US 10437976

I'll run targeted searches to verify the patent data and check for any CAFC 2026 docket activity.

Now let me check for litigation and CAFC 2026 docket activity specific to this patent.

Let me search specifically for litigation asserting the '976 patent and any 2026 CAFC activity.

Summary of U.S. Patent No. 10,437,976 (literal: 10437976)

Search scope note: I searched for the exact number 10437976 / 10,437,976. All authoritative hits return the same single document, and I found no results for look‑alike numbers that I could confirm as the object of your request. I also searched for CAFC 2026 docket activity naming this number and did not find any — see the litigation/uncertainty section below.

Bibliographic data (as confirmed by the patent text and secondary databases)

Sources: https://patents.google.com/patent/[US10437976B2](/patent/US10437976B2) ; https://portal.unifiedpatents.com/patents/patent/US-10437976-B2 ; https://www.patents-review.com/a/20190065721-biometric-personal-data-key-pdk-authentication.html

Uncertainty flag on term: Google Patents lists the anticipated expiration as 2025‑12‑20 (20 years from the 2005‑12‑20 filing of the earliest non‑provisional parent, 11/314,199). The patents-review.com page instead lists an "adjusted expiration" of 2038‑10‑25 (20 years from this application's own 2018 filing). These are inconsistent, and the correct term for a continuation claiming §120 benefit back to 2005 should be measured from the earliest non‑provisional parent — i.e., consistent with the 2025‑12‑20 figure. Do not treat either date as a legal conclusion.

Abstract (verbatim, per the fetched patent text)

"Systems and methods verifying a user during authentication of an integrated device. In one embodiment, the system includes an integrated device and an authentication unit. The integrated device stores biometric data of a user and a plurality of codes and other data values comprising a device ID code uniquely identifying the integrated device and a secret decryption value in a tamper proof format, and when scan data is verified by comparing the scan data to the biometric data, wirelessly sends one or more codes and other data values including the device ID code. The authentication unit receives and sends the one or more codes and the other data values to an agent for authentication, and receives an access message from the agent indicating that the agent successfully authenticated the one or more codes and other data values and allows the user to access an application."

Plain-language overview of the independent claims

Claim 1 — Method (reader-device side). The claim is drafted from the perspective of the reader device, not the biometric key:

1. Detect a user's computing device (the "integrated device"/biometric key) in proximity of the reader.
2. Establish a secure communication link between reader and computing device.
3. If the user passes biometric verification on the computing device itself (the device compares scan data from its own biometric scanner against stored biometrics of the legitimate user), receive over the secure link an identifier code stored in the device that uniquely identifies the computing device.
4. Send that identifier code to a trusted authority server for authentication.
5. Receive a message back from the trusted authority server confirming the code was successfully authenticated.
6. In response to that confirmation, allow the reader device to complete a transaction.

Claim 9 — System. The same six-step logic, claimed as a "computerized device including a reader device and a memory including instructions that, when executed, causes the system to…" perform the steps of claim 1.

Claim 17 — Computer program product. The same six-step logic, claimed as a computer-readable storage device containing a program that, when run on a computerized device including a reader device, performs the steps of claim 1.

Claim-drafting observation (technical, not legal advice): The abstract and specification emphasize a "plurality of codes and other data values comprising a device ID code … and a secret decryption value," but the issued independent claims recite only "an identifier code … uniquely identifying the computing device." The secret decryption value is not positively recited in the independent claims, and the older "access message received by an application" language that appears in sibling family members is likewise absent here — claim 1 instead recites "allowing the reader device to complete a transaction."

Dependent claims (brief):

• 2 / 10 / 18: reader sends a request for biometric verification to the computing device upon detecting proximity.
• 3 / 11 / 19: the secure link is wireless and established using encryption/decryption keys.
• 4 / 12 / 20: the identifier code is registered with the trusted authority server.
• 5 / 13: receipt of the identifier code indicates biometric verification succeeded.
• 6 / 14: biometric data may be fingerprint, palm print, retinal scan, iris scan, hand geometry, facial recognition, signature recognition, or voice recognition.
• 7 / 15: computing device may be a mobile phone, tablet, laptop, mp3 player, mobile gaming device, watch, or key fob.
• 8 / 16: reader device may operate on the same system as a casino machine, keyless lock, ATM, computer, or point-of-sale register.

Litigation and CAFC 2026 docket check — what I found and did not find

No CAFC 2026 docket entry naming U.S. Patent No. 10,437,976 was found. I want to be explicit about this rather than infer it. What the searches did surface is family-level and sibling-patent activity for Proxense, LLC:

• Google Patents flags this family (Darts‑IP family ID 36602354) as "Family has litigation." That is a family-level flag; it does not establish that the '976 patent itself was asserted in a case. (https://patents.google.com/patent/US10437976B2)
• The asserted patents I could positively identify in Proxense's district-court campaigns are siblings and cousins — not the '976: '730 (8,352,730), '954 (8,886,954), '905 (9,298,905), '989 (10,698,989), '042 (8,646,042), and '188. Sources include Proxense's complaints and claim-construction papers in Proxense v. Samsung, 6:21‑cv‑00210 (W.D. Tex.); Proxense v. Google, 6:23‑cv‑00320 (W.D. Tex.); Proxense v. Microsoft, 6:23‑cv‑00319 (W.D. Tex.); Proxense v. Apple, 6:24‑cv‑00143 (W.D. Tex.).
• 2026 Federal Circuit activity exists for Proxense — but for sibling patents. An ip fray report (April 2026) states Proxense filed its opening appellate brief arguing PTAB/APA error in IPRs of U.S. Patent No. 8,886,954 ("Biometric personal data key (PDK) authentication") — a different patent in the same family with the same title. (https://ipfray.com/ptab-abused-discretion-by-staying-[reexamination](/resources/patent-reviews)-that-favored-patentee-proxense-tells-federal-circuit/)
• Related appeals surfaced: Federal Circuit 25‑2130 (from Microsoft v. Proxense, IPR2024‑00405, '988/related), notice of appeal in IPR2024‑00782 filed 2025‑12‑22, and notices of appeal in IPR2024‑00573 / IPR2024‑01398. These again concern sibling/related Proxense patents, not '976.

Bottom line: I have high confidence in the bibliographic and claim information for US 10,437,976 because it comes from the patent's own full text. I have no authoritative evidence that the '976 patent is itself the subject of an IPR, a district-court suit, or a CAFC 2026 appeal; the litigation and 2026 appellate activity I located belongs to other members of the same Proxense patent family. If you need a definitive docket answer for the '976 specifically, that should be verified against USPTO Patent Center / PTAB APIs and the CAFC CM‑ECF docket directly — my searches (Google Patents, Unified Patents, Docket Alarm, Justia, UniCourt, ip fray) did not return such a record, and I will not assert one exists.

I'll run targeted searches on patent-litigation databases for the specific patent number 10,437,976.

Let me search the litigation-specific databases with the exact patent number.

Let me check the litigation case-list databases specifically and the Stanford NPE database.

Answer: Litigation involving U.S. Patent No. 10,437,976

Short answer: On the record I was able to retrieve, I found NO litigation — district court, PTAB/IPR, or Federal Circuit — in which U.S. Patent No. 10,437,976 ("the '976 patent") is itself asserted, challenged, or named. I am reporting that as a negative finding, not an inference, and I flag the limits of the search below.

What I searched and the literal-number discipline applied

Per the strict rule, I searched the exact string 10437976 / 10,437,976 and did not substitute look‑alike numbers. The patent number is a continuation in the Proxense "Biometric PDK" family (application 16/170,234, granted 2019‑10‑08, priority 2004‑12‑20). The full text confirms the family members are US 8,352,730; 8,886,954; 9,298,905; 10,698,989; 10,437,976; and the abandoned 16/551,243 (US 2019/0384903).

Positive results (the parent family IS heavily litigated — but with different numbers)

Every complaint, infringement contention, and PTAB paper I retrieved names sibling/cousin patents, never the '976. Representative record evidence:

PTAB/IPR activity likewise targets siblings, not the '976:

• Microsoft Corp. (and Apple) v. Proxense, IPR2024‑00573 — U.S. 8,646,042 ('042); Final Written Decision 2025‑08‑04 holding claims 1, 5, 6, 8–11, 13, 14 unpatentable under §103.
• Samsung Elecs. America v. Proxense, IPR2021‑01438 (U.S. 9,049,188) and IPR2021‑01349/01439 (U.S. 9,235,700).
• Google LLC v. Proxense, IPR2024‑00233 and IPR2024‑00846 — directed to U.S. 8,886,954 ('954).
• Other related appeals/petitions surfaced for '954, '989, and '730 family members (IPR2024‑00782, IPR2024‑01398, IPR2024‑00405).

None of the above names U.S. 10,437,976. The Google Patents banner "Family has litigation" (Darts‑IP family ID 36602354) is a family‑level flag and does not establish that the '976 was asserted. That is consistent with the summary already generated for this patent, which flagged the same distinction — I found no contradiction with that earlier section.

Bottom line

• Plaintiff(s)/defendant(s)/jurisdiction/case number/filing date/status for the '976: none found. There is no case record to report, so I will not manufacture case numbers or outcomes. The '976 patent's expected term has also lapsed ("Expired – Lifetime," anticipated expiration 2025‑12‑20 per Google Patents), which further reduces the practical likelihood of a new assertion.
• If you need a definitive docket answer for the '976, it should be confirmed directly against USPTO Patent Center, the PTAB API/P-TACTS, and the CAFC CM‑ECF / PACER dockets, and the Unified Patents litigation case list and Stanford NPE database filtered on patent number 10437976. My tool access was capped before completing those last targeted lookups, so I am expressly noting that limitation rather than asserting the negative as absolute.

(Distinction to keep sharp: "Proxense is a prolific litigant" ≠ "the '976 patent has been litigated." The litigation I can document is real, but it belongs to other numbers in the same family.)

I'll verify the structured "no proceedings" block against live PTAB sources before drafting.

PTAB / AIA Trial Proceedings — U.S. Patent No. 10,437,976 ("Biometric personal data key (PDK) authentication")

Verification note: I treated the structured "PTAB proceedings on file" block as the canonical source (USPTO Open Data Portal: no AIA trial proceedings for this patent), then ran independent web searches across Google Patents, Docket Alarm, UniCourt, Unified Patents, the PTAB public hearing schedules, PTAB Final Written Decisions surfaced in the record, and ip fray's Proxense coverage. Nothing in those searches names U.S. Patent No. 10,437,976 as the subject of an IPR, PGR, or CBM. The heavy Proxense PTAB activity that dominates search results is directed at other members of the same family. Details and caveats below.

---

Proceedings overview

Total AIA trial proceedings on U.S. Patent No. 10,437,976: zero (0). No IPR, PGR, or CBM — active, instituted, denied, settled, or decided — is on file for this patent as of 2026-09-30, which means there is no PTAB claim-level disposition, no estoppel running against any challenger, and all 20 claims (1–20, including independent claims 1, 9, and 17) remain UNTESTED at the Board; a defendant facing assertion of the '976 has no PTAB cancellation order to point at, but equally no PTAB precedent upholding it.

What I checked, and what I found instead

Search terms used included the exact numbers 10437976, 10,437,976, and '976 combined with "Proxense," "IPR," and each major petitioner name. I could not confirm any hit. Do not treat the absence as proof of non-existence — my searches were not a live PTAB E2E / PTAB API docket query, and very recently filed petitions (within the last ~30 days) can lag both the ODP ingest and web indexing. The correct negative-confirmation step is a direct E2E / PTAB API query by patent number (see Recommended next steps).

Important: I found no proceeding to format under the per-proceeding template. I will not manufacture proceeding numbers, panels, or claim lists for a proceeding that does not exist. What follows is family-adjacent context, explicitly labelled as not proceedings against the '976.

Family-adjacent proceedings (situational awareness only — these do NOT dispose of '976 claims)

These matter because the '976 shares a specification, an inventor, and a 2004‑12‑20 priority chain with the patents below, and the same prior-art families (Ludtke, Burger, Giobbi '139/'157, Broadcom, Dua) have been litigated at the Board.

Sources for the table: IPR2024-00573 FWD, IPR2024-00846 institution decision, IPR2024-00407 FWD, ip fray on the '730 reexam affirmance, ip fray on the Federal Circuit appeal of the '954 IPRs.

Two flags on that table:

1. A Bloomberg Law PTAB report surfaced dated 2026‑09‑21 stating the Board held claims of a Proxense "hybrid device having a personal digital key and receiver-decoder circuit" patent unpatentable as obvious in an Apple IPR. On the record I have, that description maps to the '042 family (Apple's IPR2024-01398/‑01399 joinders), not the '976. I could not positively confirm the patent number from the snippet, so treat it as a '042-adjacent data point with unverified identification, not as an '976 outcome. (Bloomberg Law)
2. The '954 FWD and the '730 reexamination certificate point in opposite directions on overlapping art; that conflict is the subject of Proxense's pending APA appeal and creates unusual uncertainty about how the Board would treat the same prior-art families against the '976.

---

Strategic summary

Claim status for the '976 itself: no claim is CANCELED, none is SUSTAINED, and all 20 are UNTESTED at the PTAB. Independent claims 1, 9, and 17 and dependents 2–8, 10–16, and 18–20 stand exactly as issued on 2019‑10‑08. That is an unusual posture for an NPE patent this heavily asserted across a family: the '976 sits inside one of the most IPR-attacked Proxense families and yet, on everything I can verify, was never itself petitioned. The plausible explanations — none of which I can confirm — are (a) the '976 issued after the waves of 2024 petitions and was not among the patents asserted in the W.D. Tex. complaints against Samsung/Google/Microsoft/Apple, (b) the '976 expires under the 20-year term running from the 2005‑12‑20 parent filing before a new IPR could realistically be tried to FWD (see the term/expiration inconsistency flagged in the earlier bibliographic section: Google Patents shows anticipated expiration 2025‑12‑20; patents-review.com shows an "adjusted expiration" of 2038‑10‑25), or (c) challengers chose the broader '730/'954/'905 claims for their petitions and simply never needed the '976. A petition filed now would face the March 2025 "Interim Process for PTAB Workload Management" settled-expectations screen, and the '976's 2004 priority and near-term expiration are exactly the profile that produced the Intel v. Proxense discretionary denial. That is a real, non-merits headwind for any new challenger.

Estoppel landscape: there is none against the '976. 35 U.S.C. § 315(e)(2) estoppel is claim-specific and attaches only to a petitioner (or its real parties in interest/privies) that obtained a § 318(a) FWD on that claim. Because no IPR reached an FWD on any '976 claim, no party is estopped from raising any § 102/§ 103 ground against claims 1–20 in district court. Conversely, a defendant here also gets no free ride: the '976's claims were not construed at the PTAB, and the only constructions in the record come from the W.D. Tex. Markman orders in Proxense v. Google (2024‑01‑23) and Proxense v. Microsoft (2024‑05‑24) and the Proxense v. Apple joint claim construction statement (2024‑12‑30) — which the PTAB panels in the sibling IPRs used as evidence. Note the reversal risk: because the '954 claims were held unpatentable while the '730's materially similar amended claims were held patentable in reexam, an '976 challenge built on Ludtke/Burger is exposed to the same "the Office already said the other way" attack Proxense is running on appeal.

Pattern signals. (i) The same three petitioners — Microsoft, Google, Apple — filed serial and joinder-style petitions across this family, frequently with the same expert (Andrew Wolfe, Ph.D.) and near-identical grounds, drawing discretionary-denial pushback from Proxense (IPR2024-01319 denied on copycat grounds; IPR2025-00074 denied). (ii) Proxense is pursuing PTAB appeals aggressively — the '954 APA appeal at the Federal Circuit, plus CAFC No. 25‑2130 from the '960 IPR and notices of appeal in IPR2024‑00573 and IPR2024‑01398/‑01399. (iii) No defensive aggregator is in the chain on the '976. Unified Patents has a portal page for US‑10437976‑B2 but I found no Unified-filed petition; the Proxense challenges are all Big-Tech defendant-filed. (iv) The campaigns are unwinding by settlement, not judgment: Samsung (2023), Google (2025‑01‑24), Intel (2026‑03‑24), with the Apple and Microsoft suits stayed.

---

Recommended next steps

1. Confirm the negative directly, before relying on it. Query PTAB E2E by patent number (10437976) at ptacts.uspto.gov and pull the PTAB API trial-proceedings endpoint by patent number. The ODP block and my searches are consistent, but only the E2E record is dispositive, particularly for anything filed in the last ~30 days. Search PTAB decisions at uspto.gov/patents/patent-trial-and-appeal-board/decisions and CourtListener's PTAB/Federal Circuit collections (courtlistener.com) for any "10,437,976" hit.
2. Because there is no FWD to cite, your invalidity case must be built, not borrowed. There is no PTAB disposition to quote and no estoppel to leverage. Two consequences: (a) you will need your own § 102/§ 103 analysis and expert declaration for claims 1–20, and (b) you can freely use art the Big-Tech petitioners used elsewhere in the family, since § 315(e)(2) does not bar you on '976 claims. The Ludtke/Burger line that killed the '954 and the Giobbi '139/'157 + Broadcom/Dua combinations that killed the '289 and '042 are the obvious starting templates — but they are not dispositive on the '976, whose independent claims recite an "identifier code … uniquely identifying the computing device," a "trusted authority server" round-trip, and "allowing the reader device to complete a transaction" — different limitations than the hybrid-device and access-message claims the Board has construed. Read the IPR2024-00407 FWD § 112(f) analysis and the '954 panel's "trusted authority" construction (discussed in Proxense's appellate brief) before you pick a claim-construction position — Proxense has litigated both and knows the terrain.
3. Mind the clock and the discretionary-denial screen. Two gating items if you are considering a petition: (a) § 315(b) — a one-year bar runs from service of a complaint alleging infringement of the '976, and the sibling-patent suits were filed in 2021–2024, so verify service dates and RPI/privy status carefully; (b) the 2025‑03‑26 Interim Process memo and Intel v. Proxense, IPR2025‑00327, Paper 12 (2025‑06‑26), which applied "settled expectations" to deny institution against a Proxense patent — anticipate the same argument here given the 2004 priority date, and prepare evidence that the '976 was not commercialized, asserted, marked, or licensed in your technology space, which is the factor that overcame the doctrine in later Director decisions.
4. If you are defending a demand letter today, the accurate posture to state is: "No IPR, PGR, or CBM has ever been filed against U.S. 10,437,976; no claim has been canceled or confirmed by the Board; all 20 claims are untested at the PTAB. Sibling patents in the same family have been both invalidated (the '954, the '289, the '960, the '042) and upheld (the '730 in reexamination), and those outcomes are on appeal." Do not assert that any '976 claim has been held invalid — no such holding exists — and do not cite the '954 FWD or the '730 reexam certificate as disposing of '976 claims.

Confidence statement: High confidence that no AIA trial proceeding names U.S. Patent No. 10,437,976, based on the ODP block plus corroborating searches; moderate confidence regarding very recent filings that may not yet be indexed. All statements above about the specific panels, grounds, and claim outcomes of the sibling proceedings reflect the source documents linked, not independent verification of the underlying PTAB files. I have not asserted the existence of any proceeding number that I did not find in a source.

I'll research the assignment record for US 10,437,976, starting with the USPTO assignment data and cross-referencing ownership sources.

Let me dig into the assignment records and legal events.

Let me try to locate the specific reel/frame records and any security agreements or later transfers.

US 10,437,976 B2 — Ownership Chain Reconstruction

Verification note (read first). The instructions ask me to work from USPTO Assignment Center (https://assignmentcenter.uspto.gov/), keyed on reel/frame, assignor, assignee, execution date, recording date, and correspondent of record. In this session my only retrieval channel was web search — I could not query the Assignment Center's record-level API directly, and the search index does not expose PROXENSE's reel/frame entries for this patent. I therefore confirm the existence, parties, and recording date of the one assignment below from the patent's own Google Patents legal-events record (which mirrors the USPTO assignment record), but I cannot report reel/frame numbers or the recording correspondent with confidence, and I will not invent them. That gap is flagged at each point where it matters. To close it, open https://assignmentcenter.uspto.gov/ → Search by Patent Number → enter 10437976.

---

Inventors

Observations:

• Single-inventor patent. The entire Giobbi family is inventor-controlled; the parent filings here name only Giobbi. (Contrast sibling PDK patents such as US 10,374,795, which name Giobbi + David L. Brown + Fred S. Hirt — those co-inventors are not on the '976.)
• No inventor-departure pattern. The classic fire-sale precursor ("all inventors leave the assignee within 12 months of filing") is absent. Giobbi is the assignee's control person (Manager of Proxense, LLC per the Oregon business registry), and he has continued to prosecute and assert the portfolio for two decades.
• The inventor→company transfer here is therefore a founder assign-back, not an arm's-length sale. That cuts against, not toward, an NPE signal on the assignment chain itself.

---

Original assignee

Proxense, LLC, Bend, OR (the entity named on the issued patent and the recorded assignee).

• Entity status. Active Oregon LLC. Oregon registry file no. 785675-92; registration date on file 2011-07-12 (an Oregon re-registration of a business founded in 2001); current registered address 689 NW Stonepine Dr, Bend, OR 97703; Manager: John Giobbi; status ACT (active). Source: https://www.city-data.com/business-entities/OR/PROXENSE-LLC-785675-92-OR.html
• Primary line of business. In its own litigation filings Proxense describes itself as founded in 2001, developing mobile-payment technologies and commercial products from roughly 2004–2012, employing "over thirty engineers," with a portfolio of "over 65 patents" covering proximity detection, multi-factor authentication, secure mobile data storage/exchange and access control. (Proxense press release, Mar. 8, 2021, reproduced in IPR exhibit materials; complaint in Proxense v. Microsoft, 6:23-cv-00319.)
• Did it ship a product embodying the claims? Unclear / historically asserted, not currently verifiable. Proxense claims product development in the 2004–2012 window, but I found no current commercial product, SKU, or revenue-generating offering in commerce; the company's visible 2021–2026 activity is licensing-and-litigation. Its own press materials characterize the portfolio as the asset, and Proxense's counsel describes the company as one that "will vigorously defend its IP." Treat "ships products embodying the claims today" as not established.
• Current status. Operating as a private, single-principal patent holding/enforcement company (not dissolved, not in bankruptcy). It has openly monetized via infringement suits since 2021.

---

Assignment timeline

Recorded chain as verified for application 16/170,234 (US 10,437,976):

• 2018 (execution date not confirmed) / recorded 2018-12-06 — Reel not retrieved (Assignment Center lookup required)
  • Conveyance: Assignment (assignment of assignors' interest — text recorded as "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
  • Assignor: GIOBBI, JOHN J.
  • Assignee: PROXENSE, LLC (Bend, OR)
  • Correspondent: Not confirmed from the record. I could not retrieve the recorded correspondent for this assignment. For context only — and not a substitute for the assignment correspondent — the prosecution agent of record on this family is Patent Law Works LLP, the same firm of record on sibling members (e.g., US 9,990,628 and US 10,374,795 both list "Attorney, Agent, or Firm — Patent Law Works LLP"). That is a prosecution recurrence across the Giobbi/Proxense family, not evidence of an assignment-recording repeat player.
  • Context: Routine inventor → operating-company transfer, recorded 42 days after the 2018-10-25 filing of the continuation. This is the standard chain-of-title record a continuation needs to keep standing clean; it is not an acquisition, fire-sale, securitization, or transfer-to-asserter.

No other records surfaced for this patent number. Specifically, I found no Security Agreement, Merger, Change of Name, License, Release, or Correction records for US 10,437,976 / 16/170,234 in the sources I could reach. Given Proxense's litigation program, a security interest (if any) would be an important find, but I have no evidence of one and will not assert one.

Family context (for the chain, not separately recorded against the '976): each continuation in the family carries its own Giobbi→Proxense record — 11/314,199 (2005-12-20) → US 8,352,730; 13/710,109 (2012-12-10) → US 8,886,954; 14/521,982 (2014-10-23) → US 9,298,905; 15/049,060 (2016-02-20) → US 10,698,989; 16/170,234 (2018-10-25) → US 10,437,976; and 16/551,243 (2019-08-26) → US 2019/0384903 A1, abandoned. There is no change of ultimate owner anywhere in the family — every link terminates at the same Proxense, LLC.

If the Assignment Center returns no records for 10437976 when you search it, then per the task framing the finding is simply that the original assignee retains ownership — which is exactly what the Google Patents legal-events entry indicates regardless.

---

Timeline diagram

timeline
    title Ownership of US 10437976
    2001 : Proxense founded in Bend Oregon
    2004 : Provisional applications filed by Giobbi
    2005 : Parent application 11 314 199 filed
    2008 : Parent patent 8352730 issues
    2012 : Continuation 13 710 109 filed
    2014 : Continuation 14 521 982 filed
    2016 : Continuation 15 049 060 filed
    2018 : Application 16 170 234 filed
         : Giobbi assigns to Proxense LLC
    2019 : US 10437976 issues
         : Continuation 16 551 243 filed
    2021 : Proxense sues Samsung
    2023 : Proxense sues Google and Microsoft
    2024 : Proxense sues Apple
    2025 : Anticipated expiry

---

NPE / troll-pattern signals

---

Verdict

NPE — moderate confidence.

The determination does not rest on the assignment record, which is a clean single founder→company transfer (GIOBBI, JOHN J. → PROXENSE, LLC, recorded 2018-12-06) with no shell entities, no cascading LLCs, and no pre-suit transfer. It rests on signal 2: the sole owner is a non-practicing enforcement entity — Proxense, LLC, Bend, OR, Manager John Giobbi — that has been a repeat Patent Owner in PTAB proceedings (Microsoft v. Proxense IPR2024-00405/-00776/-00782/-00846/-01320/-01328; Google v. Proxense IPR2024-00232/-00234) and a repeat district-court plaintiff (Samsung 6:21-cv-00210; Google 6:23-cv-00320; Microsoft 6:23-cv-00319; Apple 6:24-cv-00143; Hyundai/Kia/Genesis, Jan. 2025), with no currently verifiable product in commerce embodying the claims and revenue visibly derived from portfolio monetization rather than sales. Because the assignment chain itself shows none of the classic NPE transfer mechanics — and because the '976 specifically has not been identified in any Proxense suit or IPR I could locate — this is a moderate, not high-confidence, call on an entity-level basis.

Verification link: https://assignmentcenter.uspto.gov/ (search by patent number 10437976; the Google Patents mirror of the same legal event is https://patents.google.com/patent/[US10437976B2](/patent/US10437976B2)).

Open items to close with primary sources: (a) reel/frame and execution date of the 2018-12-06 Giobbi→Proxense record; (b) the recorded correspondent for that assignment; (c) whether any security agreement (a common encumbrance for litigating NPEs funded by litigation finance) is recorded against the '976; (d) USPTO PEDS/ODP and Patent Center to confirm no later assignee change. None of these appear in the sources I could reach, and I have deliberately not supplied placeholder reel/frame values.

I'll search the USPTO and related databases for this exact patent and its cited prior art.

Let me get the full cited-references list and verify the key biometric prior-art references.

Prior-Art Analysis — U.S. Patent No. 10,437,976 (literal "10437976")

Scope / method note. Everything below keys off the exact number 10,437,976 ("10437976"). I did not substitute look-alike numbers. The reference set is drawn from the authoritative full text you supplied (the patent's own "References Cited" section, which Google Patents totals at 273 citations), cross-checked by search against Google Patents / USPTO. I could not retrieve the complete 273-entry list (the fetched text truncates after US6667684), so this is the most relevant subset I could positively verify, not an exhaustive enumeration.

Important caveat on §102 vs §103. The '976 was allowed as a §120 continuation, so I found no examiner §102 rejection record for it specifically. Several references below are near-misses on claim 1's key limitation — the separate "trusted authority server" — so most of them function as §103 obviousness references, not clean §102 anticipations. I flag each accordingly and treat none as a legal conclusion.

---

1. The reference point: what claim 1 actually requires

Recall from the prior section that independent claim 1 (and claims 9, 17 in system/CPP form) requires, from the reader-device perspective:

• (a) detecting a user's computing device in proximity of the reader;
• (b) establishing a secure communication link;
• (c) responsive to successful biometric verification on the device itself (device compares its own scan data to stored biometric data of a legitimate user), receiving over that link an identifier code stored in the device that uniquely identifies the computing device;
• (d) sending the identifier code to a trusted authority server;
• (e) receiving a message back from the trusted authority server confirming authentication;
• (f) responsive to that message, allowing the reader device to complete a transaction.

The most relevant prior art must therefore touch biometric-verified portable devices AND an external authority that authenticates a code. That is the axis I used to rank references.

---

2. Most relevant cited prior art

Foreign patent documents also cited in the '976 record include WO 00/62505 A1, WO 01/22724 A1, WO 01/75876 A1, and WO 01/77790 A1 (all pre-2004 and thus candidate prior art), plus the family's own WO 2006/069330 A2 (which is the PCT counterpart of this invention, not prior art).

---

3. §102 anticipation assessment (reference-by-reference)

No single cited reference appears to disclose every limitation of claim 1, primarily because claim 1 requires the code to be sent to a "trusted authority server" distinct from the reader — a limitation the older fob/card art satisfies only by analogy (the "door" or terminal).

• §102 candidates (need element-by-element check):

  • US 6,041,410 (TRW fob) — maps to almost all of claim 1 except the separate trusted-authority-server step; the "door" is the verifier. If a court read "trusted authority server" broadly to cover the registered "door," it would be the strongest anticipatory candidacy for claims 1, 6, 7 (and system/CPP analogs 9/14 and 17). Its doubly-encrypted transmission (door public key + fob private key) is squarely on point for claim 3 / 11 / 19.
  • US 6,325,285 (AT&T smart card) — strong on claim 1 steps (a)–(c) and claims 6/7, weaker on the trusted-authority-server element.
  • US 2002/0140542 (Prokoski "Personal biometric key") — thematic name match; anticipation would depend on its actual disclosure (I did not verify element coverage), so I will not assert it.

• §103 references (to be combined): #1–#7 above are the natural combination set — a biometric-verifying portable device (#1, #2) in view of a networked/central authentication authority (#6 Bionetrix, #11 Sensar, #15 Swisscom) and a certificate/registered-credential model (#4 IBM). That combination reads on the full scope of claim 1 and its dependents.

• Not prior art / should not be used: the family's own WO 2006/069330 A2 and the sibling U.S. patents (8,352,730; 8,886,954; 9,298,905; 10,698,989) share the same 2004-12-20 priority and are not available as prior art against the '976.

---

4. Cross-check against the family's litigation/IPR art (context, not literal '976 citations)

The prior section established that the '976 itself has no confirmed IPR or CAFC-2026 docket, and that the Proxense IPRs target siblings ('954, '730). For completeness on what art the family has been attacked with:

• Search results show Google LLC IPR2024-00233 against US 8,886,954 (Exhibit 1001) and Microsoft IPR2024-01326 against US 8,352,730 (Exhibit 1001) — both siblings, same title. Their front pages list the same early references as the '976 (e.g., 5,187,352; 5,296,641; 5,392,453; 5,416,780).

Contradiction flag (important): The "Bradford, Foss, Yamane" combination that surfaces in search results belongs to CPC Patent Technologies' U.S. 8,620,039 ("biometric card pointer system") — a different patent and different family, litigated against Assa Abloy (CPC Patent Techs. v. Assa Abloy, Fed. Cir. Nos. 24-1492/24-1493, decided Nov. 10, 2025). It should not be imported into the '976 analysis as a cited reference. Do not conflate "Bradford" (WMS/IGT gaming-style references, US 6,612,928 / 6,709,333) with this family's art.

---

5. Confidence and residual gaps

• High confidence: the citations in §2 that I could tie to the authoritative full-text "References Cited" list (all US numbers above appear in the '976 record) and their dates/titles as returned by Google Patents.
• Not verified / explicitly open: (i) the complete 273-reference list (fetched text truncated at US6667684); (ii) the examiner's actual §102/§103 rejections for the '976, if any exist in the file history; (iii) element-by-element disclosure mapping for US 2002/0140542 (Prokoski) and the foreign WO references, which I could not fully read.

To close these gaps definitively, the next step is the USPTO Patent Center file-wrapper for application 16/170,234 (for any examiner citations/rejections), and the EPO Espacenet / Google Patents "Cited By" and "Citations" tabs for the complete reference inventory. I will not assert a specific anticipatory mapping beyond what is grounded above.

I'll analyze obviousness under § 103 using the prior art cited on this patent's own face. Let me first pull the key references from the citations list so I can ground each mapping.

Good material. Now let me pull the server-side "trusted authority" references and the transaction/wireless-payment references from the citations list.

Obviousness Analysis — U.S. Patent No. 10,437,976 B2 under 35 U.S.C. § 103

Scope note. This analysis is technical, not a legal validity opinion. It is built only on the prior art that appears in the Citations (273) section on the face of US 10,437,976 ("the '976 patent") plus the references I could independently confirm via search. I flag confidence levels per reference. Where I only have a title from the citation list (no retrieved disclosure), I say so explicitly rather than inventing content. Consistent with the earlier section, I treat the priority date as 2004-12-20 (provisional 60/637,538). This analysis is unaffected by the term/expiration discrepancy flagged previously (2025-12-20 vs. 2038-10-25); §103 is assessed as of the priority date regardless.

---

1. Legal framework and the person of ordinary skill

§103 standard (post-KSR). The claims are obvious if the differences between the claimed subject matter and the prior art are such that the subject matter as a whole would have been obvious at the time of the invention to a PHOSITA. Under Graham v. John Deere the analysis is: (1) scope and content of the prior art; (2) differences between the prior art and the claims; (3) level of ordinary skill; (4) secondary considerations. Under KSR Int'l v. Teleflex, a combination of known elements is obvious where it does "no more than yield predictable results," where there is a "design need or market pressure," or where "a person of ordinary skill can implement a predictable variation." A teaching, suggestion, or motivation (TSM) to combine — from the references themselves, the nature of the problem, or common knowledge — supports the conclusion; a rigid TSM test is not required.

PHOSITA (proposed). As of December 2004: a person with a bachelor's in electrical engineering or computer science (or equivalent) and ~2–4 years of experience in one or more of (a) access control / physical-security credentials, (b) biometric authentication (fingerprint/iris), and (c) short-range wireless/RFID or Bluetooth systems, or equivalent work experience. This is a routine, non-inventive skill level; the '976 claims are drafted to such a person.

Time-qualification. All references relied on below published or issued before 2004-12-20, so each qualifies as prior art under §102(a)/§102(b) (pre-AIA, consistent with the 2005 filing of parent 11/314,199).

---

2. Claim 1 decomposed into limitations (the analysis anchor)

Claim 1 is drafted from the reader side. Independent claims 9 (system) and 17 (computer program product) recite the identical six steps, so the same mapping disposes of all three independent claims.

The heart of the claim is the architecture: biometric match happens locally on the portable device → device ID (not a biometric template) is released → a third-party/remote trusted-authority server — not the reader itself — validates the device ID → on a positive response the reader completes a transaction.

---

3. Prior-art inventory (all from the '976 Citations list unless noted)

Confirmed by search (high confidence):

• US 6,041,410 B1 — Hsu et al. (TRW Inc.), "Personal identification fob," granted 2000-03-21 (cited by examiner). Fob with fingerprint sensor + correlator comparing sensed image to a reference image stored in the fob; on a match, generates a numerical value (CRC) from the stored reference image, encrypts it using a door public key + device private key, and wirelessly transmits it with user ID to a "door" (car/building door, ATM, computer). The door compares the received value to one previously stored at registration and grants access. Discloses public/private-key exchange and double encryption for the link. URLs: https://patents.google.com/patent/[US6041410A](/patent/US6041410A) ; PDF text at https://patentimages.storage.googleapis.com/c1/eb/77/3067f9e988bfef/EP0924656A3.pdf ; https://uspto.report/patent/grant/[6041410](/patent/6041410)

• US 2002/0140542 A1 — Prokoski & Prokoski (Mikos, Ltd.), "Personal biometric key," published 2002-10-03 (granted as US 6,850,147 B2). A portable personal biometric key: user-selectable biometric sensor (fingerprint, iris/retinal, face, hand geometry, voice, signature), template stored in the key, a composite code generated from the biometric code + a unique output code (clock/GPS), Bluetooth transmitter broadcasting the code to a compatible "universal biometric lock" at a portal to which the user is enrolled; Bluetooth handshaking + GPS "to ensure that the PBK is physically near the biometric lock." Critically, FIG. 4 encodes the "device ID and transmission counter," and FIG. 6 discloses a Central Authority 650: "The sender 600 transmits biometric code, date/time/location, device ID to the central authority…"; "Recipients of a personal biometric code which wish to do so may check the authenticity by transmitting it to the Central Authority 650… The Central Authority 650 confirms the authenticity … and sends authorization back to the recipient." Also: the CA "can serve as a clearing house when a person loses his personal biometric key … would transmit or maintain that information for all participating systems." URLs: https://patents.google.com/patent/US20020140542 ; https://patents.justia.com/patent/[6850147](/patent/6850147)

• US 6,490,443 B1 — "Communication and proximity authorization systems," granted 2002-12-03. Multiple-channel wireless transceiver that detects/recognizes nearby wireless devices and interacts with proximity service units — expressly including ATM systems, store checkout systems, vending machines, building/hotel/home access, garage doors, and networked entertainment facilities, and it describes "a biometrics recognition system for ID and credit transaction services." URL: https://patents.google.com/patent/[US6490443B1](/patent/US6490443B1)

• US 6,325,285 B1 — Baratelli (AT&T Corp.), "Smart card with integrated fingerprint reader," granted 2001-12-04. Card stores a fingerprint template and compares an acquired fingerprint to it; if they match the card is enabled, giving the user access to "information and/or services that require cooperation of the smart card"; expressly applied to access to sensitive data/software and the point-of-sale purchaser-confirmation use case. URL: https://patents.google.com/patent/US6325285 ; PDF: https://patentimages.storage.googleapis.com/f6/88/aa/19d8112f159bb0/US6325285.pdf

• US 6,256,737 B1 — Bianco et al. (BioNetrix Systems), "System, method and computer program product for allowing access to enterprise resources using biometric devices," granted 2001-07-03. A biometric server stores templates and "the engine and collections of data required by the system to authenticate users"; client devices present biometric data; the server executes policies and returns an authentication result that gates access to enterprise resources; includes enrollment and administration stations and digital certificates. This is a textbook remote/central authority authenticating a credential presented by a device. URL: https://patents.google.com/patent/US6256737

• US 2002/0098888 A1 (IGT), "Wireless gaming environment," 2002-07-25 and US 2002/0103027 A1 (IGT), "Gaming environment including portable transaction devices," 2002-08-01 — wireless portable devices used with gaming machines / wagering transactions; bear on dependent claims 8/16 (reader "operable on a same system as … a casino machine") and on "complete a transaction."

Title-level grounding only (in the Citations list; I did not retrieve disclosures — verify before relying):

• US 4,993,068 (Piosenka et al., Motorola), "Unforgeable personal identification system," 1991-02-12.
• US 5,598,474 (Johnson), "Process for encrypting a fingerprint onto an I.D. card," 1997-01-28.
• US 5,623,552 (Cardguard), "Self-authenticating identification card with fingerprint identification," 1997-04-22.
• US 6,012,476 (IBM), "Portable information and transaction processing system and method utilizing biometric authorization and digital certificate security," 2000-01-18.
• US 6,219,439 B1 (Burger), "Biometric authentication system," 2001-04-17.
• US 6,363,485 B1 (Entrust), "Multi-factor biometric authenticating device and method," 2002-03-26.
• US 2002/0073042 A1 (Maritzen et al.), "Method and apparatus for secure wireless interoperability and communication between access devices," 2002-06-13.

---

4. Combination A (primary): Prokoski '542 + US 6,490,443 (+ optionally the TRW fob)

This is the strongest §103 combination because Prokoski alone covers limitations 1.1–1.5 almost verbatim, and the proximity/payment reference supplies the tailored "detect in proximity" + "complete a transaction" framing.

Result: Claim 1, and the parallel limitations of claims 9 and 17, are rendered obvious over Prokoski in view of US 6,490,443. Even without US 6,490,443, Prokoski alone anticipates/obviates most of claim 1; the payment/proximity reference merely supplies explicit "transaction" language and a detection trigger.

Motivation to combine: Both references are in the same field (proximity access + biometric authorization) and solve the identical problem — releasing access/transaction rights only to a biometrically-verified holder of a portable credential. US 6,490,443's explicit enumeration of "ATM systems, store checkout systems … [and] biometrics recognition system for ID and credit transaction services" provides an express reason to graft Prokoski's portable biometric key onto a point-of-sale/ATM/transaction unit — i.e., predictable use of the key for the very transactions the secondary reference is about.

---

5. Combination B (alternative): TRW fob '410 + BioNetrix '737 (+ US 6,490,443 for proximity)

This combination is useful if a tribunal reads claim 1's "identifier code uniquely identifying the computing device" and "trusted authority server" narrowly.

Motivation to combine: The fob's own disclosure identifies the problem the server solves — the user must register at every door, and the door stores the user's CRC and public key. Replacing per-door registration/verification with a central authority that already holds the enrollment record is the classic, predictable network-architecture improvement motivated by (i) scalability across many readers, (ii) single-point revocation for lost/stolen keys, and (iii) reduced per-reader storage/administration. BioNetrix '737 teaches exactly that architecture (biometric server + enrollment/administration stations + certificates). Prokoski FIG. 6 independently confirms that a central authority is the expected design choice in a portable-biometric-key system (it even describes the lost-key clearing-house function). This is a textbook KSR "known technique to improve similar devices in the same way" combination.

---

6. Combination C: Smart-card '285 + BioNetrix '737 + IGT '888/'027 (for the system/POS and casino limitations)

For dependent claims 8/16 (reader on the same system as a casino machine, keyless lock, ATM, computer, or POS register):

• US 6,325,285 ties on-card fingerprint verification to POS and computer-software/access use cases;
• US 6,490,443 enumerates ATM, checkout, vending, home access;
• IGT '888/'027 place wireless portable devices in the wagering/gaming-machine environment (casino machine);
• The fob '410 names ATM and computer as its "doors."
  The enumerated list in claims 8/16 is a mere recitation of known applications of the same portable-credential system, which is the very definition of obviousness when each application was independently known and the adaptation is mechanical.

---

7. Dependent claims 2–8, 10–16, 18–20

• 2 / 10 / 18 — reader sends a biometric-verification request to the device upon detecting proximity. Obvious over any of Combos A–C. Prokoski's lock "cues" interaction via Bluetooth handshake; the fob's verification is initiated by the device/door exchange; US 6,490,443 teaches interrogation/recognition triggering service. Adding a request step is a predictable two-way messaging design in a handshake the art already uses.
• 3 / 11 / 19 — wireless secure link using encryption/decryption keys. Directly taught by fob '410 (public/private key exchange; double encryption) and by Prokoski (Bluetooth baseband authentication/encryption). Prokoski/US 6,490,443 also support wireless. Strong §103 showing.
• 4 / 12 / 20 — identifier code registered with the trusted authority server. Directly taught by Prokoski FIG. 6 (Central Authority enrollment/registry; sender's device ID transmitted and enrolled) and by BioNetrix '737 (enrollment station populating the biometric server); fob '410's per-door registration is the analog.
• 5 / 13 — receipt of the identifier code indicates biometric verification succeeded. Inherent/express: fob '410 only transmits the code upon a match; Prokoski only broadcasts after the biometric code is produced. A printed matter/functional-language recitation adding no structural step.
• 6 / 14 — biometric type list (fingerprint, palm, retina, iris, hand geometry, facial, signature, voice). Prokoski expressly lists fingerprint, iris/retinal, face, hand geometry, voice, signature; BioNetrix '737 lists "finger and hand geometry, retina and facial images, weight, DNA, breath, voice, typing stroke and signature." A enumerated genus wholly within the prior art.
• 7 / 15 — device is a mobile phone, tablet, laptop, mp3 player, mobile gaming device, watch, or key fob. Prokoski expressly integrates the key into a cell phone / PDA and describes a keyring fob form factor; the fob '410 is a fob. Substituting known portable hosts is a predictable design choice.
• 8 / 16 — reader on same system as casino machine, keyless lock, ATM, computer, POS register. See Combination C; each item is independently known in the cited art.

Conclusion for dependents: Every dependent limitation is either expressly taught or a predictable variation of a known element; none introduces an unexpected result.

---

8. Motivation to combine (KSR factors, consolidated)

1. Same field / same problem. All primary references target releasing a secured resource only to a biometrically-verified holder of a portable credential — no field-switch is required.
2. Express use-case overlap. US 6,490,443 names ATM/checkout/credit-transaction units and a biometrics ID system; the fob names ATM and computer; smart-card '285 names POS; IGT names gaming machines. These are direct invitations to apply the biometric-credential system to those readers/transactions.
3. Known architectural improvement. Offloading validation to a central server (BioNetrix; Prokoski CA 650) is the standard scale/revocation answer to per-reader registration (the fob's admitted burden). Predictable, and beneficial for revocation of lost/stolen keys.
4. Interchangeable, known wireless security. Bluetooth/RF with built-in authentication/encryption, and public/private-key exchange, were commodity techniques — combining them with the credential is "arranging old elements with each performing the same function."
5. Design need / market pressure. The '976 patent's own Background concedes the two driving needs: (a) avoid memorized credentials and (b) bind a credential to the legitimate user (not just a valid object). Both cited references exist precisely to meet those needs.

---

9. Counter-considerations (why a §103 rejection is not a foregone conclusion)

I want to be balanced, not one-sided:

• Prosecution outcome. The claims issued over this art. That means the examiner evidently did not treat Prokoski '542 + proximity art as a complete §103 case. The most plausible reason is the specific combination of (i) an identifier code that uniquely identifies the computing device (as opposed to the fob's CRC derived from the biometric image, or Prokoski's composite biometric+clock/GPS code) with (ii) a trusted authority server that authenticates that device-specific code and returns a message authorizing a transaction. A patentee would argue Prokoski's "composite code" is not "an identifier code uniquely identifying the computing device," and that the fob verifies locally at the door rather than at a remote authority. That is the strongest non-obviousness position, and it should be tested against the actual claim language and any file-wrapper statements.
• "Authorization at the lock" language. Prokoski states an object of the invention is that "authorization decisions [be] made not by the personal biometric key device but by a biometric lock at the portal being accessed." This is consistent with claim 1 (device doesn't authorize; reader/server does) and is not a teaching away from a reader-plus-server split. If anything it supports the claimed division of labor.
• Secondary considerations. I found no evidence in the material reviewed of unexpected results, industry skepticism, commercial success attributable to these specific claim elements, or praise. Absent such evidence, the objective-indicia prong does not rebut the §103 case. (If a patentee has such evidence — e.g., licensing nexus or copying — it would need to be evaluated separately.)
• Claim-construction sensitivity. The outcome of 1.3 turns on how "identifier code uniquely identifying the computing device" is construed; reading it to require a device-unique ID distinct from user/biometric data makes Prokoski FIG. 4/FIG. 6 ("device ID … to the central authority") the key disclosure.

---

10. Bottom line

• Claim 1 / 9 / 17 are prima facie obvious over US 2002/0140542 A1 (Prokoski) in view of US 6,490,443 B1, and over US 6,041,410 (TRW fob) in view of US 6,256,737 (BioNetrix) — with US 6,490,443 supplying explicit proximity detection and transaction completion. Prokoski, a single reference on the '976 face, discloses nearly every limitation including the previously "hard" one (a central authority that receives the device ID and returns an authentication result).
• All dependent claims 2–8, 10–16, 18–20 are obvious as express teachings or predictable variations of the same references.
• The principal vulnerability to a §103 attack is the breadth of Prokoski FIG. 6 combined with the fob's on-device biometric verification; the principal defense is the specific "device-unique identifier code + trusted-authority-server-returns-message + reader completes transaction" architecture, which the patentee will argue is not squarely disclosed, plus any secondary-considerations evidence.

Confidence: High for the Prokoski, TRW-fob, BioNetrix, US 6,490,443, and smart-card mappings (I retrieved their disclosures). Moderate/Low for the title-only references (US 4,993,068; US 5,598,474; US 5,623,552; US 6,012,476; US 6,219,439; US 6,363,485; US 2002/0073042) — I grounded those on titles in the '976 Citations list only and did not verify their disclosures; do not rely on the specific uses I attributed to those without pulling their full texts from the URLs on the patent page.

Flagged inconsistency (from the prior section, carried forward): the two expiration dates for '976 (2025-12-20 vs. 2038-10-25) remain unresolved; it does not affect this §103 analysis but should be reconciled if term/expiry matters to your purpose.