US 5623552

I'll verify the patent details and check for any litigation or CAFC docket activity.

Let me check specifically for any PTAB/reexamination or litigation activity referencing this patent.

US 5,623,552 — Verification Summary

Identification. I found one and only one record for the literal number 5,623,552 (published as US5623552A). No near-number variants were substituted. Sources cross-checked: Google Patents (patents.google.com/patent/US5623552A/en), the official USPTO facsimile (patentimages PDF), EveryPatent's front-page transcript (everypatent.com/comp/pat5623552.html), and Searchlight (patents.searchlight.law/doc/US5623552).

Bibliographic data (as printed on the face of the patent)

Abstract (verbatim)

"A self-authenticating identification card includes a fingerprint sensor for authenticating the identity of a user. An ID card memory permanently stores information related to a fingerprint of the card user. An on-card authenticator compares information related to a sensed fingerprint with the stored fingerprint information and produces an authentication signal if the sensed fingerprint information matches the stored fingerprint information. An audio generator is preferably included in the card for transmitting a predetermined audio signal to a speaker if a match is sensed. A second memory, in the form of a programmable magnetic stripe, may be included on the card for storing account information related to the card user. A magnetic stripe programmer is then used for loading predetermined account information into the programmable magnetic stripe if the sensed fingerprint information matches the stored fingerprint information. Account information is automatically cleared from the programmable magnetic stripe after a predetermined elapsed time."

Plain-language overview of the independent claims

There are four independent claims: 1, 10, 17, and 26. Claims 2–9 depend from 1; 11–16 depend from 10; 18–25 depend from 17; 27–29 depend from 26.

Claim 1 — the card (enrollment-centric). An ID card built on a substrate that carries:

• a fingerprint sensor on the card itself;
• memory inside the substrate for holding the cardholder's fingerprint data;
• an activator that, when triggered for the first time, starts internal storage of the user's fingerprint;
• a controller inside the substrate that "independently program[s]" that internal memory by issuing an internal control signal — i.e., the card writes its own template from on-card sensing, without an external programmer;
• a verification device that confirms the fingerprint was successfully stored (this is the "did my enrollment work?" indicator); and
• an authenticator that compares a later sensed fingerprint to the stored one and emits an authentication signal on a match.

The practical thrust: this is a single-card, self-contained identity system — no external optical scanner or terminal is required, and the card performs its own enrollment loop and confirms it.

Claim 10 — the self-authenticating card (verification-centric). Same core architecture (substrate, on-card fingerprint sensor, internal memory for the user's fingerprint, an internal controller that independently programs that memory via an internal control signal), but here the activation is responsive to the fingerprint sensor (touching the sensor itself triggers the sequence rather than a separate pad), and the claim goes straight to the authenticator producing a match signal. Claim 10 notably does not require the verification/storage-confirmation element of claim 1 — it reads on the authenticate-and-match function.

Claim 17 — method of using the card. Steps:

1. sense fingerprint information on the card's own fingerprint sensor;
2. independently generate at least one control signal internal to the card to move the sensed data internally from the sensor to an internal card memory and electronically store it there; and
3. produce a verification signal confirming the data was successfully stored in that internal memory.
   This is the method counterpart to claim 1, again emphasizing that enrollment/transfer is done by the card's own internal control, not an outside device.

Claim 26 — the card with a programmable magnetic stripe and auto-expiry. This is the most feature-loaded independent claim. It requires everything in the card architecture above plus a programmable magnetic stripe on the substrate, a verification device, and magnetic stripe programming means that (a) load predetermined account information into the stripe only if the sensed fingerprint matches the stored fingerprint, and (b) clear the account information after a predetermined elapsed time. The claim thus captures the commercial core of the disclosure: the account number is not sitting vulnerable on the stripe; it appears only after a successful biometric match and then self-destructs on a timer.

Representative dependent-claim coverage (for scope context): yellow/green/red LED visual signals (claims 2, 4, 27; and the red no-match LED in the spec), speaker/audible indication (claims 3, 5, 28), audio generator emitting a predetermined coded audio signal on a match (claims 6, 13, 29 — used to authenticate the card to a phone-based approval center), second memory for account data (claims 7, 14), programmable magnetic stripe + programmer (claims 8, 15), and automatic clearing on elapsed time (claims 9, 16, 25).

Prosecution, family, and docket check

• Family: the only related U.S. application is the parent Ser. No. 08/184,686 (Jan. 21, 1994, abandoned); US5623552A is the sole publication and the sole family member. Google Patents lists a single "Family Applications (1)" entry. No continuation-in-part, no foreign counterpart in the same family.
• Prior art of record (23 U.S. references) includes Lemelson (3,812,461; 3,940,795; 4,471,343), Szwarcbier (4,253,086), Tsikos (4,353,056, expressly incorporated by reference for the capacitive ridge sensor), Dreifus (4,575,621), Froelich (4,983,036), Lofberg (4,298,825 – printed as 4,582,985), Costello (5,321,765), plus JP 0313288.
• Litigation / CAFC: My searches of public web indexes returned no CAFC 2026 docket entry, and no district-court or ITC case, naming 5,623,552. I want to be explicit about the limits of that finding: I checked general web sources, not a PACER/CM-ECF or the Federal Circuit's own docket system, so this is "nothing found," not a certified negative. It is also consistent with the expiration of the patent on 2014-01-21, which makes live 2026 appellate activity implausible absent a very unusual collateral proceeding.
• One USPTO document I did surface is a PTAB/USPTO petitions artifact (ptacts.uspto.gov … petitions/1476836) that reproduces claim 1 of this patent in the exact language shown above. I could not confirm from the retrieved snippet which proceeding (or which party) that petition belongs to, so I am flagging it as an unverified lead rather than asserting a reexamination, IPR/PGR, or interference exists. The claim text in that document matches the issued claims as reproduced by Google Patents and EveryPatent, so at minimum it corroborates the claim wording.

Notes on fidelity and uncertainty

1. OCR artifacts in the source text — preserved, not corrected. The specification summary paragraph contains "fingerpaint sensor" (an evident scanning error for "fingerpaint"→fingerprint), and claim 7 reads "second memory means an said substrate." I have quoted/reported these as they appear and note them as transcription errors, not substantive claim content.
2. Claim language is consistent across independent transcripts. The "controller means internal to said substrate … generating an internal control signal" and "activation … for a first time" limitations in claim 1 appear identically in the Google Patents claim set, the EveryPatent front-page transcript, and the USPTO petition document — so I treat that language as the issued claim text, and not as a post-issuance (e.g., reexamination) amendment. I found no B1/B2/C1 reexamination certificate variant of this number.
3. Assignee field requires care. The face of the patent names CardGuard International, Inc.; Google Patents separately records a 2004 license agreement to Bob Barker Company, Inc. and a 2004 assignment to UNIQARD, LLC, with current-assignee fields listing both UNIQARD LLC and CardGuard International Inc. Google's own disclaimer applies — it performs no legal analysis of assignment accuracy. Treat current ownership as needing confirmation from USPTO Assignment records.
4. "Self-authenticating" is used in the patent's own idiosyncratic sense. The patent claims the card performs biometric enrollment, storage, and matching on-card without external equipment — the card is self-authenticating in that it does not need a separate fingerprint scanner. It is not claiming cryptographic self-authentication in the modern sense.
5. Sibling patents in the same commercial family (e.g., WO02065375A3 / "Self-authenticating identification substrate with encoded packet output," ID Technologies Corp.) cite or descend from this work; if you need the full portfolio picture I would check those separately rather than assuming they share this specification.

I'll research litigation involving US patent 5623552 across litigation databases and other sources.

Let me search specific litigation databases and docket sources.

Let me search the specific litigation databases mentioned and related case repositories.

Litigation Search Results for US Patent 5,623,552

Summary: No litigation involving US Patent 5,623,552 was identified in the sources I was able to search.

I want to be candid about the limits of this result, since a negative is only as good as the search behind it.

What I searched and what I found

I queried for the exact patent number (5,623,552), its unformatted form (5623552), and its title/inventor/assignee combinations against the web. The results returned were prior-art and patent-family references, not litigation:

• Patent office search reports citing it as prior art — e.g., the European search report for EP 0 924 657 A3 (TRW Inc.) lists "US 5 623 552 A (LANE WILLIAM F) 22 April 1997" as a category "A" background reference; the PCT search report for WO 2002/075665 A3 cites it as an "X" reference against claims 1, 3, 11, 12, 14; WO 2008/066792 A3 and GB 2 420 098 A similarly cite it as prior art. These are examination citations, not lawsuits.
• Patent database entries (Google Patents, Justia, EveryPatent) describing the patent itself.
• Unrelated hits for the same number — notably a Japanese patent "特許 5623552" (JP registration 5623552) for an Fe-Pt ferromagnetic sputtering target. That is a different patent number in a different jurisdiction and has no relationship to the Lane card patent.

None of the returned results were dockets, complaints, verdicts, or litigation summaries naming US 5,623,552.

Important limitations on this finding

I was unable to directly retrieve pages from the specific litigation-oriented databases you named (Unified Patents litigation portal, CAFC, PACER, Docket Alarm) — those tool calls hit a step limit before returning results, so I did not get a clean direct query of those databases itself. Given that, please treat the following as a caveat rather than a guarantee:

1. I cannot affirmatively state that litigation never existed. I can only report that no litigation surfaced in the searches I ran.
2. Procedural/contextual facts that bear on any litigation search:
   • Google Patents records the patent as "Expired – Lifetime," with anticipated expiration on 1994-01-21 + 20 years = 2014-01-21. Any infringement suit would have had to be filed before that date (a § 271(e)(2) ANDA-type action aside, which is implausible for this subject matter).
   • The chain of title shows the patent moving CardGuard International, Inc. (original assignee, 1995/1996) → licensing to Bob Barker Company, Inc. (2004) → assignment to UNIQARD, LLC (2004). These transactions are recorded, but recorded assignments/licenses are not litigation, and the Google Patents front matter explicitly disclaims any representation about assignee accuracy.
   • The extraordinarily large "Cited By (609)" list reflects subsequent patents citing this one as prior art (the normal forward-citation family for a seminal card/biometric reference), not lawsuits.

Conclusion

Based only on the sources I was able to search, I found no known litigation involving US Patent 5,623,552. If you need a confirmatory answer, the definitive next steps would be a direct query of:

• Unified Patents litigation portal — https://portal.unifiedpatents.com/litigation/caselist (search 5,623,552)
• CourtListener / RECAP (free PACER mirror) and PACER Case Locator, using both 5623552 and 5,623,552
• Docket Alarm / Lex Machina / Darts-ip for a patent-number-keyed litigation report
• USPTO Patent Center "Litigation" / assignment records for the patent

I did not fabricate any case names, numbers, jurisdictions, or dates for this patent; a fabricated docket entry would be worse than a documented "none found."

I'll verify the structured data against live sources before reporting.

PTAB / AIA Trial Activity — US 5,623,552

Verification basis: The structured "PTAB proceedings on file" block (USPTO Open Data Portal, most recent ingest) returns no AIA trial proceedings. I ran independent web sweeps (Patent Trial and Appeal Board dockets via PTAB E2E / Docket Alarm indexing, Federal Circuit / CourtListener, and general query strings on the patent number and the "self-authenticating identification card" title) and found nothing to contradict the ODP result. I did not locate a single IPR, PGR, or CBM petition, institution decision, Final Written Decision, or Federal Circuit appeal involving this patent. Confidence: high for PTAB; moderate for district-court history (see caveat at the end).

Proceedings overview

Total AIA trial proceedings: 0. No active proceedings, no claims invalidated by the Board, no claims sustained by the Board, no settlements, no institution denials — because no petition was ever filed. That is not a "hardened patent" signal; it is a signal that the patent is dead on arrival. US 5,623,552 lists an anticipated expiration of 2014-01-21 and a legal status of "Expired – Lifetime." There is no live infringement exposure to defend against, and therefore no defensive reason to have petitioned. A demand letter today citing this patent is asserting claims that expired over twelve years ago.

Per-proceeding detail

None. There are no proceedings to itemize. I am deliberately not constructing placeholder entries — inventing a proceeding number or a fabricated FWD disposition would be the single most damaging thing I could do in this memo, because a defendant relies on claim-level validity status. The canonical list is empty; the absence is the finding.

What the record does show (the useful part for a defendant)

Although there is no PTAB paper trail, the patent's prosecution, assignment, and licensing record is instructive — and materially more important than any IPR would be.

Chain of title / monetization posture. The '552 patent issued 1997-04-22 to CardGuard International, Inc. (inventor William F. Lane; application 08/515,151, filed 1995-08-15, as a continuation of Ser. No. 08/184,686 filed 1994-01-21, now abandoned). The record then shows:

• 2004-01-15 — assignment to Bob Barker Company, Inc., recorded as a License Agreement, assignor ID Technologies Corporation.
• 2004-07-23 — assignment to UNIQARD, LLC, assignor ID Technologies Corporation.
• Current assignee of record: UNIQARD LLC (with CardGuard International Inc. also listed).

This is a classic small-portfolio, license-driven posture — a card-guard/licensing entity rather than a practicing manufacturer. Notably, an Amended & Restated License Agreement (effective February 2001) surfaced in public SEC-filing archives defines "Licensed Product" expressly by reference to infringement of "the 552 Patent," with a Field of License limited to FAA-approved airline/airport identification and security cards, and expressly excluding general-purpose credit/debit cards and Internet security. That tells you the patent owner historically monetized by carving out narrow fields rather than litigating broadly. I found no filed infringement action, no ITC complaint, and no appellate opinion naming this patent.

Terminal disclaimer / family. The '552 patent has essentially no family: the parent 08/184,686 was abandoned, there are no continuations or divisionals issuing from it, and no reissue. Its foreign counterpart record shows "NONE." Everything downstream of it is citation traffic, not family: it appears as a cited reference in 609 later US patents and in PCT search reports (e.g., WO 2008/066792 cites it as a "Y" reference against claims 1–10), and in the prosecution file history of unrelated patents (e.g., US 7,747,636). Being cited as prior art does not confer scope on the '552 claims.

Term analysis. Priority/filing chain runs from 1994-01-21; the '552 continuation was filed 1995-08-15, which places it under URAA twenty-year term measured from the earliest claimed filing date. Expiration = 1994-01-21 + 20 years = 2014-01-21, matching the ODP-adjacent record entry "2014-01-21 Anticipated expiration." The patent is expired and has been for over a decade. (For completeness: no § 154(b) patent term adjustment or § 156 extension appears anywhere in the record, so there is no lurking extension.)

Strategic summary

Claim status. All 29 claims — independent claims 1, 10, 17, and 26, and their dependents 2–9, 11–16, 18–25, and 27–29 — are expired, but note the precise reason they are untested: none were ever canceled by the Board, and none were ever adjudicated in a PTAB FWD, because no AIA petition was filed against this patent. "Untested before the PTAB" and "enforceable" are not the same thing. The claims were never invalidated administratively; they simply ran out of term. There is no set of "surviving claims" to worry about, because there is no surviving term. Do not let opposing counsel blur that distinction — if a demand letter or complaint recites claims 1, 8, or 26 (the independent claims with the on-card controller/programmable-magnetic-stripe limitations, which are the commercially interesting ones), the correct response is not an invalidity argument; it is a term-expiration argument.

Estoppel landscape. Section 315(e)(2) estoppel is irrelevant here — it binds only petitioners, real parties in interest, and privies who filed an IPR that reached a FWD, and no one did. There is no petitioner estopped from anything, and correspondingly no prior-art ground that has been "used up." If you were somehow facing a live assertion (you aren't, on the current term), every § 102/§ 103 ground would be on the table, plus § 112 and § 101. The prior-art universe is unusually rich: the patent's own background section concedes U.S. Pat. No. 4,253,086 (Szwarcbier), U.S. Pat. No. 4,575,621 (Dreifus), U.S. Pat. No. 4,983,036 (Froelich), and — critically — the fingerprint sensor itself is expressly incorporated from U.S. Pat. No. 4,353,056 (Tsikos). That incorporation is a standing admission that the sensing element is not the inventive contribution, which would have been the obvious attack vector had anyone bothered to petition.

Pattern signals. No repeat petitioner, no serial filer, no defensive aggregator (no Unified Patents petition), no patent-owner appeal activity, no IPR certificate ever issuing. The reason is structural and worth internalizing as a general heuristic: defendants petition PTAB against asserted patents. This patent's monetization era (roughly 1999–2004, per the license and assignment record) largely predated the AIA — IPR only became available on 2012-09-16 and CBM on the same date under the transitional program. By the time AIA trials existed in volume, the '552 patent had roughly two years of term left. That two-year window never generated a petition, which itself is evidence that no one was being meaningfully sued on it in 2012–2014.

Recommended next steps

1. If you are a defendant and have received an assertion of US 5,623,552: do not budget for an invalidity case. Confirm the expiration date from USPTO Patent Center (application 08/515,151) and lead with § 286. All causes of action for infringement require the act to occur during the patent term; the last possible infringing act was 2014-01-21. Section 286's six-year damages lookback means that, as of today (2026-09-30), only suits filed on or before roughly 2020-01-21 could reach even a sliver of pre-expiration conduct. A suit filed now reaches nothing. Check the docket for any prior action and its dismissal date before assuming no earlier case exists.

2. If opposing counsel cites PTAB outcomes: there are none, and say so plainly, with the ODP result. Anyone representing that this patent was "upheld by the PTAB" or "survived IPR" is misrepresenting an empty docket as a validity win. There is no FWD to link, no institution decision to distinguish, and no CourtListener / Federal Circuit opinion to cite.

3. If you nonetheless need an invalidity position on the merits (e.g., a declaratory-judgment posture, a licensing diligence question, or a portfolio-acquisition review): the strongest invalidity story is the Tsikos incorporation plus the Szwarcbier/Dreifus/Froelich background art, with Froelich (U.S. Pat. No. 4,983,036) being the closest on the "on-card comparison of a stored exemplar" concept. Because no § 315(e)(2) estoppel attaches, all of it remains available in any forum.

4. Do not file an IPR as a defensive move. Although the Board has instituted IPRs on expired patents where claim construction required it (see, e.g., Sony Corp. v. Yissum Research Dev. Co., IPR2013-00327), there is no practical benefit here: no live damages, no injunctive exposure, and fees that buy nothing.

Caveat on scope, stated honestly. My sweeps were deliberately targeted and did not exhaustively reconstruct every 1990s–2000s district-court docket that might have named the '552 patent. I found no litigation and no PTAB filing, but I am reporting "none found," not a certified negative on district-court history. The PTAB negative, by contrast, is sourced to the canonical ODP block and corroborated by independent search — treat "0 AIA trials" as reliable.

I'll research the assignment record using USPTO Assignment Center, Google Patents legal events, and related sources.

Good — found a key cross-reference. Let me dig deeper into the assignees and assignment record specifics.

I hit the tool-call limit partway through, so some primary-source fields (reel/frame, correspondent) could not be pulled directly. Everything below is either from the authoritative patent text in this thread or from sources I retrieved, and I flag every gap explicitly rather than filling it by inference.

Inventors

• William F. Lane, Wilson, North Carolina — sole named inventor. Per the issued patent front page (US5623552A PDF, "[75] Inventor: William F. Lane, Wilson, N.C.").
• Employer at filing: CardGuard International, Inc., also of Wilson, N.C. — same city, same entity as the assignee named on the face of the patent ("[73] Assignee: CardGuard International, Inc., Wilson, N.C."). This is an inventor-assigns-to-his-own-employer pattern, which is corroborated by the 1996-10-21 record in which the assignor of record is "LANE, WILLIAM F." and the assignee is CARDGUARD INTERNATIONAL, INC.
• Unusual patterns: The patent is a continuation of Ser. No. 08/184,686, filed 1994-01-21, now abandoned, filed by the same inventor. So the effective priority date (1994-01-21) precedes the CardGuard assignment record by ~2.75 years, and the patent did not issue until 1997-04-22. I found no evidence that the sole inventor left CardGuard within 12 months of filing, and I have no employment-history data for Lane at all beyond the assignee match. I am not treating this as a pre-fire-sale signal — I simply lack the data either way.

Original assignee

• CardGuard International, Inc., Wilson, North Carolina — named on the issued patent (1997-04-22) and the assignee of the 1996-10-21 recorded assignment from Lane.
• Product evidence: None retrieved. The specification describes the card as a "preferred embodiment" and the sensor as simulated/tested ("A model of the system has been tested on real and simulated fingerprint data"), which is protocol language, not commercialization language. The patent itself is not evidence that CardGuard shipped a card embodying the claims, and I found no product record for CardGuard.
• Business/status: Not determinable from the sources I retrieved. The name "CardGuard" is reused in unrelated fields (a NIC-based intrusion-detection research system, and a Swiss crypto/security vendor), so I deliberately did not attribute those to this CardGuard International, Inc. Its current corporate status (operating / acquired / dissolved / bankrupt) is unverified in my searches.
• Key downstream fact: by 2000–2004 the patent rights were being held and licensed by ID Technologies Corp. ("IDTEK") of Raleigh / N.C. State's Centennial Campus, which a contemporaneous trade-press item describes as a "biometric technology licensing company … The company owns the rights to patents that allow identification to be made by self-authenticating credit and smart cards" (Triangle Business Journal, 2000-12-04). That press characterization is a documented, non-inferred description of the intermediate holder as a licensing (non-practicing) entity.

Assignment timeline

Important data limitation, stated up front: the Google Patents legal-events block in the authoritative text for this patent lists the three recorded conveyances with dates and parties but no reel/frame numbers and no correspondent of record. My attempts to open USPTO Assignment Center / assignment.uspto.gov for this patent did not return before the tool limit, so I could not capture Reel/Frame or the recording attorney. I am not inventing those fields. Where the format below asks for them, I mark them [not retrieved].

1. Executed [date not retrieved] / recorded 1996-10-21 — Reel [not retrieved]/Frame [not retrieved]

   • Conveyance: Assignment of assignors' interest (per the legal-events label "ASSIGNMENT OF ASSIGNORS INTEREST").
   • Assignor: LANE, WILLIAM F. (sole inventor)
   • Assignee: CARDGUARD INTERNATIONAL, INC.
   • Correspondent: [not retrieved] — cannot assess recurrence.
   • Context: Invention-capture / founder assignment — the sole inventor assigning his application to the company that was already named as assignee on the eventual patent.

2. Executed [date not retrieved; press indicates the underlying deal dates to ~Dec 2000] / recorded 2004-01-15 — Reel [not retrieved]/Frame [not retrieved]

   • Conveyance: LICENSE AGREEMENT (recorded as a citable "reassignment" event, but the conveyance type is a licence — not an assignment of title).
   • Assignor: ID TECHNOLOGIES CORPORATION
   • Assignee: BOB BARKER COMPANY, INC.
   • Correspondent: [not retrieved] — cannot assess recurrence.
   • Context: Field-limited commercial licence, not a title transfer. Cross-referenced to the Triangle Business Journal report: IDTEK "issued a prepaid exclusive license for detention systems to Bob Barker Company, the largest prison supplier in the United States. The license is valued in excess of $300,000 with a 10 percent royalty and a 15 percent manufacturing surcharge to IDTEK." So Bob Barker was a licensee in the correctional/detention field only. Note the ~3-year gap between the reported deal (Dec 2000) and recordation (Jan 2004) — an execution-vs-recordation lag, not two separate transactions.

3. Executed [date not retrieved] / recorded 2004-07-23 — Reel [not retrieved]/Frame [not retrieved]

   • Conveyance: Assignment of assignors' interest ("ASSIGNMENT OF ASSIGNORS INTEREST").
   • Assignor: ID TECHNOLOGIES CORPORATION
   • Assignee: UNIQARD, LLC
   • Correspondent: [not retrieved] — cannot assess recurrence.
   • Context: Transfer-to-current-holder. Same assignor (IDTEK) moved the patent to an LLC six months after recording the Bob Barker licence. This is the last recorded title event; UNIQARD LLC is listed by Google Patents as the current assignee.

⚠️ Contradiction with the previously generated litigation section

The earlier Litigation summary states the chain as "CardGuard International, Inc. (original assignee, 1995/1996) → licensing to Bob Barker Company, Inc. (2004) → assignment to UNIQARD, LLC (2004)." That is imprecise on the record:

• The assignor on both 2004 records is ID TECHNOLOGIES CORPORATION, not CardGuard International and not Bob Barker. There is therefore a missing link in the recorded chain — CardGuard International → ID Technologies Corporation — that does not appear among the legal events on the Google Patents page. Its execution and recordation dates are unknown to me.
• Bob Barker never held title; its record is a licence, so the "→ assignment to UNIQARD" arrow should not be read as passing through Bob Barker.

I flag this rather than silently reconciling it, per the cross-reference instruction.

Corroborating the "IDTEK as licensor" reading, the recorded licence excerpts I retrieved reference the "'552 Patent" (i.e., this patent) as the licensed property, name "IRE" as a manufacturing partner, and define the field of licence as mass-transit cards and related systems (Law Insider exhibit). I could not confirm the counterparty names in those excerpts, so I use them only as evidence that the '552 patent was field-of-use licensed across multiple industries (detention, mass-transit) by its then-owner — consistent with a licensing-monetization strategy, not a product business.

Timeline diagram

timeline
    title Ownership of US 5623552
    1994 : Priority application filed
    1995 : Continuation filed by CardGuard
    1996 : Lane assigns to CardGuard
    1997 : Patent issued
    2004 : Bob Barker license recorded
         : Assigned to UNIQARD LLC
    2014 : Patent expired

NPE / troll-pattern signals

1. Shell-entity transfer — unclear. The terminal holder is UNIQARD, LLC (recorded 2004-07-23), an LLC-style entity reached directly from a licensing company (ID TECHNOLOGIES CORPORATION, described in trade press as a "biometric technology licensing company"). That is a suggestive sequence, but the required tells are not documented: I could not confirm (a) that UNIQARD has no products in commerce, (b) a registered-agent-only address, or (c) single-member Delaware/Texas formation. "UNIQARD" also lacks the "IP / Patents / Holdings / Ventures" suffix pattern. Not enough to call it. Unclear.

2. Known asserter in the chain — not present. Neither ID TECHNOLOGIES CORPORATION nor UNIQARD, LLC appears on the enumerated lists (Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities). No Unified Patents or RPX high-frequency-plaintiff hit surfaced for either name. IDTEK is a documented licensing-only entity, but licensing-only ≠ named-asserter, so this signal is scored not present for the listed directories. Not present.

3. Repeat correspondent across the chain — unclear (no data). Correspondent of record was not retrieved for any of the three recordings. I therefore cannot test whether one attorney/firm filed all of IDTEK's recordings, which is exactly the tell that would matter here. Unclear — flagged as the single highest-value missing datum.

4. Cascading transfers — weak / unclear. Two conveyances left the same assignor (ID TECHNOLOGIES CORPORATION) within roughly six months of each other in 2004 (Jan 15 licence to Bob Barker; Jul 23 assignment to UNIQARD LLC). That same-assignor, short-window pattern (skimming field licences, then pushing residual title to an LLC) is a mild monetization tell. But there is no chained LLC-to-LLC cascade, no shared correspondent address demonstrated, and I could not establish common principals. Weak/unclear.

5. Pre-litigation transfer — not present. The prior litigation section found no litigation naming US 5,623,552. With no suit to anchor against, there is no pre-litigation-transfer window to measure. (The patent's anticipated expiration was 2014-01-21, so any infringement action would have had to predate that.) Not present.

6. Bankruptcy fire-sale — not present / unclear. No bankruptcy record for CardGuard International, Inc. or ID Technologies Corporation surfaced, and the 2004 transfers show no Chapter 7/11 sale mechanism. CardGuard's ultimate corporate fate is simply unknown to me; absence of evidence here is not evidence of absence. Not present (with data caveat).

7. Privateering — not present. Privateering requires an operating-company-to-NPE transfer with the NPE asserting against the operating company's competitors on its behalf. Neither the IDTEK→Bob Barker licence (an operating prison-supply company taking a field licence) nor the IDTEK→UNIQARD assignment has any documented backer-asserts-for-a-patron structure, and no SEC/EFF/Patent Progress coverage surfaced. Not present.

8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN; it terminates at UNIQARD, LLC. No neutralization signal. Not present.

Verdict

Insufficient data.

The recorded ownership chain is short and does end in an LLC — CardGuard International, Inc. (1996-10-21) → [unrecorded link to] ID Technologies Corporation → UNIQARD, LLC (2004-07-23), with a field-limited licence to Bob Barker Company, Inc. recorded 2004-01-15 — and a licensing-only intermediary (ID Technologies Corp.) is documented by contemporaneous press. That is enough to show a licensing-monetization lineage, but no strong NPE signal is satisfied: UNIQARD is not on any enumerated asserter list, no litigation naming this patent was found (so no pre-litigation-transfer window exists), and the corroborating tells that would convert suspicion into a finding — USPTO reel/frame numbers and the correspondent of record on each recording — were not retrieved and are recorded above as gaps rather than guessed. If UNIQARD, LLC can be confirmed to be a non-practicing, single-purpose licensing entity (no products, registered-agent address), the correct re-grade would be NPE — moderate confidence, driven by the IDTEK→UNIQARD sequence; I did not make that call on present evidence.

Verify independently at the USPTO Assignment Center search page (search 5623552): https://assignment.uspto.gov/patent/index.html — and pull the Reel/Frame and correspondent for each recording, which are the decisive fields this analysis is missing.

I'll search for patent 5623552 and its prior art citations.

I have the complete references-cited list. Let me pull details on the remaining key references (Dreifus, Froelich, Tsikos, Lofberg, Elmes, Costello).

Prior Art Analysis — US Patent 5,623,552

Patent under analysis: US 5,623,552 A — "Self-authenticating identification card with fingerprint identification"

• Inventor: William F. Lane (Wilson, NC)
• Original assignee: CardGuard International, Inc.
• Appl. No. 08/515,151; filed Aug. 15, 1995
• Priority/effective filing date: Jan. 21, 1994 (continuation of Ser. No. 08/184,686, filed Jan. 21, 1994, now abandoned)
• Granted: Apr. 22, 1997; 29 claims; Art Unit 266; Primary Examiner Couso, Jose L.
• Status: Expired – Lifetime (anticipated expiration Jan. 21, 2014)

Identifier disambiguation (per the "do not auto-correct" rule): Searching the literal string 5623552 also surfaces JP registration 5623552 (an Fe-Pt ferromagnetic sputtering target) and the unformatted US number 5623552. Those are different rights; only US 5,623,552 A (Lane/CardGuard) is analyzed here. (This was flagged in the earlier litigation section and is repeated so the distinction is not lost.)

Methodological caveat: I was unable to open USPTO Patent Center / PAIR directly in this session. The citations list below is taken from the authoritative front-page "References Cited" of the patent itself (the granted-PDF front page, as reproduced via Google Patents / EveryPatent / Searchlight), which is the controlling record of the examiner's and applicant's citations. Where I could not verify a reference's title or disclosure from a primary source in this session, I say so explicitly rather than guessing.

0. Threshold point: which "citations" are prior art here

"Citations" for this patent run in two directions, and only one direction is prior art:

• Backward citations (23 items) — the U.S./foreign patents cited on the face of the '552 patent. These are the prior art. They are analyzed below.
• Forward citations ("Cited By (609)") — the ~609 later patents/publications that cite the '552 patent as prior art. These are not prior art to the '552 patent (they post-date it). They are relevant only to the patent's impact/value, not its validity. I do not treat them as anticipatory art here.

1. The independent claims that drive the §102 analysis

The anticipation question turns on the four independent claims. The recurring, distinguishing elements are:

• Claim 1 — card substrate + fingerprint sensing means carried by the substrate + memory means internal to the substrate + activation means + controller means internal to the substrate that independently programs the internal memory by generating an internal control signal to transfer fingerprint information to the internal memory (i.e., on-card enrollment) + verification means + authentication means on the card producing an authentication signal on a match.
• Claim 10 — same as claim 1 but without a verification means, and activation is responsive to the fingerprint sensor.
• Claim 17 (method) — sense fingerprint on the card's sensor; independently generate ≥1 control signal internal to the card to transfer/store the fingerprint internally; produce a verification signal.
• Claim 26 — the "kitchen-sink" independent claim: all of claim 1's elements plus a programmable magnetic stripe, a magnetic stripe programmer that (a) loads account data on authentication and (b) clears account data after a predetermined elapsed time.

The critical novelty hook is on-card, self-contained sensing + internal enrollment via an internal controller + on-card comparison, i.e., no external fingerprint scanner. That framing matters because, as the patent's own Background states, the cited art "need[s] to operate in conjunction with some form of separate fingerprint scanner."

2. The complete backward-citation list (23 items)

Ordered as they appear on the face of the patent. Class numbers are the patent's own classifications.

§102 date framework. With a Jan. 21, 1994 effective filing date:

• §102(b) statutory-bar art = published/patented before Jan. 21, 1993. Items 1–21 all satisfy this (all issued 1972–1991).
• §102(a)/(e) art = before Jan. 21, 1994. Item 22 (Costello, Jun 1994) is the only listed reference that does not predate the priority date on its face; treat it as §102(e) only if its effective filing predates Jan. 21, 1994.

3. Substantive §102 analysis of the key references

3.1 US 4,253,086 — Szwarcbier (Feb. 24, 1981)

• Full citation: U.S. Pat. No. 4,253,086, Szymon Szwarcbier, "Process and apparatus for positive identification of customers," granted Feb. 24, 1981. Appl. No. 06/002,316, filed Jan. 10, 1979; Canadian priority Jul. 28, 1978 (CA 308,362). Class 382/126, 340/5.83, 356/71.
• What it discloses: A credit/ID card that carries a master fingerprint of the owner on the card (in mirror image, on a transparent window), plus a separate fingerprinting device and a transparent "fresh-print" card on which the presenter's print is taken at the transaction. The two prints are compared by superimposition with color change, or via a viewer/episcope/projector, or by a computer preprogrammed with the master print (claim 18–21). Match/no-match is signaled by a green light/buzzer or red light/bell (see the patent's own description and the issued claims).
• §102 assessment: This is the reference the '552 patent itself characterizes in its Background. It discloses the concept of fingerprint-on-card verification, but the sensing/reading is done by external apparatus and the stored item is a physical/optical print, not electronic fingerprint information in an internal memory. It therefore does not anticipate claim 1, 10, 17, or 26 (each requires fingerprint sensing means carried by the substrate and an internal memory/controller).
• Claims it potentially anticipates, if at all: None of the independent claims. It is properly characterized as §102 background art / §103 art that establishes fingerprint-on-card verification was known — relevant to motivation-to-combine and to the scope-and-content inquiry, not as an anticipating reference.

3.2 US 4,353,056 — Tsikos (Oct. 5, 1982)

• Full citation: U.S. Pat. No. 4,353,056, Constantine Tsikos, "Capacitive fingerprint sensor," granted Oct. 5, 1982. Appl. No. 156,571, filed Jun. 5, 1980. Class 340/146.
• What it discloses: A two-dimensional array of microscopic capacitors on a sensing member; pressing a finger against the sensing surface locally changes capacitance according to the ridge/valley pattern; the capacitance distribution is converted into an electrical signal (series/parallel capacitor networks read by multiplexing).
• Relationship to the '552 patent: The '552 specification expressly incorporates Tsikos in its entirety by reference as the fingerprint sensor 102 ("One sensor of this type is disclosed in U.S. Pat. No. 4,353,056 to Tsikos, the disclosure of which is incorporated in its entirety herein by reference").
• §102 assessment: Tsikos is a component-level reference only. It discloses the sensing means but no card, no memory, no controller-programmed enrollment, no verification, and no authentication signal. It cannot anticipate any independent claim standing alone.
• Claims it potentially anticipates: None. It bears on the "fingerprint sensing means" element of claims 1, 10, 17, 26 only as §103 art (or as incorporated subject matter).

3.3 US 4,575,621 — Dreifus (Mar. 11, 1986) — the closest on-card-electronics reference

• Full citation: U.S. Pat. No. 4,575,621, Henry N. Dreifus, "Portable electronic transaction device and system therefor," granted Mar. 11, 1986. Appl. No. 586,938, filed Mar. 7, 1984; assignee Corpra Research, Inc. Class 235/380, 235/492.
• What it discloses: A credit-card-form portable transaction device containing a central data processor (CPU), a fixed memory (ROM), an adaptable memory (RAM), an on-board battery, encryption/decryption, a clock, and optical I/O (LED/phototransistor). It operates in a stand-alone mode (self-monitoring/intrusion detection) or an operating mode communicating with a terminal. The '552 patent's Background states that Dreifus's memory "may include retinal patterns or fingerprints for comparison to user characteristics by a terminal at the point of sale."
• §102 assessment: Dreifus supplies the "internal controller + internal memory + self-contained card" architecture, and even the idea of biometric data residing on the card — but the comparison is performed by a terminal at the point of sale, not on-card, and Dreifus has no on-card fingerprint sensor. Every independent claim of '552 requires fingerprint sensing means carried by the substrate. Dreifus therefore does not anticipate claims 1, 10, 17, or 26 as issued.
• Claims it potentially anticipates: None on its own. Dreifus is the primary §103 reference for the "controller means internal to said substrate … independently programming said internal memory means" limitations of claims 1, 10, and 26 and the method step of claim 17. (Note: Dreifus is also a heavily litigated/IPR-tested reference in unrelated patents, which corroborates its breadth as a §103 workhorse rather than a §102 knockout against '552.)

3.4 US 4,582,985 — Lofberg (Apr. 15, 1986) — the reference most worth verifying

• Full citation: U.S. Pat. No. 4,582,985, Lofberg, granted Apr. 15, 1986. Class 340/825 (on the '552 face). Title appears to be "Identification card."
• What it discloses (verification caveat): Löfberg's '985 is widely cited in later fingerprint-sensor/ID-card art alongside Tsikos — e.g., it appears in the "Referenced Cited" list of Veridicom's capacitive fingerprint sensor patent US 6,016,355, sitting next to US 4,353,056 (Tsikos). That placement is consistent with an identification-card reference involving fingerprint sensing/verification.
• §102 assessment — flagged: I could not retrieve the Löfberg full text in this session, so I will not assert what it does or does not disclose. If Löfberg discloses an ID card carrying an integrated fingerprint sensor plus on-card storage/comparison, it would be the single most dangerous §102 anticipation candidate for claim 1/10, because it would supply the "sensor carried by the substrate + internal memory + comparison" combination in one reference. I do not have high confidence on this point and I am not treating Löfberg as an anticipatory reference. It should be pulled and compared element-by-element before any §102 conclusion is drawn.
• Claims it potentially anticipates: Undetermined — cannot be stated without the full text. Listed here as the top verification priority.

3.5 US 4,983,036 — Froelich (Jan. 8, 1991) — closest on-card memory/ROM reference

• Full citation: U.S. Pat. No. 4,983,036, Ronald W. Froelich, "Secure identification system," granted Jan. 8, 1991. Appl. No. 286,214, filed Dec. 19, 1988. Class 356/71 / G06K 9/20.
• What it discloses: An identification card bearing an exemplar fingerprint compared with the presenter's print using optical autocorrelation (the presenter's print is taken on a treated paper carrier and an image of it is raster-scanned over the exemplar). To defeat exemplar forgery, the card contains a 128-cell shift register (read-only, fusible-link-locked) holding a representation of the match waveform, which is compared at presentation. A microprocessor lights a "pass" lamp only when both the coincident peak and the stored waveform correspondence are present. Froelich's system builds on his earlier US 4,690,554 (the '554 patent).
• §102 assessment: Froelich has card-resident electronic memory and a comparison/microprocessor, but the fingerprint is sensed externally (imaged onto the card; presenter's print taken on treated paper). It lacks fingerprint sensing means "carried by the substrate." It therefore does not anticipate claim 1, 10, 17, or 26. Its most transferable teaching is the card-resident verification/authentication signaling, which maps onto '552 claim 1's verification means and the authentication signal.
• Claims it potentially anticipates: None as issued. Strong §103 reference for claims 1 and 10 (on-card memory + comparison + signal), and, in combination, for claim 26's on-card result signaling.

3.6 US 5,321,765 — Costello (Jun. 1994)

• Full citation: U.S. Pat. No. 5,321,765, Costello, granted Jun. 1994. Class 382/4 (on the '552 face). Title/disclosure not verified in this session.
• §102 assessment: Because its grant date post-dates the Jan. 21, 1994 priority date, Costello can only be art under §102(e) (earlier U.S. filing). If its effective filing post-dates Jan. 21, 1994 it is not prior art at all to '552. It should be checked for effective filing date before being relied upon. Class 382/4 (fingerprint/pattern recognition) suggests it concerns fingerprint comparison, but I cannot state its disclosure.

4. The remaining cited references (grouped, with §102 posture)

For the references whose full text I did not retrieve, I group them by the patent's own classification fields, which is grounded on the record, and state the §102 posture. None of these is a candidate to anticipate an independent claim on its own, because none of them (on the face of the record) combines an on-substrate fingerprint sensor with internal memory and an internal controller-programmed enrollment for on-card comparison.

Group A — transaction-card / IC-card electronics (classes 235/380, 235/379, 235/492): US 4,256,955 (Giraud et al., Mar 1981); US 4,614,861 (Pavlov et al., Sep 1986); US 4,667,087 (Quitana, May 1987); US 4,760,245 (Fukaya, Jul 1988); US 4,782,455 (Morinouchi et al., Nov 1988); US 4,864,115 (Imran et al., Sep 1989).
→ Portable card data processors/memories akin to Dreifus. §103 art only for the "internal memory/controller" limitations; no anticipation of claims 1, 10, 17, 26.

Group B — fingerprint-optical / print-comparison references (class 356/71, 340/146): US 4,202,120 (Engel, May 1980); US 4,455,083 (Elmes, Jun 1984); plus US 4,353,056 (Tsikos) covered above.
→ Print-capture/comparison optics. §103 art for the sensing-comparison element; no anticipation.

Group C — access/pattern-recognition and "general" references (classes 340/149, 340/571, 340/825, 179/15, 360/101, 380/23, 364/408): US 3,673,331 (Hair et al., Jun 1972); US 3,812,461 (Lemelson, May 1974); US 3,940,795 (Lemelson, Feb 1976); US 4,471,343 (Lemelson, Sep 1984); US 4,582,985 (Lofberg, Apr 1986 — see §3.4); US 4,677,657 (Nagata et al., Jun 1987); US 4,802,218 (Wright et al., Jan 1989); US 4,926,325 (Benton et al., May 1990); US 4,995,086 (Lilley et al., Feb 1991).
→ Access control, cryptography, and financial data processing. §103 art (context; possibly motivation to combine), no anticipation.

Group D — foreign: JP 63-313288 (Dec 1988, class 902/3).
→ Japanese published application in the card/token area; §102(a)/(b) art to the extent of its disclosure, but no independent-claim anticipation on the record.

5. Bottom line

1. The searched number resolves to US 5,623,552 A (Lane / CardGuard International) — the self-authenticating fingerprint ID card. JP 5623552 (sputtering target) is unrelated.
2. 23 backward citations are on the face of the patent (22 U.S. patents + JP 63-313288). The four the specification discusses substantively are Szwarcbier (4,253,086), Dreifus (4,575,621), Froelich (4,983,036), and Tsikos (4,353,056, incorporated by reference).
3. No cited reference appears to anticipate independent claim 1, 10, 17, or 26 on its own, because each independent claim requires an on-substrate fingerprint sensor plus internal memory plus internal, controller-driven enrollment and on-card comparison — a combination aimed precisely at the deficiency the patent identifies in Szwarcbier, Dreifus, and Froelich (all of which require external sensing/reading).
4. The references function as §103 art, not §102 art: Dreifus for the on-card processor/memory/battery architecture; Froelich for card-resident memory + comparison + result signaling; Tsikos for the capacitive sensor element; Szwarcbier for fingerprint-on-card verification generally.
5. Highest-value verification gap: US 4,582,985 (Lofberg). If it discloses an ID card with an integrated fingerprint sensor and on-card comparison, it would be the leading §102 candidate against claims 1/10. I did not verify its disclosure and am not asserting anticipation. A secondary gap is US 5,321,765 (Costello), whose effective filing date must be checked before it can be relied on at all (only possibly §102(e)).

Recommended next steps for a definitive §102/§103 position: pull the full texts of Lofberg '985, Costello '765, and the six Group-A card-electronics references; confirm Costello's and Lofberg's effective filing dates; and construct an element-by-element chart against independent claims 1, 10, 17, and 26. If a single-reference anticipation is the goal, Lofberg is the only plausible candidate among the 23; absent that, the cited art supports a §103 combination, not §102.

I'll ground this analysis in the actual prior-art references, verifying their content before reasoning about combinations.

I have solid content on Szwarcbier, Dreifus, Froelich, and Tsikos. Let me pin down the two other references of record that matter most for the card-with-onboard-memory-and-display theory.

Obviousness Analysis — US 5,623,552 under 35 U.S.C. § 103

Analyst note on scope. You asked me to use "the results from the Prior Art section of this page." That section was not carried into the material provided to me (I received the Patent summary and Litigation summary sections only). I therefore built this analysis on the two authoritative enumerations that are available and consistent with the patent text itself: (a) the references the specification's own "Background of the Invention" expressly discusses and characterizes (Szwarcbier '086, Dreifus '621, Froelich '036, plus Tsikos '056, which is expressly incorporated by reference), and (b) the references of record listed in the earlier Patent Summary (Lemelson ×3, Szwarcbier, Tsikos, Dreifus, Froelich, "Lofberg," Costello, JP 0313288). Where I could retrieve the reference itself, I quote it; where I could not, I say so. I flag a numbering discrepancy at § 6.

1. Legal framework and premises

Governing law. The application's effective filing date is 21 January 1994 (parent Ser. No. 08/184,686, filed 1994-01-21, now abandoned; the '552 patent is a straight continuation with no new matter). This predates the AIA, so pre-AIA § 103 governs. Graham v. John Deere Co., 383 U.S. 1 (1966), supplies the four factual inquiries; KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), supplies the operative standard for motivation ("a court must ask whether the improvement is more than the predictable use of prior art elements according to their established functions") and for combination (design incentive, market demand, known problem, finite number of predictable solutions). KSR applies to pre-AIA claims.

Because all lists of art are old, the § 102(b) critical date for the parent's 1994-01-21 filing is 21 January 1993; every reference below (Tsikos 1982, Szwarcbier 1981, Dreifus 1986, Löberg 1986, Froelich 1991) comfortably predates it. Costello is the only date-sensitive one (§ 6).

Person of ordinary skill in the art (PHOSITA). A worker with a B.S. in electrical engineering or computer science (or equivalent) and roughly two to three years of experience designing battery-powered, card-format electronic security/identification devices with stored-template biometric comparison, including familiarity with thin-film or capacitive sensor integration. This is a combination art (smart-card electronics + fingerprint recognition), which matters: the relevant question is whether a card-scale electronics designer and a fingerprint-verification designer would have been motivated to merge their fields.

Claims under analysis. Four independents — 1, 10, 17, 26 — with dependents 2–9, 11–16, 18–25, 27–29. I take the claim language from the issued claims as reproduced in the Patent Summary and cross-checked against Google Patents/BigPatent/EveryPatent.

2. What each piece of prior art contributes (limitation map)

3. The combinations

Combination A (primary): Löberg '985 + Dreifus '621 + Tsikos '056, optionally + Froelich '036

This is the combination the Examiner most plausibly would have run, and it is the strongest.

• Löberg '985 alone discloses the entire verification architecture on the card: card-mounted fingerprint sensor → on-card signal processing forming an identification bit sequence → on-card permanently-stored reference bit sequence → on-card comparator → acceptance signal. Löberg even states the design rationale the '552 patent later claims as its contribution: "Due to the fact that the verification is carried out on the very data carrier, the risk of the above related wardrobe case is eliminated." That is the '552 patent's "self-authenticating" concept, four years earlier.
• Dreifus '621 supplies the missing "controller internal to the substrate that independently programs the internal memory" limitation in substance: a credit-card-format device with its own processor, fixed and adaptable memory, internal battery, and a stand-alone mode. Whether a template is written into the on-card adaptable memory by the card's own CPU or fetched from a terminal is a routine implementation choice once Dreifus puts a programmable processor and non-volatile memory on the card.
• Tsikos '056 removes any doubt that the sensor could be an on-card element rather than an external optical scanner: an array of small capacitors producing a bit-mapped electrical signal. The '552 patent itself concedes this by incorporating Tsikos in haec verba and by describing its own sensor as built from Tsikos-type capacitance meters.
• Froelich '036 fills the indicator gap: an on-card microprocessor driving a lamp to signal the outcome of verification (lamp 73).

Result: claims 1, 2, 3, 4, 5, 7, 10, 11, 12, 14, 17, 18, 19 read on this combination with only the "verification means for indicating successful storage" element requiring argument. Claim 10 and 17 do not even require a storage indicator (10 has no verification means; 17 requires a "verification signal," which is closer to Szwarcbier/Froelich's lamps), making them weaker still.

Combination B: Szwarcbier '086 + Dreifus '621 + Tsikos '056 (+ Froelich '036)

Szwarcbier supplies the comparison-and-indicate teaching in the exact form the '552 dependent claims recite (green light/buzzer on match; red light/bell on mismatch; a computer preprogrammed with the card's master print; remote comparison over telephone lines). Dreifus supplies the card-resident processor/memory/clock. Tsikos supplies the integrable sensor. This combination is stronger than A on the audio and remote-approval dependent claims (6, 13, 22, 23, 29) because Szwarcbier expressly discloses returning a comparison result by signal and expressly contemplates the comparison devices being wired or dialed into a central computer — the disclosed genesis of the '552 patent's coded-tone/approval-center feature.

Combination C: for claim 26 (programmable magnetic stripe + auto-clear)

Claim 26 requires (i) a programmable magnetic stripe on the substrate, (ii) loading account data into it only on a fingerprint match, and (iii) clearing that data after a predetermined elapsed time.

• (i) Löberg's own background describes cards with "a loop of magnetizable material … a read-out device … and a read-in device for a read-in of new information" — i.e., a re-writable stripe is old in the data-carrier art.
• (ii) The match-gated write is the only function claim 26's authenticator performs; gating a write on an authorization signal is the established function of Dreifus's adaptable memory.
• (iii) The timer-based clear is supplied by Dreifus's "real-time clock … interrupt control unit for effecting various 'watchdog' functions," which exists precisely to "incapacitate" the device on abnormal conditions and prevent "unauthorized use."

So claim 26 is Löberg '985 + Dreifus '621 + the conventional re-writable magnetic stripe. I regard the auto-clear feature as highly vulnerable: it is the predictable application of a known card-resident clock to the known problem of a lost/stolen card — the very problem Szwarcbier's specification opens with ("thieves and forgers … defraud the credit company"; "hijackers, terrorists, narcotics smugglers … slip through … with falsified identity papers").

4. Motivation to combine (why this is not hindsight)

Four independent KSR rationales are available on the face of these references, without the Examiner needing any hindsight:

1. The problem is recited by the references themselves. Szwarcbier spends its entire background on credit-card fraud and the inadequacy of signatures; the '552 patent's own Background adopts that framing almost verbatim ("The credit card industry, for example, loses tens of millions of dollars every year in credit card fraud"). Same field, same problem, same solution space.
2. The patent itself names the deficiency of the prior art as the design goal. The '552 specification states flatly: "The above cited references using fingerprint identification need to operate in conjunction with some form of separate fingerprint scanner. Separate scanners … dramatically increase the costs and complexity associated with an identification card system." A motivation to eliminate the external scanner is therefore not imputed by the Examiner — it is the applicant's own stated motivation, which is fatal to a "no motivation" argument.
3. The solution was a finite, identified set. By 1991–1994 the only viable on-card sensors were the capacitive/electrode types of Tsikos and its progeny — the '552 patent picked the only real option and said so.
4. Predictable function. Placing a processor, non-volatile memory, clock, and battery on a credit-card substrate, and having the processor write and later clear a memory location, is precisely "the predictable use of prior art elements according to their established functions" (KSR). No new principle of operation is required; the enrollment routine is card firmware.

There is also a teaching-away check to run: Löberg '985 worried that self-selected PINs weaken security, and Dreifus '621 worried that metal contacts and clear-text transfers invite intrusion. Neither criticizes on-card biometric enrollment or on-card template storage; Löberg's entire thrust is on-card storage and on-card verification. Nothing in the set teaches away from Combination A.

5. Where the claims are least and most vulnerable

Most vulnerable. Claim 26 (magnetic stripe + auto-clear) and claim 17 (method of sensing, internally storing, and signaling success). Both sit squarely on old elements used for their known purposes.

Strongest non-obviousness hooks, in descending order of defensibility:

1. The "verification means … indicating that the information related to a fingerprint … has been successfully stored … upon activation … for the first time" (claim 1; and the "verification signal" of claim 17). None of the references I retrieved discloses the storage-confirmation indicator as such — Löberg, Szwarcbier, and Froelich all indicate a match, not a completed enrollment. This is the element I would expect to have carried the allowance. Counter-argument available to a challenger: user feedback confirming enrollment is an obvious ergonomic necessity, and once Froelich's card-resident lamp exists, driving it from the enrollment routine is trivially predictable. The "for a first time" limitation also maps onto Dreifus's one-shot configuration/time-window concepts.
2. "Controller means internal to said substrate … for independently programming said internal memory means … by generating an internal control signal" (claims 1, 10, 26). The word "independently" is doing the work — it means "without an external programmer." But Löberg already performs verification "on the very data carrier" and programs its memory on-card ("the corresponding control information is programmed into the memory 6"). A challenger should press hard that "independently" is a statement of capability, not a structural difference. I would expect this limitation to have been the subject of prosecution argument; its dense, means-plus-function phrasing (repeated near-verbatim in three independent claims) has the earmarks of claim language added or narrowed by amendment to distinguish a card-with-external-enrollment reference. I cannot confirm that from the file wrapper — see § 6.
3. Coded multi-tone audio output (claims 6, 13, 29), as distinct from a buzzer. Szwarcbier's buzzer/bell is not a coded signal carrying identity to a remote machine, and Dreifus's optical LED I/O is not acoustic. If a "predetermined coded audio signal" is construed narrowly (multi-note, machine-decodable, identifying), there is a colorable argument. But Szwarcbier's telephone-linked central-computer comparison and result signaling brings this within the range of obviousness.

6. Confidence, caveats, and what I could not verify

• Numbering discrepancy to flag explicitly (per your no-autocorrection rule). The earlier Patent Summary lists "Lofberg (4,298,825 – printed as 4,582,985)." My searches return US 4,582,985, "Data carrier," inventor Löberg, issued 1986-04-15, as the operative smart-card-with-embedded-fingerprint-comparison reference; I found no Lofberg reference at 4,298,825. I have used 4,582,985 throughout and flag the "4,298,825" string as an unexplained artifact rather than correcting it silently.
• Costello is date-sensitive. The earlier summary lists Costello at 5,321,765. Search results show two Costello patents: 4,947,443 (Aug. 7, 1990 — clearly prior art) and 5,321,765 (issued 1994-06-14, i.e., after the 1994-01-21 priority date). Whether 5,321,765 is § 103 prior art depends on its own filing date under pre-AIA § 102(a)/(e), which I did not verify. I have not relied on Costello for any element.
• Lemelson (3,812,461; 3,940,795; 4,471,343) and JP 0313288 are listed as of record but I could not retrieve and read them within this session, so I have not mapped any claim limitation to them. They are not load-bearing in this analysis.
• Depth of reading. My limitation-mapping rests on the patent's own Background characterization of Szwarcbier/Dreifus/Froelich and on the abstract, summary, and claim text of the references I retrieved. I did not perform a full element-by-element reading of every column of every reference. A filing-grade validity opinion must do that and must check the file wrapper for the amendment history and the Examiner's stated reasons for allowance.
• I did not fabricate a prosecution history. I have no verified record of the specific rejection, the Examiner's art, or the reason for allowance. The observation in § 5(2) that the "internal control signal / independently programming / for a first time" language looks amendment-driven is an inference from claim drafting style, expressly labeled as such.
• Practical posture. The patent is expired (anticipated expiration 2014-01-21, per Google Patents' legal-status field), so this is a retrospective § 103 assessment. It is nonetheless the correct frame if the question is whether the disclosure is valid-as-issued art for a later patent's obviousness fight — in which case, as a seminal reference, its broad claims are the more likely target.

Bottom line. Claims 1, 10, 17, and 26 — and essentially all dependents save the coded-audio ones — would most likely have been held obvious over Löberg '985 in view of Dreifus '621 and Tsikos '056, with Froelich '036 or Szwarcbier '086 supplying the indicator and comparison-signal features and Szwarcbier '086 supplying the remote/telephone-approval features. The single limitation that is genuinely not met by the art I retrieved is claim 1's (and claim 17's) storage-confirmation indicator tied to first-time enrollment — and even that is a predictable feedback feature once Froelich's card-resident indicator is on the table.