US 7512806

I'll run targeted searches for the exact patent number and any CAFC 2026 docket activity.

Let me search specifically for any 2026 litigation or CAFC activity involving this patent, and verify the continuity family.

US Patent 7,512,806 — Findings

Important note on scope: I searched via web/USPTO-mirror sources (Google Patents, Justia, uspto.report) and the Federal Circuit's public opinions/orders feed. I found no CAFC 2026 docket or opinion involving 7,512,806, and no litigation at all surfaced for this patent. I could not query PACER/Docket Navigator directly, so treat the "no 2026 CAFC activity" statement as a negative search result, not a certification.

Bibliographic data (from the authoritative patent record)

Ownership chain (per recorded assignments): Palm, Inc. (2001-03-07) → Palm Source Inc. (2003-03-18) → PalmSource, Inc. (2004-05-20) → Access Systems Americas, Inc. (change of name, effective 2006-10-17) → Access Co., Ltd. (2011-03-04).

Family: PCT/US2001/044718 (WO 2002/044873 A2, "Computer related access control"); AU 2002/228670 A1. Two continuation children claim the same 2000-11-30 priority: US 12/364,529 → US 7,774,613 B2 (2009-02-03) and US 12/853,250 → US 8,225,381 B2 (2010-08-09). The earlier publication US 2002/0066041 A1 is the pre-grant publication of this same application — not a separate case.

Abstract

The patent covers protecting a computer network accessed by wireless devices (PDAs) using user-specific biometric data to identify users, prevent use by unauthorized persons, and prevent unauthorized changes to the biometric data. Control of both the biometric data and network access is retained by the network administrator, which also provides anti-theft security. If a device is lost/stolen or the authorized user leaves, the reader stays with the device but the biometric data can be removed remotely, rendering the device inoperable; for a new authorized user, the network manager can remotely reprogram the device.

Plain-language overview of the independent claims

Claim 1 — Method for controlling access to a computer network (read with the spec, which contemplates PDAs/palmtops):

1. Gain network access via a portable computing device.
2. Read user-specific biometric data with a reader coupled to that device.
3. If no biometric data is yet stored on the device, verify whether the user is authorized for network use.
4. If authorized, acquire new biometric data and store it.
5. Compare the read biometric data against previously stored biometric data held both on the portable device and in the computer network, to identify the user — and the on-device stored data can be removed on instruction from a remote station on the network, which retains its own copy.
6. If the comparison fails to identify an authorized user, deny further access to both the network and the portable device.

Dependent claims 2–6 add: wireless connection (2); fingerprint (3); iris scan (4); measured electrical characteristics (5); and remote programming of the device's biometric data (6).

Claim 7 — Method for controlling access to a portable computing device (device-side lockout, independent of network access):

1. Read user-specific biometric data.
2. If nothing is stored on the device, verify the user is authorized for network use; if so, acquire and store new biometric data.
3. Compare against previously stored biometric data to identify the user — again, the stored data can be remotely removed on instruction by a remote station on a computer network that keeps a copy.
4. If the user is not identified as authorized, prevent access to both the network and the device, and power the device down.

Dependent claims 8–11 add fingerprint (8), iris scan (9), electrical characteristics (10), and remote control of the device's stored biometric data (11).

Claim 12 — Apparatus for controlling access to a computer network:
A system comprising (a) a network of one or more workstations that grant access upon biometric authentication; (b) a portable computing device providing wireless network access, powered down on unsuccessful biometric authentication; (c) a biometric reader coupled to the portable device; (d) a storage device storing biometric data "capable of identifying one and only one user" (i.e., single-user binding); and (e) a wireless communication device enabling loading and removing of the on-device biometric data, where removal can be instructed by one of the network workstations, which retains a copy — plus the same bootstrap enrollment logic (verify authorization, acquire and store new data) if no data is yet stored.

Dependent claims 13–21 add: a bus/memory/storage/biometric-reader/communication/processor architecture that reads and compares biometric data (13); PDA implementation (14); reader integrated into the device (15); fingerprint (16); iris scan (17); "any electronically storable identifying biometric data" (18); measured electrical characteristics (19); a remote station on the network (20); and that remote station performing upload, download, and erase of the device's biometric data (21).

Points worth flagging

• The "no previously stored biometric data" /
  "acquire and store new biometric data" language in claims 1, 7 and 12 reads like enrollment/bootstrap logic grafted into the independent claims. The public claim text is unusual for a 2000-filing-date application (priority 2000-11-30), and the file wrapper shows no reexamination, reissue, or certificate of correction in the record I retrieved. I cannot authoritatively explain the provenance of that phrasing from the available data — flagging it as an open question rather than asserting a cause.
• Expired status matters for 2026 appellate activity. With an adjusted expiration of 2023-04-20 and "Expired – Lifetime" status, a live 2026 Federal Circuit appeal on this patent is unlikely (though not impossible if arising from a pre-expiration damages case, IPR, or DJ). Nothing surfaced.
• Do not confuse with "Access Advance LLC" (the HEVC/VVC patent-pool administrator active in E.D. Tex. in 2026) — it is a different entity from Access Co., Ltd., the assignee here.
• Two "controlling access to a network" patents to keep distinct: 7,512,806 (this one) and its two continuations, 7,774,613 and 8,225,381 — same title, same priority date. Claim-scope questions should specify which of the three is at issue.

I'll search for litigation involving US patent 7512806 across patent litigation databases.

Let me search more targeted litigation databases.

Let me check a few more specialized sources.

Litigation Search Report — U.S. Patent No. 7,512,806

Prepared: April 26, 2026 | Subject: Known litigation involving U.S. Patent No. 7,512,806 ("the '806 patent")

Bottom line

I found no known litigation involving U.S. Patent No. 7,512,806. In the sources I was able to reach, the '806 patent does not appear as an asserted patent in any district court action, Federal Circuit appeal, ITC investigation, or PTAB proceeding. No plaintiff, defendant, jurisdiction, case number, filing date, or outcome can be reported, because no case was identified.

I want to be explicit about the confidence level here rather than present this as an exhaustive clearance: this is a "not found," not a verified "none exists." See the limitations section below.

What I searched and what came back

Important disambiguation note: one search hit referenced a publication number "7512806 A" in a 1976–1979 patent-family table (apparently a non-US chemical patent document). That is not U.S. Patent No. 7,512,806 and I have not treated it as such. It is an unrelated alphanumeric coincidence.

Patent-record cross-check (from the authoritative full text)

The Google Patents record for US 7,512,806 was fetched and contains no litigation section, no "Litigation" tab entries, and no litigation history in its prosecution/legal-events data. The events of record are exclusively:

• Assignment chain: Palm, Inc. (2001-03-07) → Palm Source Inc. (2003-03-18) → PalmSource, Inc. (2004-05-20) → Access Systems Americas, Inc. (change of name, 2010-08-27, eff. 2006-10-17) → Access Co., Ltd. (2011-03-04)
• Continuations/related family: US 7,774,613 (from App. 12/364,529) and US 8,225,381 (from App. 12/853,250)
• Foreign/PCT equivalents: WO 2002/044873 A2 and AU 2002/228670 A1
• Fee events and a status of "Expired – Lifetime," adjusted expiration 2023-04-20

Notably, the "Cited By" lists on the family page consist of later patent documents, not litigation. The closest thing to a case record anywhere in the family data is the sibling continuations citing the '806 patent — which is normal patent-family citation, not litigation.

Practical / strategic observations

1. The patent is expired. With an adjusted expiration of 2023-04-20, the '806 patent can no longer be asserted, and any hypothetical past infringement claim is now time-barred. This materially lowers the present value of a litigation search and explains why continued enforcement activity would be unlikely even if it once existed.
2. Ownership sat with non-practicing-adjacent entities. After the Palm → PalmSource → Access Co., Ltd. chain, the asset ended up with a Japanese software company (Access). Given the vintage (priority 2000) and the subject matter (remote biometric provisioning/erasure for wireless network access), this would be a plausible candidate for a later monetization campaign — but I found no evidence one occurred, and I decline to infer one.
3. The related continuations (US 7,774,613 and US 8,225,381) share the same priority date and claimed subject matter. If you are assessing assertion risk in this family broadly, they should be checked individually — I did not confirm any litigation on them either.

Limitations — please read before relying on this

• No direct PACER, Docket Navigator, RPX Insight, or Docket Alarm case-index query was completed. Those are the authoritative sources for a comprehensive litigation check, and several are paywalled. My searches stopped at the search-step limit before I could run every planned query.
• I could not open the Unified Patents litigation caselist for a direct patent-number query (e.g., portal.unifiedpatents.com/litigation/caselist?patents=7512806).
• "No results" from a general web search is weak negative evidence for district court litigation, because the great majority of patent suits generate no broadly indexed web coverage.

Recommended next steps to convert this into a verified negative

1. Query PACER / Docket Navigator / Lex Machina directly by patent number 7512806.
2. Query the Unified Patents portal caselist with ?patents=7512806.
3. Search FreeLaw / CourtListener RECAP for "7,512,806" and for "7,774,613" / "8,225,381".
4. Check PTAB (PTABE2E / Docket Alarm) for any IPR/PGR/CBM naming the '806 patent.
5. If desired, cross-check Access Co., Ltd. as plaintiff in the relevant period (roughly 2011–2023) for any portfolio assertion that might have included this patent under a different asserted number.

If you can authorize access to a litigation index (or want me to continue searching specific databases), I can extend this and convert the current "not found" into a documented negative with case-level citations.

I'll verify the structured data against live sources before stating a conclusion.

No substantive hits. Let me search more targeted terms before concluding.

Let me run a few final checks on litigation history and any defensive-aggregator activity.

PTAB proceedings on file — US 7,512,806 B2

Sourcing note: The structured "PTAB proceedings on file" block in this prompt is the canonical USPTO Open Data Portal (ODP) list and it is empty. I attempted to corroborate or contradict that with targeted web searches (patent number alone, "IPR"/"PTAB" combos, petitioner names, Federal Circuit appeals, and defensive-aggregator queries). Every hit returned was unrelated — e.g., a PTAB termination order for a different patent family (IPR2024-00626) and an IPR petition quote addressing US 5,828,751. No document referencing US 7,512,806 or application 09/727,984 in any AIA-trial context surfaced. I found no IPR, PGR, or CBM number, no institution decision, no FWD, and no CAFC appeal for this patent.

Proceedings overview

Total AIA trial proceedings on US 7,512,806: zero (0 active, 0 with claims invalidated, 0 with claims sustained, 0 settled, 0 institution denials). The patent has never been through an AIA trial — so there is no claims-invalidated finding to lean on, and equally no IPR-driven narrowing. The bottom-line defensive posture is the opposite of "hardened by surviving IPRs": this is an untested, but also now-expired, patent whose practical exposure to you runs almost entirely through its expiration date rather than through the PTAB. The dominant defensive fact is not PTAB at all — per the structured data, the patent's legal status is "Expired - Lifetime," adjusted expiration 2023-04-20 — which caps any damages theory to pre-expiration conduct within the § 286 six-year lookback (i.e., roughly 2017-04-20 through 2023-04-20) and eliminates injunctive relief.

Because there are no proceedings, the per-proceeding template below has no entries to populate. I am not fabricating proceeding numbers to fill it.

Strategic summary

Claim status: all 21 claims UNTESTED (none canceled, none sustained). Claims 1 and 7 are the two independent method claims; claim 12 is the independent apparatus claim; claims 13 and 21 are dependent claims of independent claim 12. Nothing in the structured data or my searches shows any of these claims being invalidated, disclaimed, or confirmed in a contested AIA proceeding. Note that the patent's "related child" continuations — US 7,774,613 B2 and US 8,225,381 B2 — are in the same family and also carry no AIA trials in the structured data. If you are mapping the family's patentability risk, treat the whole family as PTAB-uncontested.

Estoppel landscape: § 315(e)(2) estoppel is a non-issue here because no IPR was ever instituted. There is no petitioner, no privy, and no "ground raised or reasonably could have been raised" that binds anyone. That cuts both ways for you as a defendant: (a) you are not blocked from filing an IPR, but (b) you get none of the benefit of someone else having already killed or narrowed claims. Any invalidity ground you want to assert — in district court under §§ 102/103/112 or in a seriatim IPR — is fully available, subject to the ordinary § 315(b) one-year bar running from service of a complaint. Practically, the strongest § 102/103 art is likely to be pre-2000 biometric-access art (the patent's own cited references are a useful starting set: US 5,712,912; US 5,963,908; US 6,016,476; US 6,087,956; US 6,189,099; US 6,442,286; and WO 1998/012670). The prior-art date of interest is the 2000-11-30 priority date.

Pattern signals — none. No repeat petitioner, because there is no petitioner. No PTAB appeal activity by Access Co., Ltd. (or its predecessors Palm, Inc. / PalmSource / Access Systems Americas). No defensive aggregator (Unified Patents or similar) appears in the chain. The patent's commercial history instead runs through Acacia Research Corp., which held licensing rights to Access/PalmSource patents and, per 2010 WSJ reporting, asserted a portfolio including Palm-origin patents against Apple, RIM, Samsung, Motorola and others in a Tyler, Texas action — with Microsoft taking a license instead of litigating. That is a licensing/enforcement narrative, not a PTAB one. I could not tie a specific Acacia or Access complaint asserting US 7,512,806 to a docket within my searches, so I flag that as not confirmed.

Recommended next steps

• There is no PTAB activity to work from, and no FWD to cite. Do not build a defense narrative around a cancellation that does not exist. If a demand letter asserts this patent, the single highest-value fact is the adjusted expiration of 2023-04-20 — verify it directly against the USPTO Patent Center / Patent Public Search record, and confirm there was no terminal disclaimer or intervening extension. Any past-damages theory is confined to the § 286 six-year window ending 2023-04-20.
• Confirm the negative on the record. Pull the patent's full transaction history and the PTAB E2E "Patent Trial" tab for US 7,512,806 (and family members US 7,774,613 and US 8,225,381) at the USPTO PTAB E2E system, plus a CourtListener / PACER docket check for any Access Co. or Acacia enforcement involving this number. The absence of IPRs is itself informative: a patent in a portfolio that was actively licensed and litigated in the 2010-2012 period would normally have attracted challenges if it were being asserted against deep-pocket defendants — its non-appearance in the PTAB trends toward it either never being squarely asserted or being asserted only against parties who took licenses.
• If you nevertheless consider an IPR/CBM: note that (i) CBM is unavailable — the patent claims a biometric/network-access method, not a "financial product or service," and CBM has in any event sunset; (ii) IPR of an expired patent is permissible but the patent owner cannot amend claims, and the Board will apply Phillips-style construction rather than broadest reasonable interpretation; and (iii) the § 315(b) one-year clock runs from service of a complaint alleging infringement. Given expiration, weigh whether a district-court invalidity case (or a motion practice on damages) is the more efficient path than a post-grant petition.
• What I could not verify: any PTAB or CAFC activity for this patent, any settlement, any judge panel, any claim-level outcome. All of those fields are empty because no proceeding exists, not because the data is merely missing from my sources — but I state that as my conclusion on the evidence available, not as a certified search of the PTAB's complete paper file.

Constraint compliance: No proceeding number has been invented; no FWD or panel has been described because none exists; all dates are given as YYYY-MM-DD.

I'll verify what I can against live sources before writing, since the Google Patents legal-events view I was given omits correspondents.

Good — the reorg chain is confirmed. Now checking for any assertion/licensing activity by the current assignee and any litigation naming this patent.

Ownership & Assignment Analysis — US 7,512,806 B2

"Security technique for controlling access to a network by a wireless device"
App. 09/727,984 · filed 2000‑11‑30 · granted 2009‑03‑31 · inventor Steve Lemke · current assignee of record ACCESS Co., Ltd.

Verification caveat up front. The reel/frame numbers below come from the grant‑level legal‑events record (Google Patents, fetched 2026‑09‑30). That view exposes reel/frame, execution date, recording date, conveyance type, assignor and assignee for this patent — but not the correspondent of record. My attempts to pull the underlying USPTO Patent Assignment Center detail records and any litigation/NPE‑directory cross‑checks were cut off by tool limits, so I have no correspondent data for this chain and I am not going to guess names. Everything else below is grounded in the record shown.

Inventors

• No co‑inventors. Single‑inventor patent, which means there is no co‑inventor‑fracture risk in the chain.
• No departure pattern. Lemke continued filing Palm‑family applications well after the split — e.g. US 10/754,969 (filed 2004) and US 11/520,873 (filed 2006), later associated with Palm, Inc. / PalmSource, Inc. / HP — and his inventor profile lists PEM/PalmSource/Palm/HP/Access as successive assignees. That is the opposite of the "all inventors gone within 12 months → portfolio fire‑sale" tell. Weak‑confidence, since it rests on secondary inventor‑index sites rather than assignment records for those other applications.

Original assignee

Palm, Inc. (Sunnyvale, CA), as recorded at reel 011602/0555 — the assignee at grant, re‑recorded through the Palm corporate split (see timeline).

• Primary line of business: handheld computing — the PalmPilot / Palm series PDAs and, later, the Treo smartphone line; the Palm OS software was pushed down into the spun‑off PalmSource/PalmOne structure after 2003.
• Did they ship a product embodying the claims? Unclear, leaning no. The claims (1 and 7) require a portable device with an integrated biometric reader whose data is remotely removable by a network station. Palm's shipping handhelds of the 2000–2003 window authenticated with passwords and (from the m500 series) were not commercially fitted with fingerprint sensors. I found no evidence of a Palm product practicing the '806 claims. Treat this as an unresolved point, not a proven negative.
• Current status of the original assignee: no longer an independent operating company. Palm, Inc. was acquired by Hewlett‑Packard (2010) after the palmOne → Palm rename; the Palm brand was subsequently separated out and licensed onward. The original assignee entity is effectively absorbed — but that has no bearing on this patent, which had already left the Palm hardware line in the 2003–2004 split.

Assignment timeline

Five recorded assignments. All are corporate‑structure events inside one corporate family; there are no third‑party or LLC transferees anywhere in the chain.

1. 2001‑02‑20 (executed) / recorded 2001‑03‑07 — Reel 011602/0555

   • Conveyance: Assignment (of inventors' interest)
   • Assignor: Steve Lemke
   • Assignee: Palm, Inc. (California)
   • Correspondent: Not available in the source record (Assignment Center detail page not retrievable — see caveat). Cannot assess recurrence.
   • Context: Standard employment‑driven inventor assignment at filing; patent is held by the operating company.

2. 2003‑02‑12 (executed) / recorded 2003‑03‑18 — Reel 013850/0913

   • Conveyance: Assignment
   • Assignor: Palm, Inc.
   • Assignee: Palm Source Inc. (California)
   • Correspondent: Not available.
   • Context: Internal reorg — carve‑out of the OS/software IP into the newly created PalmSource subsidiary ahead of the 2003 Palm → palmOne / PalmSource split.

3. 2004‑04‑05 (executed) / recorded 2004‑05‑20 — Reel 014682/0572

   • Conveyance: Assignment
   • Assignor: PalmOne, Inc.
   • Assignee: PalmSource, Inc. (California)
   • Correspondent: Not available. Note the record lists PalmSource, Inc. twice as owner and is captured as a double entry in the legal‑events feed (reel 014682/0572). That duplication looks like a re‑recorded/clean‑up instrument, not two separate transfers.
   • Context: Internal reorg cleanup — hardware‑side entity (palmOne) conveying residual IP back to the software‑side entity (PalmSource) post‑split. Note the direction is unusual versus entry 2 and is worth reading in the actual instrument if a clean chain is needed.

4. 2006‑10‑17 (effective) / recorded 2010‑08‑27 — Reel 024892/0374

   • Conveyance: Change of Name (no change in legal owner)
   • Assignor: PalmSource, Inc.
   • Assignee: ACCESS Systems Americas, Inc. (California)
   • Correspondent: Not available.
   • Context: Pure name change following ACCESS Co., Ltd.'s $324M acquisition of PalmSource (announced 2005‑09‑09; ~83% premium, all‑cash). Recording lag of ~4 years — a docketing‑laxity tell, but not an NPE tell. Note: no separate assignment instrument conveying PalmSource to ACCESS appears in this patent's record; the acquisition is captured only as this change of name.

5. 2011‑02‑25 (effective) / recorded 2011‑03‑04 — Reel 025898/0852

   • Conveyance: Assignment
   • Assignor: ACCESS Systems Americas, Inc. (California)
   • Assignee: ACCESS Co., Ltd. (Japan)
   • Correspondent: Not available.
   • Context: Upstream intra‑group transfer to the Japanese parent — consolidating the acquired PalmSource portfolio (including this patent) in the publicly traded parent. This is the current assignee of record.

Dormant‑patent note: two continuations of this family were filed by the same owner during this period — US 12/364,529 (filed 2009‑02‑03 → US 7,774,613) and US 12/853,250 (filed 2010‑08‑09 → US 8,225,381) — showing the owner was still actively prosecuting the family in 2009–2010. Both are now Expired – Fee Related; the '806 patent itself is Expired – Lifetime, adjusted expiration 2023‑04‑20 (12‑year maintenance fee paid 2020‑09‑17).

Timeline diagram

timeline
    title Ownership of US 7512806
    2000 : Application filed by Steve Lemke
    2001 : Inventor assigns to Palm Inc
    2003 : Palm Inc to Palm Source Inc
    2004 : PalmOne to PalmSource Inc
    2006 : PalmSource renamed ACCESS Systems Americas
    2011 : Transferred to ACCESS Co Ltd Japan
    2013 : Apple licenses the PalmSource portfolio
    2023 : Patent term expires

(The 2013 entry is the portfolio‑level Apple licence reported by IT‑Times / AppleInsider — approximately ¥1bn, ~US$10M, covering patents originating with Palm and PalmSource. It is not confirmed to name US 7,512,806 specifically.)

NPE / troll‑pattern signals

1. Shell‑entity transfer — NOT PRESENT. All five recorded links are between named operating entities: Palm, Inc. → Palm Source Inc. → PalmSource, Inc. → ACCESS Systems Americas, Inc. → ACCESS Co., Ltd. (reels 013850/0913, 014682/0572, 024892/0374, 025898/0852). No "IP / Holdings / Ventures / Licensing" suffix, no registered‑agent address, no single‑purpose LLC. Terminal assignee ACCESS Co., Ltd. is a Tokyo‑listed software company.

2. Known asserter in the chain — NOT PRESENT. None of Acacia, Marathon, IV, IPNav, Wi‑LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation, or any Spangenberg entity appears. Caveat on this call: ACCESS Co., Ltd. does monetise the legacy PalmSource portfolio by licence (the reported ~US$10M Apple licence in April 2013). That is licensing‑adjacent behaviour by an operating company — worth recording, but it is a licence deal, not an assertion campaign, and I found no infringement suit naming this patent.

3. Repeat correspondent across the chain — UNCLEAR / INSUFFICIENT DATA. The correspondent field is the single most probative tell and it is not exposed in the grant‑level record I have; I could not retrieve the Assignment Center detail pages, so I have no attorney names to compare against Unified Patents / RPX / Patent Progress lists. Structurally, the five instruments span 10 years and four successive corporate identities, which pushes against a single repeat‑player attorney running the chain — but that is inference, not a finding. Recommend re‑running this leg directly in Assignment Center.

4. Cascading transfers through chained LLCs in <24 months — NOT PRESENT. Gaps between links are ~2 years, ~1 year, ~2.5 years and ~4.5 years; all transferors/transferees are corporations in one family, not a chain of LLCs sharing a correspondent address.

5. Pre‑litigation transfer — NOT PRESENT. No infringement action naming US 7,512,806 was surfaced, and the last transfer (reel 025898/0852) predates the 2013 Apple licence by two years. There is no observable suit for the chain to have been arranged around.

6. Bankruptcy fire‑sale — NOT PRESENT. The PalmSource exit was a cash acquisition at an 83% premium ($18.50/share, ~$324.3M), not a Chapter 7/11 disposition. No bankruptcy conveyance appears on any of the five reels.

7. Privateering — NOT PRESENT. The final move is upstream into the acquirer's own parent (ACCESS Systems Americas, Inc. → ACCESS Co., Ltd., reel 025898/0852), the opposite of placing patents into an assertion vehicle. ACCESS shipped its own products (NetFront browser; the PSP browser; Access Linux Platform / Garnet OS), so it has genuine commercial exposure of its own.

8. Defensive aggregator — NOT PRESENT. The chain does not terminate at RPX, AST, LOT, Unified Patents or OIN, so the patent has not been formally neutralised by a defensive buy‑in. Its practical neutralisation comes from expiry instead (adjusted expiration 2023‑04‑20, confirmed by the 2020‑09‑17 12‑year maintenance‑fee payment).

Verdict

Defensive / non-asserting (closest available fit — see caveat)

Justification: the chain is a textbook single‑family corporate reorg — Lemke → Palm, Inc. (reel 011602/0555, executed 2001‑02‑20), then the Palm/PalmSource split (reels 013850/0913 and 014682/0572, 2003–2004), then the name change to ACCESS Systems Americas (reel 024892/0374, effective 2006‑10‑17) after ACCESS's $324M cash acquisition of PalmSource, and finally consolidation into the Japanese parent (reel 025898/0852, effective 2011‑02‑25). There is no shell transferee, no listed NPE, no bankruptcy and no assertion event, and the patent expired 2023‑04‑20, so nothing here reads as NPE acquisition behaviour.

Caveat on the pick: the rubric's "Defensive / non‑asserting" bucket is defined by termination at a defensive aggregator, which is not what happened — the terminal owner is an operating company. I am using that category as the "no assertion / not an NPE" bucket because it is materially closer than "Operating‑company assertion," which requires a current assignee shipping products embodying these claims and suing competitors — I found evidence of neither. I would move to that category only if evidence emerged that ACCESS asserted the '806 claims (or the PalmSource portfolio generically) against a device maker.

Evidence gaps to close before relying on this: (a) correspondent of record for all five reels — the recurring‑attorney test is untested here; (b) whether the 2004 instrument at reel 014682/0572 is a genuine transfer or a corrective re‑recording; (c) whether US 7,512,806 was ever pleaded in any action, including as a portfolio member in ACCESS‑related licensing demands.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search patent number 7512806) — mirror index: https://assignment.uspto.gov/patent/index.html — alongside the grant‑level legal events at https://patents.google.com/patent/US7512806/en.

I'll research US patent 7512806 and its cited prior art references.

I have the patent's citation data. Let me research the key cited references to assess their content against the claims.

Prior-Art Analysis — US 7,512,806 B2

1. Confirmation of the target patent (interpreted literally)

Note on identifier handling: I treated "7512806" literally and did not merge it with the near-neighbor family members listed on the face of the document (US 7,774,613 B2 from 12/364,529; US 8,225,381 B2 from 12/853,250; WO 02/044873 A2). Those are separate continuation/grant documents and are not prior art to 7,512,806 — they share the same 2000-11-30 priority. Grounding: the Google Patents record at https://patents.google.com/patent/[US7512806B2](/patent/US7512806B2)/en shows these as "Family Applications."

2. The controlling limitations (what prior art must meet)

Because anticipation under 35 U.S.C. § 102 requires every limitation in a single reference, the analysis hinges on the distinctive limitations of the independent claims:

• Claim 1 — (a) network access via a portable computing device; (b) reading biometric data peculiar to the user via a reader coupled to the portable device; (c) if no stored biometric data is on the portable device, determining authorization and enrolling new biometric data; (d) comparing the read data to stored data located both on the portable device and on the computer network; (e) the stored data is removable from the portable device on instruction by a remote station on the network, which retains a copy; (f) denying access to both the network and the device on a failed comparison.
• Claim 7 — same core, but framed as device access control, and expressly powering down the device on failure.
• Claim 12 — apparatus: network of workstations authenticating biometrics; portable device with wireless network access, powered down on failed authentication; biometric reader; storage for biometric data "capable of identifying one and only one user"; wireless link enabling loading and removing of biometric data by a workstation that retains a copy; plus the no-stored-data enrollment branch.
• Claim 13 — internal architecture (bus, memory, data-storage, biometric reader, communication device, processor comparing read vs. stored data).
• Claim 21 — remote station that uploads, downloads, and erases the biometric data from the portable device.

The novel hinge across all independent claims is the network-side remote control of on-device biometric data (including remote erasure, with the network retaining a copy) combined with the on-device comparison. That is also exactly what the specification admits was previously unknown — the Background states that biometric security had been "inhibited" for wireless access precisely because "the biometric data and its applicable reader would remain with the wireless device unless there were a means for remotely reprogramming the data resident in the wireless device."

Important admission to note: the specification states "The software to accomplish this comparison is well established in the prior art," and that digitized print storage "is well established in the prior art as is the comparison software." These are applicant admissions that compress the § 102 attack onto the remote-removal/enrollment limitations rather than the biometric-matching core.

3. Cited references (face-of-patent "Patent Citations (8)")

These eight are the references listed under Patent Citations on the patent. All were cited by the examiner (asterisked "Cited by examiner"). All predate the 2000-11-30 priority date either by publication or by § 102(e) filing date.

3.1 US 5,712,912 A

• Full citation: Mytec Technologies Inc.; "Method and apparatus for securely handling a personal identification number or cryptographic key using biometric techniques"; priority 1995-07-28; granted 1998-01-27.
• Description: Uses a biometric (biometric-derived key/bioscrypt) to securely handle a PIN or cryptographic key — biometric data used as the gate for releasing/using protected data.
• Potential § 102 relevance: Discloses the generic concept of gating access to protected data on a biometric match. Touches the biometric-comparison element of claims 1, 7, 12, 13. It does not disclose a portable/wireless device, network-side storage of the biometric, the dual-storage element, or remote removal. Cannot anticipate any independent claim.
• Confidence: High on citation/date (face of patent); moderate on internal disclosure (derived from title and the field).

3.2 WO 98/012670 A1

• Full citation: Dew Engineering And Development Limited; "Biometric identification system for providing secure access"; priority 1996-09-18; published 1998-03-26.
• Description: A biometric identification system that grants secure access on an identity match.
• Potential § 102 relevance: Element (b) and (d) biometric matching for access control → relevant to claims 1, 7, 12. Silent on portable wireless network access, dual-location storage, remote removal, and power-down-on-failure. Cannot anticipate.
• Confidence: High on citation/date; moderate on content.

3.3 US 5,963,908 A — Intel

• Full citation: Intel Corporation; "Secure logon to notebook or desktop computers"; priority 1996-12-23; granted 1999-10-05.
• Description (grounded): Biometric circuitry (voice pattern per the preferred embodiment; also "fingerprint or retinal scanners") internal to an electronic system, verifying the user before allowing access to stored content and to sites on a publicly accessible network (e.g., the Internet); a captured digital voice pattern is compared with a pre-stored master pattern; failure bars access. Source: https://uspto.report/patent/grant/[5963908](/patent/5963908).
• Potential § 102 relevance:
  • Claim 1/7/12: discloses (a) an electronic system that is a laptop/notebook (portable) controlling access to both stored content and network sites, (b) biometric reading coupled to the system, (d) comparison against pre-stored biometric data, (f) denial of access to content/sites.
  • Missing for anticipation: It is a notebook/desktop with a modem/NIC — the disclosure frames it around hardwired/public-network logon, not a dedicated portable-wireless device of the claimed type; and, critically, it has no on-device + network dual storage, no remote removal of biometric data, and no network-retained copy. So it cannot anticipate claim 1 in full, but it is the strongest single-reference teaching for the biometric logon core of claims 1, 7, 12, 13 and the dependent biometric-type claims.
• Confidence: High (retrieved abstract/description).

3.4 US 6,016,476 A — IBM (Maes et al.)

• Full citation: International Business Machines Corporation (inventors Maes, Stephane; Sedivy, Jan); "Portable information and transaction processing system and method utilizing biometric authorization and digital certificate security"; priority 1997-08-11; granted 2000-01-18. Family publication WO 99/08238 A1 (1999-02-18). Sources: https://ie.espacenet.com/publicationDetails/biblio?...WO9908238A1 ; https://uspto.report/patent/grant/[6016476](/patent/6016476).
• Description (grounded): A portable client PDA with touch screen/CPU performing biometric verification, memory storing the user's data, wireless communication (RF and IR) as well as modem/serial, operating in a client/server mode in which a central server verifies the user biometrically (or by PIN/password) and downloads a temporary digital certificate; a local mode performs local verification. Source: EP 1 004 099 B1 description (same family) — https://patentimages.storage.googleapis.com/6d/8d/11/099f3efa4759c6/EP1004099B1.pdf.
• Potential § 102 relevance: This is, among the eight, the closest structural match to claims 1, 7, 12, 13: portable computing device ✓; biometric reader coupled to the device ✓; network-side server that participates in user verification ✓; wireless access ✓; local comparison against stored data ✓.
  • Missing for anticipation: the server downloads a digital certificate, not biometric data; there is no teaching that the biometric data itself is stored on both the device and the network, and no remote removal/erasure of on-device biometric data with the network retaining a copy. Therefore it cannot anticipate the independent claims, but it is highly probative (and would be a primary obviousness reference) for the platform + network-verification architecture.
• Confidence: High.

3.5 US 6,087,956 A — Helferich

• Full citation: Helferich, Richard J.; "Paging transceivers and methods for selectively erasing information"; priority 1997-09-19; granted 2000-07-11.
• Description: Paging/wireless transceivers with the ability to selectively erase information (e.g., messages) — i.e., remote-over-the-air data deletion controlled from the network side.
• Potential § 102 relevance: This reference is the best cited teaching for the "removable on instruction by a remote station" limitation in claims 1, 7, 12, 21. However, it discloses erasure of paging/message information, not biometric data, and does not disclose biometric comparison or device power-down. It cannot anticipate, but it is the key secondary reference supplying the remote-erasure element.
• Confidence: High on citation/date; moderate on content (derived from title).

3.6 US 6,189,099 B1 — Durango Corporation

• Full citation: Durango Corporation (inventors Rallis, William N.; Behar, Yancov); "Notebook security system (NBS)"; filed 1998-02-11; granted 2001-02-13. Source: https://patentimages.storage.googleapis.com/61/79/24/25ead9e575004f/US6189099.pdf.
• Description (grounded): Each user carries a hardware key device holding a serial number and encryption key; a validation record on the computer's hard disk holds validation data; at boot-up, before OS download, the user is validated with no manual password entry; a failure at any step immediately powers down the computer, "rendering it useless to a thief," and the key device is removed after validation.
• Potential § 102 relevance: Discloses power-down-on-failed-validation and anti-theft device disabling — elements of claims 7 and 12 ("powered down upon unsuccessful authentication"). But it is not biometric (hardware key), not wireless, and has no network-side biometric store or remote removal. Cannot anticipate; it is a supporting reference for the power-down limitation.
• Confidence: High.

3.7 US 6,442,286 B1 — STMicroelectronics

• Full citation: STMicroelectronics, Inc.; "High security flash memory and method"; priority 1998-12-22; granted 2002-08-27.
• Description: Security architecture for flash memory (controlling access to and protection of data stored in non-volatile memory).
• Potential § 102 relevance: Bears on the on-device storage of security data recited in claims 1, 7, 12 (data storage device) and 13, and generally on tamper-resistant storage. Not a biometric, network, or wireless reference. Cannot anticipate.
• Confidence: Moderate — I did not independently retrieve its full text; assessment relies on the title and field.

3.8 US 6,496,928 B1 — Microsoft

• Full citation: Microsoft Corporation; "System for transmitting subscription information and content to a mobile device"; priority 1998-01-07; granted 2002-12-17.
• Description: Server-side delivery/updating of subscription information and content to a mobile device.
• Potential § 102 relevance: Relevant to the remote provisioning/loading capability in claims 6 and 21 (network station programming data into the portable device). It does not involve biometrics, comparison, or denial of access. Cannot anticipate.

4. References cited family-wide ("Family Cites Families (5)")

These appear on the face of the document as family-level citations (they were cited in the related applications). They are relevant because they bear directly on the novel hinge.

5. Anticipation matrix (analyst assessment — not a legal conclusion)

Legend: ✓ element disclosed · ~ partially/obliquely disclosed · ✗ absent

Finding: No single cited reference discloses every limitation of claim 1, 7, 12, 13, or 21. In particular, the combination of (i) biometric data stored on both the portable device and the network, and (ii) remote removal of the on-device biometric data by a network station that retains a copy, is not met by any one reference. Under a strict § 102 reading, therefore, none of the cited references anticipates the independent claims. The remote-removal limitation would instead have been attacked via § 103 combination (e.g., US 6,016,476 or US 5,963,908 for the biometric-login/network architecture + US 6,087,956 for remote data erasure + US 6,189,099 for power-down).

6. Most relevant prior art (ranked)

1. US 6,016,476 A (IBM — Maes/Sedivy) — closest overall architecture: portable PDA, on-device biometric verification, wireless, and a network server participating in user verification. Primary reference for claims 1/7/12/13.
2. US 5,963,908 A (Intel) — biometric-gated access to both on-device content and networked sites; supports the "deny network + device" limitation.
3. US 6,087,956 A (Helferich) — the only cited reference squarely teaching remote selective erasure of data, supplying the "removable by a remote station" element.
4. US 6,189,099 B1 (Durango) — power-down on failed validation / anti-theft disabling (claims 7, 12).
5. US 6,038,666 A (TRW) and US 6,111,977 A (Cross Match) — family citations supporting remote identity verification and hand-held fingerprint reading.
6. US 5,727,202 A (Palm Computing) — supports the device/network dual-storage concept.

7. Confidence and caveats

• High confidence in the patent's bibliographic data, the list of cited references, their titles and dates: these come from the authoritative full text you supplied (Google Patents record) and were cross-checked against the Google Patents page (https://patents.google.com/patent/US7512806B2/en).
• High confidence in the content descriptions for US 5,963,908 (uspto.report), US 6,016,476 (Espacenet WO 99/08238 A1 + EP 1 004 099 B1), and US 6,189,099 (patent PDF).
• Moderate confidence in the internal content of US 5,712,912, WO 98/012670, US 6,087,956, US 6,442,286, US 6,496,928, and the four family citations — I did not retrieve full text for these; descriptions are inferred from titles/field and should be verified against the actual documents before any legal use.
• The "Cited By" and "Families Citing this family" lists on the patent page are later, post-dating documents (e.g., the Proxense and AirWatch families) and are not prior art to US 7,512,806; I excluded them from the § 102 analysis.
• "Potentially anticipates" above reflects a technical element-mapping assessment only — it is not a legal conclusion of anticipation or invalidity. Any actual § 102/§ 103 determination requires full-text review of each reference (including all figures and dependently incorporated limitations) and application of the exact claim language.

If you'd like, I can pull and map the full text of US 6,016,476 and US 5,963,908 line-by-line against each limitation of claims 1, 7, 12, 13, and 21 to produce a formal element-by-element anticipation chart.

I'll verify the cited prior-art references and their teachings before building the § 103 combinations.

Obviousness Analysis — U.S. Patent No. 7,512,806 under 35 U.S.C. § 103

Prepared: April 26, 2026
Scope note: Per the task instruction, this analysis uses the prior-art references appearing in the "Prior Art" / "Citations (8)" section of the authoritative patent record for US 7,512,806. I verified the substantive disclosures of five of the eight references by direct retrieval (US 5,963,908; US 6,016,476; US 6,189,099; US 6,087,956; WO 98/12670). For the remaining three (US 5,712,912; US 6,442,286; US 6,496,928) my search budget was exhausted before I could retrieve full text, so I rely on the bibliographic data in the patent record and flag those characterizations as lower-confidence. This builds on the earlier summary and litigation sections; I do not repeat their content.

1. Legal framework and framing assumptions

1.1 Prior-art status of the eight cited references (pre-AIA)

Practical point: the two references that supply the core novelty of the '806 claims — Helferich '956 (remote erase while the remote retains the copy) and Durango '099 (power-down on failed validation) — are § 102(b) and § 102(e), respectively. Only the § 102(e) references (Durango '099, STMicro '286, Microsoft '928) carry the pre-AIA restriction that they may only be relied on for § 103 as of their own filing dates, which they comfortably satisfy. There is no apparent common-ownership disqualification under pre-AIA § 103(c): the '806 patent's chain (Palm/PalmSource/Access) is unrelated to Durango, STMicro, Microsoft, IBM, Intel, Helferich, Mytec, or Dew.

2. What each reference supplies (element inventory)

US 6,016,476 — IBM (Maes et al.), "Portable information and transaction processing system… utilizing biometric authorization and digital certificate security"

Retrieved: https://patents.google.com/patent/[US6016476A](/patent/US6016476A)/en ; https://www.docketalarm.com/cases/[PTAB](/ptab)/IPR2016-01109/.../Exhibit-1004-Exhibit_1004.pdf

• Portable device + biometric reader: a "portable client PDA with a touch screen… microphone and local CPU for processing voice commands and for processing biometric data to provide user verification." Biometric modalities expressly include voice, and dependent claims recite retinal/vascular.
• Storage of user-specific data on device: "a memory for storing financial and personal information of the user."
• Wireless network access: "capable of transmitting or receiving information through wireless communications such as radio frequency (RF) and infrared (IR) communication."
• Network-side control of the credential used to unlock the device: client/server mode downloads a temporary digital certificate from a central server; "In the absence of an unexpired digital certificate… the selected card information will not be written… notwithstanding that the user may have passed local biometric verification."
• Enrollment/pre-enrollment: "the user selects one of the pre-enrolled credit cards that are stored on the PDA."

This is the single best primary reference: it discloses a biometric-verifying PDA, wireless network access, device-resident user data, and the concept that an authority elsewhere on the network periodically provisions (and by expiry, withholds) the data that gates the device's operation.

WO 98/12670 A1 — Dew Engineering, "Biometric identification system for providing secure access"

Retrieved: https://patents.google.com/patent/WO1998012670A1/en

• Portable biometric input device, battery-powered, with fingerprint imaging ("the present invention… may detect fingerprint ridges").
• Wireless transmission of the biometric signal ("transmission means comprises a biometric data encoder and an infrared transmitter").
• On-device storage + comparison: "storage means for storing data related to said biometric data"; "processor means for comparing the biometric data with previously stored biometric data to provide comparison results."
• Encryption of the biometric data (public/private key, session key).
• Host-side authorization: a "biometric security identification system comprising: a portable transmitting module… and a receiving module comprising… means for comparing the encoded biometric data with predetermined reference values, and means for authorizing access to a host system."
• Enrollment/registration: the Dew family (e.g., WO 98/25227 and US 6,259,805) describes in detail the registration process — capturing multiple biometric instances, characterizing them, selecting and storing a template, and validating the registration against the database of existing templates. This is the enrollment teaching.

US 5,963,908 — Intel, "Secure logon to notebook or desktop computers"

Retrieved: https://uspto.report/patent/grant/[5963908](/patent/5963908)

• Biometric-controlled logon on a portable personal computer center/laptop.
• Controls access both to on-board stored content and to networked sites: "for internally performing biometric operations to verify… an individual before allowing him or her access to stored content within the electronic system as well as access to various sites on a publicly accessible network."
• Comparison against "pre-stored master digital voice pattern."
• Notes that "other types of biometric access control systems (e.g., fingerprint or retinal scanners) may be utilized."

US 6,087,956 — Helferich, "Paging transceivers and methods for selectively erasing information"

Retrieved: https://patents.google.com/patent/[US6087956A](/patent/US6087956A) ; https://uspto.report/patent/grant/[6087956](/patent/6087956)

• A portable wireless device (paging transceiver) and a remote system that each store information.
• Remote command → erasure of data on the portable device: the transceiver sends/"[t]he system may transmit… an erase signal…, whereupon receiving the erase signal the system erases the information" (claim 1); complementary claim 92 recites "message erasing means for erasing the message identifier from the memory in response to the receiver receiving a second selective call signal comprising an erase signal."
• The remote system retains its copy and can redeliver: "By erasing the information at the system, the user is able to conserve memory space at the system"; "the system can redeliver the message"; "means for continuing to store the message in the storage unit after transmitting the message to the selective call transceiver" (claim 130).
• Purpose: "provides a user with remote control over remotely stored information," and preserves a system-side copy so the user can recover it.

This is the reference that maps onto the '806 patent's central "removal" limitation: credential/data taken off the device on instruction from a remote station that itself keeps a copy.

US 6,189,099 — Durango, "Notebook security system (NBS)"

Retrieved: https://patents.google.com/patent/[US6189099B1](/patent/US6189099B1)/en

• Pre-boot validation before the OS loads, with the validation record stored on the device (reserved hard-disk sector / BIOS ROM).
• Power-down on failure: "A failure at any step in the user-validation procedure will immediately power down the computer, thereby rendering it useless to a thief not possessing the required key device."
• Expressly contemplates adaptation to PDAs in the sibling patent (US 6,425,084: "the system can be adapted for use with other computers, such as desktops or Personal Digital Assistants (PDA)").
• The specification of the '806 patent itself adopts precisely this logic for the device-lockout embodiment ("the device would not power up without an authorized user's identification").

US 6,442,286 — STMicroelectronics, "High security flash memory and method"

• Title/assignee from the '806 citation table. Supports tamper-resistant non-volatile storage of access-control data on the device — relevant to the on-device biometric store and the anti-tampering rationale. (Content not independently verified; lower confidence.)

US 6,496,928 — Microsoft, "System for transmitting subscription information and content to a mobile device"

• Title/assignee from the '806 citation table. On its title and claim-1-style framing, this is a remote provisioning / remote load-and-remove teaching for a mobile device — the closest match to '806 claims 6, 11, 12, and 21 ("loading and removing," "uploading/downloading/erasing"). (Content not independently verified; lower confidence.)

US 5,712,912 — Mytec Technologies, "Method and apparatus for securely handling a personal identification number or cryptographic key using biometric techniques"

• Title/assignee from the '806 citation table. Supports binding a biometric to a secret/credential so that the biometric gates access to a stored secret, and secure handling of the biometric-derived data. (Content not independently verified; lower confidence.)

3. Element-by-element mapping to the independent claims

3.1 Claim 1 (network-access method)

3.2 Claim 7 (device-access method)

Same mapping as claim 1, minus the network-access preamble, plus the added final step "powering down said portable computing device" — supplied squarely by Durango '099 ("immediately power down the computer"), which the '806 specification mirrors almost verbatim in its own disclosure. No new non-obvious subject matter beyond claim 1.

3.3 Claim 12 (apparatus)

Claim 13 (bus/memory/storage/reader/comm/processor architecture) is met by Dew WO 98/12670's portable module (processor, storage, sensor, wireless transmitter) and IBM '476's PDA architecture. Claim 21 (remote station uploads, downloads, and erases the device's biometric data) is met by Microsoft '928 in view of Helferich '956.

4. The combinations, with articulated motivations

Under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), and MPEP § 2143, a combination is obvious where an articulated reasoning with rational underpinning supports it. I set out four combinations of increasing breadth; Combination I alone disposes of the independent claims, and Combinations III–IV are needed only for the enrollment/bootstrap limitations.

COMBINATION I — IBM '476 + Helferich '956 (+ Durango '099 for claim 7) → claims 1, 7

Rationale (KSR (A) + (F) + (C)): combining prior-art elements according to known methods to yield predictable results; and "the desire to enhance… security" plus "the recognized need for centralized control of portable-device credentials."

1. IBM '476 supplies everything but the remote-removal step. It discloses a biometric-verifying PDA wirelessly accessing a network, where the network-side authority provisions (by certificate) the data that gates device operation. Motivation to improve upon it is supplied by IBM '476's own vulnerability: a temporary certificate that merely expires does not let an administrator take back the credential from a device that is in a thief's hands, nor immediately render that device inert.
2. Helferich '956 supplies the remote-removal mechanism. It expressly teaches a remote system transmitting an erase command that deletes device-resident information while the system continues to store it and can redeliver it. The '806 specification's own stated rationale — "the biometric data would be inaccessible until authorized by the network manager, rendering the device inoperable," with the network manager retaining a copy - is the very benefit Helferich '956 articulates ("remote control over remotely stored information," "conserve memory space at the system," retain-copy-for-redelivery).
3. Why a POSITA would combine them: both are in the same field (remote/system control of data resident on a portable wireless device); Helferich '956's mechanism is a known technique (remote addressing of a device to erase data) applied to a known device ready for improvement (an IBM-'476-type PDA whose operation is gated by a network-issued credential). Substituting a biometric credential for Helferich's message/identifier, as the data to be remotely erased, is a simple substitution of one known data element for another — Helferich itself already erases "identifiers" and message data generically (independent claim 92).
4. Durango '099 clinches claim 7's power-down step: pre-boot validation and immediate power-down on failure for a portable computer (expressly adaptable to PDAs), which the '806 specification adopts as its own device-lockout mechanism.

Conclusion: claims 1 and 7 are obvious over IBM '476 in view of Helferich '956, and (for claim 7's power-down) further in view of Durango '099.

COMBINATION II — Dew WO 98/12670 + Intel '908 + Helferich '956 (+ Microsoft '928) → claims 1, 7, 12, 13, 20, 21

This is the alternative primary-reference route, useful if a patentee argues IBM '476's digital-certificate embodiment is too far from "biometric data stored on the device and in the network."

• Dew WO 98/12670 supplies the portable fingerprint reader, wireless transmission, on-device storage, on-device comparison to "previously stored biometric data," and host-side reference-value comparison plus authorization.
• Intel '908 supplies the "computer network" dimension: biometric logon controlling access to both on-board content and networked sites, compared against a pre-stored master template.
• Helferich '956 supplies the remote-erase-with-retained-copy limitation.
• Microsoft '928 supplies express remote loading and removing of data on a mobile device (claims 6, 11, 12, 20, 21).
• Motivation: Dew WO 98/12670 itself frames the goal as authorizing access to "a host system" and flags the portability of the personal biometric device; Intel '908 supplies the network-access use case; Helferich '956 supplies the revocation mechanism; and the combination is a predictable application of a known security technique to a known portable-device architecture (KSR (D)).

Conclusion: claims 1, 7, 12, 13, 20, and 21 are obvious over Dew WO 98/12670 in view of Intel '908, Helferich '956, and Microsoft '928.

COMBINATION III — Adding the enrollment/bootstrap teaching (Dew registration family) → the "if no previously stored biometric data… acquire and store new biometric data" clauses of claims 1, 7, 12

The earlier summary correctly flagged this enrollment language as the most unusual feature of the '806 claims. It is also the limitation for which the § 103 case is most dependent on a specific secondary reference — but it is still vulnerable:

• Dew's registration family (WO 98/25227; US 6,259,805 / US 6,038,334 / US 6,072,891, all sharing the WO 98/12670 priority lineage and filed 1997-1998) discloses the complete enrollment workflow: capture multiple biometric instances → characterize → select and store a template → validate the registration against the existing template database to prevent false registration. That is literally "acquiring new biometric data… storing said new biometric data."
• IBM '476 independently discloses "pre-enrollment" of the user's credentials on the PDA.
• Microsoft '928 supplies the alternative that the enrollment data be pushed from a remote station.
• Motivation: any biometric system must be initialized for a new user; Dew's registration requirement is inherent to the art, and conditioning initialization on an authorization check (Dew's operator/validation step; IBM '476's service-provider central server) is the obvious way to prevent an unauthorized person from self-enrolling — which is the stated security goal of both the '806 patent and Dew's "reduce false registration" teaching.

Conclusion: the enrollment clauses introduce no patentable distinct invention over the already-known template-registration steps of Dew's family, applied within Combination I or II. This limitation is the weakest link in the claims, not the strongest.

COMBINATION IV — Secondary references for secure on-device storage and credential binding

• US 6,442,286 (STMicro, high-security flash memory) and US 5,712,912 (Mytec, biometric-bound PIN/crypto key) supply (a) tamper-resistant storage of the biometric/credential on the portable device and (b) binding the biometric to a cryptographic secret. Both are cited on the face of the '806 patent and are directly on point for the "data storage device… capable of storing said biometric data" and the security rationale. Their inclusion merely corroborates what Durango '099 and IBM '476 already make obvious, so they are not load-bearing. (Content not independently verified — lower confidence.)

5. Dependent claims

None of claims 2–6, 8–11, or 14–21 adds an element that rescues the independent claims.

6. Anticipated patentee counterarguments and their weaknesses

1. "No motivation to combine / different fields." Weak. All eight references are in computer/network access security using biometrics or remote device-data control; the Federal Circuit and KSR treat "familiar elements according to known methods" as obvious when the references address the same problem. Dew WO 98/12670 and IBM '476 are squarely in portable biometric access control; Helferich '956 is squarely in remote erasure of portable-device data — the exact function the '806 claims require.
2. "Helferich's erased data is a message, not biometric data." Weak. Helferich's independent claims are drafted generically ("information," "identifiers") and expressly contemplate remote erase of device data on command; applying a known remote-erase mechanism to a different category of stored data is a simple substitution, not an inventive leap.
3. "The remote station must retain a copy of the biometric data." This actually helps the obviousness case: Helferich's whole point is that the system keeps the message and redelivers it; the '806 patent's own specification justifies the retained copy as a backup "available to reprogram a new device should the first one be lost, damaged or stolen." That is a design choice with a predictable benefit.
4. "Unexpected results / secondary considerations." No evidence of unexpected results is presented. The patent is expired (adjusted expiration 2023-04-20 per the record), and no litigation, IPR, or CAFC activity surfaced in the earlier searches — so there is no adjudicated record of commercial success, copying, or industry praise to rebut the prima facie case. Absent such evidence, no nexus-based rebuttal is available.
5. The real exposure, if any, is not § 103 but § 112. The "if no previously stored biometric data… acquire and store" language flagged in the earlier summary reads like grafted-on enrollment language; if it lacks written-description support in the as-filed application it is vulnerable under § 112(a), and if "capable of identifying one and only one user" (claim 12) is indefinite it is vulnerable under § 112(b). These are separate from, and additional to, the § 103 problem.

7. Bottom line

Claim 1 — Reasonably expected to be obvious over IBM '476 in view of Helferich '956; alternatively over Dew WO 98/12670 in view of Intel '908 and Helferich '956. The enrollment clause is covered by Dew's registration family.

Claim 7 — Same combinations, further in view of Durango '099 for the "powering down said portable computing device" step (which Durango expressly teaches and the '806 specification adopts almost verbatim).

Claim 12 — Obvious over IBM '476/Dew WO 98/12670 in view of Durango '099 (power-down), Helferich '956 (remote removal with retained copy), and Microsoft '928 (remote load/remove to a mobile device).

Dependent claims 2–6, 8–11, 14–21 — Add only conventional, art-recognized features (fingerprint/iris/electrical biometrics, wireless links, PDA form factor, remote provisioning), all of which are disclosed in the cited references themselves.

Overall: there is a strong, well-documented § 103 case against all 21 claims. The only limitation requiring forensic attention is the bootstrap/enrollment clause, and a secondary reference from the same Dew Engineering family that is already cited on the face of the patent (WO 98/12670) supplies it.

8. Limitations on this analysis

• Three references were not independently full-text verified (US 5,712,912; US 6,442,286; US 6,496,928) because I reached the search-step limit. Their mapping rests on the patent record's bibliographic data and titles and should be confirmed against full text before being relied on in a formal invalidity contention.
• No claim-construction analysis of disputed terms ("peculiar to a user," "removed… by a remote station," "capable of identifying one and only one user") was performed; claim construction could narrow the mapping of the remote-removal limitation, though Helferich '956's generic "information" language is broad enough to survive most constructions.
• This is a technical analyst's prima facie assessment, not a legal opinion, and it does not account for the '806 patent's expired status (which affects remedies, not the § 103 merits) or for any post-issuance disclaimer or certificate of correction not visible in the retrieved record.