US 7788633

I'll search for the specific patent number and any related litigation/docket activity.

Let me run more targeted searches for litigation and docket activity.

Let me run additional targeted searches to check for any litigation or CAFC activity specifically involving this number.

US Patent 7,788,633 — Analyst Summary

Scope note: I was able to ground the bibliographic and claim information in the full patent text (patents.google.com/patent/US7788633/en, fetched 2026‑09‑24). For the litigation portion of your request I must flag a limitation: my searches did not surface any CAFC 2026 docket entry tied to patent 7,788,633 specifically. Details and caveats are given below the claim summary.

---

1. Bibliographic data (from the authoritative patent record)

Note on the number: 7,788,633 is confirmed literally in the record as US 7,788,633 B2 / US7788633B2. This is a different patent from the similarly numbered US 4,788,338, US 8,078,336/337 (Rawls‑Meehan bed patents) or US 7,788,638 that appeared incidentally in my searches — none of those are the subject here.

---

2. Abstract (verbatim)

"A bank note processing machine includes a plurality of sensors, a transport system, an input/output device, a control device and an interface. The control device has a memory configured to control elements of the bank note processing machine by means of software and/or data stored in the memory. The interface is arranged to couple memory systems of different kinds to the bank note processing machine in order to alter, supplement or replace software and/or data stored in the memory."

---

3. Claims — plain-language overview

The patent has 5 claims: one independent claim (claim 1) and four dependent claims (2–5).

Independent Claim 1 — the bank note processing machine

A bank note processing machine that has:

• Sensors, a transport system that includes a singling unit and at least one stacking unit, and an input/output device;
• a control device with an associated memory that controls the machine's elements using software and/or data held in that memory;
• an interface that lets memory systems of different kinds be coupled to the machine, so the stored software/data can be altered, supplemented, or replaced;
• the coupled memory system itself contains a drive plus a storage medium suitable for optical and/or magnetic recording (i.e., the memory system is a drive‑plus‑removable‑media unit, not just a socketed chip);
• the transport system transports singled bank notes along the sensors to the stacking unit(s) based on the control device's evaluation of the sensor data; and
• data obtained inside the machine during operation are stored back onto the memory system.

Gist: the core inventive point is the open, multi‑type interface combined with an optical/magnetic drive‑based memory system that both supplies software/data to the machine and receives operating data back from it.

Dependent Claim 2

The interface is a standardized interface, in particular a PCMCIA interface. Effect: limits claim 1 to recognized standard bus interfaces (PCMCIA named as the preferred example).

Dependent Claim 3

The machine's memory has a non‑volatile area, and after the memory system is coupled, the software/data from the memory system is written into the non‑volatile area. Effect: covers the "load‑and‑keep" model (the machine retains the update after the memory system is removed).

Dependent Claim 4

The machine's memory has a volatile area, and after coupling, the software/data is loaded into the volatile area. Effect: covers the "run‑from‑external‑media" model (the update exists only for the session while the memory system is connected).

Dependent Claim 5

The software/data stored on the memory system is in encoded form, and the controller is configured to decode it. Effect: adds an encryption/encoding requirement for the transferred software and data.

---

4. Specification highlights (context for the claims)

• Problem addressed: Known bank note machines were inflexible because they required one particular storage medium (e.g., EPROM/EEPROM chips, or a flash card as in US 5,909,502). The invention's stated aim is to let any suitable memory system be used, chosen by criteria such as memory volume, size, robustness, and price.
• Supporting embodiments: interface may be contacting or contactless (e.g., infrared); PCMCIA called out as especially advantageous. Memory system may be remote and linked via wire (LAN, Internet) or wireless (GSM). Explicit example: the Iomega "Click Drive" optical/magnetic drive‑plus‑media system (corroborated by the cited Chroust 1999 review).
• Operating modes described: a test/service mode (test software and parameters loaded from the memory system; sensor data stored back for off‑site evaluation, allowing remote servicing without on‑site personnel); a user‑specific mode (sorting/denomination/currency defaults stored on the memory system); and use of the memory system purely for accounting data and per‑customer "deposits."
• Partial update: the spec notes that only some software/data elements (e.g., transport‑control code or the characteristic data of selected bank notes) need be replaced, rather than the whole non‑volatile memory.

---

5. Litigation / CAFC 2026 docket — what I found and what I could not confirm

What I found: A CAFC 2026 appeal involving an entity with "Giesecke+devrient" in its name exists — No. 26‑1855, Lumenco LLC v. Giesecke+devrient GmbH, filed 05/21/26, an appeal from PTAB Originating Case IPR2024‑00836. However, I could not connect this case to US 7,788,633, and on the available evidence it appears unrelated (it is a PTAB appeal concerning a different, more recent patent family involving security/diffraction‑type technology). I am not asserting any link between 26‑1855 and the '633 patent.

What I could not confirm (explicit uncertainty):

• I did not identify any CAFC 2026 appeal, argument calendar entry, or docket naming US 7,788,633 (or US 09/917,947) as the patent in suit.
• I did not identify any pending U.S. district court litigation or USPTO post‑grant proceeding (IPR/PGR) specifically targeting the '633 patent.
• I was unable to complete a direct USPTO PatentCenter / PACER query within this session, so the absence of found litigation is not proof that none exists. Treat the litigation findings as inconclusive rather than as a negative confirmation.

Consistent with no active enforcement: the patent is recorded as Expired – Fee Related, having lapsed for non‑payment of maintenance fees effective 2018‑08‑31 (the last recorded fee payment was the 4‑year payment on 2014‑02‑24; the 12‑year payment was not made). An expired patent cannot be asserted for ongoing infringement, which is consistent with the absence of recent enforcement activity.

---

6. Bottom line

• US 7,788,633 is a Giesecke & Devrient (now Giesecke+Devrient Currency Technology GmbH) patent, filed 2001‑07‑31, priority 2000‑07‑31 (German), granted 2010‑08‑31, naming Kersten and Buntscheck.
• It claims a bank note processing machine with a multi‑type memory‑system interface (PCMCIA preferred) coupled to a drive‑plus‑optical/magnetic‑media memory system that both supplies software/data to and receives operational data from the machine; dependents add the PCMCIA limitation, non‑volatile loading, volatile loading, and encoded‑software/decoding.
• The patent is legally expired for fee non‑payment (2018), with an adjusted‑expiration flag of 2028‑02‑13.
• I found no authoritative CAFC 2026 docket entry for this patent; the only 2026 Giesecke‑related CAFC appeal I located (26‑1855) appears unrelated, and my litigation search should be regarded as incomplete.

I'll search litigation-tracking sources for this specific patent number.

Let me run more targeted litigation searches on the specific number.

Let me check Justia's patent page and search for any IPR or infringement suits naming this number.

Litigation Search — US Patent 7,788,633

Result: No litigation identified

Searching litigation-tracking sources for the specific number 7,788,633 (and the application number 09/917,947), I found no known litigation — no U.S. district court infringement suits, no declaratory judgment actions, no ITC §337 investigations, and no PTAB post‑grant proceedings (IPR/PGR/CBM) naming this patent. I also found no CAFC appeal.

Required disclosure of search limitations (as an attorney's diligence note, this matters):

• Every exact‑match hit for the string "7788633" that was not the patent itself was a coincidental match in an unrelated dataset — a Brazilian commercial registry minute (JUCIS‑RS), a Missouri unclaimed‑property list, and a patent‑office document referring to different patents. None of these are litigation involving this patent.
• My searches surfaced CAFC and district court activity involving other patents and other parties whose numbers or names resemble this one (e.g., unrelated Giesecke+Devrient matters, unrelated "7,78…" and "'788" patents). Per your instruction, I am not returning those as results for 7,788,633. They are not the same patent.
• I could not complete a live, direct query of the Unified Patents litigation portal, PACER, or PatentCenter within this session. An absence of found results is therefore not proof that no suit was ever filed — treat this as "none identified," not "none exists."

---

Corroborating context (why "no litigation" is plausible here)

Although this is contextual rather than docket evidence, it is consistent with a zero‑litigation finding:

A patent that lapsed for fee non‑payment in 2018 cannot support an ongoing infringement claim, and this patent shows no record of having been asserted while in force. Note the "adjusted expiration" flag of 2028‑02‑13 in the bibliographic record — that is an administrative term calculation, not a sign of current enforceability, given the 2018 lapse.

---

Bottom line

• Litigation involving US 7,788,633: none identified. No plaintiff, defendant, jurisdiction, case number, filing date, or outcome can be reported, because no case was found.
• This is a negative finding subject to the search limitations stated above (no direct PACER/PatentCenter portal query was completed).
• If you need a definitive negative clearance, the next step is a paid Docket Navigator / Lex Machina / Unified Portal query on patent number 7,788,633, plus a PACER docket search across all districts — I recommend wording the conclusion as "no litigation located" rather than "no litigation exists" until that is run.

I'll first verify the structured "no PTAB activity" default against live sources before writing the analysis.

Let me run additional targeted checks for any proceeding, reexamination, or litigation involving this patent.

Proceedings overview

There are zero AIA trial proceedings on file for US 7,788,633. The canonical structured source — the USPTO Open Data Portal "PTAB proceedings on file" block supplied for this analysis — returns no IPR, no PGR, and no CBM for this patent, and the targeted searches I was able to run did not surface any petition, institution decision, or termination order naming US 7,788,633 or application 09/917,947.

Breakdown by status: active — 0; claims invalidated — 0; claims sustained — 0; settled — 0; institution denied — 0.

Bottom-line defensive posture: there is no IPR win to inherit and no IPR loss to fear — because nobody ever filed. Claims 1–5 stand exactly as they issued on 2010-08-31, un-narrowed by any Board proceeding. The "hardened vs. dead" framing does not apply here; the correct framing is "never tested at the Board," which cuts both ways: no claim is canceled, but equally, no petitioner is estopped, so a defendant today faces no § 315(e)(2) ground-suppression and is free to run any § 102/§ 103 theory it can assemble. The decisive practical fact for a defendant is not the PTAB docket at all — it is that the patent lapsed for non-payment of maintenance fees effective 2018-08-31 (see cross-reference to the earlier section).

---

Proceeding-by-proceeding analysis

None to enumerate. Because the count is zero, I am deliberately not generating placeholder entries — the constraint against inventing proceeding numbers is absolute, and a synthesized "IPR2024-XXXXX" would be fabrication.

For completeness, here is what the absence does not mean, and the adjacent records I did find so you can rule them out with confidence:

Adjacent records checked and ruled out

• CAFC No. 26-1855, Lumenco LLC v. Giesecke+devrient GmbH (filed 2026-05-21, appeal from PTAB Originating Case IPR2024-00836) — this is an appeal from an IPR on a different, more recent G+D/Giesecke family (micro-optic security-feature technology, involving micro-mirrors and ambient-light security features, as reflected in the petition excerpts returned in my searches). It is not an appeal involving US 7,788,633 and not an appeal from any proceeding on the '633 patent. I am not asserting any link. This confirms the earlier section's flag.
• US 7,783,788 / IPR2022-00211 (Hewlett Packard Enterprise v. Intellectual Ventures II) and US 7,788,338-type numbers — numerically near-miss references that surfaced repeatedly in search results. Different patents, different owners, different technology. Not the subject patent.
• Ex parte reexamination control numbers (e.g., 90/012,783; 90/013,183; 90/012,957; 90/019,676) that appeared in search results all belong to unrelated patents (VoltStar '833, Cao Group '770, Rochette '058). I found no reexamination request — ex parte or inter partes — directed at US 7,788,633. Note that pre-AIA inter partes reexamination would have been the only non-AIA third-party option during the 2010–2013 window, and none is on file either.

I must be explicit about the limits of this: my searches were not a substitute for a direct PTAB E2E / PatentCenter / PACER query, and I was unable to complete one within this session. Treat the zero-count as the structured-source default, corroborated but not independently exhaustively verified. What raises my confidence considerably is the combination of (a) the ODP block returning nothing and (b) the patent having expired for fee non-payment in 2018 — an economically dead patent with four years of remaining nominal term attracts no IPR filers, which is exactly the pattern observed.

---

Strategic summary

Claim status. There is no § 318(b) certificate and no adverse judgment, so no claim of US 7,788,633 has been canceled and none has been "sustained" in the PTAB sense. All five claims remain as issued: independent claim 1 (multi-kind memory-system interface; drive-plus-optical/magnetic storage medium; transport of singled notes along sensors to stacking unit(s) on the basis of sensor-data evaluation; operational data written back to the memory system) and dependents 2 (standardized/PCMCIA interface), 3 (load into non-volatile area), 4 (load into volatile area), 5 (encoded software/data with controller-side decoding). Anyone characterizing claim 1 as "dead" or as "survived IPR" would be misstating the record. The only narrowing of this patent is whatever the examiner applied during 2001–2010 prosecution.

Estoppel landscape. This is the single most defendant-favorable consequence of the empty docket: because no IPR reached a final written decision, § 315(e)(1) and § 315(e)(2) estoppel is not triggered against anyone. There are no barred petitioners, no barred privies, and no grounds that were "raised or reasonably could have been raised" and are now off the table. Every prior-art ground — including the 18 references already of record (e.g., US 5,909,502 to Cummins-Allison, US 6,248,069, US 6,078,012, EP 0 987 627, and the non-patent citations: Chroust's Apr. 1999 Iomega Click! Drive review, Fokes in Design & Elektronik (1989-04-04), and Rankl/Effing Handbuch der Chipkarten, 2nd ed. 1996) — remains fully available in district court, subject only to § 282 and the usual IPR-estoppel-free posture. Note the mirror-image risk: the Click! Drive NPL citation is on the face of the patent, so a § 325(d)-style "same art already considered" argument could be raised by a patent owner if you tried to build an IPR on it — but that concern only bites if you file at the Board, which in practice you would not need to.

Pattern signals. There is no petitioner with multiple IPRs on this patent (there is no petitioner at all), no PTAB appeal activity by the patent owner on this patent, and no defensive aggregator in the chain — I found no evidence of Unified Patents or any similar entity ever challenging the '633 patent. There is also no sign of active assertion: consistent with the 2018 lapse, I found no district court complaint or ITC action naming US 7,788,633. G+D's related later filings in this space (notably DE 10360860 A1, "Method of loading and operating bank note processing machine with software, authenticates software source prior to permanent storage in memory," priority 2003-12-23) show the patent owner continued to develop the software-loading concept but on a separate filing, not via continuation of the '633 family.

---

Recommended next steps

If you are a defendant and a demand letter or complaint cites US 7,788,633:

1. Lead with the lapse, not the PTAB. The patent is recorded Expired – Fee Related; maintenance fees were last paid on 2014-02-24 (the 4-year payment) and the patent expired for failure to pay the 12-year maintenance fee, effective 2018-08-31 (lapse event docketed 2018-10-08; final lapse 2018-10-30). A lapsed patent cannot support an ongoing-infringement claim for conduct after 2018-08-31. Ask opposing counsel, in writing, for the specific pre-2018-09-01 acts and the six-year § 286 damages window they intend to pursue. (Caveat: the record also carries an "adjusted expiration" flag of 2028-02-13; that is a term-adjustment entry, and it does not revive a patent lapsed for fee non-payment. Do not let the flag be misused against you — but do preserve the point that the legal expiration date and the flag are different things.)

2. There is no FWD to cite, so do not pretend there is. The honest and still-strong statement is: "US 7,788,633 has never been the subject of any IPR, PGR, CBM, or reexamination proceeding; no claim has been canceled; and no petitioner is estopped." The defensive leverage comes from step 1, not from a Board disposition.

3. If (and only if) pre-2018 damages are genuinely in play, the invalidity runway is wide open. With zero estoppel, you may raise any § 102/§ 103 ground in district court, and you may file an IPR — but weigh three timing realities: (a) § 315(b) bars an IPR filed more than one year after service of an infringement complaint on you or a privy/RPI; (b) PGR is unavailable — it applies only to patents with an effective filing date on or after 2013-03-16, and this patent's priority is 2000-07-31, filed 2001-07-31; and (c) CBM review is foreclosed — the transitional program sunset on 2018-09-16, so no CBM petition can be filed now regardless of how the "payment circuits / cheque" G06Q 20/04 classification might otherwise have supported covered-business-method standing. Realistically, for a fee-lapsed patent the correct next step is usually a § 282 invalidity and/or laches/statute-of-limitations posture in litigation, not a Board filing.

4. Verify the zero myself before relying on it. Because I could not complete a direct database query, pull the PTAB E2E docket and PatentCenter "Transactions" tab for 09/917,947 and confirm zero AIA trials and zero reexaminations. Points of contact for the record: the patent's own prosecution/publication record at https://patents.google.com/patent/US7788633/en, USPTO PTAB E2E at https://ptacts.uspto.gov, and CourtListener for any CAFC activity. If your E2E pull surfaces a proceeding I did not find, treat the live USPTO record as controlling over this analysis — the operating rules for this task prefer live results over prior data.

Uncertainty I am flagging rather than papering over: the absence of PTAB activity is asserted on the strength of the structured ODP block plus corroborating searches, not a completed first-party database query. The absence of litigation is likewise inconclusive for the same reason. Neither gap is a negative confirmation — it is an incomplete one.

I'll verify the assignment chain against the USPTO Assignment Center records and look for the correspondent of record on each recording.

Assignment & Ownership Chain — US 7,788,633 B2

Source caveat, stated up front: I could not complete a live query of the USPTO Assignment Center (https://assignmentcenter.uspto.gov/ / the mirror at https://assignment.uspto.gov/patent/index.html) within this session. The reel/frame data below is taken from the Google Patents legal-events record, which mirrors the USPTO assignment data (reel/frame strings and conveyance types). Google Patents does not expose the correspondent-of-record field for these recordings, so I have not supplied correspondent names — I will not fabricate them. That gap directly limits signal #3 below, and I flag it as an explicit hole rather than guessing a firm.

What the record shows is a two-entry chain, both entries sourced from the same corporate family.

---

Inventors

• Basis for employer: both inventors are assignors on Reel 012298/0389 to Giesecke & Devrient GmbH, and the German priority application DE 10037177.9 / DE 10037177 A1 (2000‑07‑31) was a G&D filing. The assignment was executed ~13 months after the priority date and ~5 weeks after the US filing (2001‑07‑31) — i.e., a routine post-filing employment assignment, not a divestiture.
• Unusual-pattern check — no flag. There is no record evidence of either inventor departing G&D within 12 months of filing. The only recorded documents naming them as assignors are the original employment assignments, and no individual-to-third-party conveyance exists in the chain.

---

Original assignee

• Entity on the issued patent: Giesecke & Devrient GmbH, Munich, Germany. (Google Patents lists original assignee as "Giesecke+Devrient GmbH" using the post‑2017 stylization; the issued-patent assignee of record from the 2001 deed is Giesecke & Devrient GmbH — I am reading the entity literally as it existed at recording, not auto-correcting it.)
• Primary line of business: security technology — banknote printing paper and features, banknote processing machines, cash‑center systems, smart cards / mobile security, identity systems.
• Did they ship a product embodying the claims? Yes — with high confidence on the product line, moderate confidence on claim-level correspondence. G&D ships the BPS family of banknote processing systems (e.g. BPS 200, BPS C1, BPS M7) and the Numeron line, which are listed as G+D machines in the ECB banknote-authentication-device registers (manufacturer address Prinzregentenstraße 159/161, 81677 Munich). A G+D BPS M7 installation is described in the G+D Annual Report 2017 (cash center of a central bank). These are exactly the class of machines described in the '633 specification (singling unit → sensors → stacking units, controller, I/O device). I did not find a public teardown confirming that a specific shipping BPS unit implements claim 1's drive-plus-optical/magnetic-media memory system, so the "embodies the claims" statement rests on product-line overlap plus the specification, not on a verified BOM.
• Current status: operating, restructured — not acquired, not dissolved, not in bankruptcy. Per G+D's April 2017 press release and the 2017 Annual Report, G&D moved to a holding structure in 2017, splitting into four legally independent subgroups: Giesecke+Devrient Currency Technology GmbH (formerly the Banknote unit), Giesecke+Devrient Mobile Security GmbH, Veridos, and secunet. The German trade register (North Data, HRB 148256, Munich) shows the Currency Technology entity acquired the "Currency Technology" business unit and the G&D Shenzhen participation from Giesecke & Devrient GmbH (HRB 4619) by spin‑off agreement dated 2017‑06‑07, registered July 2017. A 2017‑07‑18 register filing confirms the old control/profit-transfer agreement migrated to the new entity. Employee-representation coverage (ver.di, April 2017) describes ~900 employees in Munich plus the Leipzig security printing plant in Currency Technology — i.e., a real operating business, not a paper entity. Current ownership per a June 2021 Maryland Lottery & Gaming Control Commission filing: Giesecke & Devrient Currency Technology GmbH ← wholly owned by Giesecke & Devrient GmbH ← wholly owned by MC Familiengesellschaft mbH (MCF).

---

Assignment timeline

Two recorded entries. Both are intra-family; no third party ever held this patent.

1. 2001‑09‑06 to 2001‑09‑10 (executed) / recorded 2001‑11‑05 — Reel 012298/0389

• Conveyance: Assignment
• Assignor: Guido Kersten; Hans Wilhelm Buntscheck (individuals)
• Assignee: Giesecke & Devrient GmbH, Germany
• Correspondent: Not obtainable from the sources available to me (Google Patents legal events expose the reel/frame string only; Assignment Center was not queryable in-session). No recurrence flag can be made.
• Context: Initial inventor-to-employer assignment of the priority/US filing — standard employment assignment, not a fire‑sale, securitization, or asserter transfer.

2. Effective/executed 2017‑11‑08 / recorded 2017‑11‑27 — Reel 044809/0880

• Conveyance: Assignment (recorded as an Assignment, not as a Merger or Change of Name — worth noting, because substantively this is a German Ausgliederung spin‑off, not an arm's-length sale)
• Assignor: Giesecke & Devrient GmbH, Germany
• Assignee: Giesecke+Devrient Currency Technology GmbH, Germany
• Correspondent: Not obtainable from the sources available to me. No recurrence flag can be made.
• Context: Internal corporate reorganization — statutory spin‑off of the Currency Technology business unit (spin‑off agreement 2017‑06‑07; entity HRB 148256, Munich) into a newly created, wholly owned subgroup. Zero change in ultimate beneficial ownership (MC Familiengesellschaft mbH throughout).

Is there any post‑2018 entry? No. The final legal events are fee-related, not ownership-related: 2018‑04‑16 maintenance-fee reminder, 2018‑10‑08 "Lapse for failure to pay maintenance fees," 2018‑10‑30 FP lapse effective 2018‑08‑31. Per the record, the last fee payment was the 4‑year payment on 2014‑02‑24; the 12‑year payment was never made. Ownership never left the G+D family.

Note: I found no separate "Release," "License," "Security Agreement," or "Correction" recording for this patent. If any exist, they were not surfaced by the sources I could reach — treat the two-entry list as complete-as-mirrored, not as a PACER/Assignment-Center-verified negative.

---

Timeline diagram

timeline
    title Ownership of US 7788633
    2000 : DE 10037177 priority filed
    2001 : Inventors assign to Giesecke Devrient
         : Recorded reel 012298 frame 0389
    2010 : US 7788633 B2 issued
    2017 : Spin-off to G+D Currency Technology
         : Recorded reel 044809 frame 0880
    2018 : Lapsed for unpaid maintenance fees

---

NPE / troll-pattern signals

---

Verdict

Insufficient data — with a taxonomy caveat I want to be explicit about.

Justification: the ownership chain consists of exactly two recordings — the original inventor assignment (Reel 012298/0389, recorded 2001‑11‑05) and a purely intra-family spin-off (Reel 044809/0880, effective 2017‑11‑08, recorded 2017‑11‑27) — after which the patent lapsed for non-payment of maintenance fees effective 2018‑08‑31 and was never asserted. That functional profile ("original assignment plus a same-family successor transfer, no third-party owner, no assertion") is what the Insufficient data category is defined to cover, so I am selecting it rather than manufacturing an NPE verdict.

Honest flag on the taxonomy: none of the four substantive verdicts cleanly fits. There is no evidence whatsoever of an NPE pattern (all eight signals are not present or unassessable), so "NPE — high/moderate" would be unsupported; "Operating‑company assertion" is unsupported because I found no suit by the current assignee against competitors; and "Defensive / non-asserting" is unsupported because the chain does not terminate at a defensive aggregator. If the taxonomy were extended, the accurate characterization would be "operating-company-owned, never asserted."

Verification link: search patent number 7788633 at https://assignment.uspto.gov/patent/index.html (or https://assignmentcenter.uspto.gov/) — this is the specific query that would close the one remaining gap, namely the correspondent of record on Reels 012298/0389 and 044809/0880.

Cross-reference vs. earlier sections: consistent with the prior summary's finding that the 2017 Giesecke+Devrient CAFC appeal (No. 26‑1855, Lumenco LLC v. Giesecke+devrient GmbH) is unrelated to the '633 patent; nothing in the ownership record creates a link to it. One minor item in the earlier summary that this section corrects: the earlier table compressed the prior-art string as "US 5,907,?/5,909,502" — the clean reference is US 5,909,502 (Cummins‑Allison, Software loading system for a currency scanner), which is the closest art and is worth noting here because it is the very flash-card-update system the '633 specification distinguishes itself from.

I'll research the key cited references to ground the prior-art analysis.

Let me research the remaining cited references.

Prior-Art Analysis — US Patent 7,788,633

Scope confirmation and method

I searched for the exact identifier 7788633. The record retrieved is US 7,788,633 B2, "Bank note processing machine and method for operating bank note processing machine," Giesecke+Devrient Currency Technology GmbH, filed 2001‑07‑31, priority 2000‑07‑31 (DE 10037177), granted 2010‑08‑31. No similarly-numbered patent was substituted; I did not auto-correct the number.

Anticipation caveat up front. 35 U.S.C. § 102 anticipation requires that a single reference disclose every element of the claim. On the face of the record below, none of the 18 cited references fully anticipates independent claim 1, because claim 1 requires the conjunction of (a) a multi‑kind memory-system interface, (b) a memory system comprising a drive plus an optical/magnetic storage medium, (c) sensor‑guided transport of singled notes to a stacking unit, and (d) operating data written back to the memory system. The citations are best characterized as element‑level art that the examiner used to build the § 103 obviousness case, with US 5,909,502 as the closest single reference. I flag anticipation only where a reference is genuinely single‑reference dispositive, and otherwise say so.

---

A. The 18 patent citations on the face of US 7,788,633

Tier 1 — Closest art (software/exchange loading into currency machines)

1. US 5,909,502 A — "Software loading system for a currency scanner"

• Citation: Mazur, R. A.; Cummins‑Allison Corp.
• Dates: filed (US 08/715,029) 1996‑09‑17; granted 1999‑06‑01.
• Description: A flash card pre‑programmed with updated software is removably coupled to a currency discrimination machine. In a first embodiment the resident code is erased and replaced by the flash‑card code (card may then be removed and reused); in a second embodiment the machine runs from the flash card and reverts to resident code on removal. This is the reference the '633 specification itself names and distinguishes.
• § 102 relevance: This is the strongest single reference and the natural § 102 candidate for the "alter, supplement or replace software/data stored in the memory" function of claim 1, and directly to claims 3 and 4 (non‑volatile overwrite vs. volatile/run‑from‑removable‑media execution). It does not disclose the claimed drive‑plus‑optical/magnetic medium, the sensor/transport/stacking architecture, or writing machine operating data back to the memory system, so it does not anticipate claim 1 as a whole.

2. US 5,940,623 A — "Software loading system for a coin wrapper"

• Dates: filed 1997‑08‑01; granted 1999‑08‑17; Cummins‑Allison Corp.
• Description: Same software‑loading architecture as the '502 patent applied to a coin‑wrapper peripheral.
• § 102 relevance: Cumulative to US 5,909,502; relevant to the software‑loading element of claim 1 and to the volatile/non‑volatile loading models of claims 3–4. Different machine class (coins, not bank notes) weakens it as anticipatory art.

3. US 6,039,645 A — "Software loading system for a coin sorter"

• Dates: filed 1997‑06‑24; granted 2000‑03‑21; Cummins‑Allison Corp.
• Description: Same family pattern as the '623/'502 references, applied to coin sorting.
• § 102 relevance: Same as above — element‑level art against the software‑update limitations of claims 1, 3, 4.

4. US 5,759,102 A — "Peripheral device download method and apparatus"

• Dates: filed 1996‑02‑12; granted 1998‑06‑02; International Game Technology (Pease et al.).
• Description: Reprogr ams peripheral memories in gaming terminals by identification, negotiation, block‑wise downloading with error checking, and verification from an external information source; optionally over a reusable communication link (including optical link / portable computer).
• § 102 relevance: Element‑level art for the "interface allowing memory systems of different kinds to be coupled to alter/replace software" element of claim 1, and for remote/contactless coupling. Not anticipatory — it is not a bank note processing machine.

Tier 2 — Interface / memory-controller art (relevant to the multi-kind interface and claim 2)

5. US 6,148,347 A — "Mode selectable memory controller for PCMCIA standard memory cards and non-standard memory cards"

• Dates: filed 1996‑11‑20; granted 2000‑11‑14; Cisco Technology, Inc. (Finch, Collins, Rowett, Youngman).
• Description: A memory controller with a type register that accepts both PCMCIA‑standard cards and "non‑standard" cards in the same socket, generating control and acknowledge signals at the timings appropriate to each card type. The specification expressly describes PCMCIA cards used as a "removable code store."
• § 102 relevance: The most pointed art against the "memory systems of different kinds" element of claim 1 and against claim 2's standardized/PCMCIA interface. It is not a value‑document machine, so it cannot anticipate a claim that also recites sensors, transport, singling/stacking, and back‑writing of operating data.

6. US 2001/0043432 A1 — "Flexible recording system, flexible disk drive and recording disk"

• Dates: priority 1997‑01‑31 (JP); published 2001‑11‑22; Kazunobu Hamatsu.
• Description: A drive‑plus‑removable‑recording‑disk flexible recording system with a flexible disk drive and recording disk.
• § 102 relevance: This is the citation most directly on point for claim 1's "memory system has a drive and a storage medium suitable for optical and/or magnetic recording" element. Date flag: its publication (2001‑11‑22) postdates the '633 filing (2001‑07‑31), so it can only be prior art via its earlier effective date (e.g., § 102(e) on the underlying US filing / its 1997 JP priority if perfected). I would not treat it as clean § 102(a)/(b) art without confirming its US filing date in PatentCenter.

Tier 3 — Currency/value-document handling art (machine architecture, sensor evaluation, data recording)

7. US 5,467,405 A — "Method and apparatus for currency discrimination and counting"

• Dates: priority 1990‑02‑05; granted 1995‑11‑14; Cummins‑Allison (Raterman et al.).
• Description: Foundational currency discrimination/counting method and apparatus using optical sensing with comparison to stored denomination parameters.
• § 102 relevance: Element‑level art for the sensor‑and‑evaluation/transport architecture of claim 1 and for the "characteristic features stored as data" concept. Not anticipatory for the interface/removable‑media limitations.

8. US 6,241,069 B1 — "Intelligent currency handling system"

• Dates: filed 1999‑02‑05 (continuation‑in‑part chain to 1990‑02‑05); granted 2001‑06‑05; Cummins‑Allison (Mazur, Csulits, Graves).
• Description: Document handling system with an input device that obtains information from authentic documents, a processor that generates "master information" and a memory that stores it — i.e., the machine learns/records processing information.
• § 102 relevance: This is the cited reference closest to claim 1's final limitation, "data obtained in the bank note processing machine during operation are stored" — but note it stores master/reference information internally, whereas claim 1 requires the operating data to be stored in the memory system. Element‑level art; not anticipatory.

9. US 2006/0210137 A1 — "Method and apparatus for currency discrimination"

• Dates: priority 1990‑02‑05; published 2006‑09‑21; Raterman.
• Description: Later pre‑grant publication in the same Cummins‑Allison currency‑discrimination family as US 5,467,405 / US 6,241,069.
• § 102 relevance: Cumulative; relevant to the sensor/transport/evaluation architecture of claim 1. Date flag: published well after the '633 filing, so its prior‑art effect depends on the earlier family filing date; treat as cumulative rather than independent art.

10. US 6,621,919 B2 — "Customizable international note counter"

• Dates: priority 1998‑03‑17; granted 2003‑09‑16; Cummins‑Allison Corp.
• Description: A note counter that can be customized (e.g., to operator/currency/denomination preferences) — i.e., user‑specific operating configuration of a value‑document counter.
• § 102 relevance: Element‑level art for the user‑specific mode described in the '633 specification (settings/parameters loaded to configure sorting, denominations, currencies) underlying claim 1; and generally for the "software and/or data alterable to configure the machine" concept.

11. US 2001/0006556 A1 — "Currency evaluation and recording system"

• Dates: priority 1997‑10‑31; published 2001‑07‑05; Graves, B. T.
• Description: Currency evaluation combined with a recording function — capturing data about evaluated currency.
• § 102 relevance: Element‑level art for claim 1's "data obtained during operation are stored" limitation (though again, storage location is not the removable memory system), and for accounting/deposit‑style record keeping described in the '633 specification.

12. US 5,680,721 A — "Currency operated modular display device"

• Dates: priority 1992‑05‑15; granted 1997‑10‑28; HS Industries, Inc.
• Description: A modular, currency‑operated display device with removable/modular components.
• § 102 relevance: Weak; peripheral art for the modularity/input‑output aspects of the claimed machine (claim 1, input/output device). Not anticipatory.

Tier 4 — General software-update / memory-change art (relevant mainly to claims 3–5)

13. US 6,078,012 A — "Method and apparatus for upgrading software subsystems without interrupting service"

• Dates: filed 1998‑05‑22; granted 2000‑05‑30; Nortel Networks Corp.
• Description: Live/background upgrade of software subsystems without taking the system out of service.
• § 102 relevance: Element‑level art for the "alter/replace software" concept of claim 1 and for the specification's stated advantage that normal operation continues with only a short interruption during servicing. Not machine‑specific.

14. EP 0 987 627 A1 — "Method and system for performing an update"

• Dates: priority 1998‑09‑15; published 2000‑03‑22; Eastman Kodak Company.
• Description: Method and system for performing a software/data update.
• § 102 relevance: General update‑methodology art; element‑level against the "alter, supplement or replace" limitation of claim 1. Not machine‑specific.

15. DE 195 34 528 A1 — "Method for changing the data of an electronic franking machine loaded in memory cells" (family: EP 0 762 335 B1; US 5,734,571 A)

• Dates: DE published 1997‑03‑13 (priority 1995‑09‑08; EP filed 1996‑08‑30; parallel US 5,734,571 granted 1998‑03‑31); Francotyp‑Postalia (Pilz, Zarges).
• Description: An electronic franking machine changes data loaded in its memory cells by plugging in an external memory module carrying a program to be interpreted; a check routine in a secure internal region validates the loaded program (MAC/authentication), and only then are machine data modified. This is also the '633 examiner's model for "encoded software/data + decoder" (Rankl/Effing chip‑card citation).
• § 102 relevance: The most relevant cited art for claim 5 — encoded/authenticated software transferred from an external storage medium and decoded/validated by the machine controller — and element‑level art for the external‑medium/different‑memory‑device coupling of claim 1. It is a franking machine, not a bank note machine.

16. DE 43 12 894 A1 — "Method for inputting data into a franking machine"

• Dates: published 1994‑10‑20; Francotyp‑Postalia GmbH.
• Description: Data input into a franking machine (input of data that configures the machine, as opposed to replacing an EPROM).
• § 102 relevance: Background art for the "data alterable/inputtable at the machine" concept underlying claim 1; cumulative to the Francotyp‑Postalia line above.

Tier 5 — Peripheral/supporting art

17. US 6,766,056 B1 — "Image pattern detection method and apparatus"

• Dates: priority 1998‑11‑30; granted 2004‑07‑20; Canon Kabushiki Kaisha.
• Description: Image‑pattern detection method/apparatus.
• § 102 relevance: Remote/image‑processing art; at most background for the sensor‑data evaluation of claim 1. Not anticipatory.

18. EP 0 854 456 A1 — "Self service financial terminal"

• Dates: priority 1996‑12‑24; published 1998‑07‑22; NCR International Inc.
• Description: Self‑service financial terminal architecture (value‑document/currency handling terminal with control and I/O).
• § 102 relevance: Background art for the machine context of claim 1 (financial value‑document terminal with input/output device and control device); not anticipatory.

---

B. Non‑patent literature cited (all under § 102(b) as printed publications)

---

C. Element‑by‑element mapping of the closest art to the claims

---

D. Bottom line

1. No single cited reference anticipates claim 1 as a whole. The novel conjunction — an open multi‑kind memory‑system interface plus a drive‑based optical/magnetic medium plus the bank‑note sensor/transport/singling‑stacking architecture plus writing operating data back to that medium — is not found in any one of the 18 citations. The examiner's citations read most heavily as a § 103 combination.
2. US 5,909,502 (Cummins‑Allison, 1999‑06‑01) is the closest single reference and is the only citation the '633 applicants themselves distinguished by name in the specification. It maps to the software alter/replace function and, in its two embodiments, to claims 3 and 4 specifically.
3. Claim 2 (PCMCIA) is most directly met by US 6,148,347 (Cisco, 2000‑11‑14), which couples PCMCIA and non‑standard cards through one socket.
4. Claim 5 (encoded software + decoder) is most directly met by the Francotyp‑Postalia line — DE 195 34 528 A1 / EP 0 762 335 B1 / US 5,734,571 — which loads an authenticated/encoded program from an external module and validates it in a secure internal region before modifying machine data.
5. The drive‑plus‑optical/magnetic‑medium element of claim 1 rests on the Chroust (1999) Iomega Click! Drive review (NPL) and secondarily on US 2001/0043432 A1 (Hamatsu), for which the publication date (2001‑11‑22) postdates the '633 filing, so its prior‑art status requires verification of its effective US filing date.
6. The "operating data stored back into the memory system" limitation is the least well‑covered by the cited art — the closest, US 6,241,069 and US 2001/0006556, store data internally or record evaluation results generally rather than writing them onto the coupled drive/medium. That limitation is the strongest basis for the claim's validity over this particular citation set.

Verification caveat: the reference details above are taken from the cited‑reference table on the '633 record and from the corresponding US/EP/DE documents surfaced in my searches. For any assertion intended for a validity opinion, each reference's actual US filing/priority and publication dates should be re‑confirmed in USPTO PatentCenter, particularly the two date‑flagged items (US 2001/0043432 A1 and US 2006/0210137 A1).

I'll ground the key prior-art references before constructing the § 103 analysis.

Obviousness Analysis — US 7,788,633 B2 under 35 U.S.C. § 103

Scope and caveats. This builds on the claim summary and litigation sections already produced for this patent; I do not repeat them. The analysis uses the 18 patent citations and 3 non-patent citations in the "Prior Art" section of the Google Patents record, together with the specification's own admissions. Where I could not obtain the full text of a reference, I say so and treat its teaching as limited to what its title/abstract/family record support. This is a technical obviousness assessment, not a legal opinion.

Two threshold observations, flagged because they frame everything below:

• The patent admits the closest art. The specification expressly discusses US 5,909,502 (Cummins-Allison) as known, and distinguishes it only on the ground that it requires a particular storage medium (a flash card). The stated "problem" is therefore not a technical impossibility — it is the generalization of a known removable-media software-loading scheme to a wider class of media.
• The claimed "invention" is largely an aggregation of standard, commercially available building blocks — a PCMCIA socket, a removable-media drive, and a value-document handling section that were each independently well known. That posture is highly favorable to a § 103 challenge under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007).

---

1. Legal framework applied

Under Graham v. John Deere Co., 383 U.S. 1 (1966), I assess (a) scope/content of the prior art, (b) differences between art and claims, (c) PHOSITA level, and (d) secondary considerations. Under KSR, a claimed combination is obvious where the elements were known, the combination does no more than yield predictable results, and there is an articulated reason (design incentive, market demand, "obvious to try," or substitution of a known equivalent) to make the combination. Under In re Keller, the test is what the combined teachings suggest, not bodily incorporation.

Note also the effective filing date: 2001-07-31 (priority 2000-07-31). Every reference dated on/before that date is prior art; US 2001/0043432 A1 (published 2001-11-22) is only usable as § 102(e) art via its earlier filing/priority date (1997-01-31), and US 6,766,056 B1 (filed 1998-11-30) likewise.

---

2. Level of ordinary skill in the art (PHOSITA)

A POSITA here would be a controls/software engineer or electromechanical systems engineer with ~2–4 years' experience in program-controlled document/value-document handling equipment (currency scanners, sorters, postage meters), familiar with:

• sensor-based bank-note discrimination and transport/stacking mechanisms;
• embedded control (microprocessor + non-volatile/volatile memory, software & stored characteristic data);
• field-updating of embedded software/data via removable media; and
• standard PC-card interfaces, including PCMCIA, then a mature, ubiquitous standard (see Cisco '347's discussion of the PCMCIA/PC Card Standard).

That skill level is important: coupling any compliant removable-memory peripheral to a PCMCIA bus was, by 2000, a routine integration task, not an inventive one.

---

3. The prior-art references and what each supplies

---

4. Claim 1 — element-by-element and the combination

Claim 1 is a machine claim; every element reads on the art, and the two elements the applicant would point to as novel (the multi-kind interface and the drive+optical/magnetic media) are each supplied by a single secondary reference that is squarely combinable.

The two-step combination that renders claim 1 obvious:

• Combination A (primary): US 5,909,502 alone or with Cummins '405/'069/'919 and Graves '556. This yields the entire bank-note-machine + software/data-loading architecture, both loading behaviors, and the store-operating-data function. '502 alone gets the machine, the sensors/transport, the control-with-memory, the removable-memory-system interface, and the replace-vs-run modes.
• Combination B (secondary): Cisco '347 added to the '502-based machine to convert '502's single-medium (flash-card) interface into an interface that accepts memory systems of different kinds, and to specify the PCMCIA standard.
• Combination C: Chroust/Iomega Click! Drive (or Hamatsu '432) substituted for the flash card as the removable memory system, providing the drive + optical/magnetic medium.

Because '502's flash card and the Click! Drive are both removable stores of machine code/data that couple to the machine's bus, substituting one for the other is the paradigm substitution of a known equivalent yielding predictable results — a KSR-recognized rationale. The Click! Drive being PCMCIA-native means it drops directly into the Cisco '347 interface, so the three references are mechanically and functionally compatible (In re Keller).

Motivation to combine (articulable reasons)

1. Same field, same problem. '502, the Cummins family, Graves, IGT '102, Nortel '012, Kodak EP '627, and the Francotyp DE references all address updating/extending the software or data of a program-controlled machine from an external medium. All are reasonably pertinent to the problem the '633 patent states (keeping machine software/data current).
2. The patent's own stated problem supplies the motivation. The specification criticizes '502 because a specific medium is required and states the object as selecting "the most suitable" memory system by criteria such as memory volume, size, robustness, price. That is an explicit design incentive to render the interface medium-agnostic — exactly what Cisco '347 teaches.
3. Recognized interchangeability / "obvious to try." PCMCIA was a published, ubiquitous standard (Cisco '347 summarizes it); a POSITA seeking medium flexibility would predictably adopt the standard socket that a whole market of drives/cards already used, and would reasonably expect success.
4. Market/cost and availability pressures. '502 itself is motivated by cost/service burden of EPROM handling; Chroust's review presents the Click! Drive as an inexpensive, high-density, reusable medium. Combining cheap reusable drive+media with a standard socket is a cost-driven, predictable improvement, not an unexpected result.
5. Feature complementarity, no change in principle of operation. The multi-kind interface and drive+media only improve '502's loading scheme; they do not alter its principle of operation (load software from removable store; optionally retain or run-from-media; optionally record operating data).

---

5. Dependent claims 2–5

Claim 5 note: even if a PHOSITA did not treat Francotyp as anticipatory of "encoded/decoded" per se, combining it with '502/Cisco supplies the element, and the motivation is manifest — externally supplied code/data for a secure value-handling machine must be authenticated/decoded before use.

---

6. Strongest combinations and residual considerations

Strongest primary combination for claim 1:
US 5,909,502 + US 6,148,347 + Chroust/Iomega Click! Drive, optionally with US 2001/0006556 (Graves) for the "store operating data in the memory system" element and Cummins US 5,467,405 / US 6,241,069 for the sensor/transport/stacking architecture.

• '502 → bank-note machine + removable-media software loading (both modes).
• Cisco '347 → interface coupling different kinds of memory systems + PCMCIA.
• Click! Drive (Chroust) / Hamatsu → drive + optical/magnetic storage medium.
• Graves / Cummins → machine details + operating-data recording.

This combination maps all elements of claim 1 and the substance of claims 2–4; adding Francotyp/MAC-security and Rankl & Effing covers claim 5.

Anticipated applicant rebuttals:

• "'502 is limited to flash cards." Correct, and that is precisely why the patent is vulnerable: the only difference is the type of removable memory system, and '502's own teaching plus Cisco '347 (multi-kind) and the Click! Drive (drive+media) close that gap with a predictable substitution.
• "No motivation to combine." The specification itself states the motivation (choose the medium by volume/robustness/price); Cisco '347 supplies the structural means; the references share the same problem.
• "Unexpected results / secondary considerations." No nexus evidence is apparent in the record; the patent lapsed in 2018 for fee non-payment, and I identified no litigation or licensing suggesting commercial recognition attributable to the claimed subject matter. Absent such evidence, secondary considerations do not rebut the Graham/KSR showing.

Residual limitations on this analysis (stated honestly):

• I obtained full or near-full text for US 5,909,502, US 5,940,623, US 6,148,347, US 5,734,571/DE 195 34 528/EP 0 762 335, and US 2001/0006556; for several others (e.g., EP 0 854 456, US 5,759,102, US 6,078,012, EP 0 987 627) I relied on title/abstract/family data, so their mapping should be confirmed against full text before being relied on as primary references.
• The non-patent Chroust Click! Drive review (Apr. 1999) is central to the drive+media element; its status as a printed publication should be verified (it is dated before the priority date and is offered as prior art by the examiner).
• This is not a claim-construction opinion; I assumed the plain meaning of claim terms. If "memory systems of different kinds" were construed narrowly (e.g., requiring simultaneous heterogeneous media), that construction would need to be tested against Cisco '347's dual-mode controller.

---

7. Conclusion

Under § 103, claim 1 of US 7,788,633 would likely have been obvious over US 5,909,502 in view of US 6,148,347 and the Iomega Click! Drive (Chroust, 1999) — with US 2001/0006556 (Graves) and the Cummins sensor/transport patents supplying the machine and data-recording elements — because each claim element was independently known, the references are in the same field and address the same software/data-updating problem, and the patent's own stated objective (medium flexibility by volume/size/robustness/price) supplies the explicit motivation to substitute a standard, multi-kind PCMCIA interface and a drive-plus-magnetic-media memory system for '502's single flash-card medium, with predictable results and no unexpected effect. Claims 2–4 fall with claim 1 and are independently met by '502's two modes and Cisco '347; claim 5 is met by the Francotyp MMAC/encoding art (DE 195 34 528 / US 5,734,571 / EP 0 762 335) and Rankl & Effing's chip-card encoding teaching.