US 10073960

I'll verify the patent details against live sources, including checking for any 2026 CAFC activity.

Let me check for the specific 2026 CAFC docket activity and the IPR outcomes.

US Patent 10,073,960 B1 — Verification Summary

I searched for this exact number (not similar ones). Here is what authoritative and live sources show, with confidence levels noted.

Bibliographic data (high confidence)

Source: Google Patents US10073960B1; the issued patent PDF (Ex. 1001 in PTAB filings).

Abstract (as issued)

A hybrid device includes a personal digital key (PDK) and a receiver-decoder circuit (RDC), coupled for communication with each other. In one embodiment it also provides a physical interconnect to other devices for control signals/data and power. It operates in PDK-only, RDC-only, or PDK-and-RDC modes, enabling mixed configurations (PDK/RDC, RDC/RDC, PDK/PDK). Disclosed system configurations include: use in a cell phone; simultaneous PDK+RDC use; multiple links of the hybrid device to generate an authorization signal; multiple PDK links to the hybrid device to generate an authorization signal; and authorization inheritance.

Independent claims (plain language)

The patent has two independent claims — claim 1 (device) and claim 14 (method); claims 2–13 depend from claim 1 and 15–20 depend from claim 14.

Claim 1 — a device comprising:

• a battery;
• an integrated, secure memory storing local, secured information; and
• an integrated, wireless communication interface for communicating wirelessly with at least one external device within a proximity zone, coupled to the secure memory and drawing power from the battery;
• wherein one or more of (a) the interface communicating with the external device within the proximity zone enables an application, function, and/or service, and (b) the local secured information enables an application, function, and/or service.

Plain-language: A battery-powered gadget with (i) locked-down memory holding user/personal data and (ii) a short-range wireless radio. Either the radio detecting/connecting to a nearby external device, or the data in the secure memory, "unlocks" the use of some app, function, or service.

Claim 14 — a method comprising:

• creating a first wireless link between an integrated wireless communication interface of a device and an external device (the device has a battery, an integrated secure memory storing local secured information, and the interface powered by the battery);
• receiving a first signal at the interface via the first link from the external device; and
• generating an enablement signal enabling an application, function, and/or service.

Plain-language: The method counterpart — establish a short-range wireless link, receive a signal from the other device, and in response produce a signal that turns on/enables a capability.

Note on dependent claims: I do not have the verbatim text of every dependent claim in front of me. From PTAB pleadings I can confirm claim 9 relates to an external authentication database separate from the merchant, and claims 12/13 (and 20) relate to form factors such as jewelry or a watch. I would verify the remaining dependents against the official claim set before relying on them.

Claim-construction/validity context

• In Proxense v. Microsoft, No. 6:23-cv-00319 (W.D. Tex.), Judge Albright's claim construction order (ECF 66) addressed terms such as "access message" (construed "no construction necessary") and previously-construed terms including "receiver-decoder circuit (RDC)," "personal digital key (PDK)," and "enabling one or more of an application, a function, and a service."
• Microsoft's IPRs argued the '960 claims are obvious over Giobbi-157 (US 2007/0245157) + Giobbi-139 (US 2004/0255139), with Dua (US 9,042,819), and over Broadcom/Buer (EP 1536306) + Giobbi-157.

Litigation and 2025–2026 docket status (this is where it matters)

PTAB — the '960 patent is the subject of three proceedings:

• IPR2024-00405 (Microsoft Corp. v. Proxense): filed Jan. 16, 2024; instituted July 22, 2024; Final Written Decision July 17, 2025 — all challenged claims (1–20) held unpatentable. Status: appealed. (Panel: APJs Dang, McKone, Beamer.)
• IPR2024-00784 (Google LLC v. Proxense): filed April 19, 2024; instituted Nov. 18, 2024. This proceeding is now reported as Settlement (Google and Proxense settled in Jan. 2025).
• IPR2024-01320 (Google joinder petition): Not instituted (procedural).

CAFC — the specific appeal:

• Appeal No. 25-2130, Proxense LLC v. Microsoft Corp, filed September 22, 2025, appealing from PTAB case IPR2024-00405. As of the most recent snapshots I retrieved, the docket shows "Pending Judge Assignment" and no decision or outcome.

⚠️ Important distinction — do not conflate: The widely reported 2026 Federal Circuit activity in Proxense (opening brief filed ~April 2026 arguing PTAB/CRU conflict and an Appointments Clause issue) concerns U.S. Patent 8,886,954 ("Biometric personal data key (PDK) authentication"), consolidated from IPR2024-00233, IPR2024-01334, and IPR2024-00846 — a different patent and different appeals than 25-2130. I found no authoritative 2026 docket entry showing an oral argument date, briefing completion, or decision specifically in 25-2130 for the '960 patent. I would not state that a 2026 CAFC decision exists for 10,073,960.

District court:

• Proxense v. Microsoft, No. 6:23-cv-00319 (W.D. Tex.) — the '960 patent is one of the asserted patents (complaint Exhibit 6); the case has been stayed pending the IPRs, and there was a motion to dismiss the '730 patent with prejudice (May 2025).
• Proxense v. Google, No. 6:23-cv-00320 (W.D. Tex.) — '960 patent also asserted there; settled (Jan. 2025).
• Google Patents also flags a Texas Western District Court case 6:25-cv-00016 and a Federal Circuit case 25-2130 against this patent family.

Explicit uncertainties

1. Priority date is an assumption, not a legal conclusion. Google Patents labels the prior-art date 2007‑12‑05 (the 12/329,329 filing); the provisional (60/992,953) is dated 2007‑12‑06.
2. Expiration values differ across sources (2028‑12‑04 vs 2028‑12‑05) and are "anticipated."
3. No 2026 outcome exists (as far as I can verify) for CAFC 25-2130 on this patent. If a 2026 decision has since issued, I could not confirm it with the sources available to me — flag for re-check.
4. I have not independently re-read every dependent claim's verbatim text; only the independent claims and the dependent-claim notes above are grounded in what I retrieved.

Primary URLs: https://patents.google.com/patent/US10073960B1/en • https://portal.unifiedpatents.com/patents/patent/US-10073960-B1 • https://ai-lab.exparte.com/case/cafc/25-2130/proxense-llc-v-microsoft-corp • https://ai-lab.exparte.com/case/ptab/IPR2024-00405/microsoft-corp-v-proxense-llc

I'll research the litigation record for US Patent 10,073,960 specifically, verifying against live sources.

Let me verify each case specifically and check which actually assert the '960 patent.

Litigation Involving U.S. Patent No. 10,073,960

Patent: US 10,073,960 B1 — Hybrid Device Having a Personal Digital Key and Receiver-Decoder Circuit and Methods of Use
Owner/Plaintiff: Proxense, LLC (Bend, OR)
Issued: September 11, 2018 (so any suit asserting it must post-date that issue)

I searched for this exact patent number and discarded hits for related-but-different members of the same Proxense family (e.g., 8,646,042; 9,679,289; 8,352,730; 8,886,954; 9,298,905; 9,049,188; 10,698,989). Below, I separate cases where the '960 patent was actually asserted from related Proxense cases where it was not, so nothing is mis-attributed.

---

I. District court cases asserting the '960 patent

If your question is simply "has the '960 patent been litigated?" — yes, at least three times, all in the Western District of Texas.

---

II. PTAB proceedings on the '960 patent (administrative, not district-court litigation, but part of the same dispute set)

---

III. Federal Circuit appeal

---

IV. Related Proxense cases that do NOT assert the '960 patent (do not attribute)

These are frequently bundled with '960 coverage but the '960 patent is not among the asserted patents:

• Proxense, LLC v. Samsung Electronics Co., Ltd. et al., No. 6:21-cv-00210-ADA (W.D. Tex.) — filed March 5, 2021; settled/dismissed with prejudice in January 2023 on the eve of trial. Centered on the PDK/biometric family ('730, '954, '905, '042). The '960 patent does not appear among the asserted patents in the record I retrieved.
• Proxense, LLC v. Apple, Inc., No. 6:24-cv-00143-ADA (W.D. Tex.) — filed March 18, 2024; still pending (claim construction ongoing). Complaint exhibits list 9,298,905; 8,352,730; 8,886,954; 10,698,989; 9,049,188; 8,646,042 — the '960 patent is not among them. (Apple's related IPRs, e.g., IPR2024-01398/01399, target the '042 patent, not '960.) ⚠️ Do not merge the '960 with the '042/‘289 hybrid-device family for citation purposes even though they share a specification lineage.
• Proxense v. LG and Proxense v. Intel (No. 6:24-cv-00283) — reported as pending Proxense campaigns in W.D. Tex., but I could not verify that either asserts the '960 patent; I found no confirmation, so I do not attribute '960 to them.

---

V. Caveats and explicit uncertainties

1. Hyundai dismissal scope — flagged. The Proxense v. Hyundai docket (6:25-cv-00016) shows a Notice of Voluntary Dismissal on April 14, 2025 and the case as terminated/closed. I could not confirm whether the dismissal covered all defendants/patents, whether it was with or without prejudice, or whether Proxense refiled. Verify the April 2025 papers on PACER before relying on this outcome.
2. '960 asserted-claim set varies by case. In the Microsoft case, Proxense's final infringement contentions asserted claims 14 and 16 of '960; the Hyundai complaint charted at least claim 1. Do not assume a single asserted-claim set.
3. PTAB ≠ district-court litigation. I list the three IPRs and the CAFC appeal separately because your request targeted litigation, but they are the reason the Microsoft case was stayed.
4. Patent-number discipline. I did not auto-correct any identifier. Every case above was verified as involving 10,073,960 specifically, except where I expressly say the patent is not asserted (Section IV). Note that the related 8,646,042 case caption ("Hybrid device having a personal digital key and receiver-decoder circuit…") is nearly identical to the '960 title — a common source of confusion; the IPRs IPR2024-01318/01398/01399 target the '042 patent, not the '960.
5. Date basis. This is current as of the sources retrieved; the most recent docket verification dates in the retrieved records were 2025. I found no litigation naming the '960 patent outside the W.D. Tex. cases above.

Sources: CourtListener dockets (Proxense v. Microsoft, 6:23-cv-00319 — Dkt. 1 (Ex. 6 = '960), Dkt. 66 Claim Construction Order, Dkt. 95); Justia dockets (6:24-cv-00143 Apple; 6:25-cv-00016 Hyundai); PacerMonitor and RPX Insight case pages; ip fray (Proxense campaign reports); Unified Patents patent page for US-10073960-B1; Docket Alarm PTAB dockets (IPR2024-00405 / -00784 / -01320); Law360 case tracker for Proxense, LLC v. Google LLC et al.

I'll verify the proceeding list and fill in claim-level detail from live sources. Note upfront: the structured "PTAB proceedings on file" block says the USPTO ODP API returns no AIA trials — that conflicts with the patent text and the earlier summary, both of which document three proceedings. Let me resolve that contradiction with live search.

Strong signal on all three proceedings. Let me nail down the FWD reasoning, the joinder disposition, and any 2026 appellate activity.

PTAB / AIA Trial Proceedings — US 10,073,960 B1

Hybrid Device Having a Personal Digital Key and Receiver-Decoder Circuit and Methods of Use (Proxense, LLC)

---

⚠️ Data-source conflict — flagged up front

The structured "PTAB proceedings on file" block in this prompt states that the USPTO ODP API returns no AIA trial proceedings for this patent and instructs that the "default should be 'no PTAB activity on file.'"

That is wrong for this patent, and I am not treating it as authoritative. Every other source in the record contradicts it:

• The full patent text supplied in this prompt lists three AIA proceedings in its litigation block: IPR2024‑00405 (Final Written Decision), IPR2024‑00784 (Settlement), IPR2024‑01320 (Not Instituted – Procedural).
• The previously generated summary (treated as authoritative here) documents the same three, with an FWD date and a CAFC appeal number.
• Live web sources confirm all three with docket-level detail (below).

Conclusion: the ODP ingest for this patent is stale or incomplete. The absence of PTAB activity is a false negative, not a signal. Do not rely on that block for this patent.

---

Proceedings overview

Three AIA trials have been filed against US 10,073,960 — zero claims secured a patentability holding. Breakdown: 1 proceeding reached a Final Written Decision and invalidated all challenged claims (now on appeal); 1 was instituted and then terminated by settlement; 1 was a conditional joinder petition that was never instituted. The bottom line for a defendant: this patent is not "hardened" — it is a liability. On 2025‑07‑17 the Board held claims 1–20 unpatentable, and the only reason you cannot yet call any claim "canceled" is that Proxense's appeal (CAFC 25‑2130, consolidated with 25‑2131) is still pending. If you receive a demand letter citing this patent, the FWD is your opening exhibit.

---

IPR2024-00405 — Microsoft Corporation v. Proxense, LLC

• Type: Inter Partes Review (pre‑AIA §§ 102/103 applied; the application is a continuation of pre‑2013 filings)
• Filed: 2024‑01‑16
• Status: Final Written Decision – Appealed (structured/live status); Outcome: Unpatentable
• Judge panel: Thu A. Dang, David C. McKone, Norman H. Beamer
• Petition grounds (all § 103, pre‑AIA; claims 1–20 challenged):
  • Ground 1A — claims 1–20 obvious over Giobbi‑157 (US 2007/0245157) in view of Giobbi‑139 (US 2004/0255139)
  • Ground 1B — claims 1–20 over Giobbi‑157 + Giobbi‑139 + Dua (US 9,042,819)
  • Ground 2A — claims 1–8, 10–11, 14–19 over Broadcom (EP 1 536 306 A1)
  • Ground 2B — claims 1–20 over Broadcom + Giobbi‑157
  • Supporting testimony: Declaration of Dr. Patrick Traynor (Ex. 1003), later supplemented (Ex. 1032).
  • Petitioner represented that the grounds were "similar to those advanced by Samsung in instituted and terminated IPRs (IPR2021‑01438 and 01439) directed to related US Patent Nos. 9,049,188 and 9,235,700" — i.e., the same art family Proxense had already faced and settled.
• Institution decision: Instituted 2024‑07‑22. Petitioner argued against discretionary denial under § 325(d) (Examiner never considered the asserted art) and under Fintiv, stressing the parallel litigation's uncertain schedule (pending transfer motion, cancelled Markman hearing) and the July 2025 statutory FWD deadline. The Board instituted.
• Final Written Decision: 2025‑07‑17 — all challenged claims (1–20) held unpatentable. Independent claims 1 (device) and 14 (method), and every dependent claim from 2–13 and 15–20, fell together.

  Caveat / honesty note: I retrieved the FWD's disposition and date from aggregated docket sources, but not the verbatim text of the FWD. I therefore cannot state which of Grounds 1A/1B/2A/2B carried the day, nor quote the panel's reasoning, and I will not guess. The claim-level statement "1–20 unpatentable" is what the aggregators report (Law360: "all the claims across two patents … were invalid"); I have no indication any claim survived, but treat the granular reasoning as unverified pending the FWD PDF.

• Appeal: Yes. Notice of appeal filed 2025‑09‑22, docketed as Proxense, LLC v. [Microsoft Corp.](/litigations/by-plaintiff/Microsoft%20Corp.), CAFC No. 25‑2130, from IPR2024‑00405. On 2025‑10‑07 the Federal Circuit consolidated 25‑2131 as a member case under 25‑2130 as lead (all filings in the lead appeal). Certified list received 2025‑11‑04. As of the latest docket snapshots: "Pending Judge Assignment," no decision, no confirmed 2026 activity.
  • ⚠️ Do not conflate with the widely reported 2026 Proxense Federal Circuit brief (Appointments Clause / PTAB–CRU arguments). That appeal arises from a different patent — US 8,886,954 — consolidated from IPR2024‑00233 / ‑00846 / ‑01334. It is not 25‑2130 and has nothing to do with the '960 patent.
  • I could not confirm which FWD the member appeal 25‑2131 challenges; I infer it is a companion Microsoft IPR on a related family patent, but I will not assert it.
• Defensive value: This is the single most valuable document in the file. A fully briefed, judicial finding that every claim of the '960 patent is obvious over Giobbi‑157/Giobbi‑139/Dua/Broadcom. Any infringement theory built on claims 1–20 must now contend with a Board decision agreeing the claims are unpatentable. Confirm the FWD text and the appeal status before relying on it in a filing.

---

IPR2024-00784 — Google LLC v. Proxense, LLC

• Type: Inter Partes Review
• Filed: 2024‑04‑19
• Status: Terminated – Settled (settlement after institution; terminated 2025‑02‑05)
• Judge panel: Thu A. Dang, David C. McKone, Norman H. Beamer (same panel as 00405)
• Petition grounds (all § 103, pre‑AIA; all claims 1–20 challenged):
  • Ground 1 — claims 1–6, 8–11, 14–19 over Dua (US 2006/0258289) and Giobbi '157
  • Ground 2 — claims 1–7, 10, 11, 14–19 over Buer (= Broadcom, EP 1 536 306)
  • Ground 3 — claims 4, 8–10, 12, 13, 17, 20 over Buer + Giobbi '157
  • Ground 4 — claim 4 over Buer + Nishikawa (EP 1 600 885)
  • Supporting testimony: Declaration of Andrew Wolfe, Ph.D. (Ex. 1003).
  • Note: Google's art substantially overlapped Microsoft's (Giobbi‑157 and Buer/Broadcom appear in both), which is why the ranking papers call Google's separate joinder petition a "copycat."
• Institution decision: Instituted 2024‑11‑18 on all challenged claims 1–20 (Paper 12). This decision is unusually useful to a defendant because it contains express claim-construction rulings adverse to Proxense:
  • "local, secured information" — the Board refused Proxense's proposed narrow definition ("information stored within a service block of a PDK and unlocked with a corresponding access key"). It held the term "encompasses information stored locally on an integrated, secure memory, wherein the information is secured," and that the specification's "in this example" passage was "merely an exemplary embodiment and not a definition."
  • "enabl[ing] one or more of an application, a function, and a service" — the Board rejected any requirement of "receiving information from a PDK in exchange for an access key," finding no lexicography or disavowal with the requisite "clarity, deliberateness, and precision."
  • Proxense's fallback § 112 ¶ 6 (means‑plus‑function) theory was rejected: no "means for" language, and the terms were not "generic terms or black box recitations."
  • These constructions are the analytical backbone that likely produced the 00405 FWD. They are the terms to attack on.
• Settlement / termination: Joint Motion to Terminate under 35 U.S.C. § 317(a) filed 2025‑01‑23; Board termination decision 2025‑02‑05. The parties settled both this IPR and Proxense's district‑court assertion of the '960 patent in Proxense v. Google LLC, No. 6:23‑cv‑00320 (W.D. Tex.), and represented they "do not anticipate further litigation between them concerning U.S. Patent No. 10,073,960." The settlement agreement was filed as Ex. 1028 and granted business‑confidential treatment — terms are confidential. ⚠️ Critical: because termination came before any FWD, no § 315(e)(2) estoppel attached to Google.
• Defensive value: Google cannot be enlisted as an invalidity ally (it settled and licensed), but it also carries no estoppel, so Google's art is not "burned." More importantly, the institution decision is public and quotable — a defendant can cite the Board's own rejection of Proxense's narrowing constructions without paying for an IPR.

---

IPR2024-01320 — Google LLC v. Proxense, LLC

• Type: Inter Partes Review (filed with a conditional motion for joinder under § 315(c) to Microsoft's IPR2024‑00405)
• Filed: 2024‑08‑22
• Status: Not Instituted – Procedural (structured status). Plain English: a "me‑too" joinder petition that never matured.
• Judge panel: Not separately published to me; managed alongside Google's related family petitions (IPR2024‑00782/00783) before APJs Dang, Turner, McKone, Beamer.
• Petition grounds: Substantively identical to the Microsoft petition — Grounds 1A/1B/2A/2B: Giobbi‑157 + Giobbi‑139 (claims 1–20), + Dua (1–20), Broadcom (1–8, 10–11, 14–19), Broadcom + Giobbi‑157 (1–20). Google's own ranking paper concedes the joinder petition "is a 'copycat' petition … substantially the same as the instituted Microsoft Petition."
• Institution decision: Not instituted. Google filed the petition conditionally, asking the Board to hold it in abeyance "until, and only if, the Board declines to institute" Google's earlier‑filed original petition (IPR2024‑00784). Because the Board did institute 00784 on 2024‑11‑18, the condition never triggered and the joinder petition had no independent path — consistent with a procedural, non‑merits non‑institution.
  • Caveat: I did not retrieve the actual non‑institution paper for 01320. The characterization above is reconstructed from Google's ranking/abeyance filings and the structured status; verify against the docket before quoting.
• Settlement / termination: N/A (never instituted).
• Appeal: None.
• Defensive value: Zero estoppel, zero merits content. Its only value is as a pattern signal — it shows Google tried two bites at the '960 (a standalone petition plus a joinder to Microsoft's), which underscores how exposed the patent was.

---

Strategic summary

Claim status. There is no "narrowed but surviving" set of claims here — there is no surviving claim set at all. In IPR2024‑00405 the Board held claims 1–20 unpatentable (2025‑07‑17). Independent claim 1 (device) and independent claim 14 (method), and all sixteen dependents (2–13, 15–20), were included in that disposition per aggregator reporting. No proceeding has ever held any claim of the '960 patent patentable. The only caveat is procedural, not substantive: the claims are not yet cancelled — statutory cancellation under § 318(b) awaits exhaustion of the appeal. Until CAFC 25‑2130 resolves, Proxense can still assert the claims, and an infringer cannot yet say "claim 1 is canceled." Which specific ones the FWD reasoned about individually — and on which ground — I could not verify, and I do not assert it.

Estoppel landscape.

• Microsoft is estopped under § 315(e)(2) from asserting, in the W.D. Tex. case or any later civil action, any ground it raised or reasonably could have raised in 00405 — i.e., the Giobbi‑157 / Giobbi‑139 / Dua / Broadcom(Buer) combinations. That bar runs to Microsoft and its privies/RPIs, not to the patent.
• Google is NOT estopped. IPR2024‑00784 died by settlement before an FWD, and estoppel under § 315(e)(2) attaches only after a final written decision; IPR2024‑01320 never instituted. Google's art therefore remains legally available to others.
• A new defendant is not bound by anyone else's estoppel. You may file your own IPR on the same or different art — the Giobbi and Broadcom/Buer references are not off the table for you. Two cautions: (1) the § 315(b) one‑year clock runs from service of a complaint alleging infringement of the '960 patent; (2) expect the Board to weigh § 325(d) and General Plastic given the crowded family history — though note Microsoft successfully argued the Examiner never considered this art, which cuts against § 325(d) denial.

Pattern signals. (i) Multiple independent petitioners — Microsoft (00405) and Google (00784 + 01320) — attacked the same claims with overlapping art (Giobbi‑157 and Broadcom/Buer in both), a strong signal of a weak patent. (ii) Proxense litigates, then settles — it settled Google (family-wide, Jan 2025) and previously Samsung, while appealing the one loss it could not settle (Microsoft). (iii) The same three‑judge panel (Dang/McKone/Beamer) sat on 00405 and 00784, giving unusually consistent claim constructions adverse to Proxense. (iv) No defensive aggregator (Unified Patents, RPX, etc.) appears as petitioner — these were operating‑company defendants defending themselves. Unified Patents is not in the chain for this patent.

---

Recommended next steps

1. Pull the IPR2024‑00405 FWD (2025‑07‑17) and quote its disposition verbatim. Every defense brief citing this patent should open with the Board's own words that claims 1–20 are unpatentable. Get the PDF from PTAB E2E / PTAB Decisions (https://ptab.uspto.gov/) or the case portal: https://portal.unifiedpatents.com/ptab/case/IPR2024-00405. Also pull the IPR2024‑00784 Institution Decision (Paper 12, 2024‑11‑18), https://portal.unifiedpatents.com/ptab/case/IPR2024-00784 — the "local, secured information" and "enabling …" constructions are directly usable against Proxense's infringement contentions.
2. Track CAFC 25‑2130 (lead) / 25‑2131 (member). Docket via PACER / CourtListener: https://dockets.justia.com/docket/circuit-courts/cafc/25-2131 and https://litigation.rpxcorp.com/federal_circuit/[3015487](/patent/3015487)-proxense-v-microsoft. Disposition = affirmance ⇒ all claims 1–20 cancel and Proxense's assertion of the '960 ends. Key milestones: briefing following the 2025‑11‑04 certified list; watch for oral‑argument scheduling. Verify independently — I found no 2026 decision or argument date as of this analysis.
3. Consider a stay / leverage the parallel proceedings. The W.D. Tex. Microsoft case (6:23‑cv‑00319) was stayed pending the IPRs; the Google case (6:23‑cv‑00320) settled. If you are newly sued, the FWD is a powerful basis for a stay and, if the appeal affirms, for summary judgment of invalidity or dismissal.
4. Your own IPR window. Calendar the § 315(b) deadline from service of the complaint. The Giobbi‑157/Giobbi‑139/Dua and Broadcom/Buer combinations are available to you notwithstanding Microsoft's estoppel — but be prepared to distinguish on § 325(d)/General Plastic, and note the very short runway: the patent's anticipated expiration is 2028‑12 (Google Patents: 2028‑12‑05; Unified Patents: 2028‑12‑04), and the patent is subject to a terminal disclaimer, so file promptly to get a decision before expiry.
5. Do not cite the ODP "no PTAB activity" result. For this patent it is a false negative. Use the three proceeding numbers and the live sources above.

---

Source links

• IPR2024‑00405 (Microsoft): https://portal.unifiedpatents.com/ptab/case/IPR2024-00405 • https://ai-lab.exparte.com/case/ptab/IPR2024-00405/microsoft-corp-v-proxense-llc
• IPR2024‑00784 (Google): https://portal.unifiedpatents.com/ptab/case/IPR2024-00784 • Institution Decision (Paper 12, 2024‑11‑18) • Joint Motion to Terminate (Paper 14, 2025‑01‑23)
• IPR2024‑01320 (Google joinder): https://portal.unifiedpatents.com/ptab/case/IPR2024-01320 • Petition (2024‑08‑22)
• CAFC 25‑2130 / 25‑2131: https://dockets.justia.com/docket/circuit-courts/cafc/25-2131 • https://litigation.rpxcorp.com/federal_circuit/3015487-proxense-v-microsoft
• Patent record: https://patents.google.com/patent/[US10073960B1](/patent/US10073960B1)/en • https://portal.unifiedpatents.com/patents/patent/US-10073960-B1

Explicit uncertainties

1. The "PTAB proceedings on file" block is contradicted by every other source and is treated as stale. (Flagged above.)
2. I did not retrieve verbatim FWD text for IPR2024‑00405. The "claims 1–20 unpatentable" disposition, the 2025‑07‑17 date, and the panel are well-sourced; the ground-by-ground and claim-by-claim reasoning is unverified and I have not asserted it.
3. I did not retrieve the actual non‑institution paper for IPR2024‑01320. Its "Not Instituted – Procedural" status comes from the structured data and the patent text's litigation block.
4. The underlying FWD for CAFC member appeal 25‑2131 is unidentified in my sources.
5. No 2026 outcome exists as far as I can verify for 25‑2130. If one has issued since these snapshots, I could not confirm it.

I'll research the assignment history for US 10,073,960 before constructing the record.

Let me dig into the specific assignment record details and Proxense's corporate history.

Let me try to pin down the specific reel/frame and correspondent details for the 2017 recording.

Inventors

Notes / unusual patterns:

• This is a lone-inventor patent. The rest of the immediate family (e.g., US 8,646,042; 9,049,188; 9,235,700; 9,679,289; 11,080,378) and the underlying provisional 60/992,953 "Reverse Prox" name Brown, Giobbi, and Hirt; the '960 continuation names only Brown. Prosecution-address block confirms "Patent Law Works/Proxense — Greg Sueoka, 201 South Main Street, Suite 250, Salt Lake City, UT 84111," with applicant "Proxense, LLC, Bend, OR."
• No evidence of inventor departure within 12 months of filing. Brown was a long-tenured Proxense inventor (PatentBuddy lists 20+ Proxense filings spanning 2007–2017). No pre-fire-sale inventor-attrition signal.
• Caution — name collision: a different "David L. Brown" is a prolific KLA/KLA-Tencor inventor (image-sensor art). Do not merge the two. The Proxense David L. Brown is the Jupiter, FL / Bend, OR inventor.

Original assignee

• Entity on the issued patent: Proxense, LLC, Bend, OR (Delaware LLC per Proxense v. Intel, No. 24-cv-283, ¶2). Recorded address in Proxense assignment filings: 689 NW Stonepine Drive, Bend, Oregon 97701.
• Corporate history: incorporated 2001 as Margent Development LLC; renamed Proxense LLC in 2005 (per Proxense v. Microsoft, 6:23-cv-00319, complaint ¶36 n.4). Founder/CEO John Giobbi.
• Did they ship a product embodying the claims? Historically yes, but not now. The complaints allege that from ~2004–2012 Proxense "developed… commercial products, employing over thirty engineers," including a "ProxPay" mobile-payment/PDK proximity product line. Independent press (DARKReading 2009: "in the process of bringing to market a proximity-based communications device"). However, contemporaneous reporting (Biometric Update, May 2023) states Proxense "does not have a working website, and has not publicly announced any customers." So the historical operating business has effectively wound down into a licensing/assertion posture; Unified Patents lists "Parent Company: Proxense LLC" as current assignee.
• Current status: operating as a patent owner/assertion entity — ~80 patents in the portfolio (per complaint ¶31), no known current product sales. Not in bankruptcy and not recorded as acquired or dissolved on the assignment record.

Assignment timeline

USPTO Assignment Center / Google Patents legal-events record shows exactly ONE recorded assignment for this patent (App. 15/595,739):

• 2017-07-20 (executed) / recorded 2017-07-20 — Reel/Frame not retrievable from the sources I could access (flag below)
  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)"
  • Assignor: Brown, David L. (Jupiter, FL)
  • Assignee: Proxense, LLC (Bend, OR)
  • Correspondent: could not retrieve the recording correspondent. The prosecution firm of record for this patent is Patent Law Works LLP (Greg Sueoka, Salt Lake City, UT) — and Patent Law Works is the agent of record on essentially the entire modern Proxense portfolio (e.g., US 11,080,378; 11,551,222; 12,597,030). Flag: treat Patent Law Works as the repeat-player firm for prosecution, not a confirmed assignment-recording correspondent — I could not verify the latter.
  • Context: inventor-to-employer assignment (standard prosecution-time recordation; no third party, no shell entity).

Comparator (same assignee, different patent — shows Proxense's recorded-assignment mechanics): U.S. 9,298,905 — Reel/Frame 034843/0145, recorded Jan 29, 2015, Assignor Giobbi, John J. → Assignee PROXENSE, LLC, 689 NW Stonepine Drive, Bend, Oregon 97701. This confirms Proxense records inventor assignments to itself at that Bend address; it does not establish the reel/frame for the '960 recording.

⚠️ Verification gap — stated plainly: Google Patents legal events confirm a 2017-07-20 assignment from Brown to Proxense, but I could not retrieve the specific reel/frame number or the recorded correspondent for that entry from the sources available to me. Do not treat any reel/frame I might imply for the '960 as verified. Verify at the Assignment Center: https://assignmentcenter.uspto.gov/ (or https://assignment.uspto.gov/patent/index.html — search patent 10073960).

What the record does not contain

• No post-issuance transfer. No assignment after the 2018-09-11 issue date — no move to an "IP Holdings/Licensing/Ventures" vehicle, no security agreement, no merger, no change of name, no release.
• No chain. Ownership has been continuous in Proxense from the inventor assignment through today. The chain terminates at the original assignee.

Timeline diagram

timeline
    title Ownership of US 10073960
    2007 : Provisional filed by Brown Giobbi Hirt
    2008 : First non provisional filed
    2017 : Continuation application filed
         : Assignment recorded Brown to Proxense
    2018 : Patent issued to Proxense
    2023 : Proxense sues Microsoft and Google
    2024 : Microsoft and Google file IPRs
    2025 : PTAB finds all claims unpatentable
         : Appeal to Federal Circuit

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. There is exactly one assignment (2017-07-20, inventor → Proxense) and no transfer into a separate licensing-only LLC. Ownership stays with the original operating entity. No "IP/Licensing/Holdings/Ventures" vehicle appears.

2. Known asserter in the chain — PRESENT (moderate). The current (and only) assignee, Proxense, LLC, is a high-frequency patent plaintiff: suits against Microsoft (6:23-cv-00319), Google (6:23-cv-00320), Apple (6:24-cv-00143), Samsung (6:21-cv-00210), Intel (6:24-cv-00283), plus the W.D. Tex. case flagged on Google Patents (6:25-cv-00016). It is tracked on Unified Patents (https://portal.unifiedpatents.com/patents/patent/US-10073960-B1) and its litigation documents are indexed by RPX Insight (insight.rpxcorp.com). This is the conduct-based signal, not a name-based one — Proxense does not appear to be a listed classic NPE fund (Acacia/Marathon/IV etc.); it is the original operating company monetizing its own portfolio.

3. Repeat correspondent across the chain — UNCLEAR / NOT PRESENT as a chain signal. With only one assignment link, there is no recurrence within this chain to measure. Patent Law Works LLP (Greg Sueoka) is the recurring firm across the broader Proxense family (prosecution of record), but I could not confirm it as the assignment-recording correspondent for the '960 entry. Flagged as unverified rather than counted.

4. Cascading transfers — NOT PRESENT. No chained LLC hops; one assignment in ~11 years of ownership.

5. Pre-litigation transfer — NOT PRESENT. The only assignment (2017-07-20) precedes the first suit naming this patent (May 2, 2023) by ~5.8 years. No <6-month assignment-to-suit sequencing.

6. Bankruptcy fire-sale — NOT PRESENT. No Chapter 7/11 record; ownership uninterrupted in Proxense.

7. Privateering — NOT PRESENT. No operating-company-to-NPE transfer; the plaintiff is the owner of record, not a proxy funded by a competitor.

8. Defensive aggregator — NOT PRESENT. The chain does not terminate at RPX/AST/LOT/Unified/OIN; it terminates at the asserting entity itself.

Net: the assignment record is a clean, single-link inventor→original-assignee chain — the opposite of the classic shell-transfer pattern. The NPE character here comes from the assignee's conduct (serial assertion against large operating companies, no current products, no website), not from any transfer mechanics.

Verdict

NPE — moderate confidence.

Justification: The assignment record contains only one link — the 2017-07-20 recording of Brown→Proxense (Google Patents legal events) — and no post-issuance transfer, so the classic shell-entity/cascading-transfer signals (#1, #4, #5, #6, #7, #8) are all absent, which by itself would read as "Insufficient data." The NPE call rests on a single strong conduct signal (#2): the sole/current owner, Proxense, LLC, is a high-frequency patent plaintiff (Unified Patents- and RPX-tracked; suits vs. Microsoft, Google, Apple, Samsung, Intel) with no product in commerce (no working website, no announced customers per Biometric Update, May 2023). Because the NPE character derives from the assignee's own assertion conduct rather than from any assignment-chain transfer, I stop at moderate — a stricter, assignment-record-only reading would be "Insufficient data."

Verify at: https://assignmentcenter.uspto.gov/ (search 10073960) — confirm the 2017-07-20 entry's reel/frame and recording correspondent, which I could not retrieve.

Sources: https://patents.google.com/patent/[US10073960B1](/patent/US10073960B1)/en • https://portal.unifiedpatents.com/patents/patent/US-10073960-B1 • https://insight.rpxcorp.com/litigation_documents/15827032 • https://www.biometricupdate.com/[202305](/patent/202305)/having-slayed-samsung-in-patent-dispute-biometrics-firm-targets-google-microsoft • https://www.courtlistener.com/docket/67312721/1/proxense-llc-v-microsoft-corporation/

I'll research the prior-art citations for US 10,073,960 B1. Let me run several targeted searches.

I have strong data on the IPR grounds. Let me now retrieve the references cited on the face of the '960 patent itself and the prosecution references (Finn, Hochstein).

Prior Art Analysis — US 10,073,960 B1

Patent under analysis: US 10,073,960 B1, Hybrid Device Having a Personal Digital Key and Receiver-Decoder Circuit and Methods of Use, David L. Brown, assignee Proxense, LLC. App. 15/595,739 filed May 15, 2017; issued Sept. 11, 2018. Earliest priority: provisional 60/992,953 ("Reverse Prox"), Dec. 6, 2007 (root non-provisional 12/329,329 filed Dec. 5, 2008).

Important framing caveat before the reference-by-reference list. You asked which references anticipate under §102. I must flag that every substantive prior-art ground actually presented against the '960 patent was an obviousness ground under §103, not a standalone §102 anticipation ground. The Microsoft and Google IPR petitions each listed the legal basis column as "35 U.S.C. § 102/103," but the Grounds themselves are expressly worded "Claims 1–20 are rendered obvious by…" (Bd. Pet., IPR2024-00405, §III). Where a §102 posture is nevertheless plausible for a reference, I say so explicitly and label it as my inference, not as something the Board or the petitioner asserted. I also flag what I could not independently verify.

Note on pre-AIA vs. AIA: Because the '960 family's root application (12/329,329) was filed Dec. 5, 2008 — before the March 16, 2013 AIA date — the pre-AIA §102 framework (including §102(e), "secret prior art") governs prior-art qualification. This is why a 2007 Giobbi publication and a 2004 Giobbi publication both qualify notwithstanding the '960 patent's 2017 filing date.

---

A. References applied during original prosecution of the '960 patent

From the Microsoft petition (MS‑1002) and Dr. Traynor's declaration (MS‑1003, ¶28), the application was allowed after a single office action that contained a §103 rejection over two references plus a double-patenting rejection. The applicant amended claim 1 (replacing "reader decoder circuit (RDC)" with "wireless communication interface"; adding "a battery," "coupled to draw power from the battery," and "enables one or more of an application, a function, and a service") and added 19 claims, then filed a terminal disclaimer; the Examiner allowed the claims as overcoming the rejection.

Confidence flag: I could retrieve the fact and role of these two references (they are recited in the petition and in Ex. 1003 ¶28 as the Finn‑Hochstein §103 rejection) but I could not verify their titles, full bibliographic dates, or subject matter from the sources available to me. I am deliberately not guessing. Note also that common names "Finn" and "Hochstein" appear on unrelated later documents (e.g., an IBM "virtual universe" appeal by a Peter G. Finn; various "Hochstein" LED-lighting patents), so I will not attribute content to these publications without confirmation.

Separately, the '960 specification incorporates by reference three of applicant's own earlier applications: Ser. No. 11/292,330 ("Personal Digital Key And Receiver/Decoder Circuit System And Method," filed Nov. 30, 2005); Ser. No. 11/620,581 ("Wireless Network Synchronization Of Cells And Client Devices On A Network," filed Jan. 5, 2007); and Ser. No. 11/620,577 ("Dynamic Real-Time Tiered Client Access," filed Jan. 5, 2007). These are background/incorporated material, not adverse prior art.

---

B. Prior art relied on in the PTAB proceedings (the art that actually mattered)

Three IPR petitions were filed on the '960 patent (IPR2024‑00405 Microsoft; IPR2024‑01320 Google; IPR2024‑00784 Google). The same core four references drive the Microsoft and joinder petitions.

3. US 2007/0245157 A1 — "Giobbi‑157"

• Full citation: U.S. Patent Application Publication No. 2007/0245157 A1 (Giobbi et al.), "Giobbi‑157." Exhibit MS‑1005 / Ex. 1006.
• Dates: Filed May 5, 2007; published Oct. 18, 2007 (per petition Ground table). Prior art under pre‑AIA §102(e)/(b) relative to the Dec. 6, 2007 priority date.
• Description: Directed to "efficient, secure and highly reliable authentication for transaction processing and/or access control applications." Discloses a PDK storing profiles (e.g., a biometric profile) in tamperproof memory; the PDK may be integrated into a cell phone; the PDK communicates with an RDC that authenticates the PDK and enables a "transaction"; the PDK may store credit/debit card, bank, or personal information; multiple registries (Central Registry 114, private registry 116, validation database 112) administered by a trusted third party; and the external PDK may be embodied in jewelry — "watches, rings, necklaces or bracelets."
• Claim mapping (as petitioners used it): Fits claim 9 (external authentication database separate from the merchant) and claims 12, 13, 20 (form factor "jewelry" / "a watch"), and supplies the integrated-PDK-with-secure-memory element for claim 1. This corroborates the dependent-claim notes in the previously generated summary (claim 9 = external auth DB separate from merchant; claims 12/13/20 = jewelry/watch).
• §102 vs §103: Used in §103 combinations (Giobbi‑157 + Giobbi‑139 [± Dua]; Broadcom + Giobbi‑157). Standalone anticipation of claim 1 is not something petitioner asserted.

4. US 2004/0255139 A1 — "Giobbi‑139"

• Full citation: U.S. Patent Application Publication No. 2004/0255139 A1 (Giobbi), "Personal Digital Key Digital Content Security System," "Giobbi‑139." Exhibit MS‑1006 / Ex. 1007.
• Dates: Filed May 17, 2004; published Dec. 16, 2004 (per petition Ground table). Prior art under pre‑AIA §102(b)/(e) relative to Dec. 6, 2007.
• Description: Shares Giobbi‑157's goal of securing physical/digital access. Teaches that an RDC can be incorporated into a cell phone / PDA; the RDC "is an integrated circuit able to process PDK-Key information, as well as encrypt/decrypt"; an integrated RDC can "enable/disable" the device based on communication with a PDK (and "lock and disable" the device when the PDK-Key is absent); and describes wired PDK↔RDC coupling (serial cable; smart/magnetic card into a reader slot).
• Claim mapping: Supplies the integrated wireless communication interface that enables/disables a function/service (claim 1 / claim 14) and the absent-PDK-disables-device behavior.
• §102 vs §103: Used in §103 combination with Giobbi‑157. Standalone §102 was not asserted.

5. US 9,042,819 — "Dua" (and its publication US 2006/0258289)

• Full citation: U.S. Patent No. 9,042,819 (Dua) — Exhibit MS‑1007 / Ex. 1007; and, in Google's IPR2024‑00784, U.S. Patent Application Publication No. 2006/0258289 (Dua) (Ex. 1004).
• Dates: Patent filed Sept. 30, 2014; issued May 26, 2015. The petition notes Dua "relies upon previous applications which claim priority to May 12, 2005," and the published application (US 2006/0258289) issued from that earlier chain.
• Description: A wireless media player / cell phone establishing wireless connections to enable functions and exchange data; employs an RFID system with an RFID "electronic key" (tag) and RFID reader, for point-of-sale transactions.
• Role: Presented as corroborating evidence — an RFID tag (analogous to the PDK) and RFID reader (analogous to the RDC) integrated into a single mobile device for secure transactions.
• §102 vs §103 caveat (important): The patent (2014/2015) post-dates the '960 priority date; it is only citable via its earlier priority/published application (US 2006/0258289; priority May 12, 2005). It was used as an obviousness adjunct (Ground 1B), never as a standalone §102 reference.

6. EP 1 536 306 A1 — "Broadcom" / Buer et al.

• Full citation: European Patent Application Publication No. EP 1 536 306 A1 (Buer et al.), "Broadcom." Exhibit MS‑1008 / Ex. 1005 ("Buer").
• Dates: Filed Sept. 30, 2004; published June 1, 2005 (per petition Ground table). Prior art under pre‑AIA §102(b)/(e).
• Description: An "access device" (e.g., a mobile phone) containing a secure "authentication component" (read as the claimed integrated secure memory) and a "wireless proximity reader" (read as the claimed wireless interface), which communicates with an external "wireless proximity token" to enable access to a secured service. FIG. 3 token 316 shows an antenna 334, an RF interface 336, and credentials 328.
• Claim mapping (as petitioners used it):
  • Ground 2A — Broadcom alone: asserted against claims 1–8, 10–11, and 14–19. (Note: framed as an obviousness ground over a single reference.)
  • Ground 2B — Broadcom + Giobbi‑157: against claims 1–20, with Giobbi‑157 supplying dependent-claim features (jewelry/watch → claims 12, 13, 20; third-party auth DB → claim 9).
• §102 vs §103: Broadcom-alone (Ground 2A) is the closest thing to a §102-style, single-reference showing in the record, but it was pleaded as obviousness. If you were building a strict §102 theory, Broadcom (alone) would be the candidate to test first against claims 1–8, 10–11, 14–19.

7. Additional references in Google's IPR2024‑00784 exhibit list (dates/content unverified)

Google's updated exhibit list (Sept. 10, 2024) shows the following additional asserted exhibits. I have the citations but could not confirm their dates or technical content from my sources, so I present them as-is and do not attribute subject matter to them:

(Petitioner's Ex. 1003 was the Declaration of Andrew Wolfe, Ph.D.; Ex. 1009/1020 were the Samsung and Google claim-construction orders; the remainder were litigation documents. Those are not prior art.)

Cross-reference / consistency check: The previously generated summary correctly identified Giobbi‑157 + Giobbi‑139 + Dua (Ground 1) and Broadcom/Buer + Giobbi‑157 (Ground 2), the claim-9 and claim-12/13/20 dependent-claim anchors, and the two Examiner‑cited references (Finn + Hochstein). No contradiction found between that summary and what I retrieved here. The one refinement: the earlier summary listed these as "obvious over Giobbi‑157 + Giobbi‑139 … and over Broadcom/Buer + Giobbi‑157"; the fuller record shows a Broadcom-alone ground (2A) in addition to the Broadcom + Giobbi‑157 combination (2B), and adds the Google‑only references in row 7.

---

C. Quick reference table (citation → date → role → claims)

---

D. Bottom line and explicit uncertainties

1. The operative invalidity art against the '960 patent is Giobbi‑157, Giobbi‑139, Dua, and Broadcom/Buer. On July 17, 2025 the Board issued a Final Written Decision in IPR2024‑00405 holding all challenged claims (1–20) unpatentable on these grounds; that decision is currently on appeal at the Federal Circuit (No. 25‑2130). Google's paral- lel petition (IPR2024‑00784) instituted and then settled; the joinder petition IPR2024‑01320 was not instituted (procedural). (These are consistent with the previously generated summary.)
2. No reference was applied as a strict §102 anticipation reference. Even the single-reference Broadcom showing (Ground 2A) and the Giobbi‑157-inclusive showings were pleaded as §103 obviousness. If the task requires a §102 theory, Broadcom-alone is the most natural candidate to test against claims 1–8, 10–11, 14–19, and Giobbi‑157 would be the candidate for the jewelry/watch dependents (12/13/20) — but I am labeling these as inferences, not record facts.
3. Unverified items I will not fabricate: the titles/dates/subject matter of US 2006/0208066 A1 (Finn) and US 2006/0113381 A1 (Hochstein); and the dates/content of the Google IPR2024‑00784 references (Lee, Jei, Bella, Nishikawa, Kotola, Von Hoffmann, Bluetooth spec). Each needs confirmation against the primary documents (Ex. 1005/1006 in IPR2024‑00405; Ex. 1008–1021 in IPR2024‑00784).
4. The formal "References Cited" block on the patent face (Unified Patents lists "Patent Art (20)") was not retrievable in full text from my sources; the list above covers the references that were substantively applied, which is the analytically relevant set.

Primary sources used: Microsoft IPR2024‑00405 Petition and Ex. 1003 (Traynor Decl.) — docketalarm.com / ai‑lab.exparte.com; Google IPR2024‑01320 Petition and IPR2024‑00784 Updated Exhibit List (Sept. 10, 2024) — docketalarm.com; Google Patents US10073960B1; Unified Patents portal.

I'll verify the Final Written Decision reasoning and the motivation-to-combine findings before building the §103 analysis.

Let me pull the actual '960 Final Written Decision text and the institution decision constructions.

Let me try to pull the '960 FWD itself and the dependent claim text for claims 12/13/20.

Obviousness Analysis Under 35 U.S.C. § 103

US 10,073,960 B1 — Hybrid Device Having a Personal Digital Key and Receiver-Decoder Circuit and Methods of Use

Scope note: This section builds on the Prior Art section above and does not re-describe the references. It takes as given the reference set (Giobbi‑157, Giobbi‑139, Dua, Broadcom/Buer, Nishikawa) and the procedural posture (IPR2024‑00405 FWD unpatentable; appeal pending at CAFC 25‑2130). What it adds is the legal §103 architecture: which combinations work, on which claims, and why a POSITA would have been motivated to make each one — grounded in the record evidence I could retrieve.

---

1. Framework, and what "obvious" has to mean here

Governing law. The '960 application (15/595,739) is a continuation of pre‑March 16, 2013 filings, so pre‑AIA §§ 102/103 govern — the Board confirmed this framing for the sibling '289 patent (IPR2024‑00783 Inst. Dec., n.1, "[b]ecause the application from which the '289 patent issued was a continuation of applications filed before this date, the pre‑AIA versions of §§ 102, 103 apply"). This matters because it makes Giobbi‑157 (2007), Giobbi‑139 (2004), and Broadcom (2005) all available as §103 art against a Dec. 6, 2007 invention date.

Critical date used throughout: December 6, 2007, per Dr. Traynor: "I have applied a date of December 6, 2007, as the date of invention in my obviousness analyses, although in many cases the same analysis would hold true even at an earlier time."
https://www.docketalarm.com/cases/PTAB/IPR2024-00405/Microsoft_Corporation/01-16-2024-Petitioner/Exhibit-1003-Declaration_of_Dr_Patrick_Traynor/

Level of ordinary skill (POSITA). Proxense's own formulation: "a bachelor's degree in computer or electrical engineering (or an equivalent degree) with at least three years of experience in the field of encryption and security (or an equivalent)." That is a mid-level, non-esoteric POSITA — which is the single most important input to the obviousness call, because it means a systems-integration motivation suffices.
https://www.docketalarm.com/cases/PTAB/IPR2024-00407/Microsoft_Corporation_v._Proxense_LLC/docs/04-29-2024-Patent_Owner/Exhibit-2006-Responsive_Claim_Construction_Brief.pdf

Graham/KSR frame applied. Every combination below satisfies at least three KSR rationales: (i) combining prior-art elements according to known methods to yield predictable results; (ii) use of a known technique to improve a similar device in the same way; and (iii) applying a known technique to a known device ready for improvement. The petitions expressly invoked all three (MS Pet. excerpt quoting KSR v. Teleflex, 550 U.S. 398, 415–421 (2007), and MPEP 2143).

⚠️ Flag on my own sourcing. I retrieved the verbatim claim text, the petition's element mapping and motivation analysis, and a sibling Final Written Decision — but I did not retrieve the verbatim text of the '960 FWD. I therefore state the disposition (claims 1–20 unpatentable) as reported, and I do not assign ground-by-ground reasoning to the '960 panel. See §10.

---

2. The claim set that must be invalidated (and a breadth point that matters)

From the listing of challenged claims in the IPR2024‑00405 petition, claim 1 reads:

Source: https://ai-lab.exparte.com/documents/ptab/ptacts-uspto/1555152/.../IPR2024-00405_doc_2_exparte.pdf

🔎 Cross-reference refinement to the earlier "Patent summary." That section described claims 2–13 as depending from claim 1 and 15–20 from claim 14, and flagged that it could not verify every dependent claim verbatim. I can now confirm and complete that picture (dependent claim text below, §8), and I can now confirm the structural point that the earlier summary implied but did not state: '960 claim 1 is broader than the claim 1 of its sibling patents '042 and '289. The '289 claim 1 recites "an integrated reader-decoder circuit (RDC)…," whereas '960 claim 1 recites "an integrated, wireless communication interface." The '960 specification and the prosecution history show the applicant replaced "RDC" with the broader "wireless communication interface" during prosecution (per the earlier "Prior art" section and MS‑1002/Ex. 1003 ¶28).

That substitution is legally decisive for §103 purposes: the broader claim is harder to defend, not easier. A reference that discloses any wireless interface (Bluetooth, RFID reader, proximity reader) reads on [1.3] — it need not disclose the applicant's particular RDC. For the same reason, if the narrower '042/'289 claims fell, the '960 claims fall a fortiori.

---

3. The combinations, ranked

The single strongest combination is A, because it is a two-reference (really two-flush-family) showing that covers all limitations of all claims 1–20 and carries an unusually well-documented motivation. The single cleanest combination is B, because one reference (Broadcom) supplies every element of the independent claims on its own.

---

4. COMBINATION A — Giobbi‑157 + Giobbi‑139 [± Dua]

4.1 Element-by-element mapping (claim 1)

Claim 14 (method) maps to the same disclosure via Giobbi‑157's Fig. 4 process flow: (i) establishing the wireless link between the integrated interface and the external device; (ii) receiving a signal (the profile/credential exchange); (iii) generating an enablement signal — the step the petition treats as the RDC "complet[ing] the transaction" / enabling access. Note: "enablement signal" was a term Google proposed to construe as plain and ordinary meaning ('960 claim 14) — https://gaeflexstaging-dot-docketupdate.appspot.com/cases/PTAB/IPR2024-00784/.../Exhibit-1015-Google’s_Proposed_Constructions.... I have not verified the petition's verbatim mapping of this specific phrase; treat the third step as inferential but secure given [1.4]'s mapping.

4.2 Motivation to combine — the key passage

The petition's motivation analysis is unusually complete and is quoted/paraphrased in the record:

1. Same field of endeavor and similar purposes. "Giobbi‑157 and Giobbi‑139 are directed to similar operations having similar purposes (e.g., 'provide efficient, secure, and highly reliable authentication for … access control applications' and 'protect[] any data stored on such devices')."
2. Similar structures for performing those operations — "e.g., PDKs and RDCs."
3. Compatibility. "the techniques of Giobbi‑157 and Giobbi‑139 are compatible with each other, given that both were filed by the same applicant and describe similar techniques."
4. Predictable operation. "combining the techniques of Giobbi‑157 and Giobbi‑139 would have worked in the same manner already described in the references."
5. Express teaching of the very integration claimed. "Giobbi‑157 expressly teaches the integration of a PDK into a mobile device[;] Giobbi‑139 expressly teaches the integration of a RDC into a mobile device."
6. A concrete, two-way benefit. Dr. Traynor: a POSITA "would have further recognized the benefit of using the integrated PDK to enable secure access to assets external to the mobile device" (e.g., external hard drive, room, safe) in addition to "using the integrated RDC to enable secure access to assets internal to the mobile device."

Sources: MS petition excerpt at https://ai-lab.exparte.com/documents/ptab/ptacts-uspto/1555152/.../IPR2024-00405_doc_2_exparte.pdf; Ex. 1003 ¶52.

Why this is legally sufficient motivation. The rationale is a textbook KSR "predictable results" case: two known security primitives (PDK, RDC) that were already designed to interoperate, applied to a known platform (a cell phone), producing the expected benefit of a single device that can both (a) be authenticated to external systems and (b) authenticate others at itself. There is no need for the references to suggest the '960 invention specifically; it is enough that they, or the knowledge of the POSITA, suggest the combination of elements — which they do expressly. See KSR, 550 U.S. at 418–421 (predictable variation; "known technique to improve similar devices in the same way").

4.3 What Dua adds (Ground 1B) — corroboration and the corroboration trap

Dua is used not as a missing element but as corroborative evidence of feasibility: "Dua further corroborates the teachings of the Giobbi references and provides implementation details for integrating an RDC and PDK into the same device." The petition's syllogism: (i) Dua's RFID Tag ≈ the PDK; (ii) Dua's RFID Reader Unit ≈ the RDC; (iii) Dua integrates both a reader unit 304 and a tag unit 306 into one mobile device (FIG. 4A) and uses the tag as an "electronic key" and the reader as the "electronic lock" for point‑of‑sale transactions. Because "like RFID technology, Giobbi's PDK and RDC architecture require both components within proximity of each other," a POSITA would recognize the analogy and be confirmed in the integration.

Practical consequence for a defendant: even if Proxense successfully argued that Giobbi‑157 and Giobbi‑139 alone do not expressly integrate both components, Dua is independently probative that such integration was known and routine — that is, Dua defeats any "no reasonable expectation of success" rebuttal whether or not it is needed to supply an element.

---

5. COMBINATION B — Broadcom (Buer) alone (Ground 2A)

Ground 2A — Buer Renders Obvious Claims 1‑7, 10‑11, and 14‑19 (Google) / "Claims 1‑8, 10‑11, and 14‑19 are rendered obvious by Broadcom" (Microsoft).

Broadcom teaches an "access device" (e.g., a mobile phone) containing:

• a secure "authentication component" → mapped to the claimed "integrated, secure memory" [1.2]; and
• a "wireless proximity reader" → mapped to the claimed "integrated, wireless communication interface" [1.3];

which communicates with an external "wireless proximity token" to enable access to a secured service → mapped to [1.4]. The token figure (FIG. 3, token 316) shows an antenna 334, an RF interface 336, and credentials 328.

Motivation to modify/apply. No combination is even required for Ground 2A — it is pleaded as a single-reference obviousness ground. The rationale is the third KSR category: applying a known technique (proximity-token access control) to a known device (a mobile phone) that is ready for improvement, to yield predictable results. Broadcom's own disclosure supplies the architectural motivation: its access device is already described as a mobile phone, so moving the proximity-reader function into the phone (rather than leaving it at a fixed reader) is precisely the "improvement" the reference invites.

Why this ground is strategically valuable. Because it is a single-reference ground covering independent claims 1 and 14 and most dependents, it is structurally immune to the standard multi-reference rebuttals (non-analogous art, incompatible teachings, no bodily incorporation). It is the closest thing in this record to a §102‑style showing without being pleaded as one (see prior art section, §C‑2).

---

6. COMBINATION C — Broadcom + Giobbi‑157 (Ground 2B) — closes claims 9, 12, 13, 20

Broadcom alone does not supply:

• claim 9 — external authentication database separate from the merchant; and
• claims 12, 13, 20 — the jewelry / watch form factor.

Giobbi‑157 does, expressly:

• Registries administered by a trusted third party (Central Registry 114, private registry 116, validation database 112) → claim 9; and
• an external PDK embodied in "watches, rings, necklaces or bracelets" → claims 12, 13, 20.

Motivation. "A POSITA, starting with Broadcom's secure access system, would be motivated to incorporate the teachings of Giobbi‑157 to add further implementation details and features. Giobbi‑157's disclosures on alternative form factors (e.g., watches) for external tokens and specific transaction architectures (e.g., third‑party authentication) were argued to be well‑known, compatible, and advantageous enhancements to Broadcom's system." Expected success: "augmenting Broadcom's system with conventional and compatible features from Giobbi‑157… incorporating known design choices and system configurations into an existing, suitable framework."

This is the "KSR — obvious design choice / market pressure" rationale: alternative form factors (wristwatch, jewelry) and third‑party authentication are predictable, commercially motivated variations of a token‑based access system, not inventive leaps. Note also that the '960 specification itself treats jewelry/watches as a mere embodiment ("a PDK 102 can be integrated into … jewelry items such as watches, rings, necklaces or bracelets"), which undercuts any argument that the form factor was non‑obvious.

---

7. COMBINATION D — Dua + Giobbi‑157 (Google's independent route)

Google's IPR2024‑00784 restructured the primary reference:

• Ground 1: Claims 1‑6, 8‑11, and 14‑19 would have been obvious over Dua in view of Giobbi157.
• Dua's U.S. 2006/0258289 publication (priority May 12, 2005) — not the later patent — is the operative art.

Why a second, independent route matters. Microsoft's route starts from Giobbi‑157 (primary) and adds Giobbi‑139; Google's route starts from Dua (primary) and adds Giobbi‑157. These are independent obviousness theories with a different primary reference. If Proxense were to succeed in disqualifying one primary reference (e.g., by a common‑ownership/§103(c) argument aimed at the Giobbi applications — see §9), the other route survives. Independent, PTAB‑instituted routes starting from different primaries is a strong indicator that no single reference is load‑bearing for the invalidity case.

Google also listed Dua + Giobbi‑157 + Kotola in the sibling '289 proceeding (IPR2024‑00783), supplying Kotola as a further dependent‑claim reference.

---

8. Dependent claims — element-by-element

🔎 Cross-reference check on dependent-claim numbering. The earlier "Patent summary" flagged claim 9 (external auth DB separate from merchant) and claims 12/13/20 (jewelry/watch) and said it could not verify the rest. This table confirms all three and now supplies verbatim text for claims 2–11 and the method counterparts. I found one discrepancy to flag rather than auto-correct: the earlier "Prior art" section attributes Google's Ground 4 (Buer + Nishikawa) to claim 4, but the finger‑scan limitation is claim 5 (claim 4 is biometric information generically, already covered by Giobbi‑157 in Ground 3). Since Nishikawa is a fingerprint/bio‑scanning reference, I would expect it to be aimed at claim 5. Verify against the IPR2024‑00784 petition's Ground 4 heading before citing it — the "claim 4" attribution may be a transcription slip in the aggregated sources.

Composite point. Every dependent claim adds either (a) a use case the references already disclose (biometric auth, financial transaction, third‑party auth), (b) a form factor the references already disclose (cell phone, watch, jewelry), or (c) an access‑granularity feature the references already disclose (different keys → different profiles). None adds an unexpected structural or functional element. There is no dependent claim that can survive independently of claim 1 on this record.

---

9. Proxense's counterarguments — and why they fail

9.1 The two constructions Proxense actually argued

Proxense's POPR attacked Grounds 1A/1B and 2B on exactly two claim‑construction theories:

1. "local, secured information" — Proxense urged it means "information stored within a service block of a PDK and unlocked with a corresponding access key." The Board rejected this: the term "encompasses information stored locally on an integrated, secure memory, wherein the information is secured," and the specification's "in this example" passage is "merely an exemplary embodiment and not a definition."
2. "enabl[ing] one or more of an application, a function, and a service" — Proxense urged it requires "receiving information from a PDK in exchange for an access key." The Board rejected any such requirement, finding no lexicography or disavowal with the required "clarity, deliberateness, and precision."

https://portal.unifiedpatents.com/ptab/case/IPR2024-00784 (Institution Decision, Paper 12, 2024‑11‑18) — and POPR at https://www.docketalarm.com/cases/PTAB/IPR2024-00405/Microsoft_Corporation_v._Proxense_LLC/docs/04-29-2024-Patent_Owner/POPR_filed-6-Patent_Owners_Preliminary_Response.pdf

Why these constructions matter to §103. Both were narrowing constructions designed to insert the "service block + access key" architecture as a missing limitation that only the '960 specification (not Giobbi‑157) discloses. If either construction had been adopted, Combination A would have failed on [1.2]/[1.4]. Because the Board refused to read the specification's service‑block embodiment into the claims, Giobbi‑157's "profiles in tamper‑proof memory" reads directly on [1.2] — the doctrinal hook is Phillips / Comark ("particular embodiments and examples appearing in the specification will not generally be read into the claims").

9.2 Proxense's fallback § 112 ¶ 6 theory

Proxense also argued, in the alternative, that certain terms were means‑plus‑function. Rejected: the claims contain no "means for" language, and the terms are not "generic terms or black box recitations." Consequence: no §112 ¶ 6 construction that could narrow the corresponding structure to Proxense's disclosed RDC/hybrid architecture — i.e., no route to limit [1.3] to an RDC proper. That failure is why Microsoft's substitution of "wireless communication interface" for "RDC" cannot be counteracted by a §112 ¶ 6 argument. (Microsoft later retained Dr. Traynor specifically to opine on §112 ¶ 6 — see Ex. 1032, filed 2025‑01‑07.)
https://www.docketalarm.com/cases/PTAB/IPR2024-00405/Microsoft_Corporation_v._Proxense_LLC/docs/01-07-2025-Petitioner/Exhibit-1032-Supplemental_Declaration_of_Dr_Traynor.pdf

9.3 The argument Proxense did not make — and a caution for defendants

⚠️ Flag — unflagged issue worth stress-testing. Giobbi‑157 and Giobbi‑139 are the patentee's own family / a commonly owned applicant's applications, and the petitioner used that common ownership affirmatively as a motivation-to-combine ("both were filed by the same applicant"). That is a double-edged sword: it invites the pre‑AIA §103(c) response that commonly owned §102(e) art is disqualified as prior art. I found no indication that Proxense raised §103(c) in the '960 IPR POPR (its arguments were confined to the two constructions plus §112 ¶ 6), and the Giobbi references are attributed to Giobbi, a different inventive entity than the '960's sole named inventor Brown — which is the predicate §103(c) requires to be missing. But a defendant should proactively brief the inventive‑entity distinction and confirm ownership dates rather than assume the point is waived. This is my own analytical inference, not a record finding.

---

10. Secondary considerations (objective indicia)

Graham factor four. I found no evidence in the record that Proxense asserted — or the Board credited — any secondary consideration of non‑obviousness (long‑felt need, commercial success, unexpected results, copying, industry praise, licensing). Two structural reasons make such evidence unlikely to succeed here:

1. No nexus. The '960 patent is subject to a terminal disclaimer and, per third‑party reporting, Proxense "does not have a working website, and has not publicly announced any customers"; its historical products ("ProxPay") predate the 2017 continuation filing. Commercial success of a product that is not the claimed invention, and that does not practice the asserted claims, cannot establish nexus.
2. Copying cuts the other way. Microsoft and Google (and previously Samsung) as defendants were accused infringers, not copiers of the patentee's product — and the reference set (Giobbi's own earlier Proxense applications) shows the asserted architecture was the applicant's own earlier disclosure.

If a secondary‑considerations argument is to be made at the CAFC, it would have to come from the IPR record; nothing I retrieved supports it.

---

11. Corroboration: what the tribunals have actually done

Sibling patent, same panel, same art — the closest available proxy for the '960 reasoning. In IPR2024‑00573 (US 8,646,042, the closest sibling), a panel of APJs Dang, McKone, and Beamer — the same panel as the '960 proceedings — concluded:

"Based on the complete record presented, we determine that Petitioner has persuasively shown that the proposed combination of Giobbi '157, Giobbi '139, and Dua teaches the integrated RDC being 'coupled to at least one other component of the hybrid device by a second signal line' as recited in claim 1."

"Petitioner has shown by a preponderance of the evidence that claims 1, 5, 6, 8-11, 13, and 14 of the '042 patent are unpatentable under 35 U.S.C. § 103."

"Because Patent Owner does not separately address these limitations … the Petitioner's positions … stand unopposed as a part of this trial."

http://www.docketalarm.com/cases/PTAB/IPR2024-00573/Microsoft_Corporation_v._Proxense_LLC/docs/08-04-2025-Board/Final_Written_Decision__original-27-Judgment_Final_Written_Decision.pdf

⚠️ IMPORTANT — do not conflate. The quoted passages above are from the '042 FWD (IPR2024‑00573), not the '960 FWD. I flag this because the '042 FWD text is circulating under search results that also return '960 material, and the two patents share a nearly identical title. I retrieved the '042 disposition verbatim but did not retrieve the '960 FWD's reasoning. I therefore present the '042 FWD as corroborating context (same panel, same art, same family, stricter claim [the '042 recites "integrated reader‑decoder circuit") — not as the '960 holding.

What I can say about the '960 FWD: aggregators and Law360 report a 2025‑07‑17 Final Written Decision holding claims 1–20 unpatentable, with the panel recording an "Outcome: Unpatentable" and status "Final Written Decision – Appealed" (Appeal 25‑2130). Law360 (July 21, 2025): "The Patent Trial and Appeal Board has found that Microsoft was able to prove that all the claims across two patents owned by an Oregon startup that has sued the technology giant for infringement were invalid." I have not verified the ground‑by‑ground reasoning, and I do not assert it.
https://www.law360.co.uk/ptab_cases/65a7409542ad0d00fc4c0185 • https://ai-lab.exparte.com/case/ptab/IPR2024-00405/microsoft-corp-v-proxense-llc

---

12. Bottom line on obviousness

Claim 1 and claim 14 are obvious over each of at least three independent combinations, with a documented motivation to combine:

1. Giobbi‑157 + Giobbi‑139 [± Dua] — same field, same applicant, structurally compatible PDK/RDC primitives, both expressly teaching integration into a mobile device, predictable combined benefit (internal + external secure access). Covers claims 1–20.
2. Broadcom (Buer) alone — a single mobile‑phone access device with a secure authentication component and an integrated wireless proximity reader enabling an external service; covers 1–8, 10–11, 14–19.
3. Dua + Giobbi‑157 — an independent route with a different primary reference; covers 1–6, 8–11, 14–19.

Every dependent claim (2–13, 15–20) is obvious because each adds only a disclosed use case (biometric/financial authentication), a disclosed form factor (cell phone, watch, jewelry), a disclosed access‑granularity feature (multiple profiles with distinct keys), or a disclosed third‑party authentication architecture — none of which produces an unexpected result or a structural difference from the prior art.

The most probable remaining question is remedial, not merits‑based. Because the claims are not yet cancelled pending CAFC 25‑2130, Proxense can still assert them — but any assertion must contend with a Board decision that every claim is unpatentable, a sibling FWD on the same art from the same panel, and Board constructions that rejected the two narrowing theories Proxense would need to distinguish the art.

---

13. Confidence, and the verification gaps I am not papering over

One-line recommendation: open any §103 brief on the '960 patent with the '042 sibling FWD (same panel, same art, verbatim quotable) plus the IPR2024‑00784 institution decision's rejection of Proxense's two narrowing constructions, and separately obtain the '960 FWD (2025‑07‑17) PDF from PTAB E2E before quoting anything about the '960 panel's own reasoning.