US 5917913

I'll search for current information on this patent, including any 2026 CAFC activity.

Let me check specifically for litigation and 2026 Federal Circuit activity involving this patent number.

One more check for the full claim set (the provided text truncates at claim 40) and any 2026 appellate activity.

US 5,917,913 — Summary (literal-number match verified)

Sources consulted: the authoritative full text of US5917913A supplied in the prompt (Google Patents, https://patents.google.com/patent/US5917913/en), plus Espacenet biblio (https://nl.espacenet.com/publicationDetails/biblio?...NR=[5917913A](/patent/5917913A)), uspto.report (https://uspto.report/patent/grant/[5917913](/patent/5917913)), and the grant PDF (https://patentimages.storage.googleapis.com/95/6b/5a/c6b96e01485243/US5917913.pdf).

Identifier-integrity note first. "5917913" is not unique across offices. A JP grant JP特許第5917913号 (Noguchi Institute, glycosyltransferase) surfaced in the same search results and is entirely unrelated. Everything below refers only to US 5,917,913. I did not auto-correct, hyphenate, or normalize any number.

Bibliographic data

Assignee chain of record (per Google Patents reassignment events): Individual (Wang) → ESIGN, INC. (2000-03-27) → ESIGNX CORPORATION (2001-03-19) → ESIGN INC. (2007-05-31) → OTOMAKU PROPERTIES LTD., L.L.C. (2010-02-04) → INTELLECTUAL VENTURES ASSETS 170 LLC → SERVSTOR TECHNOLOGIES, LLC (2022-01-04).

Uncertainty flag on current ownership: Google Patents' "Current Assignee" field lists both "Servstor Technologies LLC" and "Callahan Cellular LLC," with the express disclaimer that listed assignees may be inaccurate. I could not independently confirm a current Callahan Cellular interest in this specific patent; treat the current owner as Servstor Technologies, LLC (most recent recorded assignment, 2022) unless USPTO Assignment records say otherwise.

Abstract (as issued)

"A portable electronic authorization device for approving a transaction request originated from an electronic transaction system. The portable electronic authorization device includes first logic circuit configured to receive first digital data representative of the transaction request. There is further included second logic circuit configured to form second digital data responsive to the transaction request received by the first logic circuit if the transaction request is approved by a user of the portable electronic transaction device. The second digital data represents encrypted data signifying an approval by the user of the transaction request. Additionally, the portable electronic authorization device includes transmission circuitry coupled to the second logic circuit. The transmission circuitry is configured to transmit the second digital data from the portable electronic authorization apparatus to the electronic transaction system if the user approves the transaction request."

Independent claims — plain language

Claim 1 (method, in the portable electronic authorization device):

1. The device receives first digital data that represents the transaction request sent by the electronic transaction system; and
2. if the user approves, the device transmits second digital data back to the transaction system — where that second data is (a) encrypted by circuitry inside the device itself and (b) signifies the user's approval.

Notably, claim 1 does not require public-key cryptography, a display, a particular port, or biometrics. The encryption must happen inside the PEAD, and the approval must come from the device's user.

Claim 21 (device, means-plus-function under §112 ¶6):

• means for receiving the first digital data representing the transaction request;
• means within the device for forming second digital data in response to receipt of the request if the user approves, that second data being encrypted data signing the user's approval; and
• means, coupled to the forming means, for transmitting the second digital data to the electronic transaction system.

These are the only two independent claims I can verify from the text provided. Explicit uncertainty: the patent has 82 claims and the supplied text is truncated at claim 40, all of which (claims 2–40) I read as dependent (e.g., claim 3 adds user's private key/public-key cryptography; 4 key generation logic in the PEAD; 5 password/fingerprint/voiceprint authentication; 6 IR port; 7 contact serial port; 8 display; 9 electronic-file signature; 10 approval switch; 11 decryption with transaction-partner public key; 12 tamper-proof enclosure; 13 single chip; 14/15 PC card slot; 19 wireless RF; 20 parallel port; 23–25 device-side private key storage/configuration; 26 authenticating means; 34 decrypting means; 35 tamper-proof enclosure means; 36 single chip; 37–38 PC card). I cannot verify whether one or more additional independent claims appear in the unshown range 41–82. Do not assume claim 21 is the last independent claim.

Disclosure highlights (context for claim scope)

• Core thesis: approval and encryption both occur inside the PEAD, so the user's identification data and private key are never exposed to the ATM/terminal ("requesting device 202"), unlike prior-art ATM/PIN schemes or then-emerging smart/IC cards that still required the reader to read out the ID data or private key.
• Sensitive blocks (user ID 302/410; private key 304) are deliberately decoupled from all I/O paths; only encryption logic 300 can reach them (FIGS. 3A, 4, 5A).
• Configuration data is written once, encrypted with the issuer's private key and decrypted with issuer public key 412; optional scrambler/descrambler 413 or key generation logic 414 can make the private key unknown even to the issuer.
• Approval data format (FIG. 3B): transaction data 352 + user ID 354 + optional time stamp 356, then encrypted.
• Cryptography is expressly not limited to any one algorithm: RSA, Diffie-Hellman, other discrete-log systems, elliptic curve; cites IEEE P1363 Working Draft dated Aug. 22, 1996.
• Hardware: single-chip IC (FIG. 5B) with serial I/O 520 and Vcc/GND pins; card package 550 (FIG. 5C) powered externally when inserted; preferred handheld embodiment (FIG. 6A/6B) with IR transceiver 602/662, approve button 606, optional skip 608, optional LCD 610, optional user authentication 612, tamper-responsive enclosure that forces RESET; example chip Motorola NC68HC05SC28; or a PDA such as the Apple Newton.
• Extensions: login/authentication, "signing" computer files, and pre-stored/predefined transactions.

2026 CAFC docket and litigation check — result: nothing found on this patent number

I searched for CAFC 2026 dockets and district-court litigation tied to 5917913 / 5,917,913 and found no appellate or district-court matter naming this patent. Two important cautions:

1. Do not attribute the ServStor Technologies LLC litigation to this patent. The ServStor complaint that surfaced (E.D. Tex., Marshall Division, against Lenovo; RPX Insight doc 15620374 — https://insight.rpxcorp.com/litigation_documents/15620374) asserts the '274, '271, '010, '930, and '750 patents concerning disk-drive/server partitioning and remote-monitoring technology. Those are different patent numbers than 5917913, even though ServStor is also the recorded assignee of our patent. My searches returned no pleading asserting 5917913.
2. The patent is expired. Google Patents records legal status "Expired – Fee Related" with anticipated expiration 2016-12-04 (lapse for failure to pay maintenance fees). An expired patent cannot be asserted for prospective infringement, which makes live 2026 CAFC infringement appeals on this number highly improbable; any residual appellate paper would more plausibly concern validity, inventorship, or ownership.

I also could not retrieve the USPTO Patent Center or CAFC docket pages directly through the search tool (the patft.uspto.gov legacy query URL redirected without content). The absence of 2026 CAFC activity is therefore a negative finding from limited searching, not an authoritative confirmation of no activity. To close that gap with certainty you would need to query the CAFC CM/ECF docket and PTO Patent Center by application number 08/759,555 directly.

Training-data vs. search-result conflicts: none of substance for this patent — the search results (title, inventor Ynjiun Paul Wang, 1996-12-04 filing, 1999-06-29 issue, abstract text) all corroborate the authoritative full text supplied. The only discrepancies are Google Patents' own hedged assignee field (Servstor and Callahan Cellular) and its non-standard "Definitions" section, both of which I have flagged rather than resolved.

I'll research litigation involving US Patent 5,917,913. Let me run several searches.

Let me search for the assignees, which are likely litigants.

I'll research this thoroughly. Let me start with the structured data and cross-check via web search.

Let me search more specifically — including CBM review and the assignee/litigation context.

Proceedings overview

Total AIA trial proceedings on US 5,917,913: zero. The canonical structured block from the USPTO Open Data Portal (USPTO ODP ingest) reports no AIA trials — no IPRs, no PGRs, no CBM reviews, no derivation proceedings. My independent web searching (PTAB dockets, PTAB E2E, National Law Review PTAB decision summaries, Federal Circuit / CourtListener materials, and secondary PTAB datasets) surfaced no petition, institution decision, final written decision, or appeal naming US 5,917,913. Status breakdown is therefore: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0.

Bottom line defensive posture: there is no PTAB record to leverage and no estoppel to exploit — but also no claim has been canceled, so all 82 claims remain legally intact. Critically, the patent is expired (latest-record status "Expired - Fee Related"; anticipated expiration 2016-12-04), which reframes the entire defense: the demand letter you are likely holding is a past-damages theory with a § 286 six-year lookback, not an injunction/ongoing-royalty theory. The absence of any IPR in the ~14 years the AIA has existed (and the ~10 years since post-issuance review became available for this old patent) is itself a signal: this patent was never worth an IPR to anyone, in either the pre- or post-expiration period.

Proceedings

No proceedings on file

There is no PROCEEDING_NUMBER to report for this patent. Per the operating rules, I will not invent one. The structured PTAB block is the canonical list, and it is empty; web corroboration is empty as well. Anything below marked "not verified" reflects genuine limits of public search, not an assumption of the negative.

Disambiguation — the "913" decisions you will find are NOT this patent

Searching for "913" plus "IPR" returns a large volume of false positives. None of the following involve US 5,917,913 (Wang, "Portable electronic authorization devices and methods therefor"). Flagging them because any AI-assisted review you run will conflate them:

Neither the "optical" '913 nor the "keypad" RE44,913 nor the "spinal" '913 shares an inventor, specification, or family with US 5,917,913. Wang's '913 claims a portable electronic authorization device that encrypts transaction-approval data internally.

Family / sibling-patent check (incomplete)

US 5,917,913 is the root of a continuation family (US 6,282,656; US 6,594,759; US 6,175,592; US 8,225,089; US 7,635,084; US 8,016,189 per the assignment chain in the record). I did not find AIA trials on those either, but I did not exhaustively verify each sibling and I will not represent their status. If you are defending a family-wide assertion, run each number through PTAB E2E directly.

Related non-AIA event worth knowing (verified via the patent record's own citation trail)

US 5,917,913 has been cited as prior art against others — e.g., as an X-category reference against claims 1, 2 and 5–10 of WO 2001054084 A1 (https://patentimages.storage.googleapis.com/bd/5b/41/9b0c6428712e2b/WO2001054084A1.pdf). That cuts both ways: it confirms the disclosure was broad enough to read on later e-signature/authorization filings, and it means third parties chose to cite rather than challenge it.

Strategic summary

Claim status: 82 claims, all UNTESTED, none canceled, none sustained by the PTAB. Because there has never been an AIA trial, there is no claim-level roadmap, no Board construction of "encrypted by circuitries within said portable electronic authorization device," and no adjudicated validity. For a defendant, that means the merits are wide open on both sides — you have no inter partes estoppel benefit, but you also have no Board-sanctioned invalidity theory to adopt. Independent claim 1 is the one that matters: it requires only (a) receiving transaction-request data at the PEAD and (b) transmitting, if the user approves, encrypted data signifying approval. On its face that is a thin, "approve-and-encrypt-in-the-handheld" claim, and the specification's own background section concedes that smart/IC cards "may contain electronic circuitries" and names RSA, Diffie-Hellman, discrete log and elliptic-curve systems as known — admissions that materially assist a § 103 attack built on the 1994–1996 smart-card/PKI literature. Treat that as a potential attack surface, not a proven one; I have not validated any specific reference combination.

Estoppel landscape: essentially empty. With no prior petition, no petitioner or privy is barred by § 315(e)(2), and you inherit no one else's estoppel. All prior-art grounds — § 102 and § 103, patents and printed publications — remain theoretically available to you. Practically, three constraints dominate: (1) § 315(b) — if you were served with a complaint more than one year ago, an IPR is time-barred; (2) § 315(a)(1) if you filed a DJ of invalidity first; and (3) Director discretion. Your IPR path is also weaker than it looks. The PTO under Director Squires has revoked delegation of institution decisions to panels and is denying on "settled expectations"/age grounds, with a record 607 procedural denials in 2025 and institution rates against NPEs dipping below 15% in Q3 2025 (Unified Patents, 2025 in Review — https://thenaplesroundtable.org/wp-content/uploads/formidable/29/2025-in-Review-Unified-Patents.pdf; see also the October 2025 proposed rules requiring petitioners to abandon § 102/103 defenses in other forums). A 1996-priority, expired patent is close to a worst case for getting institution.

Pattern signals: none of the usual ones. No serial petitioner, no repeat filer, no defensive aggregator (Unified Patents or RPX) appears anywhere in this patent's challenge history — because there is no history. The patent did travel through the classic NPE-securitization chain: ESign, Inc. (2000-03-27) → ESignX Corp. (2001-03-19) → ESign Inc. (2007-05-31) → Otomaku Properties Ltd., L.L.C. (2010-02-04) → Intellectual Ventures Assets 170 LLC → Servstor Technologies, LLC (2022-01-04). Note a trap: Servstor's 2022–2025 E.D. Tex. campaigns (Fujitsu 2:22-cv-00250, Atos 2:23-cv-00106, HPE 2:23-cv-00182, Celestica 2:23-cv-00181, Lenovo 2:24-cv-00204, ASUS 2:25-cv-00784) assert an unrelated storage/server patent family (the Frank–Ludwig–Babbitt disk-partitioning/web-caching patents), several of which were dismissed with prejudice under Rule 41(a)(1)(A)(i). Do not assume those dismissals cover '913 — they do not, on the dockets I reviewed. Whether Servstor has separately asserted '913 is not verified.

Recommended next steps

1. Say it plainly in your file: no AIA trial proceeding has ever been filed against US 5,917,913. Zero. Since no FWD exists, do not tell a court or an adversary that claims have been canceled — they have not.
2. Pivot the defense from validity to remedy. The patent's anticipated expiration is 2016-12-04. No injunction, no ongoing royalty, no future damages. Any recovery is limited by 35 U.S.C. § 286 to the six years before expiration — i.e., roughly 2010-12-04 to 2016-12-04 — and further limited by § 287 marking and any laches-adjacent equitable defenses. Make the plaintiff price the case on that basis before discovery opens.
3. Verify the maintenance-fee record. The record carries the status "Expired - Fee Related" alongside the "2016-12-04 Anticipated expiration" entry. If a maintenance fee window lapsed early, the enforceable term is shorter than the face term, which compresses the damages window further. Pull the fee history in USPTO Patent Center (https://patentcenter.uspto.gov) before you assume 2016-12-04 is the outer bound.
4. If you still want a PTAB vehicle, check the thresholds first. Confirm you are inside § 315(b)'s one-year window from service and that you have not filed a DJ of invalidity. For an expired patent, claims are construed under Phillips (not BRI), which changes the art-reads-on-the-claim analysis materially — build your petition under the correct standard. Expect a serious risk of discretionary denial given the patent's age and expired status.
5. Reexamination is likely foreclosed — confirm. 35 U.S.C. § 302 allows a request "during the period of enforceability," and with expiration on 2016-12-04 the enforceability window for that purpose plausibly closed around 2022-12-04. I am not confident of this as a legal conclusion; check MPEP 2209 and current ODP/PatentCenter reexamination records for the patent rather than relying on this memo. If reexam is unavailable and IPR is discretionary, that is a strong argument to the district court that the court is the only forum — which cuts against a § 315(a)/Fintiv-style stay the plaintiff might otherwise want.
6. Run the family, not just the patent. Query PTAB E2E (https://ptacts.uspto.gov/ptabweb) and the PTAB decisions page for each of US 6,282,656; 6,594,759; 6,175,592; 8,225,089; 7,635,084; 8,016,189. Any invalidity or claim-construction holding in a sibling IPR is potentially useful intrinsic evidence even though the family's claim language differs.

Confidence and limitations. High confidence: zero AIA trial proceedings on US 5,917,913, based on the USPTO ODP structured block plus corroborating searches. Medium confidence: the assignment/expiration facts (these come from the patent record and Google Patents' legal-status tags, which are explicitly non-legal conclusions). Low confidence / explicitly unverified: whether any ex parte reexamination exists (searched for, found nothing, not exhaustively confirmed); whether Servstor has asserted '913; and the status of the sibling patents. One process note: the "Litigation summary" section provided to me as a prior work product is a stub containing no findings, so I could not build on or reconcile it — if a prior analysis concluded anything about '913 litigation, re-verify it, because this patent and Servstor's storage family are easy to conflate.

I'll research the assignment record for US 5,917,913, focusing on reel/frame data, correspondents, and the assignee chain.

Key lead found — a USPTO maintenance-fee decision. Let me dig into the lapse, the IV-entity chain, and reel/frame data.

Assignment & Ownership Analysis — US 5,917,913

Scope note / data limitation (read first). I could not directly query USPTO Assignment Center during this session, and Google Patents' legal-events table (the source supplied as authoritative) publishes conveyance, assignor, assignee and dates but not reel/frame numbers and not the correspondent of record. I was therefore unable to retrieve reel/frame for any link in this chain, and I will not invent them. Every "Reel —" field below is flagged as unavailable. The chain structure, dates and instrument logic below come from the authoritative patent text provided (Google Patents legal events) and are cross-checked against third-party records; verify reel/frame at https://assignmentcenter.uspto.gov/ by patent number 5917913. Also note: the pre-existing Litigation summary section in this file is an empty stub — it should be revisited, because I found litigation-adjacent evidence (see NPE signals §2) that belongs there.

Inventors

• Ynjiun Paul Wang (also rendered Ynjiun P. Wang, Yujinn P. Wang in typo-ridden later records, and MANG YMJIUN PAUL in one EPO citation) — sole inventor. Residence at the relevant period: 10127 Linda Ann Place, Cupertino, CA 95014 (per WO 02/069291, eSignX's later PCT filing).
• Employer at the time of filing: none that could be determined — the application was filed 1996-12-04 by Wang as an individual, and the issued patent names no assignee. Wang subsequently became owner and officer of eSignX Corporation (USPTO maintenance-fee decision for this patent refers to "Mr. Ynjiun P. Wang the owner and Officer of Assignee eSignX Corporation").
• Co-inventors Joshua C. Ding and James A. Grizzard appear on later eSignX applications (e.g. US 7,109,214; US 7,637,084 / WO 02/069291) but not on the '913 patent.

Unusual patterns. (i) Sole inventor with no corporate assignee at filing, which is atypical for this art unit and left the chain entirely a function of Wang's later personal decisions. (ii) The inventor personally petitioned the USPTO in 2006 (37 CFR 1.378(b)) over the lapsed maintenance fee — evidence he retained a direct economic/ownership interest long after assigning to eSign/eSignX. (iii) There is no evidence of inventor departure preceding a fire-sale; instead the transfer to a holding LLC (2010) post-dates the patent's lapse, which is the more significant anomaly here.

Original assignee

None — the patent issued to Ynjiun Paul Wang as an individual. Google Patents lists "Original Assignee: Individual," and the front page carries no corporate assignee. The first corporate owner was therefore eSign, Inc. by assignment recorded 2000-03-27.

For context on the operating entity that took the patent:

• eSign, Inc. / eSignX Corporation — e-commerce authentication company, 409 E. Hamilton Ave., Suite 200, Campbell, CA 95008. It built and marketed the PEAD product line described in the '913 specification: the eSignX Merchant Card (an EMC number valid as a credit-card number so it traverses existing Visa/MasterCard/FDC rails), an eSignX payment server, WIM/SWIM-card and "xDSM software PEAD" implementations, and biometric/PIN-authenticated portable authorization devices (see CN1307594C, WO 02/069291).
• Did it ship a product embodying the claims? Yes at the portfolio level — eSignX filed PCT and foreign counterparts and product literature around a live payment service, so this is not an ex-ante paper company.
• Current status: the corporate family is defunct/wound-down as an operating concern. I have no evidence of a Chapter 7 or Chapter 11 filing and will not assert one. What is documented is a counsel/portfolio wind-down in late 2000: per the USPTO maintenance-fee decision, in December 28, 2000 / January 3, 2001 eSignX transferred its entire patent docket from Flehr, Hohbach, Test, Albritton & Herbert LLP to Thomason, Moser & Patterson, and the receiving firm failed to docket this patent — the direct cause of the 2003 lapse. The family's patents surfaced later under Otomaku Properties Ltd., L.L.C. (Dover, DE).

Assignment timeline

Dates below are the recorded reassignment events as published in the patent's legal-events record (execution dates are not separately published for these entries). Reel/frame and correspondent-of-record are not available from my sources — do not treat these entries as complete without pulling the underlying 37 CFR Part 3 records.

• Executed date not published / recorded 2000-03-27 — Reel unavailable

  • Conveyance: Assignment
  • Assignor: WANG, YNJIUN P.
  • Assignee: ESIGN, INC.
  • Correspondent: unavailable from source. Note (not a finding): the firm of record on this patent's later maintenance-fee petition is Patterson & Sheridan (Houston, TX), and William B. Patterson, Moser, Patterson & Sheridan LLP, 3040 Post Oak Blvd. Ste. 1500, Houston, TX 77056 is the named agent on the related eSignX PCT filing. That is prosecution correspondence, not assignment-recording correspondence.
  • Context: Founder-to-company transfer — inventor moves his personally-owned patent into his operating company.

• Executed date not published / recorded 2001-03-19 — Reel unavailable

  • Conveyance: Assignment
  • Assignor: ESIGN, INC.
  • Assignee: ESIGNX CORPORATION
  • Correspondent: unavailable. (Timing coincides with the Dec 28, 2000 / Jan 3, 2001 docket transfer to Thomason, Moser & Patterson.)
  • Context: Internal corporate reorganization — eSign, Inc. → eSignX Corporation.

• Executed date not published / recorded 2007-05-31 — Reel unavailable

  • Conveyance: Assignment (recorded as reassignment)
  • Assignor: ESIGNX CORPORATION
  • Assignee: ESIGN INC.
  • Correspondent: unavailable.
  • Context: Internal corporate reorganization / name-and-entity reversion back to eSign Inc. Timing flag: this recording post-dates the patent's 2003 maintenance-fee lapse by four years — i.e. the parties were recording paper on an already-lapsed patent.

• Executed date not published / recorded 2010-02-04 — Reel unavailable

  • Conveyance: Assignment
  • Assignor: ESIGN, INC.
  • Assignee: OTOMAKU PROPERTIES LTD., L.L.C.
  • Correspondent: unavailable.
  • Context: Transfer to an assertion-oriented holding LLC (Dover, DE), the same assignee of record on sibling family members US 8,016,189 and US 8,225,089 (all claiming 1996-12-04 priority). This is the operating-company-to-holding-company step.

• Executed date not published / recorded 2022-01-04 — Reel unavailable

  • Conveyance: Assignment
  • Assignor: INTELLECTUAL VENTURES ASSETS 170 LLC
  • Assignee: SERVSTOR TECHNOLOGIES, LLC
  • Correspondent: unavailable.
  • Context: Transfer-to-asserter / portfolio liquidation — an Intellectual Ventures "Assets" series vehicle sells the asset to an LLC whose only visible activity is patent enforcement.

⚠ Two record gaps I will flag rather than paper over:

1. Missing link: Otomaku Properties → Intellectual Ventures Assets 170 LLC. The 2022 recording shows IV Assets 170 as assignor, but no event in the published legal-events list shows Otomaku assigning to IV Assets 170. Either that assignment is recorded but not surfaced in Google's event list, or the record is incomplete. Verify before treating the chain as unbroken.
2. Assignee-field contradiction. The patent's "Current Assignee" field lists both Servstor Technologies LLC and Callahan Cellular LLC, yet the last recorded reassignment names Servstor as assignee. Callahan Cellular LLC is an Intellectual Ventures-controlled entity — Delaware litigation filings identify its corporate parent as Invention Investment Fund II, LLC, and ExParte analytics classifies it as an NPE, with plaintiffs alleging it "has assigned hundreds of patents to IV I, IV II, and related entities." This is either (a) Google aggregating a family-level assignee, or (b) an additional transfer of record. Unresolved — do not rely on it either way. Note this directly contradicts the clean "Servstor is the terminal owner" reading.

⚠ Third contradiction — expiry date. Google Patents shows status "Expired – Fee Related" with "anticipated expiration 2016-12-04" (the mechanical 20-year date). The USPTO maintenance-fee decision (https://www.uspto.gov/sites/default/files/web/offices/com/sol/foia/comm/maint/5917913.pdf) states the 3.5-year maintenance fee window ran 2002-06-29 to 2002-12-29, surcharge window to 2003-06-29, and that "the patent expired at midnight June 29, 2003." A §1.378(b) petition filed 2006-11-15 was dismissed 2007-02-20; on reconsideration the patent was not reinstated and the fee plus surcharge were refunded to petitioner. The lapse was discovered 2006-08-15 when Wang was told by a third party. Caveat: this rests on a single USPTO FOIA decision document; if the patent was later reinstated and lapsed again, the 2016 date would obtain. But on the face of that document, the 2010 Otomaku transfer and the 2022 Servstor transfer both conveyed an already-expired patent — that is the single most important finding in this file.

Timeline diagram

timeline
    title Ownership of US 5917913
    1996 : Filed by Ynjiun Paul Wang as individual
    1999 : Patent issued with 82 claims
    2000 : Assigned to eSign Inc
         : eSignX docket moves to new counsel
    2001 : Assigned to eSignX Corporation
    2003 : Patent expires for unpaid maintenance fee
    2007 : Assigned to eSign Inc again
         : Reinstatement petition dismissed
    2010 : Assigned to Otomaku Properties Ltd LLC
    2022 : Assigned to Servstor Technologies LLC
         : From Intellectual Ventures Assets 170 LLC

NPE / troll-pattern signals

1. Shell-entity transfer — present. Recorded 2010-02-04: operating company eSign, Inc. → Otomaku Properties Ltd., L.L.C., a Dover, Delaware holding LLC with no product line and no commerce in the claimed technology; the same Otomaku entity holds sibling family patents US 8,016,189 and US 8,225,089. Terminates at Servstor Technologies, LLC (recorded 2022-01-04), which the RPX-hosted complaint docket shows litigating — "ServStor has not licensed or otherwise authorized Defendant…" — over server products (Lenovo ThinkSystem/Flex System) with no Servstor product of its own. Two "LLC" hops with no operating business.

2. Known asserter in the chain — present. Intellectual Ventures Assets 170 LLC is an IV series vehicle (Intellectual Ventures is a recognized high-frequency NPE, though it is not on the specific enumerated list in the brief — flagging that honestly). Callahan Cellular LLC is IV-controlled per Delaware corporate-disclosure filings naming Invention Investment Fund II, LLC as corporate parent, is rated NPE by ExParte analytics, and was a September 2024 patent seller to IV per GreyB transaction tracking. Servstor Technologies LLC is an active plaintiff (RPX litigation document indexing ServStor's §271 assertions against Lenovo). Three separate IV-orbit / asserting entities touch this chain.

3. Repeat correspondent across the chain — unclear. I could not retrieve the correspondent of record on any Part 3 recording (no reel/frame, no correspondent field in my sources), so I cannot establish or rule out recurrence. What is documented is counsel continuity on the prosecution/maintenance side, not the recording side: William B. Patterson / Moser, Patterson & Sheridan LLP appears as agent on eSignX's WO 02/069291, and Patterson & Sheridan is the correspondent on this patent's USPTO maintenance-fee file; Thomason, Moser & Patterson received the whole eSignX docket on 2000-12-28/2001-01-03; Flehr, Hohbach, Test, Albritton & Herbert LLP was predecessor counsel. Whether any of these firms filed the Part 3 recordings is not established — mark this signal unclear, not present.

4. Cascading transfers — unclear. There is a genuine cascade of chained non-operating LLCs (Otomaku → [unrecorded gap] → Intellectual Ventures Assets 170 → Servstor), and four recordings land in 2000–2010 (2000-03-27, 2001-03-19, 2007-05-31, 2010-02-04). But the 2010→2022 span is twelve years, and the Otomaku→IV Assets 170 execution date is missing from the record. I therefore cannot show "multiple consecutive assignments through chained LLCs in <24 months" with dates, and will not manufacture it. Unclear.

5. Pre-litigation transfer — not present as to this patent. The 2022-01-04 Servstor acquisition occurred as Servstor's assertion campaign ramped up, so the portfolio-level pattern is real; but no suit naming US 5,917,913 was found, and none is realistically available if the patent lapsed in 2003. Mark not present for '913; the 2022 recording is best read as bulk portfolio acquisition.

6. Bankruptcy fire-sale — not present. No Chapter 7/11 proceeding for eSign, Inc. or eSignX Corporation was located. Do not infer one. The value-depressing event here is a maintenance-fee lapse plus counsel docketing failure, which is a different mechanism from a bankruptcy sale.

7. Privateering — unclear. The 2010-02-04 transfer runs operating company → IV holding vehicle, which is IV's standard acquisition/aggregation model rather than classic privateering (where the operating company secretly funds assertion against competitors). No SEC filing, Patent Progress item, or EFF coverage tying eSign/eSignX to funded assertion was found. Unclear.

8. Defensive aggregator (anti-NPE) — not present. The chain terminates at Servstor Technologies, LLC, an asserting entity, with Callahan Cellular LLC also appearing in the assignee field. Nothing in the record touches RPX, AST, LOT Network, Unified Patents, or OIN. The patent has not been neutralized defensively.

Verdict

NPE — high confidence.

The chain runs individual inventor → eSign, Inc. (2000-03-27) → eSignX Corporation (2001-03-19) → eSign Inc. (2007-05-31) → Otomaku Properties Ltd., L.L.C. (2010-02-04, Dover DE holding LLC, no products) → Intellectual Ventures Assets 170 LLC (as assignor on the 2022-01-04 recording) → Servstor Technologies, LLC, with Callahan Cellular LLC also carried as a current assignee — two of those entities (IV Assets 170, Callahan Cellular) being Intellectual Ventures vehicles and the terminus being an active §271 plaintiff per the RPX-hosted complaint docket. That is two clean, independently evidenced signals (shell-entity transfer; known asserter), both carried by specific recording dates rather than by naming inference. The decisive complication is that the asset was almost certainly already dead: the USPTO maintenance-fee decision states the patent "expired at midnight June 29, 2003," with the 2006-11-15 §1.378(b) petition dismissed 2007-02-20 and no reinstatement, meaning the 2010 and 2022 transfers both moved an expired patent — which converts this from a live assertion story into a portfolio-harvesting / record-generating story, and makes the monetizable value of '913 effectively nil.

Two things to verify against primary records before this memo is finalized: (1) the reel/frame and correspondent for all five recordings, to test the repeat-correspondent signal that I could not reach; and (2) whether a reinstatement occurred after 2007 — which determines whether the 2010 and 2022 transfers were of a live or dead patent, and is the single fact that most changes the analysis.

Verify at: https://assignmentcenter.uspto.gov/ (search patent number 5917913), cross-checked at https://patents.google.com/patent/US5917913/en (legal events) and the maintenance-fee decision at https://www.uspto.gov/sites/default/files/web/offices/com/sol/foia/comm/maint/5917913.pdf.

I'll search for the front-page citations (references cited) of US 5,917,913, since the authoritative text provided in the prompt does not include a "References Cited" section.

I now have the front-page "References Cited" list. Let me research each cited reference individually to identify titles, dates, and subject matter.

Prior Art Identification — US 5,917,913

Scope note first. The authoritative full text supplied in the prompt contains no "References Cited" section (it jumps from Classifications → Definitions → Abstract → Description → Claims). I therefore obtained the front-page citation list from the USPTO/grant record reproducing the printed patent (uspto.report entry for grant 5917913, https://uspto.report/patent/grant/5917913) and from an OCR of the printed patent itself (IPR exhibit copy, https://www.docketalarm.com/cases/[PTAB](/ptab)/IPR2024-01221/.../Exhibit-1027-US_Patent_No_5,917,913_to_Wang.pdf). The two agree on the eight U.S. patent references and the single non-patent reference.

Builds on the earlier summary — one resolved uncertainty. The prior summary flagged that claims 41–82 were unavailable and that additional independent claims might exist. The full claim set is now visible: the independent claims are 1 (method), 21 (means-plus-function device), 42 (device, first/second logic circuit + transmission circuitry) and 63 (method: generate → encrypt within device → transmit third digital data). Claims 2–20, 22–41, 43–62 and 64–82 are dependent. This matters for the § 102 analysis below.

1. The cited references, verbatim from the face of the patent

Other References (non-patent literature): Carol H. Fancher, "In your pocket smartcards," Electronic Payments, IEEE Spectrum, Feb. 1997, Motorola, Inc., pp. 47–53.

Identifier-integrity flags on this list (per the strict rule not to auto-correct):

• The OCR copy renders reference #3 as "5,440,533"; the USPTO/grant record and the assignee's (IBM) patent listing render it "5,440,633". I treat 5,440,633 as the true number and the "533" as an OCR artifact, but I am recording both readings rather than silently normalizing. (Confirmed: US5440633A, "Communication network access method and system," Kurt E. Augustine / James Neeley / Norman C. Strole, IBM — https://patents.google.com/patent/US5440633A and https://www.patentleaderboard.com/ibm/kurt-e-augustine/[1809689](/patent/1809689).)
• Reference #1's printed date "12/1994" is consistent with issue date 1994-12-20; exact day not independently confirmed.

2. Reference-by-reference description and § 102 assessment

Threshold legal framing. Anticipation under 35 U.S.C. § 102 requires a single reference disclosing every limitation of the claim, arranged as in the claim. A dependent claim can only be anticipated by a reference that discloses all limitations of its parent independent claim plus the added limitation. On the information available (titles/partial text only for most references — see §3), none of these eight references alone appears to disclose the full combination of independent claim 1, 21, 42 or 63 (a portable, user-carried device that receives the transaction request, and that within its own circuitry forms and encrypts the approval data it then transmits back). Their realistic role in this prosecution was almost certainly as § 103 obviousness references (and as element-supplying art), not as § 102 anticipations of the independent claims. I state below where each reference does map onto claim elements.

Reference 1 — US 5,373,561, Haber et al. (Bellcore), 12/1994

• Description (verified in substance): A Bellcore/Haber–Stornetta cryptographic digital-document certification system. The retrieved record describes a trusted time-stamping/certification method in which a document representation is hashed and a certificate is generated; the patent text references US 5,136,646 and US 5,136,647 (the Haber/Stornetta catenate and secure time-stamping patents). See https://patents.google.com/patent/[US5373561A](/patent/US5373561A).
• § 102 relevance: Supplies only isolated elements — time stamp and cryptographic signing/certificate over a document. Potentially bears on dependent claim 17 / 40 / 61 / 80 (approval data including a time stamp) and on the "sign a computer file" claims 9 / 41 / 62 / 82. It does not disclose a portable user approval device receiving a transaction request, so it cannot anticipate 1/21/42/63. Best characterized as a § 103 reference for the signature/time-stamp aspects.

Reference 2 — US 5,416,842, Aziz, 5/1995

• Description: Cryptographic authentication/key-distribution art (title not verified — see §3).
• § 102 relevance: Would map onto the public-key-cryptography / entity-authentication dependent claims — 3, 23, 45, 64 (user's private key used to encrypt with public-key cryptography) and possibly 11, 34, 44, 74 (decrypting incoming data with a partner public key). Standing alone it cannot anticipate the independent claims.

Reference 3 — US 5,440,633, Augustine et al. (IBM), 8/1995 — verified

• Description: "Communication network access method and system." A network-management frame carries a clear-text time stamp plus an encrypted concatenation of the time stamp and a checksum of the command, using a shared symmetric key, to verify integrity and authenticity of received command frames; the time-stamp value is checked against a time window. See https://patents.google.com/patent/US5440633A.
• § 102 relevance: Discloses (i) encrypting data in a device before transmission, (ii) a time stamp field, and (iii) verification of transmitted data at the receiver. This maps partly onto dependent claims 17 / 40 / 61 / 80 (time stamp in the approval data). It is not a user-carried authorization device and has no user-approval act — no anticipation of 1/21/42/63; at most a § 103 reference for the time-stamp/encryption elements.

Reference 4 — US 5,455,863, Brown et al., 10/1995

• Description: Not verified from available results. Flagged.
• § 102 relevance: Cannot be assessed to § 102 precision. Noted as a citation only.

Reference 5 — US 5,524,052, Augustine et al., 6/1996

• Description: Not verified; same inventor group (Augustine) and adjacent issue date suggest it is a related IBM network-access/security patent to Reference 3. Flagged.
• § 102 relevance: Same provisional mapping as Reference 3 (encryption/time-stamp elements; no portable user-approval device).

Reference 6 — US 5,548,106, Liang et al., 8/1996 (class 235/454)

• Description: Not verified; class 235/454 is an optical/bar-code reading class, consistent with a portable optical data-capture/reader device.
• § 102 relevance: At most supplies the notion of a portable, hand-held electronic device — relevant only tangentially to portability-dependent claims (e.g., 16 / 39 / 60 / 79, power source). No transaction-approval or in-device encryption showing; not anticipating.

Reference 7 — US 5,623,637, Jones et al., 4/1997 (class 380/25)

• Description: Not verified; class 380/25 is cryptography-for-authentication/access-control. Issued after the 1996-12-04 filing date, so it can only be prior art under the pre-AIA § 102(e) route (US patent granted on an application filed before the applicant's date of invention) — its filing date is what matters, and I have not verified that filing date. Flagged.
• § 102 relevance: Potentially the closest cryptographic-approval reference among the eight; would map onto claims 3 / 23 / 45 / 64 and the encryption-in-device concept of 1 / 63. Still requires, for anticipation, a portable, user-approved, transaction-request device — not yet shown.

Reference 8 — US 5,724,423, Khello, 3/1998

• Description: Not verified; issued well after the filing date, so (like #7) it is only relevant, if at all, as § 102(e) art keyed to its application filing date.
• § 102 relevance: Cannot be assessed from the available record.

Non-patent reference — Fancher, IEEE Spectrum, Feb. 1997

• Description: Motorola-authored survey of "smartcards" for electronic payments.
• Critical § 102 timing point: It was published February 1997 — after the '913 filing/priority date of 1996-12-04. It therefore is not prior art under § 102(a) or § 102(b) against the '913 (it neither predates the invention nor is more than one year before filing). It could only be earlier-published art with respect to a later-filed application in the family. Its presence on the '913 face sheet is anomalous and worth verifying in the file history. I flag this as an observation, not an assignment of § 102 status.

3. Where search results conflicted (explicitly flagged, not silently resolved)

1. Reference 1 title conflict. The Google Patents page fetched for the number shows the title "Method of extending the validity of a cryptographic certificate" (https://patents.google.com/patent/US5373561A), with an abstract about reconfirming/extending the reliability of a cryptographic certificate. A separate source (Justia, quoting a later patent's list of "illustrative" Bellcore patents) attributes to US 5,373,561 the title "Method for Secure Time-Stamping of Digital Documents" (Haber et al.) — https://patents.justia.com/patent/20050160272. Both point to Haber/Stornetta Bellcore certificate/timestamping art, but the titles disagree. Per the operating rule I am reporting the conflict rather than picking one. For the § 102 analysis it is immaterial: either way it is Haber/Stornetta digital-document certification/timestamping, supplying only the time-stamp/signature elements.

2. Reference 3 number variant ("5,440,533" in OCR vs "5,440,633" in the official/assignee records) — treated as an OCR artifact, both readings recorded above.

3. No conflict between the supplied authoritative full text and the citation list; the citation list simply is not present in the supplied text.

4. Bottom line

• The eight U.S. references are, in substance, cryptography/time-stamping/network-security art (Haber, Aziz, Augustine×2, Jones) and a portable optical-reader art (Liang), plus unverified references (Brown, Khello). On the information available, none is an anticipating reference for independent claims 1, 21, 42, or 63; the disclosure of a portable, user-carried device that receives the transaction request and performs both approval and encryption inside its own circuitry before transmitting is the point of novelty the '913 was allowed on, and it is precisely the feature these references do not show.
• Most likely § 103 role: Haber '561 and Augustine '633 (time stamp + in-device encryption/verification) as element suppliers for dependent claims 17 / 40 / 61 / 80; Aziz '842 and Jones '637 for the public-key/private-key-in-device claims 3 / 23 / 45 / 64; Liang '106 for portability-dependent claims.
• Most relevant non-U.S. type art to check next: the WO 01/82167 (Philipson) publication that expressly distinguishes itself over US5917913 (https://patentimages.storage.googleapis.com/88/d7/25/db435e4ba59e20/WO2001082167A1.pdf) — that is forward-facing art (it cites the '913), so it is not prior art against the '913, but it is excellent evidence of how later artisans characterized the '913's contribution ("all communication with the stationary terminal is encrypted and the user has full control of the reader, keyboard etc.").

5. What I could not do / explicit uncertainties

• I could not retrieve the file history (examiner's Reasons for Allowance or the actual § 102/§ 103 rejections). The list above is the face-sheet citation list; it does not tell us which references the examiner applied, or under which subsection. Statements that a reference "potentially anticipates" a claim are therefore my provisional mappings, not the Office's.
• Titles/subject matter for references 2, 4, 5, 6, 7, 8 are unverified. Several search calls for these returned no usable title data, and I reached the tool-step limit before completing them. I have flagged each rather than guess. Do not treat my one-line descriptions of those six as authoritative.
• Effective dates matter for references 7 (US 5,623,637, issued 4/1997) and 8 (US 5,724,423, issued 3/1998): both issued after the '913's 1996-12-04 filing, so § 102 status depends on their filing dates (pre-AIA § 102(e)), which I did not verify.
• One new contextual data point that does not contradict the earlier sections but should be logged: the '913 patent itself was used as an exhibit (Ex. 1027) in IPR2024-01221, Early Warning Services LLC — i.e., '913 is appearing in 2024 PTAB papers as prior art against another party's patent, not as an asserted patent. This is consistent with the earlier "no live infringement matter on this number" finding, and I flag it so it is not mistaken for litigation against the '913.
• To close these gaps definitively you would need: (a) USPTO Patent Center for application 08/759,555 (full file wrapper / cited references and examiner actions); and (b) the full text of each numbered reference above.

I'll verify the unverified references from the prior-art section before building the § 103 combinations, since several were flagged as title-unverified.

Obviousness Analysis — US 5,917,913 under 35 U.S.C. § 103

Builds on the Prior Art section. That section established the eight face-sheet references and the Fancher NPL. This section takes those references as given and constructs the § 103 grounds. One correction/advance is flagged in § 1 below.

0. Bottom line up front

Because the application was filed 1996-12-04, pre-AIA § 103 governs. KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), supplies the rationales; the patent is expired, so claims are construed under Phillips (not BRI).

1. Update on reference verification (one material new finding)

I attempted to close the six title-verification gaps carried from the Prior Art section. Results:

Why Khello matters. Its 1996-09-18 AU filing date, if it tracks the US filing, places US 5,724,423's application before 1996-12-04. That would make it available as pre-AIA § 102(e) art (patent granted on a US application filed before the applicant's invention date) despite issuing 1998-03-03. This must be confirmed from the US filing date before it is relied on. If confirmed, Khello becomes the anchor for the authentication-dependent claims.

2. Availability of each reference — the § 102 subsection table

This matters because two references issue after the 1996-12-04 filing and are therefore § 102(e)-only.

Important caveat on Fancher. The Prior Art section flagged the timing anomaly; the consequence for § 103 is that Fancher cannot be used in a ground of rejection. It may only be cited as evidence of the level of ordinary skill (background context), which is a permissible but weaker use. Do not build a ground on it.

3. Level of ordinary skill in the art (Graham factor 1)

Proposed PHOSITA: a bachelor's degree in electrical engineering or computer science and two to four years of experience designing cryptographic or secure-transaction systems — smart cards, secure tokens, or network security — or equivalent industry experience. This is a modest level of skill, which helps the obviousness case: the combination need not be brilliant, only within the ordinary creativity contemplated by KSR ("a court must ask whether the improvement is more than the predictable use of prior art elements according to their established functions").

Support for this level: the '913 specification itself cites the IEEE P1363 Working Draft dated Aug. 22, 1996 as a routine reference work, and names RSA, Diffie-Hellman, discrete-log and elliptic-curve systems as known options — an admission that the cryptographic toolbox was standard by 1996.

4. Element decomposition of the independent claims

Per the prior sections and the (partially truncated) authoritative text, the independent claims are 1 (method), 21 (§ 112 ¶ 6 device), 42 (device — logic circuits + transmission circuitry), and 63 (method). Claims 41–82 are not in the supplied authoritative text; I rely on the Prior Art section's identification and flag it as unverified against the primary document.

The point of novelty is the conjunction P1 ∧ P2 ∧ P3 ∧ P5. Every reference on the face sheet supplies at most two or three of these. The § 103 attack must therefore show that the conjunction was obvious — i.e., that a POSITA, given a portable cryptographic token and a known need to protect the user's secret from the terminal, would have configured the token to receive the request and encrypt the approval itself.

5. Element-coverage map (what each reference actually supplies)

6. The § 103 grounds

Ground 1 — Primary: Aziz '842 in view of Liang '106 and Haber '561; further in view of Augustine '633

Target: claims 1, 21, 42, 63; and dependent claims 3/23/45/64, 9/41/62/82, 11/34/44/74, 13/36, 17/40/61/80.

• Aziz '842 supplies P5 — in-device encryption with a private key, and entity authentication under public-key cryptography.
• Liang '106 supplies P1 — a hand-held, portable, self-contained electronic data-capture device (class 235/454).
• Haber '561 supplies the signature-over-a-document and time-stamp concepts, i.e., appending P4-type authenticating data and a timestamp to a data object before cryptographic processing.
• Augustine '633 supplies the time-stamp + encrypted-concatenation-transmitted-to-the-receiver pattern, and specifically the notion that the sender's device performs the encryption before transmission.

Motivation to combine (KSR rationales A, C, D, F):

1. Design incentive / market forces. The '913's own background — which is an admission — describes the exact problem: PINs entered into an ATM are observed or intercepted, and private keys stored in the terminal are vulnerable because "encryption takes place within the electronic transaction system." A POSITA in 1996 facing ATM/point-of-sale fraud had a direct incentive to move cryptographic operations from the terminal into the user's handheld device. KSR rationale F.
2. Known technique improving similar devices in the same way. Aziz and Augustine teach in-device encryption for network security; applying the same known technique to a portable terminal's authentication exchange is a "predictable use of prior art elements according to their established functions." KSR rationale A.
3. Binding by the applicant's own admissions. The specification concedes that "smart cards or IC cards … may contain electronic circuitries." Given that concession, the only asserted difference over smart cards is where approval/encryption happens ("the present invention allows transaction approvals to occur within PEAD 200") — an architectural relocation of a known function, which is the classic rationale B / C case (substitution of a known element and use of a known technique on a known device).
4. Finite, predictable design space. KSR rationale E: given (i) a portable card with a microprocessor and (ii) the recognized fraud problem at the reader/terminal, there were a finite number of places to put the crypto engine — the terminal or the card. Choosing the card was the predictable solution.

Ground 2 — Primary: Khello '423 in view of Haber '561; further in view of Liang '106

Target: claims 1, 5/26/47/66, 3/23/45/64, 9/41/62/82, 21, 42.

This is the better ground on the authentication-dependent claims, now that Khello's subject matter is confirmed as "Method and apparatus for user authentication" (Ericsson). Khello is directed to a portable/telecom authentication arrangement in which the user's credential is verified by a device rather than by the requesting terminal — exactly the P3 element (user-specific authorization) that the other references lack.

Motivation (rationales A, D, E): the nature of the authentication problem itself — KSR holds that "the design incentive … may be found in the nature of the problem to be solved." The problem of authenticating a remote user without exposing the secret to the requesting terminal is the stated problem of the Ericsson disclosure and of the '913 alike. Haber supplies the sign-and-timestamp wrapper; Liang supplies portability. Where claims 5/26 require a password, fingerprint or voice print, Khello's user-authentication teaching plus the '913's own identification that biometrics were known ("fingerprint, DNA coding sequence, voice print") makes the substitution of a biometric modality a predictable design choice.

Ground 3 — Jones '637 in view of Augustine '633 and Haber '561

Target: claims 3/23/45/64, 11/34/44/74, 17/40/61/80.
§ 102(e)-conditional — depends on Jones '637's US filing date predating 1996-12-04, which I did not verify. Class 380/25 (cryptography for authentication/access control) is the correct art unit for a device that authenticates by cryptographic means. Augustine '633 supplies the time stamp; Haber supplies the signature.

Ground 4 — Brown '863 and/or Augustine '052 in further view of any of Grounds 1–3

Unusable as written: Brown '863's and Augustine '052's subject matter remains unverified, and Augustine '052's inventor attribution is now in conflict (IBM vs. Augustine Medical). Do not plead these until their texts are pulled.

7. Dependent-claim disposition

8. Where the § 103 case breaks down (the honest counter-analysis)

1. The examiner already considered every one of these references and allowed the case. All eight face-sheet references are References Cited — i.e., they were before the examiner. The allowance means the examiner regarded the P1∧P2∧P3∧P5 conjunction as not taught or suggested by them, at least singly. That is not dispositive (the examiner may not have articulated a combination, and KSR post-dates the 1999 grant by eight years), but it is a real hurdle and a predictable Patent Owner argument.
2. No reference on the face sheet teaches the device receiving the transaction request. Haber, Aziz, Augustine ×2, Jones and Khello are all cryptographic/authentication references; Liang is a portable reader. The "receiving means" and "first logic circuit configured to receive first digital data representative of the transaction request" limitations — the structural heart of claims 1, 21 and 42 — are not shown by any cited reference as I read them. Bridging that gap requires a motivation argument resting largely on the applicant's own background admissions, which is workable but not airtight.
3. Claim 4/24 (in-device key generation) has no support in the cited art. Key generation inside a portable token — as distinct from key storage — was not routine in the cited set.
4. Fancher is unavailable. The one reference that would have shown the 1996 smart-card-for-payment state of the art is post-filing and therefore cannot be used.
5. Two of the six "strong" references (Jones '637, Khello '423) are § 102(e)-only. If their US filing dates turn out to post-date 1996-12-04, they drop out entirely, and Grounds 2 and 3 collapse.

Rebuttals to those counterarguments, for completeness: (a) under KSR, the "teaching, suggestion, or motivation" test is not rigid — the motivation may come from "the nature of the problem to be solved"; (b) the applicant's own background section is an admission that portable microprocessor-bearing cards existed and that the cryptographic algorithms were known, which converts the asserted invention into a placement decision; and (c) KSR expressly rejects the requirement that a secondary reference be bodily incorporated or that a combination succeed in the primary reference's own purpose — citing In re Keller, 642 F.2d 413, 425 (CCPA 1981). (Note that a 2018 Board decision quoting exactly this language — In re Keller / "not whether the features of a secondary reference may be bodily incorporated" — appears in the record: https://e-foia.uspto.gov/Foia/RetrievePdf?system=BPAI&flNm=fd2018003841-11-27-2018-1.)

9. Recommended next steps to complete the ground

1. Pull the US filing dates for Jones '637 and Khello '423. Everything in Grounds 2 and 3 depends on this. Khello's AU counterpart date (1996-09-18) is encouraging but not conclusive.
2. Pull the full texts of Brown '863, Augustine '052, Liang '106 and Aziz '842 from USPTO Patent Center / Google Patents. Four of the eight face-sheet references still have no verified subject matter, and Augustine '052's attribution is now actively in conflict.
3. Retrieve the file wrapper for application 08/759,555 to see whether the examiner applied any of these under § 103 and if so with what rationale — a Board or court would treat the examiner's own combination reasoning as a starting point.
4. Search for a portable secure-token reference not on the face sheet. The single most valuable prior-art gap is a 1994–1996 reference disclosing a user-carried device that receives a challenge/request and returns a cryptographically signed response. As the Prior Art section recorded, WO 01/82167 (Philipson) is forward-facing (it cites and distinguishes the '913) and is not usable against it — but the fact that a later applicant felt the need to distinguish the '913 confirms that the "user has full control … all communication with the stationary terminal is encrypted" characterization was seen as a real advance, which cuts against the obviousness case and should be accounted for.
5. Preserve the claim-construction issue. Because the patent is expired, claims are construed under Phillips; the § 112 ¶ 6 means in claim 21 must be mapped to the corresponding structure in the specification (communication port 510, logic circuitry 502, encryption logic 300). If a proposed combination does not map to that structure, the § 103 ground fails on the means-plus-function claim even if it succeeds on claim 1.

Confidence. High: the § 102 subsection availability table and the conclusion that no single face-sheet reference anticipates claims 1/21/42/63. Medium: the § 103 motivations in Grounds 1–3 (these are legally constructible arguments, not adjudicated ones). Low / explicitly unverified: subject matter of Brown '863, Augustine '052, Liang '106 and Aziz '842; the Augustine '052 inventor conflict; the US filing dates for Jones '637 and Khello '423; and the text of claims 41–82, which is absent from the authoritative source supplied.