20District%20Court/case/7%3A25-cv-00531

Note: the complaint's caption/dividend attribution could not be fully reconciled — the case number uses the "7:" (Midland–Odessa) prefix for W.D. Tex., while Proxense's prior W.D. Tex. suits were filed in the "6:" (Waco) division. I am reporting the number literally as docketed rather than "correcting" it.

Adjacent proceedings (not district court litigation, but relevant context)

• PTAB / Intel v. Proxense: Intel has been challenging Proxense patents at the PTAB. On June 26, 2025, the USPTO Acting Director denied institution in Intel Corp. v. Proxense LLC, IPR2025-00327, applying the "settled expectations" doctrine for patents in force over nine years. I could not confirm that any Intel IPR petition challenged the '919 patent specifically — the IPR/reexamination activity I located concerns other Proxense patents (e.g., the '730, '905, '954, '289, '188). Do not assume an IPR exists on the '919 without verifying the petition.
• Related Proxense assertions of the same patent family: The '919 patent is in the same family as U.S. 9,990,628 ("Two-Level Authentication for Secure Transactions"), which is asserted in the Intel case and appears in other Proxense suits. Any invalidity/licensing posture developed on the '628 may therefore be strategically relevant, but the patents have distinct claim sets.
• Google Patents "family litigation" flags: The page lists a Darts-ip family reference (family 38694426) and a Texas Western District Court case, both pointing to the same 7:25-cv-00531 action.

What I could not confirm

1. The specific claims of the '919 patent asserted against Intel (the complaint attaches an "'919 Claim Chart" as Exhibit I, but the claim numbers were not in the retrieved text).
2. Any settlement, dismissal, transfer, stay, or judgment in 7:25-cv-00531 as of the search date.
3. Any earlier assertion of the '919 (e.g., pre-2025) — none surfaced, but a complete PACER history was not run.
4. Any IPR, PGR, reexamination, or ITC proceeding specifically on the '919 patent.

If you need a litigation-grade answer, the next step would be a paid Docket Navigator / Lex Machina / Darts-ip search on the patent number plus a PACER docket pull for 7:25-cv-00531.

I'll start by checking whether any AIA trial proceedings exist for this patent, since the structured block defaults to "none on file."

The broad searches surfaced many Proxense IPRs — but mostly on related patents ('954, '730, '905, '989, '042, '700). I need to pin down whether any proceeding actually targets 8,433,919 itself.

No hit yet on the '919 itself. Let me verify directly and check the Intel IPRs (IPR2025-00327/-328/-329) which the "settled expectations" commentary flags.

Proceedings overview

Zero AIA trial proceedings are on file against US 8,433,919. The canonical structured list from the USPTO Open Data Portal returns no IPR, PGR, or CBM naming this patent, and my independent web searches surfaced none either — every Proxense PTAB proceeding I found challenges a different patent in the same family ('730, '954, '905, '989, '188, '289, '042, '700). Accordingly the breakdown is: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0. The defensive posture this creates is unusual and should be read carefully: no claim of the '919 has been tested, narrowed, or canceled at the PTAB — the patent is "untested," not "hardened." A defendant gets neither the gift of a prior FWD killing asserted claims nor the reassurance of a patent that has survived Board scrutiny; it gets a blank slate with one significant procedural headwind (the patent issued 2013-04-30 and carries a 2005-11-30 priority date, which engages the Director's current "settled expectations" discretionary-denial doctrine).

No proceeding to report

There is no IPR/PGR/CBM number to list for US 8,433,919. I checked the following before concluding:

• The structured ODP "PTAB proceedings on file" block (canonical): empty.
• Searches for "8,433,919" IPR, "Two-Level Authentication" 8433919 IPR, and Proxense-specific PTAB dockets: all hits were sibling patents.

Conflation warning. Two unrelated patents share the trailing digits "919" and pollute naive searches — do not import their outcomes into any analysis of this patent:

• US 9,978,919 (a light-emitting device / lead-frame patent, IPR2020-00151) — different patent, different owner, different art unit.
• An unnamed " '919 patent" litigated in E.D. Tex. concerning NDPA/CSI-feedback wireless sounding (claim constructions at 1/13/2023) — this is not Proxense's '919.

One unverified lead you should chase. In the Proxense v. Intel campaign, Intel filed petitions IPR2025-00327, -00328, and -00329 (Director-level decision at Paper 12, 2025-06-26). I could not confirm within my search budget which patents those three petitions challenge. Since Intel is accused on six Proxense patents — including the '919 — it is plausible but unconfirmed that one of them targets the '919. Verify each petition number against PTAB E2E before relying on the "zero proceedings" conclusion. I am not asserting they cover this patent.

Adjacent Proxense-family proceedings — context only, NOT US 8,433,919

Useful as pattern evidence; cite none of it as a holding about the '919.

Verified details worth noting because they recur: the Samsung denials were claim-construction driven — the Board found Samsung had not shown a reasonable likelihood of prevailing "based on a claim construction position advocated by Proxense in its Preliminary Response" (quoted from Samsung's own motion exhibit). The '954 FWD panel was APJs Dang, Turner, and McKone; that FWD rested on Ludtke (§ 103(a)) for claims 1, 2, 4–7, 10, 12, 13, 15, 16, 18, 19, 22–27 and Ludtke + Kon (§ 103(a)) for claims 3, 14, 17. The same panel instituted IPR2024-00846.

Strategic summary

Claim status of US 8,433,919: all 37 claims are UNTESTED at the PTAB. None canceled, none sustained, none narrowed by reexamination that I could find. The patent fronts 37 claims with independent claims at 1, 13, and 19 (method / apparatus / PDK, per RPX's claim listing); the '919 family continuation US 9,990,628 carries the identical title "Two-Level Authentication for Secure Transactions" and is asserted separately against Intel — do not let a claim chart or invalidity contention for the '628 be mistaken for one against the '919. The asserted claim is also inconsistently reported across secondary sources: the Proxense v. Intel complaint analysis (7:25-cv-00531, W.D. Tex.) says independent claim 25 is asserted, while RPX reproduces claim 1 as the first claim. Confirm from the complaint and Exhibit I ("919 Claim Chart") rather than from either secondary source.

Estoppel landscape is empty — and that cuts both ways. Because no IPR on the '919 has reached a final written decision, no petitioner or privy is estopped under 35 U.S.C. § 315(e)(2) from raising any ground against this patent in district court or the ITC. Equally, nothing has been taken off the table: every § 102/§ 103 ground is theoretically open. Note two limits: (i) IPR estoppel is patent-specific, so the FWDs against the '954, '730, and '905 create no estoppel as to the '919, even for Apple, Google, Microsoft, or Samsung; and (ii) the Samsung denials were pre-institution denials, so under SAS/post-SAS district-court practice there is live dispute about whether "non-instituted grounds" are estopped — but again, that fight concerns the '730/'905/'989, not the '919.

Pattern signals. Proxense is a serial enforcer (Samsung, Google, Microsoft, Apple, LG, Intel) whose family has drawn at least a dozen IPR petitions from five different petitioners — yet the '919 has attracted none. The most likely explanation is timing: the '919 appears to have entered the campaign late, via the Intel complaint filed 2025-11-14, meaning no § 315(b) clock had run on anyone earlier. Proxense also litigates the PTAB aggressively on the back end — it is the appellant in a consolidated Federal Circuit appeal from IPR2024-00233, IPR2024-01334, and IPR2024-00846 (all '954), arguing APA violations from the Board's stay of a parallel ex parte reexam, a claim-construction error on "a third party that operates a trusted authority," and an Appointments Clause defect from Director-review delegation to a senior APJ. In parallel it has used ex parte reexamination as an offensive hedge (90/015,052–054), which is a notable asymmetric-playbook signal for any defendant. Reported settlements with Samsung, Google (2025-01-24), and Intel (2026-03-24) show Proxense will settle — but note the patent's adjusted expiration of 2029-10-06 leaves under four years of term, which shortens both the damages runway and the value of a long IPR fight.

Recommended next steps

1. Verify the negative before relying on it. Run each of the four Intel petition numbers (IPR2025-00327, -00328, -00329, and any companion) through PTAB E2E (https://ptacts.uspto.gov/ptabweb/) and the PTAB public-information pages, and check the patent's Google Patents "Litigation"/"Trials" tabs at https://patents.google.com/patent/US8433919/en. If a petition on the '919 exists, everything below changes — prioritize reading its institution decision and scheduling order for the FWD due date.

2. If you are a defendant facing an '919 assertion, the clock is the whole ballgame. Under 35 U.S.C. § 315(b) an IPR petition must be filed within one year of service of the complaint on you. For anyone served in the Intel action (filed 2025-11-14), that window is running now. Miss it and IPR is off the table for you permanently.

3. Budget for the "settled expectations" headwind. This patent issued 2013-04-30 — roughly thirteen years of settled public expectations. Under Interim Director Coke Morgan Stewart's discretionary-denial regime (as articulated in Intel Corp. v. Proxense LLC, IPR2025-00327, -00328, -00329, Paper 12 (2025-06-26), and applied in iRhythm v. Welch Allyn, Dabico v. AXA, Kahoot! v. Interstellar, among others), an old, previously-unasserted patent is a strong candidate for § 314(a) denial. The two arguments that have actually worked are material examiner error during prosecution (by far the most successful, per the reported dockets) and, less reliably, the patentee's failure to commercialize/assert/mark in the petitioner's technology space. Build the petition around a documented prosecution lapse, and expect to have to brief discretionary denial even if the merits are strong. Note that IPR2025-00562 (Apple v. Proxense, '188) shows the Board running this referral process in the live Proxense campaign.

4. Litigate the claim-construction posture in parallel, not just the art. Samsung lost three petitions in 2022 on construction alone, and Proxense has now obtained favorable constructions from the same W.D. Tex. court formula in Samsung (6:21-cv-00210), Google (6:23-cv-00320), Microsoft (6:23-cv-00319), and Apple (6:24-cv-00143). Terms recurring across the family — "tamper proof format," "access message," "verification unit" (Apple argued § 112 ¶ 6), and the "third party that operates a trusted authority" limitation at issue on appeal — are where this family has been won and lost. For the '919 specifically, the claimed architecture (local biometric sample match, then secure channel to a remote registry for a validity decision) invites both § 112 indefiniteness on "sample"/"transformed information" and § 103 art on two-tier local/remote biometric verification.

5. Reduce reliance on PTAB outcomes in sibling cases as a proxy. The '954 FWD (IPR2024-00233) is the family's headline loss and will be cited against you in negotiation; be ready to explain that it rests on Ludtke art and claim language specific to the '954's "persistent storage / device ID code / secret decryption value" claims, and is on appeal to the Federal Circuit. Its persuasive value against the '919's "biometric profile sample + remote registry validation" claims is limited. Canonical FWD source: PTAB E2E; a hosted copy of IPR2024-00233 is at https://bannerwitcoff.com/wp-content/uploads/2025/07/IPR2024-00233.pdf. Docket tracking: https://www.docketalarm.com/cases/PTAB/IPR2025-00562/Apple_Inc._v._Proxense_LLC/.

I'll research this systematically — starting with the assignment records, the Proxense ownership history, and the litigation referenced on the patent page.

Good — I have context on Proxense's litigation posture. Now let me dig into the actual assignment records and corporate history.

Let me try to pin down the actual USPTO assignment records and any security interests or chain transfers.

I hit the tool-call limit before I could pull the raw USPTO Assignment Center record, so I'll be explicit below about what is verified versus unverified. I could not retrieve the USPTO Assignment Center reel/frame numbers for US 8,433,919 in this session, and I will not invent them.

Inventors

• The three-inventor list is confirmed by the '919 assignment record's assignor field on Google Patents legal events ("Assignors: BROWN, DAVID L., HIRT, FRED S., GIOBBI, JOHN J.") and by the sister patent US 9,542,542 (Appl. 11/745,249, filed 2007-05-07), which lists the identical inventive entity and assignee "Proxense, LLC, Bend, OR."
• Unusual pattern check — not present. There is no evidence of inventors departing within 12 months of filing. Giobbi is still identified in 2024–2026 court and PTAB filings as Proxense's CEO and controlling principal (RPX describes Proxense as "inventor-controlled"; the 2024 Microsoft complaint states "John Giobbi is the founder and CEO of Proxense"). This is the inventor-controlled-owner pattern, not the inventor-flight pattern.
• One caution on name matching: "Fred S. Hirt" also appears as an inventor on US 6,785,474 (optical communications). I cannot confirm those are the same individual, so I do not treat it as a finding.

Original assignee

Proxense, LLC — an Oregon limited liability company. Named as assignee on the face of the issued patent ("Assignee: Proxense, LLC, Bend, OR").

• Primary line of business: Secure proximity-based authentication — the "Personal Digital Key" / Reader system, biometric payment, access control, and location tracking. Founded 2001 by John Giobbi; R&D in West Palm Beach, FL, with company presence in Bend, OR.
• Did it ship a product embodying the claims? No evidence of a shipped, revenue-scale product. Contemporary reporting (The Bulletin, May 7, 2008, "A pint-sized virtual wallet") describes the technology as "ready for deployment" and notes a deal to supply tracking technology to Bally Technologies — but that is a development/OEM arrangement, not a commercial product. Proxense's own 2024 complaint pleads that from ~2004–2012 it "developed, inter alia, mobile payment technologies and commercial products, employing over thirty engineers, and investing many millions of dollars" — framing that is retrospective and litigation-oriented.
• Current status: operating as an enforcement/licensing vehicle; no product on the market. Independent third parties classify it as non-practicing: Unified Patents calls it "Proxense LLC, an NPE" and its PTAB portal tags Proxense "NPE (Small Company)"; the Stanford NPE litigation database lists Proxense with NPE status "Failed startup"; ip fray describes it as "a non-practising entity (NPE) and former operating company." RPX describes it as "inventor-controlled."
• Corporate facts (Oregon registry, entity ID 785675-92): status ACT; manager John Giobbi (689 NW Stonepine Dr, Bend, OR); registered agent Corporation Service Company, 1127 Broadway St NE Ste 310, Salem, OR; mailing address 1120 NW Couch St 10th Fl, Portland, OR. A commercial registered-agent service is a weak tell on its own — many operating companies use CSC — so I do not treat it as a finding by itself.
• No bankruptcy found. I found no Chapter 7/11 filing for Proxense. However, in July 2016 counsel for Proxense sent letters to Apple and Microsoft offering to sell "its entire patent portfolio … in a single transaction to a single purchaser," stating Proxense "has never engaged in any patent enforcement litigation" and "there are no current licenses granted." That attempted portfolio-wide liquidation is the closest analogue to a distressed sale — but it evidently did not close, and Proxense still owns the portfolio.

Assignment timeline

Important limitation, stated plainly: The USPTO Assignment Center (https://assignmentcenter.uspto.gov/ and https://assignment.uspto.gov/patent/index.html) could not be queried in this session, so I do not have reel/frame numbers or recorded correspondents for this patent. I will not fabricate them. What follows is limited to assignment events I could actually corroborate from Google Patents legal events, the face of the patent, and litigation records.

• 2007-06-26 (recorded) — Reel not verified

  • Conveyance: Assignment of assignors' interest (original assignment)
  • Assignor: John J. Giobbi; David L. Brown; Fred S. Hirt
  • Assignee: Proxense, LLC (Bend, OR)
  • Correspondent: not verified — no reel/frame or correspondent data retrieved. (Note for follow-up: Patent Law Works LLP is the attorney of record on the '919 family, and the 2016 portfolio-sale letter to Apple was signed by a "Greg T." at patentlawworks.net, tel. 801-258-9821. That is prosecution/transactional counsel, not a confirmed assignment correspondent.)
  • Context: Founder/inventor formation of the holding entity — the standard "assign to my own company" link, not an acquisition.

• No post-issuance assignment of US 8,433,919 was found. All sources — the patent's own current-assignee field, the 2016 portfolio-sale letters, the 2021–2025 complaints (which name Proxense, LLC as plaintiff and owner), and the PTAB proceedings (which identify "Proxense, LLC, as the real party-in-interest") — show Proxense, LLC as the continuing owner from 2007 to date.

  • Because the Assignment Center record itself was not retrieved, I cannot rule out a recorded security interest, license, or corrective assignment against the Proxense portfolio. Treat the single-link chain above as probable but unconfirmed at the reel/frame level.

• Portfolio-level events that are NOT recorded assignments but bear on ownership context:

  • 2016-07 — Proxense offers its entire portfolio for sale to Apple (letter dated 2016-07-25, referenced in Apple's 6:24-cv-00143 record) and to Microsoft (2016-07-29). No transfer resulted.
  • 2013-04-05 — Priority claim to US 13/857,905 (which issued as US 9,999,628), i.e. intra-family continuation activity, not a title transfer.

Timeline diagram

timeline
    title Ownership of US 8433919
    2007 : Filed 05 May by Giobbi Brown Hirt
         : Assigned to Proxense LLC 26 Jun
    2013 : Patent issued 30 Apr
    2016 : Portfolio sale offer to Apple
    2021 : First suit filed vs Samsung
    2023 : Suits vs Google and Microsoft
    2024 : Suits vs Apple and Intel
    2025 : Suits vs Hyundai Kia Genesis
         : New West Texas case filed

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. No recorded movement from an operating assignee to a licensing-only LLC. Proxense, LLC is the original assignee of record from 2007 and has never assigned the patent away. Individually suggestive facts exist (single-purpose Oregon LLC, manager = the inventor, CSC registered-agent address, no product in commerce), but the signal as defined requires a transfer, and none is in evidence.

2. Known asserter in the chain — Present. Proxense is not on the enumerated list (Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities), but the instruction extends the signal to entities "surfaced by Unified Patents or RPX as a high-frequency plaintiff," and it qualifies: Unified Patents' PATROLL page states "U.S. Patent 9,251,332, owned and asserted by Proxense LLC, an NPE"; Unified's PTAB portal lists Proxense as "NPE (Small Company)"; RPX's New Patent Litigation coverage tracks Proxense as an "inventor-controlled" campaign with active W.D. Tex. suits against Alphabet, Apple, Microsoft, Intel, and (2025) Hyundai/Kia/Genesis. Stanford's NPE database lists Proxense as a "Failed startup" NPE. Asserted patents in the campaign include US 8,352,730, 9,049,188, 9,235,700, 9,298,905, 10,698,989, 8,646,042, 9,251,332, 8,461,043, 8,457,672, 9,265,043, 11,095,640, and 8,036,152. Precision caveat: US 8,433,919 itself does not appear as a patent-in-suit in the complaints I located. It is a family member in the same portfolio, not a confirmed asserted patent.

3. Repeat correspondent across the chain — Unclear / not verified. There is no assignment chain to test for recurrence, and I retrieved no reel/frame correspondent data. The recurring professional in this story is litigation counsel David L. Hecht, Hecht Partners LLP (New York), who has represented Proxense in the Samsung, Google, Microsoft, Apple, Intel, and Hyundai/Kia matters, and Patent Law Works LLP (Salt Lake City) on the prosecution/portfolio side. Neither has been verified as a recorded-assignment correspondent. I flag them for follow-up rather than as a finding.

4. Cascading transfers — Not present. No chained LLC-to-LLC transfers. One owner since 2007.

5. Pre-litigation transfer — Not present. The patent has been held by Proxense since 2007-06-26; the first suit in the campaign was filed 2021-03-05 (Proxense, LLC v. Samsung Electronics, 6:21-cv-00210, W.D. Tex.). No title transfer precedes any suit.

6. Bankruptcy fire-sale — Not present (no evidence). No Chapter 7/11 filing or court-supervised patent sale found. The 2016 attempted portfolio-wide sale is a distress indicator (failed startup seeking a single-buyer exit) but was private, unsold, and is not a bankruptcy proceeding.

7. Privateering — Not present. No operating company transferred patents to Proxense for Proxense to assert on its behalf; Proxense is the original owner asserting its own patents against large operating companies. This is the direct-assertion case, not privateering.

8. Defensive aggregator (anti-NPE) — Not present, and in fact inverted. The chain does not terminate at RPX, AST, LOT, Unified, or OIN. Unified Patents is actively challenging Proxense's patents (PATROLL prior-art bounties on US 9,251,332 and US 11,095,640), i.e. on the opposing side.

Verdict

NPE — moderate confidence.

Justification: exactly one strong signal is present — external asserter-directory classification. Unified Patents labels Proxense "an NPE" and "NPE (Small Company)," and RPX and Stanford's NPE database track it as an inventor-controlled/failed-startup non-practicing asserter running coordinated W.D. Tex. campaigns (Samsung 2021-03-05; Google and Microsoft 2023-05-02; Apple 2024-03-18; Intel 2024-05-23; Hyundai/Kia/Genesis 2025), while no Proxense product embodying the claims is in commerce. The chain-based signals that would elevate this to high confidence — a shell-entity transfer, cascading LLC-to-LLC conveyances, a pre-suit assignment, or a recurring recorded-assignment correspondent — are not present, because this patent has apparently never left its original 2007-06-26 owner Proxense, LLC. I am therefore flagging moderate, not high, pending confirmation of the Assignment Center record.

Verification link: search patent 8,433,919 at https://assignmentcenter.uspto.gov/ (mirror: https://assignment.uspto.gov/patent/index.html) — the specific items still outstanding are (a) the reel/frame of the 2007 assignment from Giobbi/Brown/Hirt to Proxense, LLC, (b) the recorded correspondent on that document, and (c) confirmation that no security agreement, license, or later corrective assignment is of record against the '919 patent. The patent's Google Patents legal-events page (https://patents.google.com/patent/US8433919/en) shows only the 2007-06-26 Proxense reassignment and lists Proxense LLC as current assignee with adjusted expiration 2029-10-06, which is consistent with the single-link chain above.

I'll research US patent 8433919 and its prior art citations. Let me start with searches.

Let me retrieve the complete citation lists from the patent.

Let me get the complete reference lists from multiple sources.

Prior-Art Analysis — US 8,433,919 B2

1. Patent identification (confirmed)

I searched publicly accessible mirrors of USPTO data (Google Patents, FreePatentsOnline, Justia, Unified Patents) rather than PatentCenter/PatFT directly, since I have no direct USPTO portal access in this session. I did not substitute any similar number.

Two data-quality flags I will not "auto-correct":

• Google Patents lists the priority date as 2005-11-30 (via US 11/292,330), while Unified Patents' record for the same document shows a priority date of 2005-11-29. Both are reported here literally.
• Because the application was filed May 5, 2007, this is a pre-AIA patent, so pre-AIA 35 U.S.C. § 102(a), (b), (e), (g) governs, not the AIA first-inventor-to-file provisions.

Sources: https://patents.google.com/patent/US8433919/en ; https://www.freepatentsonline.com/8433919.html ; https://patents.justia.com/patent/8433919

2. What the claims require (baseline for the § 102 mapping)

The independent claims are 1, 13, 19, 25, and 34.

• Claim 1 (method) is the "two-level" core: a biometric profile sample (based on less than the entirety of a biometric profile) stored in PDK memory → receive biometric input → compare → plus establish a secure channel to a remote registry administered by a trusted third-party, transmit PDK info, receive a validation decision, and authorize based on both.
• Claim 13 — apparatus (biometric reader + receiver/decoder circuit + processor + network interface to remote registry).
• Claim 19 — the PDK itself (tamper-proof memory storing biometric profile, a sample, and a unique PDK ID; transceiver; control logic). Notably, claim 19 does not require a remote registry.
• Claim 25 — method using a portable key, again requiring registry validation.
• Claim 34 — method requiring registry validation and profile-type compatibility checking.
• Dependents 2–12, 14–18, 20–24, 26–33, 35–37 add fingerprints/other biometrics, challenge-response device authentication, PIN profile, picture profile, purchasing info, hashing, etc.

Therefore: any single reference that anticipates claim 1, 13, 25, or 34 must disclose both the biometric-sample comparison and the remote-registry validation layer. Claim 19 is the only independent claim that does not, so it is the broadest target for single-reference anticipation.

3. Backward citations (prior art cited on/of record)

The following U.S. patent documents appear in the "US Patent References" list for 8,433,919 (FreePatentsOnline record). Caveat: the retrieved list is truncated, so this is not guaranteed exhaustive, and I could not retrieve the foreign-patent-document list or non-patent literature in this session.

(Rows 1, 3, 5, 10, 14, 17, 20, and 26 are cited but, on their face, look like background/§ 103-fodder rather than standalone § 102 anticipation; the truncated record prevents certainty.)

4. Reading of the citations — most probative references

Because claim 1/13/25/34 each require both a biometric-sample match and a remote-registry validation, a single-reference § 102 anticipation is unlikely for those claims. My preliminary view:

• Not credible as single-reference anticipators of the independent claims: items 1, 3, 6, 10, 14, 16, 17, 20, 26. These appear to be cited as technological background or as secondary references for § 103 combinations.
• Element-level relevance (likely cited for specific limitations, possibly under § 102(e) or § 103):
  • US 7,574,734 (Fedronic) — the "sequentially processing a biometric sample" reference is the closest facial match to the claimed "biometric profile sample … based on less than the entirety of the biometric profile" (claims 1, 13, 19, 25, 34).
  • US 2005/0251688 (Nanavati) — "central biometric authority" is the closest facial match to the claimed trusted-third-party remote registry (claims 1, 7, 34).
  • US 6,983,882 (Cassone) and US 6,973,576 / 7,305,560 (Giobbi) — closest to the portable, tamper-proof, biometric-bearing "PDK" of claims 19 and 25.
  • US 7,529,944 (Hamid) and US 7,191,466 (Hamid) — relevant to the multiple-profile-type / multiple-authentication limitations (claims 6, 11, 34–36).
• Timing caution: several cited documents post-date the 2005-11-30 priority date (e.g., 7,644,443; 7,715,593; 7,883,417; 7,904,718; 2009/0328182; 2010/0117794; 2011/0126188). Those cannot be § 102(b) art as to the 2005 priority subject matter; if relied on, they would have to be § 102(e) art (U.S. applications/patents with earlier effective filing dates) or applied to the new matter added in the May 5, 2007 CIP.

Bottom line on § 102: on the record I can retrieve, no single cited reference discloses every limitation of independent claims 1, 13, 25, or 34. Claim 19 (PDK with tamper-proof memory storing a biometric profile and a biometric profile sample and a unique PDK ID) is the only independent claim without a remote-registry requirement, and is therefore the claim most exposed to a single-reference § 102 attack — most plausibly from the Cassone (6,983,882) or Giobbi (6,973,576 / 7,305,560) devices, depending on whether they store a sample rather than the full template.

5. Explicit limitations of this analysis

1. The FPO backward-citation list was truncated; the foreign patent documents and any non-patent literature (e.g., the "Touch Pay"/"TruProx" provisional materials) were not retrievable here.
2. I could not retrieve the full text of the cited references in this session, so the claim mappings above are based on titles, abstracts, and known subject matter — they are preliminary, not element-by-element claim charts. A definitive § 102 determination requires side-by-side comparison of each reference's disclosure against each claim limitation.
3. Some of these references were very likely cited under § 103 (obviousness) or as § 102(e) art rather than as anticipatory § 102(a)/(b) art; the face of the patent shows the references but not the statutory grounds the examiner applied.
4. Forward citations (documents citing 8,433,919, e.g., US 2015/0046707, US 8,506,336, US 8,412,949, US 9,541,542) are not prior art to this patent and were excluded, except to note they exist.

If you want, I can next pull the full text of the top candidates (US 7,574,734; US 6,983,882; US 2005/0251688; US 6,973,576; US 7,305,560) and build element-by-element § 102 claim charts against claims 1, 19, and 34.

I'll gather the prior-art record for this patent — both its face-of-patent citations and the art being asserted against it in the current campaigns — before building the § 103 combinations.

Good — the patent's own face-of-patent art and the sibling-patent IPR record are surfacing. Let me pin down the key candidate references precisely.

Scope note before the analysis

The "Prior Art section of this page" was not present in the materials provided to me. The user message contained only the (1) patent summary and (2) litigation summary already generated. I therefore could not "use the results from the Prior Art section" literally. Rather than halt, I reconstructed the prior-art record from two verifiable sources and flag exactly what is and isn't grounded:

• (a) The '919 patent's own face-of-patent reference list (as mirrored on FreePatentsOnline and Google Patents), and
• (b) The art actually adjudicated against this patent's own sibling, U.S. 8,886,954 / U.S. 9,542,542 (same Nov. 2005 priority, overlapping specification), in the Proxense IPR campaign.

Important caveat on (a): the "References Cited" list surfaced by aggregators (FreePatentsOnline/Google) mixes the IDS of this patent with the IDS of later same-title continuations (US 9,990,628, US 11,157,909). That is why items such as US 2011/0126188 (Bernstein, May 2011) and US 2010/0117794 (Adams, May 2010) appear even though they cannot be §102/§103 art against a Nov. 2005–May 2006 priority. I have segregated those out below and flagged each reference's date-sensitivity. Do not treat any item on that aggregated list as a §103 reference without confirming its own filing/priority date.

Obviousness analysis — U.S. 8,433,919 under 35 U.S.C. § 103

1. Framework and level of ordinary skill

The Graham factors govern: (1) scope/content of prior art, (2) differences from the claims, (3) level of skill, (4) secondary considerations. Graham v. John Deere, 383 U.S. 1, 17–18 (1966); KSR Int'l v. Teleflex, 550 U.S. 398 (2007).

Level of ordinary skill (POSITA). As the Board found in the sibling-patent IPRs, "a skilled artisan would have been an experienced electrical or computer engineer" (IPR2024-00846, Final Written Decision, §1). To that I would add 2–4 years of experience with one or more of: short-range wireless/RF proximity systems, biometric capture and template matching, and payment/access-control authentication. This is a low-to-moderate skill level by the standards of §103; the claims are essentially combinations of known building blocks.

Critical effective-date point. The '919 claims priority to five provisionals filed May 5, 2006 through Mar. 13, 2007 (60/798,172; 60/798,843; 60/838,788; 60/824,758; 60/894,608), with Google Patents additionally showing a "priority claimed from US 11/292,330" (2005-11-30) and Unified Patents listing the '919 priority as 2005-11-29 — the same unresolved conflict flagged in the earlier summary. This matters for §103: a later effective date (May 2006) makes more references qualify; an earlier one (Nov. 2005) narrows the field but does not exclude the backbone references (Burger, Robinson), which pre-date both.

2. The claim elements that must be mapped

Stripping the five independent claims (1, 13, 19, 25, 34) to their substantive limitations:

Dependent claims to cover: 4/5 (challenge-response device auth on entry to a proximity zone), 6 (profile-type sufficiency), 8/9 (purchasing info + remote validation database), 10 (picture profile + administrator), 11 (PIN), 12 (hashing), 14 (credit-card-terminal I/O), 15 (multiple purchasing media/default), 17 (encrypted transmission), 20 (programmer/notary/site IDs), 23–24 (form factor), 37 (registry-profile authentication).

3. Ground 1 — Burger is the primary reference and reaches most of the claim set

Reference: Burger et al., US 2005/0050367 A1 (publ. Mar. 3, 2005) — the published application of US 7,003,495 B1 ("Portable electronic authorization system and method," Chameleon Network Inc., filed Sept. 28, 2000; granted Feb. 21, 2006). This is the reference the Board relied on to hold all challenged claims of the sibling '954 patent unpatentable (IPR2024-00233, Google v. Proxense — final written decision; IPR2024-00846, Microsoft v. Proxense). KSR asks whether the reference's subject matter as a whole suggests the claimed combination, and Burger's "Pocket Vault" does.

Element mapping (Burger alone):

Motivation to combine / why obvious: Burger's own stated purpose is preventing unauthorized use of a portable wallet by ensuring "the individual… is associated with the access device" — precisely the problem the '919 background recites as unmet. Registering the chip ID with the server to enable revocation of lost/stolen devices is Burger's express anti-fraud rationale, which supplies the motivation for the remote-registry level. No physical incorporation of one reference into another is required; this is a single-reference ground for the majority of the limitations.

4. Ground 2 — Burger + Fedronic supplies the "profile sample = less than the entirety" limitation (B)

Limitation B is the most distinctive one and appears in every independent claim. Burger compares the entire stored fingerprint. The gap is filled by:

Reference: Fedronic et al., US 7,574,734 B1, "System and method for sequentially processing a biometric sample" (grant 2009-08-11; cited on the '919 face). This reference is directed to processing a biometric sample in stages/sequence — the classic coarse-then-fine matching paradigm in which a partial/derived sample is evaluated first, with full-template comparison only if the reduced test passes.

Motivation to combine: KSR recognizes that combining known techniques for their known, predictable advantages is obvious. Fedronic's staged sample processing yields exactly the advantages the '919 specification itself claims for its "sample": faster comparison and less data transmitted over the wireless link (the '919 spec: the sample "can be used for quick comparisons" and may be sent over a public/reduced-encryption channel while the full profile goes over the secure channel). Burger's Pocket Vault communicates over a wireless interface subject to bandwidth and power constraints; a POSITA optimizing that link for speed/low power — as Burger invites by making its controller "software-programmable" and its components capable of "numerous forms" — would predictably pre-screen with a subset sample. Note that this exact efficiency rationale ("unused information would be inefficient and wasteful… transmitted across a connection with limited bandwidth") was accepted by the Board in the Burger/Cheng IPRs on the sibling patents.

Result: Burger (A, C, D, E, F) + Fedronic (B) renders claims 1, 13, 19, 25, 34 obvious.

5. Ground 3 — Burger + Robinson / Nanavati reinforces the remote-registry "second level" (D, E)

If Patent Owner disputes that Burger's network server 114 is a "registry… administered by a trusted third-party organization with which the PDK is [registered]" (claim 1) — arguing it is a payment processor rather than a third-party authentication registry — the second level is independently supplied by:

• Robinson, US 2003/0177102 A1 ("System and method for biometric authorization for age verification," publ. Sept. 18, 2003), which teaches transmitting a biometric sample + a unique System ID (SID) to a central/verifier database, performing the match/authorization there, and returning an "AGE VERIFICATION APPROVED / DECLINED" decision (Figs. 3–4). Robinson thus teaches the server-side validation decision keyed to a device/user ID, with a registration step (Fig. 2) — the registry-relationship limitation.
• Nanavati et al., US 2005/0251688 A1 ("Identity verification method using a central biometric authority," publ. Nov. 10, 2005; cited on the '919 face) — squarely teaches verification at a point of transaction against a central biometric authority. Date-sensitivity: published Nov. 2005, so it qualifies only if the '919's effective date is not earlier than its own date; it is safe art if the date is the May 2006 provisional.

Motivation: Burger expressly establishes a "trust relationship" between the Pocket Vault and the network server via chip-ID registration (¶ 114). Substituting (or supplementing with) Robinson's/Nanavati's third-party authorization authority to obtain an independent, centrally-maintained revocation/validation layer is the predictable application of a known technique to a known device (KSR), improving exactly the anti-fraud property Burger already seeks.

6. Ground 4 — Burger + Cheng (+ Adams) covers claim 34's profile-type compatibility (F) and claim 6

Reference: "Cheng" (portable authentication device with selectable, graded "authentication levels"; relied on with Burger in IPR2022-01135 and IPR2024-01486 — I could not pin Cheng's publication number from the retrieved sources and do not assert one here). Cheng teaches a user selecting among alternative authentication factors, and assigning different quality/level values to biometric vs. PIN methods.

Claim 34 requires the Reader to receive profile-type information from the PDK and determine whether those types are compatible with allowable authentication types. Burger's Pocket Vault offers plural authentication methods; Cheng teaches making the method/level explicit and selectable. A POSITA would combine them to implement a capability handshake (does this token support the authentication the Reader requires?) — a routine, well-understood design step, and the claim's own "manual" (claim 6) frames it as a sufficiency check.

Adams, US 7,715,593 (biometrically-enabled multi-purpose credential management devices) is corroborating background for multi-profile credential devices. Date caution: its publication dates (2010) post-date the '919's priority; it belongs in the background/obviousness-of-design-choice discussion, not as a standalone §102/§103 reference, unless its own priority pre-dates Nov. 2005.

7. Ground 5 — Burger + Orsini covers the "both-required" combination logic and multi-factor weighting (E)

Reference: Orsini, US 2004/0049687 A1 (publ. Mar. 11, 2004) — the "trust engine" reference the Board used alongside Burger against the sibling '954 patent (IPR2024-00846, Grounds 2/3). Orsini teaches combining multiple concurrent authentication instances (fingerprint + smart card + password) and aggregating their reliabilities into a single confidence level (e.g., multiplying reliabilities; ¶¶ 238–240), plus enrollment-time trust evaluation of the data source.

Motivation: Burger already performs a sequential two-stage check (local biometric → remote ID); Orsini supplies the express teaching that two independent factors should both be satisfied / combined to reach a required confidence level. That is precisely claim 1/13/25/34's requirement that authorization turn on both the profile match and the registry decision. Orsini also supports claim 20 (trusted-initialization / enrollment provenance) via its enrollment-reliability discussion.

8. Dependent-claim coverage summary

9. Why a POSITA would have combined these references (articulated rationales)

1. Known problem, known solution. The '919 background admits the deficiencies of centralized biometric enrollment and PINs — "exposing an individual's personal biometric information to theft," and repeated re-enrollment. Burger's distributed, onboard-stored biometric template with remote device-ID validation was expressly designed to solve the same problem. Where the specification frames the invention's purpose as meeting a known need, the motivation is supplied by the art and the admission.
2. Predictable improvement in anti-fraud. Burger's own trust relationship and chip-ID registration motivate adding an independent third-party validation layer (Robinson/Nanavati/Burger's server).
3. Efficiency/engineering desiderata. The subset-sample limitation (Fedronic) is a classic speed/bandwidth optimization prompted by Burger's wireless, power-constrained device.
4. Capability matching. Burger's plural authentication methods + Cheng's authentication levels make the claim-34 compatibility check an obvious interoperability feature.
5. Combination of independent factors. Orsini supplies the express teaching that multiple factors are combined to raise confidence.

10. Strongest supporting evidence already of record

The Board has already held the sibling '954 patent (same Nov. 2005 priority, substantially overlapping specification) unpatentable over Burger-based §103 grounds — all challenged claims in IPR2024-00233 (Google), following through in IPR2024-00846 (Microsoft). Because the '919 shares that specification, a Burger-anchored attack is the highest-probability §103 posture against the '919 as well. But the mapping is not identical: the '954 claim 1 recites a "device ID code" and a "secret decryption value"; the '919's distinguishing "profile sample = less than the entirety" limitation is absent from the '954 claim set. That is why Fedronic (or an equivalent staged-matching reference) is indispensable to any '919 invalidity ground — it is the one limitation Burger does not plainly disclose.

11. Where Patent Owner will fight, and what I could not verify

• "Registry" construction. Proxense successfully pressed a narrow construction of "third party that operates a trusted authority" in the '954 appeals (arguing the third party must be separate from the user/application parties). Expect an analogous narrow reading of the '919's "registry… administered by a trusted third-party organization with which the PDK is registered." Ground 3 (Robinson/Nanavati) is directed at this.
• "Less than the entirety" — the strongest non-obviousness argument. Fedronic is the counter; confirm its text actually discloses a subset (not merely sequential) sample, and that it pre-dates the effective filing date.
• "Tamper-proof" — Burger's write-once memory 212 helps, but the Board credited expert testimony that implementing the device ID in protected write-once memory was obvious (IPR2024-00846 at limitation 1.1). Expect the same here.
• Claim 10 (picture profile + administrator) is the weakest link. I could not identify a verified pre-priority reference and will not fabricate one; this claim may survive a Burger-only ground.
• Date hygiene: several references on the aggregated "References Cited" list (Bernstein 2011; Adams 2010; some grant dates) are later-art or mixed-family IDS entries and are not §103 art against the '919. Each candidate must be verified against its own filing/priority date before use.
• Priority-date conflict (2005-11-29 vs. 2005-11-30 vs. May 2006) remains unresolved from the earlier section; it does not affect Burger/Robinson but does affect the margins of Nanavati and any mid-2006 art.
• Secondary considerations: none of the earlier sections recorded objective indicia (commercial success, long-felt need, failure of others, copying with nexus). Proxense's background narrative (industry failure to adopt conventional biometrics) is an admission of the problem, not evidence of non-obviousness; no objective evidence supporting patentability is presently of record, so this factor does not currently weigh against obviousness.

Bottom line

The '919 claims are, on the present record, highly vulnerable to §103 as combinations of known elements:

• Primary ground: Burger (US 2005/0050367 A1 / US 7,003,495 B1) alone reaches limitations A, C, D, E and F of independent claims 1, 13, 19, 25, 34 and most dependents.
• The one unmet limitation — the subset "biometric profile sample" (B) — is supplied by Fedronic (US 7,574,734), with the motivation being speed and reduced wireless-transmission overhead.
• Reinforcing references: Robinson (US 2003/0177102 A1) and/or Nanavati (US 2005/0251688 A1) for the remote-registry/central-authority level; "Cheng" for the claim-34 profile-type compatibility check; Orsini (US 2004/0049687 A1) for the "both-factors-required" and trusted-enrollment limitations.
• Decisive extrinsic support: the PTAB's Burger-based final written decisions invalidating the same-family '954 patent.

Required next step for a filing-grade conclusion: verify each reference's own filing/priority date and quote its supporting passages (especially Fedronic for the subset-sample limitation, and a picture-profile reference for claim 10) from the actual documents, since my mapping of those passages rests on the aggregated citation lists and the sibling-patent IPR record rather than on full-text review of each reference.