US 6728397

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

Let me run additional targeted searches for USPTO records and any 2026 litigation/CAFC docket activity.

Let me do a final check for any litigation or PTAB activity specifically on this patent number.

US Patent 6,728,397 — Summary

Heads-up on the search scope: I searched for the specific number 6728397 (and the literal "6,728,397") across web/patent sources and 2026 Federal Circuit reporting. I found no CAFC 2026 docket, appeal, or PTAB proceeding tied to this specific patent. I did not find any 2026 appellate activity at all for this number. Any statement that this patent is before the CAFC in 2026 would be unsupported by what I could retrieve — so I flag that as unknown / not found, not as confirmed absent. Below is what the authoritative patent record supports.

Bibliographic data (per USPTO/Google Patents record)

Note: the record lists the current assignee as Biometric Payment Solutions LLC while the "original assignee" is "Individual." I'm reporting the record literally rather than reconciling it.

Abstract (as issued)

A verification system for negotiable instruments such as checks that gathers and transmits information about the instrument plus biometric data. It can scan the magnetic number off checks, digitally encode fingerprints, scan driver's licenses/ID cards, and capture a customer signature — all at the point of sale for fund verification. The system digitizes check indicia (preferably the magnetic ink) at the point of sale and transmits the check information to a remotely located main system, which compares inputted data against an existing database to determine whether the customer is authorized to use the account and whether the account is in satisfactory condition for approval. A biometric device records/transmits biometric data such as a fingerprint taken at the point of sale, and can alternately print the biometric data on the check (actual or digitally encoded) for later bank-time checking against a database. The system alternately scans an information card carrying biometric data (e.g., a proper fingerprint and/or signature), and remotely gathered data can be compared against the card's recorded data in addition to or instead of transmission to the database(s).

The independent claims in plain language

• Claim 1 — System (device-plus-databases). A negotiable-instrument verification system for use where an instrument is presented by an individual, comprising: (a) an instrument scanning device that scans and transmits encoded instrument information data; (b) a biometric data device that scans and transmits biometric data "with" the instrument; (c) a biometric database storing biometric data for an authorized user; and (d) an instrument information database storing account information for the authorized user. The biometric device selectively transmits to the biometric database for comparison to verify identity, and the scanning device selectively transmits the encoded instrument data to the instrument information database — together with or separately from the biometric data — to determine whether the account is in condition to satisfy the instrument.

• Claim 10 — System (means-plus-function framing). The same overall architecture, but recited functionally: "negotiable instrument information means" for scanning/transmitting encoded instrument data; "biometric data means" for recording and transmitting biometric data received with the instrument; a biometric database and an instrument information database. Here the ordering is inverted relative to claim 1 — the instrument information goes to the instrument database to verify account condition, and the biometric data goes (together with or separately from the instrument data) to the biometric database to verify the presenter's identity.

• Claim 17 — Method. A method of verifying both the identity of the person tendering the instrument and the condition of the account, comprising: obtaining encoded account information from the instrument; obtaining biometric data from the person; selectively transmitting the encoded instrument information to an instrument information database and comparing it to stored account information to determine account condition; and selectively transmitting the biometric data (together with or separately from the instrument data) to a biometric database and comparing it to stored biometric information — separately from or in parallel with the account comparison — to determine whether the person is authorized to use the account.

• Claim 21 — System for use with a check processing system (two-database fraud/identity variant). A verification system for use with a check processing system, comprising: a scanning device that scans/transmits encoded instrument data; a biometric data device that records/transmits biometric data received with the instrument; a first biometric database for an authorized user; a second biometric database for an invalid user; and an instrument database for authorized-user account information. The biometric device selectively transmits to the first biometric database to verify identity, and the scanning device selectively transmits encoded data to the instrument database to verify account condition — i.e., the fraud/returned-check angle reflected in FIG. 6 of the specification.

Key dependency notes: claim 2 adds a signature scanning device and signature database; claim 3 limits the instrument to a check; claim 4 makes the biometric fingerprint data; claim 5 specifies the instrument data as the magnetic-ink check number; claim 7/8 allow printing recorded (fingerprint) biometric data on the instrument; claim 9 places the scanning/biometric devices remotely from the databases; claims 6 and 14 add customer-usage reporting; claims 19–20 add printing the biometric data on the instrument and returning authorization data to the point of sale. In the claim-21 family, claim 22 specifies MICR data, claim 23 covers digitizing and printing biometric data on the instrument, claim 24 covers use of the second (invalid-user) database, and claim 25 adds a signature-scanning device and signature database.

Family / related U.S. patents (same priority, "Electronic transaction verification system")

US 7,231,068 B2 (CIP, filed 2004-04-01); US 7,349,557 B2 (filed 2004-11-22); US 8,678,273 B2 (filed 2008-03-18); plus later publications US 2013/0002400 A1, US 2013/0002401 A1, and US 2017/0186012 A1. These share the 1998-06-19 priority.

Prior art cited on the face of the patent (examples)

US 4,213,038 (Silverman — access security); US 4,253,086 (Szwarcbier — customer identification); US 4,582,985 (Löfberg); US 4,995,086 (Lilley); US 4,993,068 (Piosenka); US 5,180,901 (Hiramatsu); US 5,484,988 (Hills — check-writing POS); US 5,745,096 (Hsumi — surface-shape fingerprint sensor); and SmartTouch/Lloyd references (US 5,764,789; US 5,802,199). The specification also credits "the devices of Digital Biometrics" for fingerprint digitization.

Confidence / caveats

• Patent text, dates, inventor, claims, and CPC classes are high confidence (taken from the full authoritative record you supplied).
• Assignee history is medium confidence — the record itself notes the assignee list "may be inaccurate," and ownership has passed through several entities.
• CAFC 2026 docket activity for 6728397: not found. My searches surfaced 2026 Federal Circuit decisions on unrelated patents and no appeal, IPR, or PGR concerning this number. I cannot confirm or deny pending 2026 activity; that would require a direct PACER/CAFC docket lookup under this exact number.

I'll search for litigation involving US patent 6728397.

Let me search more specifically for this patent's known litigation history.

I found key litigation. Let me dig deeper into the specific cases and any later assertions.

Litigation Involving U.S. Patent No. 6,728,397 ("Check verification system," Joan Tibor McNeal)

Below is what I could confirm from the searches I was able to run before hitting my tool-use limit. I was not able to query the Unified Patents litigation caselist or PACER directly, so treat this as a strong-but-incomplete picture and verify case numbers on PACER/Docket Navigator.

Confirmed litigation

1. BioPay, LLC v. Solidus Networks, Inc. (d/b/a Pay By Touch) and Indivos Corporation

• Plaintiff: BioPay, LLC (Herndon, VA)
• Defendants: Solidus Networks, Inc. d/b/a Pay By Touch (San Francisco; the then-owner of the '397 patent); Indivos Corporation
• Jurisdiction: U.S. District Court for the District of Delaware
• Case No.: 1:05-cv-00025-SLR (Judge Sue L. Robinson). The court's 35 U.S.C. § 290 notice shows docket "CA 05-0025," filed 01/18/05, D. Del., plaintiff BioPay LLC, defendant Solidus Networks Inc. — I could not independently verify the full "1:05-cv-00025-SLR" string from a second source, so confirm on PACER.
• Filing date: January 18, 2005 (some press reports say Jan. 24, 2005 — that appears to be an error; the § 290 filing form and other outlets give Jan. 18)
• Nature: Declaratory-judgment action seeking a ruling that Pay By Touch's patents are invalid and that BioPay does not infringe them.
• Patents at issue (per the court's § 290 report):
  • U.S. 6,581,042 B2 (listed holder: Indivos Corporation) — 06/17/03
  • U.S. 6,728,397 B2 (listed holder: Joan Tibor McNeal) — 04/27/04
• Outcome/status: Terminated/settled. As of Oct. 2005 the suits had "stalled"; the dispute was resolved when Solidus/Pay By Touch acquired BioPay for ~$82 million in cash and stock, announced in late 2005. Sources: Law.com, "Biometrics Deal Settles Suits"; Digital Transactions News; American Banker (Oct. 11, 2005).

2. Solidus Networks, Inc. d/b/a Pay By Touch v. BioPay, LLC (companion affirmative infringement suit / countersuit)

• Plaintiff: Solidus Networks, Inc. d/b/a Pay By Touch (with Indivos Corporation)
• Defendant: BioPay, LLC
• Jurisdiction: U.S. District Court for the District of Delaware (companion to the above)
• Case No.: Not confirmed in the material I retrieved (likely a February 2005 D. Del. companion case). Verify on PACER.
• Filing date: February 15, 2005 (American Banker reports "On Feb. 15, Solidus filed a countersuit"; press releases describe a Pay By Touch patent-infringement suit against BioPay)
• Nature: Willful patent infringement
• Outcome/status: Settled/terminated together with the case above via the BioPay acquisition.

Context on how the '397 patent came into suit

• The '397 patent issued to Joan Tibor McNeal on Apr. 27, 2004 (application 09/335,649, filed Jun. 18, 1999; priority to provisional 60/089,959, Jun. 19, 1998).
• Assignment chain (per Google Patents): Jan. 21, 2005 → Biometric Payment Solutions; Feb. 4, 2005 → Solidus Networks, Inc.; Jun. 8, 2005 → Solidus Networks, Inc. d/b/a Pay-By-Touch. This is why the '397 patent was in Solidus's bundle when BioPay sued.
• Pay By Touch publicly announced (Jan. 28, 2005) the acquisition of an issued check-cashing patent and a pending application, "serial number 335649," from Biometric Payments Systems, Atlanta. Heads-up on a numbering discrepancy: that press coverage rendered the issued patent as "USPTO number 6,278,397." Taken literally, 6,278,397 is a different patent than 6,728,397; the context (same "check-cashing" subject matter, and application serial 09/335,649 — which is the application that issued as 6,728,397) indicates the article almost certainly intended 6,728,397. I am flagging this rather than silently correcting it, consistent with your instruction to interpret identifiers literally. Digital Transactions News article: https://www.digitaltransactions.net/a-biopay-pay-by-touch-patent-tiff-takes-a-new-twist/

What I did NOT find

• No PTAB proceedings (IPR/PGR/CBM) against U.S. 6,728,397 surfaced.
• No CAFC appeal, no ITC Section 337 investigation, and no later district-court assertion by the current assignee, Biometric Payment Solutions, LLC/LLP, surfaced.
• The FusionArc, Inc. v. Solidus Networks, Inc. (N.D. Cal. 2007) case involved U.S. 6,928,546 asserted against Solidus — it did not involve the '397 patent.
• A 2004 suit by "a former employee who claimed ownership of 15 of the Pay By Touch patents," which Pay By Touch reportedly won (Digital Transactions, June 1, 2004), may or may not have implicated the '397 patent — I could not confirm which patents were involved, so I do not count it as '397 litigation. Worth checking.
• Beware unrelated "397" patents that appear in search noise: U.S. 6,278,397, U.S. 7,011,397, U.S. 7,866,397 (SRP/Transocean drilling claims-construction case), etc. — none of these is U.S. 6,728,397.

Caveats / recommended verification

• I hit my tool-use limit before I could query the Unified Patents litigation caselist (https://portal.unifiedpatents.com/litigation/caselist), PACER, or Docket Navigator, so there may be additional assertions (e.g., later Biometric Payment Solutions enforcement) that I did not surface.
• The '397 patent expired (anticipated) June 18, 2019 (term from the 1999 filing), which limits the likelihood of post-2019 assertions.
• The Feb. 2005 Solidus countersuit case number should be confirmed; I have the date and court but not a verified docket string.

I'll start by verifying the ODP-sourced proceedings list (which shows none) against public PTAB/CAFC sources.

Let me drill into whether any AIA trial ever targeted the '397 patent specifically.

Let me check litigation history and any IPRs on the related McNeal-family patents, plus Unified Patents' tracking of the '397.

PTAB / AIA Trial Report — U.S. Patent No. 6,728,397

Patent: US 6,728,397 B2 — "Check verification system"
Inventor / original assignee: Joan Tibor McNeal (Individual); current assignee listed as Biometric Payment Solutions (chain: BPS → Solidus Networks, Inc. d/b/a Pay By Touch → Bank of New York collateral agent → BPS, LLP)
Priority: 1998-06-19 (Prov. 60/089,959) · Filed: 1999-06-18 · Granted: 2004-04-27
Status: Expired – Lifetime (anticipated expiration 2019-06-18 per the structured record)
Claims: 25 total; independent claims 1 (system), 10 (means-plus-function system), 17 (method), 21 (system for use with a check processing system)

Proceedings overview

Zero AIA trial proceedings are on file against US 6,728,397 — the count is 0, so there is no active/invalidated/sustained/settled/institution-denied breakdown to report. The defensive posture this creates is unusual and worth stating precisely: the patent has not been "hardened" by surviving IPRs, because it was never tested in one — but it is also of essentially no forward-looking assertion value, because the structured record shows it expired on 2019-06-18, meaning there is no IPR victory to cite and no live infringement exposure to defend against.

Proceedings on file

None — no AIA trial has been filed against the '397

• Type: n/a (no IPR, PGR, or CBM)
• Filed: n/a — no proceeding number exists to report
• Status: The controlling structured block in this task states: "The USPTO ODP API returns no AIA trial proceedings for this patent as of the most recent ingest" and instructs that the default is "no PTAB activity on file." I adopt that.
• Judge panel: n/a
• Petition grounds: n/a
• Institution decision: n/a
• Final Written Decision: n/a — no claim of the '397 has ever been canceled, confirmed, or otherwise adjudicated at the PTAB. Claims 1–25 are all untested, not sustained.
• Settlement / termination: n/a
• Appeal: n/a
• Defensive value: You cannot build a defense on an IPR that doesn't exist. Equally, there is no § 315(e)(2) estoppel binding any party, and no PTAB record cutting either way. The real leverage here is the expiration date, not the Board.

Independent search result (flagged, not a proceeding): I ran targeted web searches for an IPR/PGR/CBM against the '397 (queries on "6728397"/"6,728,397" combined with IPR2019–IPR2021 and CBM, and "Biometric Payment Solutions" + IPR). Every hit returned either the patent's own Google Patents / FreePatentsOnline pages or IPRs on unrelated patents (e.g., IPR2020-01223 on U.S. 9,928,495; the Apple/In re Apple biometric-transaction petitions on the '010/'656/'886 patents; Proxense, USAA, and JPMorgan proceedings). I found no petition, institution decision, or FWD naming US 6,728,397. This corroborates the ODP null result. I did not independently verify the PTAB docket for the continuation members of this family (see caveat below), and I did not find a Federal Circuit appeal involving the '397.

Caveat on family members: The '397 is the root of a McNeal family including US 7,231,068, US 7,349,557, US 8,678,273, and later continuations (US20170186012A1, US20130002400A1, US20130002401A1), all claiming the 1998-06-19 priority date. The '397 itself does not inherit invalidity findings from those relatives; a PTAB outcome on a different patent in the family would not cancel the '397's claims. Treat any "the family was invalidated" claim from opposing counsel as a category error, and any claim that the '397 itself was invalidated as unverified.

Strategic summary

Claim status — the honest ledger.

• CANCELED: none. No claim of the '397, independent or dependent, has been canceled — by the PTAB, by the Federal Circuit, or (as far as I could find) by reexamination.
• SUSTAINED (patentable as adjudicated): none. "Sustained" in the IPR sense requires an instituted trial and an FWD confirming claims; neither exists here. Do not describe claims 1–25 as "surviving IPR challenges."
• UNTESTED: all 25 claims. Independent claims 1, 10, 17, and 21 — and therefore every dependent claim (2–9, 11–16, 18–20, 22–25) — have never had their validity decided in a post-grant forum. Their validity posture is exactly as the examiner left it on 2004-04-27, over the 27–28 examiner-cited references (Silverman 4,213,038; Hills 5,484,988; Szwarcbier 4,253,086; Löfberg 4,582,985; Lilley 4,995,086; Piosenka 4,993,068; Hiramatsu 5,180,901; Hsumi 5,745,096; plus SmartTouch 5,764,789/5,802,199, PenOp 5,815,895, CSP Holdings 6,032,137, and Payne 6,072,894), none of which was ever run through an IPR.

Estoppel landscape. Because no IPR was ever instituted against this patent, § 315(e)(2) estoppel attaches to no one. There is no petitioner, no privy, and no "raised or reasonably could have raised" bar. That cuts both ways: (a) a patent owner suing on the '397 faces no estoppel on the PTAB record, but (b) a defendant also faces no adverse estoppel and is free to raise any § 102/§ 103 ground in district court — including art that would have been available in an IPR, without worrying about a prior petitioner's estoppel trapping it. Practically, the § 315(e)(2) discussion is academic here; the date arithmetic is what matters.

Pattern signals. No petitioner has ever filed on this patent. In the 2005–2008 window the patent sat inside a well-funded operating/aggregation structure (Solidus Networks d/b/a Pay By Touch, with a Bank of New York patent security interest recorded 2006-02-16 and 2007-12-19, later released 2008-03-18 and the rights returned to Biometric Payment Solutions, LLP). Despite that prominence, no AIA trial was ever filed on the '397. There is no Unified Patents or other defensive-aggregator proceeding visible on this patent number. The absence of PTAB activity is consistent with a patent whose asserted value migrated to its later continuations (the family's children carry the same priority date but later expirations), rather than the '397 itself — those relatives, not this patent, are where a modern assertion would plausibly live. The most consequential structural fact on the '397 is the term: filed 1999-06-18 → expired 2019-06-18, and the AIA trial regime's practical reach (petitions filed after service, institution, 12-month trial clock) had little runway against a patent that has now been expired for over seven years.

Recommended next steps

• If you are a defendant and a demand letter cites US 6,728,397: the answer isn't "claims 1–5 were canceled" — they weren't, and there is no FWD to link to. The answer is the calendar. The structured record shows expiration on 2019-06-18. Under 35 U.S.C. § 286, damages reach back only six years from the complaint; with a complaint filed today (2026-09-27), the recovery window opens at 2020-09-27 — after the patent expired — leaving no period in which damages-bearing infringement of an unexpired patent could exist. Verify the "no term extension" assumption before relying on this in a brief: confirm via USPTO Patent Public Search / the '397 file wrapper whether any patent term adjustment or extension under 35 U.S.C. §§ 154(b)/156 moved the expiration date past 2019-06-18, and confirm no terminal disclaimer issue. That single check is the highest-value item here.
• If a live suit is asserted anyway: confirm whether the asserted claims are from the '397 or from a later family member (US 7,231,068 / 7,349,557 / 8,678,273 / the 10,360,623- and 10,664,834-generation patents). Defendants in the BPS-family campaigns have historically been answering on the later patents, not the '397. Do not assume a PTAB record exists for the patent you were actually served on — pull its own proceeding list.
• § 315(b) timing: if you were served with a complaint alleging infringement of the '397, the one-year window to petition for IPR runs from service; given the 2019 expiration, that window is likely long closed, and an IPR on an expired patent buys little. The cheaper path is a § 101/§ 112 or expired-term dispositive motion, not a PTAB petition.
• CBM is off the table. The transitional CBM program sunset on 2020-09-16; even if the '397 qualified as a covered business method patent, no new CBM petition can be filed in 2026.
• Where to verify, since the ODP result is a null: USPTO PTAB E2E / the PTAB Center (https://ptacts.uspto.gov/ptabweb/), the USPTO Open Data Portal PTAB endpoint (https://data.uspto.gov/), and — for any appeal — the Federal Circuit docket and CourtListener (https://www.courtlistener.com/). No specific PTAB or CAFC opinion is cited in this report because none exists for this patent number.
• Bottom line: there is no PTAB story on US 6,728,397. That null is not a "hardened patent" signal; it is the signature of a 1999-filed, 2019-expired patent whose assertion value, if any, moved to its continuations. Do not let an adversary characterize the absence of IPRs as evidence the claims are strong — nothing has ever tested them, and nothing needs to.

Verification limits stated plainly: I relied on the authoritative structured block (ODP: no AIA proceedings) plus targeted web searches, which surfaced no petition or decision naming the '397. I could not confirm the full litigation or PTAB docket for the family continuations within this session, and I found no Federal Circuit appeal of the '397. Anything beyond the quoted ODP status should be re-verified at PTAB E2E before it is relied on.

I'll research this assignment chain using the sources available. Let me start with several parallel searches.

Good context emerging. Let me dig into the litigation/assertion history and try to pin down reel/frame records.

US 6728397 — Ownership & Assignment Reconstruction

Patent: US 6,728,397 B2 — "Check verification system"
App. 09/335,649, filed 1999-06-18; priority to provisional 60/089,959 (1998-06-19); granted 2004-04-27; expired (anticipated) 2019-06-18.

Sourcing note (read first): I could not open the USPTO Assignment Center record set directly in this session, and Google Patents' legal-events panel does not expose the reel/frame, execution date, or correspondent-of-record fields for most of the links below. The chain below is reconstructed from (a) the Google Patents legal-events table for US 6,728,397, (b) the published PCT/foreign family documents which name the assignee and prosecution correspondent, and (c) contemporaneous and court-source news reporting on the Solidus Networks bankruptcy. Only one reel/frame is available to me (20762/0207, cited in the 2020 corrective assignment). Everything else is deliberately marked "not surfaced" rather than guessed. Verify each link at the USPTO Assignment Center: https://assignmentcenter.uspto.gov/ (search "6728397") and https://assignment.uspto.gov/patent/index.html.

Inventors

• The patent issued with no corporate assignee — Google Patents' "Original Assignee" field reads simply "Individual." Ms. McNeal is the only named inventor on the '397 patent and on every later member of the family (US 7,231,068; US 7,349,557; US 8,678,273; US 10,664,834; US 8,485,442), so there is no "co-inventor exodus" pattern to report.
• Unusual pattern noted: the inventor's small-town rural-Georgia address persists on the face of the family even after the portfolio is held by a Newport Beach, California entity ("Biometric Payment Solutions, Newport Beach, CA" on US 10,664,834; "Biometric Payment Solutions, LLP, Newport Beach, CA" on US 2008/0156866). Inventor-domicile ≠ assignee-domicile is a mild tell worth recording, though it is common in inventor-held portfolios and is not by itself evidence of anything.
• A related 2011 application in the family (13/284,728, "Method And System For Processing Transactions Using A Token") lists Kellie Rogers (Macclenny, FL) as a co-inventor under Exodus Payment Systems, LLC (Blackshear, GA) — a separate entity from the chain below. Worth noting given that Solidus/Pay By Touch was founded by John P. Rogers; I could not confirm any relationship between Kellie Rogers and John P. Rogers.
• No evidence of inventors departing the original assignee within 12 months of filing — the inventor kept filing family members for ~20 years.

Original assignee

None — the patent issued to the inventor individually (Joan Tibor McNeal). Google Patents lists "Original Assignee: Individual."

• Ms. McNeal's commercial vehicle at the time appears to have been the Atlanta-based check-verification business that contemporaneous trade press called "Biometric Payments Systems, Atlanta" — the same name as the seller in the January 2005 acquisition announced by Pay By Touch (see Context below). That business operated (it did not merely hold paper), which means the patent did have an operating commercialization path in its earliest years.
• Current status: the original assignee (the inventor, as an individual) no longer owns the patent. Current assignee of record per Google Patents is Biometric Payment Solutions LLC (Newport Beach, CA). The commercial operating business that later owned it — Solidus Networks, Inc. d/b/a Pay By Touch — filed Chapter 11 and dissolved (see below).

Assignment timeline

Dates below are the assignment/recordation dates as surfaced in the Google Patents legal-events table; execution dates were not retrievable and I have not invented them. Google Patents does not surface reel/frame for these entries — where I could not obtain a reel/frame I say so explicitly.

• 2005-01-21 (recorded 2005-01-21) — Reel not surfaced

  • Conveyance: Assignment
  • Assignor: McNeal, Joan Tibor
  • Assignee: Biometric Payment Solutions
  • Correspondent: not retrievable — the Assignment Center correspondent-of-record field was unavailable to me for this link.
  • Context: acquisition — the inventor (or her Atlanta check-verification vehicle) sells the patent; publicly announced by the buyer on 2005-01-28.

• 2005-02-04 (recorded 2005-02-04) — Reel not surfaced

  • Conveyance: Assignment
  • Assignor: Biometric Payment Solutions, LLC
  • Assignee: Solidus Networks, Inc.
  • Correspondent: not retrievable
  • Context: acquisition / two-step flip — the intermediate Biometric Payment Solutions entity (which had held the patent for roughly two weeks) passes it to Solidus Networks, the operating parent of Pay By Touch. The short 14-day gap is characteristic of an acquirer structuring a purchase through a conduit entity rather than a genuine intervening business.

• 2005-06-08 (recorded 2005-06-08) — Reel not surfaced

  • Conveyance: Assignment
  • Assignor: McNeal, Joan Tibor
  • Assignee: Solidus Networks, Inc., d/b/a Pay-By-Touch
  • Correspondent: not retrievable
  • Context: curative / confirmatory — a direct inventor-to-Solidus recording, likely covering the pending application (serial 335649) and/or curing the chain after the two-step January/February flips.

• 2005-08-09 (recorded 2005-08-09) — Reel not surfaced

  • Conveyance: Assignment
  • Assignor: McNeal, Joan Tibor
  • Assignee: Biometric Payment Solutions, LLP
  • Correspondent: not retrievable
  • Context: internal reorg / entity shuffle — a newly named LLP within the inventor's orbit takes title; it later becomes the entity that receives the patent back out of the Solidus bankruptcy.

• 2006-01-12 (recorded 2006-01-12) — Reel not surfaced

  • Conveyance: Assignment
  • Assignor: Biometric Payment Solutions, LLP
  • Assignee: Solidus Networks, Inc., d/b/a Pay By Touch
  • Correspondent: not retrievable
  • Context: internal reorg / consolidation — completes the return of title to Solidus, four months after the December 2005 settlement in which Solidus acquired Whorl LLC's biometric transaction-payments business including intellectual property (whistleblower BAP opinion, 9th Cir. BAP No. CC-08-1046, states Whorl's IP "was assigned to one of Solidus's subsidiaries … set up for that purpose").

• 2006-02-16 (recorded 2006-02-16) — Reel not surfaced

  • Conveyance: Security Agreement ("Grant of Patent Security Interest under the Amended and Restated Patent Security Agreement")
  • Assignor: Solidus Networks, Inc.
  • Assignee/Secured Party: The Bank of New York, as Collateral Agent
  • Correspondent: not retrievable
  • Context: securitization / debt financing — the portfolio is pledged as collateral for the senior lender facility. This is the first of two such grants and is the mechanism by which the lenders later control disposition of the asset.

• 2007-12-19 (recorded 2007-12-19) — Reel not surfaced

  • Conveyance: Security Agreement ("Grant of Patent Security Interest")
  • Assignor: Solidus Networks, Inc.
  • Assignee/Secured Party: The Bank of New York, as Agent, as Secured Party
  • Correspondent: not retrievable
  • Context: securitization — second lien grant recorded five days after Solidus filed its voluntary Chapter 11 (see below). This is the pivot point of the whole chain.

• 2008-03-18 (recorded 2008-03-18) — Reel not surfaced

  • Conveyance: Release (Release of Security Interest)
  • Assignor: The Bank of New York
  • Assignee: Biometric Payment Solutions, LLP
  • Correspondent: not retrievable
  • Context: bankruptcy fire-sale — the secured lender releases its lien in connection with the transfer of the collateral out of the bankruptcy estate.

• 2008-03-18 (recorded 2008-03-18) — Reel not surfaced

  • Conveyance: Assignment
  • Assignor: The Bank of New York (as secured party)
  • Assignee: Biometric Payment Solutions, LLP
  • Correspondent: not retrievable
  • Context: bankruptcy fire-sale / foreclosure transfer — the lender (not Solidus) conveys title, i.e. the secured party is exercising/deeding out its collateral. Same-day as the release, which is the normal signature pattern of a foreclosure-style disposition.

• 2008-04-03 (recorded 2008-04-03) — Reel not surfaced (a 2008-era reel, plausibly 20762 — see next entry)

  • Conveyance: Assignment
  • Assignor: Solidus Networks, Inc., d/b/a Pay By Touch
  • Assignee: Biometric Payment Solutions, LLP
  • Correspondent: not retrievable
  • Context: bankruptcy fire-sale — the debtor's confirming conveyance, two weeks after Pay By Touch ceased biometric payment processing (2008-03-19).

• 2020-06-10 (recorded 2020-06-10) — Reel 20762 / 0207 (the only reel/frame I could verify; it is cited inside the corrective document itself)

  • Conveyance: Assignment — Corrective Assignment ("Corrective Assignment to Correct the Receiving Party Entity Previously Recorded at Reel: 20762 Frame: 207")
  • Assignor: Solidus Networks, Inc.
  • Assignee: Biometric Payment Solutions (current assignee of record)
  • Correspondent: not retrievable
  • Context: change of name / entity correction only — the 2008 conveyance to "Biometric Payment Solutions, LLP" is corrected as to the receiving party's entity designation. This is a scrivener's correction, not a new economic transfer, which is why it is the only reel/frame I can cite with confidence: the document corrects reel 20762, frame 0207, a 2008-era recording.

Correspondent finding — the one thing you asked for that I must flag as a gap. I was unable to retrieve the correspondent-of-record for any of the eleven recordings above, and I will not guess at a name. The only attorney I can identify anywhere in this family is John J. Timar, Womble Carlyle Sandridge & Rice, PLLC, Post Office Box 7037, Atlanta, GA 30357-0037, who is the prosecution correspondent-of-record on the related PCT filings WO 2005/098741 and WO 2006/058039 (both naming Solidus Networks as applicant). Foreign filing lists Momsen, Leonardos & Cia as the Brazilian agent. Prosecution counsel ≠ assignment-recording correspondent, so this is not a repeat-correspondent finding under your Signal 3. If you can pull the reel/frame images, the correspondent field on reel 20762/0207 and on the 2005 recordings is the single highest-value next data point.

Also relevant to chain integrity: the recorded chain above is not clean. There are two parallel inventor→Solidus conveyances (2005-06-08 and, indirectly, 2005-08-09 → 2006-01-12), and a secured-party conveyance (2008-03-18) alongside a debtor conveyance (2008-04-03). This is the signature of a distressed portfolio being papered up quickly, not of a routine corporate conveyor.

Timeline diagram

timeline
    title Ownership of US 6728397
    1998 : Provisional filed by Joan McNeal
    1999 : Application 09 335 649 filed
    2004 : Patent issues to individual owner
    2005 : Assigned to Biometric Payment Solutions
         : Assigned to Solidus Networks
         : Assigned to Solidus Networks dba Pay By Touch
         : Assigned to Biometric Payment Solutions LLP
    2006 : Consolidation back to Solidus
         : Patent security interest to Bank of New York
    2007 : Second security interest to Bank of New York
         : Solidus Networks files Chapter 11
    2008 : Security interest released by the bank
         : Assigned to Biometric Payment Solutions LLP
         : Pay By Touch ceases operations
    2020 : Corrective assignment to Biometric Payment Solutions

NPE / troll-pattern signals

1. Shell-entity transfer — unclear. Title moves from the individual inventor to successively named vehicles (Biometric Payment Solutions → Biometric Payment Solutions, LLC → Solidus Networks → Biometric Payment Solutions, LLP, 2005-01-21 through 2006-01-12), and the terminal holder is Biometric Payment Solutions (Newport Beach, CA), an entity with no product I can document and which exists to hold and continue the McNeal family (it is the applicant/assignee on continuations filed as late as 2019, e.g. App. 16/409,609 → US 10,664,834). That is suggestive. But I could not verify a registered-agent-service address, a single-member LLC structure, or the absence of products in commerce, and the intermediate "Biometric Payment Solutions" entity in 2005 appears to have been the inventor's own operating Atlanta check-verification business. I decline to make this call on naming alone.

2. Known asserter in the chain — not present. No link in the chain matches Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg entity. Nor does the assignee appear on the RPX or Unified Patents high-frequency-plaintiff directories that I could reach. Caveat: I could not query the RPX/Unified directories directly in this session, so this is a negative finding from the sources I did reach, not an exhaustive clearance.

3. Repeat correspondent across the chain — unclear / cannot assess. The correspondent-of-record field was not retrievable for any of the eleven recordings (see the gap flagged above). I will not infer an attorney from the assignee names or from the prosecution counsel (John J. Timar, Womble Carlyle, on the PCT filings). This is the signal the record best supports and the one I could not complete.

4. Cascading transfers — present. Five recorded title transfers in ~12 months (2005-01-21, 2005-02-04, 2005-06-08, 2005-08-09, 2006-01-12), three of them touching entities sharing the "Biometric Payment Solutions" name and all of them terminating in the same operating parent. Note the 14-day gap between the 2005-01-21 acquisition and the 2005-02-04 flip. Two security-interest grants (2006-02-16, 2007-12-19) then two linked 2008-03/04 conveyances out of the secured party and the debtor.

5. Pre-litigation transfer — unclear. I could not confirm any infringement suit naming US 6,728,397, so I cannot date a first assertion and therefore cannot test the 6-month window. What I can document is an assertion-adjacent posture at the operating parent: BioPay LLC sued Pay By Touch in Delaware in January 2005 seeking a declaration of invalidity/non-infringement, and Pay By Touch's acquisition of this very patent (announced 2005-01-28) was explicitly framed by trade press as a competitive move against BioPay's check-cashing business. That is operating-company assertion, not NPE assertion.

6. Bankruptcy fire-sale — present. An involuntary Chapter 11 petition was filed against Solidus Networks on 2007-10-31, Solidus consented on 2007-12-14 and filed voluntary petitions for ten subsidiaries including Pay By Touch (9th Cir. BAP No. CC-08-1046). Pay By Touch ceased biometric operations 2008-03-19. The 2008-03-18 security-interest release and secured-party assignment and the 2008-04-03 debtor assignment to Biometric Payment Solutions, LLP fall squarely inside this liquidation window. Corroborating the wider sale: Acculink purchased 25 Pay By Touch patents and BioPay's Paycheck Secure business went to Phoenix Check Cashing in the same period.

7. Privateering — not present on the evidence I have. Solidus asserted its own portfolio (including against Whorl LLC, whose $67M+ judgment against Solidus became a prepetition claim) and used the portfolio in marketplace threats against a rival processor. That is an operating company asserting, not an operating company handing patents to a third-party NPE to assert on its behalf. I found no SEC filing or EFF/Patent Progress coverage establishing a privateering arrangement (Solidus was VC-backed and private, so there is no 10-K/8-K trail to check).

8. Defensive aggregator — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at a private Newport Beach holding entity.

Verdict

NPE — moderate confidence.

The chain is not a clean operating-company chain and it is not a defensive termination. It shows five title transfers in roughly twelve months (2005-01-21 through 2006-01-12), then a pledge of the portfolio to The Bank of New York (2006-02-16, re-recorded 2007-12-19), then exit through Chapter 11 — with the 2008-03-18 secured-party release/assignment and the 2008-04-03 debtor assignment delivering the patent to Biometric Payment Solutions, LLP and the 2020-06-10 corrective assignment at reel 20762/0207 leaving a non-operating Newport Beach holder as assignee of record while it continues to file family continuations (e.g. US 10,664,834, issued 2020). I stop short of high confidence for one specific reason: the strongest classic NPE tells — a match to a published asserter list (Signal 2, not present) and a documented pre-litigation transfer (Signal 5, unclear) — are absent, and the terminal entity's correspondent-of-record (Signal 3) could not be retrieved, so the decisive "same lawyer behind the shell LLCs" evidence is missing. Note also that the patent expired 2019-06-18, so it is unassertable today regardless of who holds record title.

Verify at: https://assignmentcenter.uspto.gov/ → search patent number 6728397; and the parallel index at https://assignment.uspto.gov/patent/index.html. The specific entries to pull and read for correspondent are reel 20762 / frame 0207 and the four 2005-era recordings.

I'll research this patent and its cited prior art. Let me start with targeted searches.

Let me research the key references most likely relevant to the independent claims.

Prior-Art Analysis — U.S. Patent No. 6,728,397 B2

1. The Patent Under Review (verified via search)

Source: https://patents.google.com/patent/US6728397/en

The search returned the correct record for 6728397 and no substitute; the citation lists in the patent text are the authoritative set analyzed below. Note a substantive point: this patent is in the same family as US 7,231,068 ("Electronic transaction verification system," continuations-in-part claiming the same 1998-06-19 priority), so several later references here also cite the '397 family.

Critical date for §102 purposes: any reference with an effective date before 1998-06-19 (provisional) or before 1999-06-18 (filing) is potential prior art. References issued/published after that date but filed earlier remain available under pre-AIA §102(e).

Independent claims at issue (element shorthand):

• Claim 1 – system: (a) negotiable-instrument scanning device transmitting encoded instrument data; (b) biometric device scanning/transmitting biometric data; (c) biometric database; (d) instrument information database; (e) selective transmission/comparison of both.
• Claim 10 – same in means-plus-function form.
• Claim 17 – method of obtaining account info + biometric data, transmitting and comparing to both databases.
• Claim 21 – check-processing system with a first biometric database (authorized user) and a second biometric database (invalid user), plus instrument database.

2. The Cited Prior Art, Ranked by Relevance

Tier A — Directly on point for the core "check + biometric + remote database" combination

US 6,032,137 A — Ballard (CSP Holdings, LLC) — Remote image capture with centralized processing and storage

• Filed 1997-08-27 (CIP of 08/917,761), issued 2000-02-29 (terminal disclaimer).
• Discloses the DataTreasury™ system: remote capture of check transaction data (payer routing number, account number, check amount, etc.), encryption, transmission to a central processor, "identification verification using signature data and biometric data," and generation of informative reports transmitted back to the remote location. Claim 3 expressly couples a card interface, signature interface and biometric interface; claim 10 has the central CPU verifying the biometric data and the signature data; claim 9 recites a report generator.
• Potential §102(e) mapping: This is the single closest anticipatory reference for the system concept. It is a strong §102(e) attack on claims 1, 3, 6 (reports), 10, 16/15 and the method of claim 17, because it discloses remote check capture + biometric/signature verification at a central site with reports returned to the point of capture. It does not clearly disclose the second biometric database of known invalid users of claim 21, so claim 21 would likely survive this reference alone.
• Source: https://patents.google.com/patent/[US6032137A](/patent/US6032137A) ; https://patentimages.storage.googleapis.com/1d/3e/12/1df7556fdef511/US6032137.pdf

US 5,484,988 A — Hills et al. (Resource Technology Services) — Checkwriting point of sale system

• Filed 1992-11-13, issued 1996-01-16. §102(b) art.
• Discloses a point-of-sale terminal with a MICR reader reading the magnetic check number "for the sole purpose of eliciting consumer bank account information," transmission to a central computer, and use of "positive and negative files" on consumer accounts to return an "Approval" or "Denial." This is explicitly discussed in the '397 background.
• Potential §102 mapping: anticipates the instrument-scan/transmit/compare-account-condition elements of claims 1 and 10, and specifically claims 5 and 13 ("check number written in magnetic ink"). Its positive/negative account files are not biometric databases, so it does not anticipate claims 4, 21. Best used for the check/MICR half of claims 1/10/17.
• Source: https://patents.google.com/patent/US5484988

US 5,764,789 A — Pare, Jr. et al. (Smarttouch, LLC) — Tokenless biometric ATM access system

• Filed 1994-11-28, issued 1998-06-09 — issued 10 days before the '397 priority date; §102(b)/§102(e) art.
• Discloses tokenless biometric (fingerprint) identification transmitted from a terminal to a remote database, comparison against stored biometric data, and transaction authorization.
• Potential §102 mapping: relevant to the biometric-device/database/compare elements of claims 1, 4, 10, 17. It lacks the negotiable-instrument scanning plus account-condition database combination, so it anticipates no full independent claim by itself; it is chiefly §103 (combination) art and §102 art against the biometric sub-elements.

US 5,802,199 A — Pare, Jr. et al. (Smarttouch, LLC) — Use sensitive identification system

• Filed 1994-11-28, issued 1998-09-01 (§102(e)).
• Same family as the '789 reference; biometric identity verification against stored data. Same limited mapping as above (claims 1/4/17 biometric elements); useful combined with Hills.

US 5,774,879 A — Custy et al. (First Data Corp.) — Automated financial instrument processing system

• Filed 1993-12-27, issued 1998-06-30. §102(b)/§102(e) art (issued just after the critical date but filed long before).
• Automated capture and centralized processing of financial instruments (checks) with remote verification.
• Potential §102 mapping: instrument-scanning/account-condition elements of claims 1, 10, 17; no biometric comparison, so not a full anticipator.

Tier B — Fingerprint-on-document / biometric identity verification (anticipate the "print biometric on the instrument" and "biometric compare" concepts)

US 4,048,618 — Hendry (The Raymond Lee Organization) — Method of identifying a check signer

• Filed 1976-11-24, issued 1977-09-13. §102(b) art.
• Claims: applying the person's fingerprint to an identification card in invisible ink; applying the same fingerprint to the check in invisible ink in the presence of the payee; scanning both fingerprints to produce electrical signals; comparing them; indicating match/no-match.
• Potential §102 mapping: This is the closest prior art to claim 19 ("printing the obtained biometric data on the negotiable instrument") and, more broadly, to the concept of capturing a check presenter's fingerprint as the basis for authenticating the check signer. It is a clean §102(a)/(b) reference against the instrument-borne-biometric concept, though it does not disclose the remote database comparisons of claims 1/17/21.
• Source: https://uspto.report/patent/grant/[4048618](/patent/4048618)

US 4,253,086 A — Szwarcbier — Process and apparatus for positive identification of customers

• Filed 1978-07-28, issued 1981-02-24. §102(b) art.
• Master fingerprint on a credit card, freshly captured fingerprint on a card during the transaction, superimposed/compared by optical or electronic means along shared coordinates. Discussed in the '397 background.
• Potential §102 mapping: biometric-capture-and-compare elements of claims 1, 4, 10, 17 (limited to a card-borne master print rather than a remote database).

US 4,213,038 A — Johnson (Silverman) — Access security system

• Filed 1976-12-20, issued 1980-07-15. §102(b) art.
• Stores a fingerprint on a card (actual print or "micropattern") read by a control means. Cited in the '397 background (Silverman).

US 4,582,985 A — Löfberg — Data carrier

• Filed 1981-03-18, issued 1986-04-15. Business-card/data-carrier bearing fingerprint data; cited as a "smart card" reference in the background.

US 4,995,086 A — Lilley et al. (Siemens) — Arrangement and procedure for determining the authorization of individuals by verifying their fingerprints

• Filed 1986-05-06, issued 1991-02-19. Cited smart-card/fingerprint reference.

US 4,993,068 A — Piosenka et al. (Motorola) — Unforgeable personal identification system

• Filed 1989-11-27, issued 1991-02-12. Stored fingerprint on a card interacting with a reader.

US 5,180,901 A — Hiramatsu (Toshiba) — IC card with individual authentication function

• Filed 1990-05-21, issued 1993-01-19. IC card with stored fingerprint and on-card/match authentication.

US 5,598,474 A — Johnson — Process for encrypting a fingerprint onto an I.D. card

• Filed 1994-03-29, issued 1997-01-28. Encrypted fingerprint recorded on an ID card.

US 5,745,046 A — Itsumi et al. (Toshiba) — Surface shape sensor, identification device using this sensor, and protected system using this device

• Filed 1993-01-08, issued 1998-04-28.
• ⚠️ Identifier caveat (per your rule, taken literally): the '397 background section refers to this sensor as "U.S. Pat. No. 5,745,096 to Hsumi, et al." The cited document in the '397 record is US 5,745,046 to Itsumi et al. I am not auto-correcting either numeral or name — the text and the citation list as published disagree, and both are reported verbatim above.
• Potential §102 mapping: fingerprint sensor element only (claim 4 context); not a full anticipator.

US 4,947,443 A and US 5,321,765 A — Costello — Method and apparatus for verifying identity

• Both filed 1986-05-07; issued 1990-08-07 and 1994-06-14 respectively (the '765 is a continuation of the '443). Biometric identity verification. (Note the '765 appears only in the fuller 28-reference list, not the 27-item list.)

US 5,815,252 A — Price-Francis (Canon) — Biometric identification process and system utilizing multiple parameters scans for reduction of false negatives

• Filed 1995-09-05, issued 1998-09-29. Multi-parameter biometric identification; §102(e) art against biometric-verification elements of claims 1/17.

US 5,852,670 A — Setlak et al. (Harris Corp.) — Fingerprint sensing apparatus with finger position indication

• Filed 1996-01-26, issued 1998-12-22 (§102(e)). Sensor hardware — supports claim 4 context only.

US 5,815,598 A — Hara et al. (Hamamatsu Photonics) — Apparatus for identifying an individual based on a fingerprint image

• Filed 1992-08-28, issued 1998-09-29. Fingerprint-image identification apparatus.

US 4,995,086 / US 4,993,068 / US 5,180,901 / US 5,598,474 / US 5,815,252 / US 5,852,670 / US 5,815,598 collectively: none discloses the negotiable-instrument account-condition database, so none anticipates claims 1/10/17/21 as a whole. They are §102 art only for the biometric sub-elements and §103 combination art.

Tier C — Signature verification (map to claims 2, 18, 25)

US 5,818,955 A — Smithies et al. (Penop Limited) — Document and signature verification system and method

• Filed 1994-08-31, issued 1998-10-06. Captures a handwritten signature as a dynamic/electrical signal, transmits it and compares to stored signature data.
• Potential §102 mapping: the signature-scan/encode/transmit/compare elements of dependent claims 2, 18 and 25. Not a full anticipator of any independent claim.

US 6,091,835 A — Smithies et al. (Penop Limited) — Method and system for transcribing electronic affirmations

• Filed 1994-08-31 (continuation; effective §102(e) date 1994), issued 2000-07-18. Same signature-capture family as the '955.

Tier D — Check imaging / document-processing infrastructure (anticipate instrument-scan elements)

US 5,444,794 A — Uhland, Sr. — Check image capture system

• Filed 1993-08-25, issued 1995-08-22. Captures check images. §102(b) art for the "encoded negotiable instrument information" element of claims 1/10/17.

US 5,668,897 A — Stolfo — Method and apparatus for imaging, image processing and data compression merge/purge techniques for document image databases

• Filed 1994-03-15, issued 1997-09-16. Document imaging/compression.

US 5,748,780 A — Stolfo — Method and apparatus for imaging, image processing and data compression

• Filed 1994-04-07, issued 1998-05-05. Companion to the '897.

The two Stolfo references and the '794 reference support the "digitize the instrument" element but contain no biometric or account-condition comparison; §102 art only for the scanning sub-element.

Tier E — Card/ID verification not specific to checks (weak; largely §103 art)

US 5,668,874 A — Kristol et al. (Lucent) — Identification card verification system and method — filed 1995-02-28, issued 1997-09-16.

US 5,799,092 A — Kristol et al. (Lucent) — Self-verifying identification card — filed 1995-02-28, issued 1998-08-25.

US 6,072,894 A — Payne — Biometric face recognition for applicant screening — filed 1997-10-17, issued 2000-06-06 (§102(e)). Face biometric, not fingerprint; remote screening.

Tier F — Not anticipating; peripheral/background art

US 5,623,522 A — Ito (Yamaha) — Asynchronous serial data receiving device — filed 1994-11-21, issued 1997-04-22. A data-communication circuit; cited presumably for the transmission link. Does not anticipate any claim.

3. Bottom-Line: Most Relevant Prior Art and Claim Mapping

Critical caveat on anticipation vs. obviousness. Under 35 U.S.C. §102 a single reference must disclose every limitation. On the record above, no single cited reference discloses all elements of independent claim 1, 10, 17 or 21. The cited art characteristically splits into (i) check/MICR + account-condition references (Hills, First Data, Uhland) and (ii) biometric references (Pare, Szwarcbier, Costello, the smart-card references). A §102 anticipation argument is strongest for US 6,032,137 against claims 1/10/17 (and for US 4,048,618 against claim 19); the remaining references are properly §103 combination art (e.g., Hills + Pare/Smarttouch, or Ballard + the smart-card references).

Claim 21 (first biometric database + second "known invalid user" biometric database): none of the 28 cited patents discloses a biometric negative/blacklist database. US 5,484,988 does disclose positive/negative account files, and US 6,032,137 discloses centralized biometric storage, but neither teaches a second biometric database of invalid users. Claim 21 is therefore the least vulnerable to §102 on the cited art, with the strongest §103 combination being Ballard + Hills.

4. Explicit Uncertainty / Data-Quality Notes

• Identifier discrepancy not auto-corrected: the '397 specification's "Hsumi" reference ("U.S. Pat. No. 5,745,096") and the citation-list entry US 5,745,046 (Itsumi) are reported exactly as published; I have not reconciled or normalized them.
• Two citation lists exist in the source: a 27-item "Citations" list and a 28-item "Patent Citations" list. The difference is US 5,321,765 (Costello), present only in the fuller list. Both are treated above.
• Dates are taken from Google Patents/USPTO records as surfaced in the search results; the priority date (1998-06-19) is stated by the source to be an assumption, not a legal conclusion.
• I did not find, and cannot assert, any §102 rejection actually made during prosecution of 09/335,649; the mapping above is analytical, based on the claim text and the references' disclosed subject matter. If you need the examiner's actual applied art, the 09/335,649 file history (USPTO PatentCenter) should be pulled directly.

Primary sources: https://patents.google.com/patent/US6728397/en · https://patents.google.com/patent/US6032137A · https://patents.google.com/patent/US5484988 · https://uspto.report/patent/grant/4048618 · https://patents.google.com/patent/US4253086

I'll ground the analysis in the actual cited references. Let me verify what several of the key cited patents disclose.

Obviousness Analysis — US 6,728,397 B2 ("Check verification system")

0. Ground rules, scope, and two literal-reading flags

Person of ordinary skill in the art (PHOSITA): For a June 1998 priority date, a POS/EFT systems engineer (or small team) with a B.S. in EE/CS and 2–3 years' experience in point-of-sale terminals, MICR/check-authorization networks, and commercially available biometric capture devices. This is a systems-integration art, which is significant: the Federal Circuit and the Board have consistently treated the combination of known transaction-authorization architecture with known biometric capture as within the skill level of such an artisan (there is no need to redesign any component).

Document facts I am treating as authoritative (from the page at https://patents.google.com/patent/US6728397/en):

• App. 09/335,649, filed 1999-06-18; provisional 60/089,959 filed 1998-06-19; granted 2004-04-27; "Expired – Lifetime," anticipated expiration 2019-06-18.
• Assignment chain: Individual → Biometric Payment Solutions → Solidus Networks, Inc. (dba Pay By Touch) → Biometric Payment Solutions, LLP, with later security interests and corrective assignments; current listed assignee Biometric Payment Solutions LLC.
• 25 claims; independent claims 1 (system), 10 (means-plus-function system), 17 (method), 21 (system "for use with a check processing system"). Slip/continuation family: US 7,231,068; 7,349,557; 8,678,273; and publications 2013/0002400, 2013/0002401, 2017/0186012.
• Because the patent expired 2019-06-18, this analysis is a retrospective/hypothetical validity assessment relevant to past-damages exposure, IPR/re-exam posture, and defensive use — not a live injunction question.

Two literal-reading flags (I am not auto-correcting either identifier):

1. The specification states: "An example of an actual fingerprint sensor is U.S. Pat. No. 5,745,096 to Hsumi, et al., which is for a surface-shaped sensor identification device." The page's citation list contains US 5,745,046 A (Kabushiki Kaisha Toshiba, "Surface shape sensor, identification device using this sensor, and protected system using this device") — a title match — and separately US 5,180,901 A to Hiramatsu. I read the identifier literally as 5,745,096 as printed. Either way the point is immaterial to § 103: the cited item is a sensor element reference, and the applicant concedes in the specification that "many devices for digitization and transmission of fingerprints are well known in the art, such as the devices of Digital Biometrics."
2. The headline date in my operating context is 2026-09-27; the request states April 26, 2026. Immaterial to the analysis; flagged for the record.

1. What the claims actually require (the § 103 gating issues)

Claim 1 is the linchpin. Stripped down it requires only:

Three construction points materially favor the obviousness case:

• "Selectively transmits, either together with or separately from." This claim language is expressly disjunctive. The patentee has claimed both parallel and serial transmission, and both remote and local architectures (claim 9 states remote; the specification says local is equally fine). A § 103 analysis therefore needs only one mode of transmission to be suggested, not the specific "unified data platform 40 for parallel transmission" of FIG. 3.
• Claim 1 does not require printing biometric data on the instrument. That limitation first appears in claims 7, 8, 15, 16, 19, 23. So the earliest check-plus-fingerprint art (Hendry, US 4,048,618) is not needed for claim 1 at all — it is needed only for the printing subgenus.
• Nothing in claim 1 requires a novel architec​ture. The specification concedes the enabling technology ("Many check scanning devices are known … such devices are commonly used at banks, sometimes referred to as bank 'proofing' machines"; "Other biometric devices such as retinal scanners are alternately used with, or in lieu of, the preferable fingerprint scanner").

Independent claim 10 is the same invention in means-plus-function form (the corresponding structures are the same scanners/databases; under § 112(f) it is no broader and adds no separate § 103 question). Independent claim 17 is the method counterpart. Independent claim 21 adds one element not in claim 1: a second biometric database of "known invalid users."

2. Combination sets

Combination A — Hills (US 5,484,988) + Szwarcbier (US 4,253,086) → claims 1, 3, 5, 9, 10, 12, 13, 17

US 5,484,988 (Hills et al., Resource Technology Services, priority 1992-11-13, granted 1996-01-16) — "Checkwriting point of sale system." Verified content (https://patents.google.com/patent/[US5484988A](/patent/US5484988A)):

• A point-of-sale terminal that reads "magnetic ink character recognition numbers appearing on a consumer check" (claim 2); note the reference's own emphasis that MICR is read "for the sole purpose of identifying and reading the consumer bank account information."
• A central computer system reached by the terminal over a communications link, receiving data from a plurality of POS terminals (claim 1) — i.e., the "remote database" architecture of claim 9.
• The central system "enabl[es] … communication with external databases for performing a consumer bank account status search" (claim 1), and the data center "stores positive and negative files concerning consumer accounts … signaling the system subscriber that said account may be reasonably relied upon … (i.e., an 'Approval') or, where listed as delinquent, indicating that the account may not be so relied upon (i.e., a 'Denial')."
• The terminal returns a decision (approval/denial) to the point of sale, and prints activity reports ("printed summations of daily activity … listed and separately total[ed] by authorization/service type").

US 4,253,086 (Szwarcbier, 1981) — "Process and apparatus for positive identification of customers." Verified content (https://patents.google.com/patent/US4253086):

• A fingerprint-based identification process at the moment of a purchase or service, expressly framed against the inadequacy of the then-current POS/credit-card routine: "there is no control which can immediately determine that the person presenting a credit card is its legitimate owner. The signature on the invoice form … cannot identify him with certainty."
• A master fingerprint on a card and, "selectively, … a master print on file" — i.e., comparison against a stored reference print, not merely against a presented card.
• A fresh print taken at the transaction and compared, with the comparison decision being accept/reject.

Mapping: (a)→Hills MICR reader; (b)→Szwarcbier's transaction-time fingerprint capture; (c)→Szwarcbier's "master print on file" biometric store; (d)→Hills' account-status file / external databases; (e)→Szwarcbier's print comparison; (f)→Hills' "positive and negative files" status search producing Approval/Denial. Claims 3, 5, 12, 13 (check / MICR number) are met by Hills claims 2 and 8 verbatim. Claim 9 (remote) is met by Hills' POS-terminal-to-central-computer topology.

Motivation (multiple, independent grounds):

1. Same field, same problem, same instant. Both references operate at the point of sale, in the presence of the consumer, at the moment of the transaction, and both are aimed at the same economic harm — merchant loss from a bad tender. Hills' own background acknowledges prior POS systems that "involv[e] copying and taking a picture of an individual whereby a dishonored check could then be traced" (citing U.S. 4,933,536 to Lindemann) and check-authorization systems "wherein a check is imprinted with a bar code and information concerning customers which are stored in a database" (citing U.S. 4,672,377 to Murphy). A POS-authorization system that already aggregates identification techniques is the natural host for one more.
2. Szwarcbier supplies the express problem statement that points to the combination. Its stated objective is to prevent, not solve, crime by removing the "no control … at the time of a sale" gap. A PHOSITA who reads Szwarcbier and Hills together sees two halves of one solution: Hills answers "is the account good?", Szwarcbier answers "is the person authorized?" Both questions must be answered to safely accept a check; neither reference answers both.
3. KSR principle (a): "if a technique has been used to improve one device, and a person of ordinary skill in the art would recognize that it would improve similar devices in the same way, using the technique is obvious." Fingerprint verification had already been applied to POS/access-control terminals; applying it to the POS terminal that already reads MICR requires no change in the principle of operation of either subsystem.
4. Architectural synergy / shared hardware. The combination reuses one terminal, one communications path, and one printer that Hills already discloses (its printer "for generating a transaction event sale slip" and report printer). That is a concrete, articulable cost/space incentive — not a hindsight rationale.
5. Predictable result. The two decision streams are independent and merge at a single accept/deny output. There is no apparent interoperability difficulty, no unexpected interaction; the patent itself treats the two comparisons as running "either in parallel with or separately."

Reasonable expectation of success: High. Off-the-shelf fingerprint digitizers existed and the specification so concedes ("the devices of Digital Biometrics"); MICR reading was routine bank practice.

Strength: strong for claims 1, 3, 5, 9, 10, 12, 13, 17.

Combination B — Combination A + Hendry (US 4,048,618) → claims 7, 8, 15, 16, 19, 23

US 4,048,618 (Hendry, "Method of identifying a check signer," 1977) — verified content (https://patents.google.com/patent/[US4048618A](/patent/US4048618A)):

• A fingerprint is applied to the check itself ("in invisible ink in the presence of a person requested to cash the check") and the fingerprint on an ID card is scanned to produce electrical signals, with the check print scanned likewise, and a comparator indicating comparison/non-comparison to establish signer authenticity.
• Critical caveat: Hendry's comparison is card-to-check, not check-to-database. It therefore does not anticipate or render obvious claim 1 (no biometric database, no account-status database). But it squarely teaches the printing of biometric data on the negotiable instrument for later authentication.

Mapping for claim 7/8/19/23: "printing recorded biometric data on the negotiable instrument" / "a fingerprint" / "digitizes a representation … and prints the digitized biometric data directly on the negotiable instrument" — Hendry teaches exactly this (its claim 1 step "applying the same fingerprint of the person to a check … in the presence of a person requested to cash the check"). Claims 16 and 19 add database comparison of the printed data, which is met by folding in Szwarcbier's "master print on file" and/or Motorola '068 (below).

Motivation: Hendry supplies the express purpose that claims 7/15/19/23 recite — the print is placed on the instrument so it can be checked later (Hendry: to give "the person requested to cash the check" certainty; the '397 specification: "such that the biometric data can be later checked against a database … at the time the check is processed at a bank"). Merely upgrading Hendry's later-comparison target from a card to a database record is a substitution of one known data source for another with a predictable change in degree, not in kind — and Szwarcbier and Motorola '068 both expressly contemplate a stored print "on file" as an alternative or supplement to the card. Adding a printing step (output 26 in FIG. 1) to the Combination A system is also a design choice with a stated security benefit (traceability if the check is returned).

Strength: moderate-to-strong for claims 7, 8, 15, 19, 23; moderate for claim 16 (requires the extra "transmit the printed imprint to the database" step, though that is a data-capture-and-forward step of the type the art already performs).

Combination C — Combination A + Szwarcbier/Motorola (US 4,993,068) + Löfberg (US 4,582,985) → claims 4, 6 (and the "biometric data" species)

US 4,993,068 (Piosenka/Chandos, Motorola, "Unforgeable personal identification system," 1991) — verified content (https://patents.google.com/patent/[US4993068A](/patent/US4993068A)):

• Identification credentials for facial photo, retinal scan, voice and fingerprints stored on a portable memory device, compared at a remote access control site against a physical trait input device.
• Crucially, it expressly discloses the remote, centralized-database variant in its background discussion: "distributed access control points are linked via a communication medium to a centralized data base … a remote access control point transmits the data … to the central repository … The central repository matches the data obtained from the remote access control point with the prestored data retrieved from the data base … [and] communicates back to the access control point to allow the requested access."
• It also expressly contemplates use at a point-of-sale terminal, with the decision affecting whether to allow the transaction, and a printer at the validation site, "applicable to validation sites involving financial transactions such as check cashing," printing onto "the transaction receipt (example—the tendered check itself)."

That last passage is worth pausing on: Motorola '068 does not merely supply the generic "biometric → central database → allow/deny" architecture; it names check cashing as the application and the tendered check itself as the printing substrate. It is arguably the single most damaging reference to claims 16/19/23's premise (though not to claim 1's combination, since '068 lacks an account-condition database and a MICR scanner).

US 4,582,985 (Löfberg, "Data carrier") and US 5,180,901 (Hiramatsu, Toshiba) and US 5,598,474 (Johnson, "Process for encrypting a fingerprint onto an I.D. card") supply the "store an enrolled print and compare at the point of use" species, with Johnson adding digitization/encryption of the print onto a carried medium.

Motivation: These references are alternative implementations of element (c) — the biometric store — in the art, and a PHOSITA selecting among a card-resident store, a "print on file," or a networked central database would do so based on cost, revocation, and coverage trade-offs, all of which are ordinary design choices. Löfberg/Motorola's card-resident approach is a teaching-away candidate (see §4) but is undercut by Motorola's own background description of the centralized-database alternative and by the '397 specification's indifference between local and remote databases.

Strength: strong for the fingerprint species (claim 4/8) and for architecture breadth; note that Löfberg/'068 do not supply claim 6 at all.

Combination D — Combination A + Hills' report teaching → claims 6, 14

Claim 6 ("selectively returns a report on customer usages") and claim 14 ("report means for transmitting a report detailing customer usage of system") are the weakest claims in the patent from a § 103 standpoint. Hills discloses report generation expressly: printed summations of daily activity, reports "identify[ing] all 'Approved' Transaction Events by service type, approval number, and amount," totals of approved/captured/voided events, and storage of "transaction event information and consumer banking account information" at the central system. The '397 report databases (FIG. 5: number of checks cashed for an account 98, locations 100, customer list 102) are aggregations of exactly the transaction-event records Hills already stores, formatted for retrieval on demand.

Motivation: Generating user-selected reports from a stored transaction log is ubiquitous data processing (In re Lintner-type routine programming; KSR's "common sense" and "market demand" rationales). The specification itself offers only a business justification — "highly informative as to customer trends such as drops in frequencies of purchases … effectively targeted by marketing efforts" — which is a demand-side motivation a PHOSITA in retail POS systems would have recognized immediately.

Strength: strong (these claims are in serious § 103 jeopardy, and arguably vulnerable on a "printed publication/known prior art" basis only if a more specific reporting reference is located; on the references cited here the motivation is common sense rather than explicit).

Combination E — Hills' positive/negative files + Szwarcbier + Motorola '068 → claims 21, 22, 23, 24, 25

This is the claim set requiring two biometric databases (authorized users; known invalid users).

(i) Negative-file logic is already in a cited reference. Hills' central data center "stores positive and negative files concerning consumer accounts," and its claim 7 recites "a database comprising information regarding consumers whose consumer banking account status is not verified as bad." The positive-vs-negative file architecture is thus squarely in the cited art — for accounts.

(ii) Szwarcbier supplies an explicit motivation to extend it to persons. Szwarcbier's deterrence argument is unusually direct: "A criminal would hardly dare make an illegal purchase when he must leave his fingerprint which may already be on file with the police department fingerprint-identification records." That sentence is, functionally, a motivation to maintain a biometric file of known wrongdoers and to check against it. A PHOSITA reading that in combination with Hills' negative file would have had every reason to build the second biometric database of claim 21 — the deterrent and detection value of a "known-bad" biometric watch list is the stated rationale, not a hindsight construct.

(iii) Claim 22 ("magnetic ink character recognition data that is printed on the negotiable instrument") is met verbatim by Hills' claims 2 and 8.

(iv) Claim 24 (transmit to the second database "to determine if the individual … is an invalid user") is the direct application of the negative-file test to the biometric modality — the same comparison logic, a different record set.

Claim 25 (second scanning device + signature database) is addressed in Combination F.

Strength: strong for 21, 22, 24; moderate-to-strong for 23; moderate for 25 depending on the signature reference relied upon.

Combination F — Combination A/E + Penop (US 5,818,955; US 6,091,835) and/or Canon (US 5,815,252) → claims 2, 11, 18, 25

Claims 2, 11, 18 and 25 add signature capture + a signature database + comparison.

• US 5,818,955 and US 6,091,835 (Penop Limited, "Document and signature verification system and method" / "Method and system for transcribing electronic affirmations") appear on the page's citation list (https://patents.google.com/patent/US6728397/en) and, per the specification, digitally-encoded-signature devices were a mature, off-the-shelf category: "There are many devices known in the art which allow a signature to be digitized for storage and comparison, and any of such devices will work satisfactorily in the present inventive system."
• US 5,815,252 (Canon, "Biometric identification process and system utilizing multiple parameters scans for reduction of false negatives") teaches combining multiple biometric modalities and expressly aims at reducing false negatives — a textbook motivation for stacking signature verification alongside fingerprint verification.
• Szwarcbier's fingerprint card contains "an area 5 for a fingerprint and an area 6 for a signature," i.e., the same reference already pairs a print and a signature captured in one transaction.

Caveat on sourcing (per my operating rules): I was able to verify Hendry, Hills, Szwarcbier, Motorola '068 and Smarttouch '789 through live search. My search budget terminated before I could independently verify the disclosures of the two Penop references; I am relying on the reference titles as printed on the page and on the specification's own admission that signature digitization was well known. The Combination F conclusion should be re-verified against the Penop full texts before being relied on in a filing. If Penop is weaker than its title suggests, US 5,815,252 plus the specification's admission is sufficient for the "signature capture + compare" element standing alone.

Motivation: The signature is the traditional authentication for a check (the specification: comparison to determine "if the signature presented is that of the account owner," decision 68 in FIG. 4). Adding a signature-verification path to a system already built around check acceptance is the most natural of augmentations; it adds redundancy against sensor failure or spoofing, and the "either together with or separately from" permissive claim language means the references need not agree on sequencing.

Strength: strong for 2, 11, 18; moderate-to-strong for 25.

Combination G — biometric sensor/implementation references → dependent claims 4, 8 and the "alternate biometric" breadth

US 5,745,046 (Toshiba, surface-shape sensor and identification device), US 5,745,096 (as literally cited — see flag §0), US 5,855,670 (Harris, "Fingerprint sensing apparatus with finger position indication"), US 5,815,598 (Hamamatsu, fingerprint image identification), US 5,745,046 and US 5,180,901 are all cited on the page and are element-level references: each improves or implements the sensor. Substituting one known fingerprint sensor for another in Combination A is the paradigm of an obvious substitution of equivalents with no change in the claimed function — and the applicant conceded the point in the specification ("Other biometric devices such as retinal scanners are alternately used with, or in lieu of, the preferable fingerprint scanner"). Claim 4 ("biometric data is fingerprint data") is therefore obvious on Combination A alone once the sensor references are considered.

3. Method claims (17–20) and means-plus-function claims (10–16)

Claims 17–20: Hills claim 8 is itself a process claim ("A checkwriting point of sale process comprising … reading the magnetic ink character recognition number information … storing … transmitting … to a central computer system"); Szwarcbier discloses a process for POS fingerprint identification; Hendry claim 1 is a method claim. Method claims 17, 19 and 20 therefore rise and fall with the system claims under the same combinations, and claim 19 ("printing the obtained biometric data on the negotiable instrument") is met by Hendry/'068. Claim 20 ("transmitting data indicative of whether the person is authorized … to the location") is met by Hills' return of the Approval/Denial message to the POS terminal (and, in Smarttouch '789, by the account-access grant/deny message — see §5).

Claims 10–16: Under § 112(f), the "means" are the scanners/databases disclosed in the specification, and the analysis mirrors claims 1–9. There is no additional narrowing structure recited (no algorithm is claimed for the comparison step beyond "comparison," which the art performs with "conventional pattern matching comparison logic" — Motorola '068).

4. Anticipated counter-arguments and how the record answers them

5. Reference that cuts closest to § 102/§ 103 for the broadest claims

US 5,764,789 (Smarttouch/Pare, "Tokenless biometric ATM access system," priority 1994-11-28, granted 1998-06-09) — verified (https://ie.espacenet.com/publicationDetails/biblio?CC=US&NR=[5764789A](/patent/5764789A)&KC=A): PIN + "at least one biometric sample" captured at an ATM/PC "or any other financial access device," transmitted to a computer system, "compared with the registration biometric samples to produce either a successful or failed identification," after which "a financial account number of the customer is retrieved." That is effectively element (e) plus the transmit-compare-returnaccount-access loop, with the biometric directly binding to the financial account rather than to a card.

It is not an anticipatory reference for claim 1, because it is expressly tokenless and contains no negotiable instrument (no scanner of a check, no MICR data, no account-"condition" file serving a paper instrument). It is, however, powerful § 103 evidence for the proposition that "capture biometric → transmit → compare to enrolled record → authorize financial account" was a known, working pipeline by 1998 — which is why the remaining delta in claim 1 (MICR scanning + account-condition database) collapses so readily into Hills.

6. Bottom line

Single strongest three-reference combination: Hills (US 5,484,988) + Szwarcbier (US 4,253,086) + Hendry (US 4,048,618), optionally with Motorola '068 (US 4,993,068) as the centralized-database/'068-prints-on-the-check bridge. This trio covers claims 1–9 and 17–20 (with Signature and Report riders supplying 2/11/18 and 6/14), and it does so with motivations written into the references rather than imported by hindsight — which matters because the June 1998 priority date sits in the pre-KSR era, when the Federal Circuit's TSM formulation nominally governed and "common sense" rationales were more contested. Notably, the references here supply explicit motivations (Szwarcbier's "no control at the time of sale" and "fingerprint … may already be on file"; Hills' positive/negative files and report printing; Motorola's check-cashing/POS embodiment), so the combination survives even under a conservative TSM-style analysis as well as under KSR.

Weakest spots in this analysis (stated candidly):

1. I could not independently verify the Penop disclosures within my search budget; Combination F's claim 2/11/18/25 conclusion rests partly on the reference titles and the specification's concession.
2. The cited-art list does not appear to contain a reference that anticipates claim 1 outright; the case is a combination case, and combination cases turn on the articulation of motivation. The motivation here is unusually good, but it is still the point of attack.
3. Claim 21's "second biometric database for … an invalid user" is the element for which the cited art provides the weakest explicit teaching — Hills gives negative files for accounts, Szwarcbier gives the deterrence rationale for a fingerprint file, but the coupling of the two into a biometric watch list is inferential. If any claim in this patent withstands a § 103 challenge on the cited references, it is most likely 21/24 — though only barely, given Szwarcbier's sentence about the criminal who "must leave his fingerprint which may already be on file."
4. This is a technical analysis, not legal advice, and it assumes claim construction favorable to the challenger on "selectively transmits, either together with or separately from" and on the absence of a printing limitation from claim 1. A narrower construction (e.g., reading claim 1 to require a single unified transmission platform) would materially strengthen the patent.

Sources: US6728397B2 page (https://patents.google.com/patent/US6728397/en) and the prior-art citations listed thereunder; US4048618A (https://patents.google.com/patent/US4048618A); US5484988A (https://patents.google.com/patent/US5484988A, incl. the #1 and #4 text views); US4253086A (https://patents.google.com/patent/US4253086); US4993068A (https://patents.google.com/patent/US4993068A); US5764789A (https://ie.espacenet.com/publicationDetails/biblio?CC=US&NR=5764789A&KC=A). Where the full online text of a cited reference was not retrievable, I have said so rather than reconstructing it.