US 5640002

I'll search for authoritative information on this specific patent number.

Let me verify the CAFC docket point and retrieve the abstract/claims text.

Summary — US Patent 5,640,002

Bibliographic data (from the fetched full text and search results)

Sources: https://patents.google.com/patent/US5640002/en ; https://FreePatentsOnline.com/5640002.html ; https://ie.espacenet.com/publicationDetails/citedDocuments?CC=US&NR=[6905072B2](/patent/6905072B2) ; https://patentimages.storage.googleapis.com/52/0f/23/c3870de25c3af3/EP1255217A3.pdf

Abstract — ⚠️ uncertainty flag

I was not able to retrieve the verbatim abstract text in the search results returned to me; the pages I hit returned the "Definitions"/description sections instead. I therefore decline to reproduce an abstract as if it were authoritative. What the record does show is that the disclosure is a handheld "Personal Information Device" / "Personal Scanner™" that combines CCD barcode scanning with RF ID tag interrogation, a touchscreen, IR and spread-spectrum RF communication, and a docking/charging base unit (specification structure, FIGS. 16–26 and 43–44). Treat any abstract quoted elsewhere as needing independent verification against the printed patent.

Overview of the independent claims (plain language)

The claim set I captured includes independent claims 1 and 6, plus dependent claims 2–5. Note: my captured copy of claims 1 and 6 is partially elided in the source fragments, so the preamble wording below is reconstructed from the visible elements rather than quoted verbatim.

• Claim 1 — handheld scanner + RFID reader with paid-item deactivation. A portable information gathering apparatus with a hand-carried housing containing a barcode scan engine, a display, a memory, a touchscreen input device superimposed over the display, and a microcomputer controlling all of them (storing decoded barcode characters, driving the display, and taking control input from the touchscreen). Added to that is an RF ID tag reader whose transmitter both powers and commands a passive RFID tag (an IC with no power supply of its own), causing the tag to transmit product information back. The microcomputer interrogates tags and stores the returned product information, and — notably — also drives the reader to write "paid for" data into the tags of purchased items so those tags will not trip store-exit RFID security alarms.

• (Dependent 2–5) Add, respectively: a thermal printer under microcomputer control; an infrared transceiver for sending/receiving modulated IR data; a rechargeable battery with contacts plus a detachable base unit with a power converter (AC→DC) to charge the battery and power the device; and a base unit having a LAN interface to a host computer plus an IR transceiver in line-of-sight with the portable unit's transceiver, so data downloads to the host over the base unit.

• Claim 6 — independent, directed to the docked/charging architecture. A portable information gathering apparatus that likewise includes the hand-carried housing, display, memory, touchscreen, and microcomputer, and that additionally requires an infrared transceiver (send and receive modulated IR) and a rechargeable battery with electrical contacts cooperating with a detachable mating base unit whose power converter supplies DC charging/power — and whose base unit carries a LAN interface circuit to a host computer with an IR transceiver positioned for line-of-sight communication, with the microcomputer programmed to download information to the host upon a download command.

Claims beyond 6 may exist; I did not retrieve the full claim set, so the total claim count is uncertain. Notably, the claim language differs in emphasis from the specification, which is largely directed to consumer self-scanning shopping (price-list download, running totals, coupon matching, fuzzy-logic shopping-list matching, budget tracking, credit-card payment, and even real-estate listing scanning).

CAFC 2026 docket check — negative result

I searched for a 2026 Federal Circuit appeal involving this patent and found none. Queries combining "5640002" with Federal Circuit/2026 appeals returned unrelated matters (e.g., PTAB discretionary-denial lists, other parties' briefs, and a GEM Products, LLC v. Rupp Marine, Inc. appeal, Fed. Cir. No. 2026-1817, which involves the unrelated "Rupp Marine," not this patent). I also found no district-court infringement assertion of US 5,640,002. This is consistent with the patent having expired 2015-08-15 for failure to pay maintenance fees and being held by an individual assignee.

Confidence: High on the bibliographic data (corroborated across Google Patents, FreePatentsOnline, Espacenet and the EPO annex). Medium on the claim-1/claim-6 preambles (source fragments were elided). Low/unknown on the abstract, which I could not verify verbatim. No CAFC 2026 docket for this patent exists in the results I retrieved — I cannot positively rule out a docket that is not indexed by public search, so treat that as a negative-search finding rather than certainty.

I'll search for litigation involving US patent 5640002.

Let me search litigation databases directly.

Let me check litigation-specific databases.

Bottom line

I found no evidence of any litigation involving U.S. Patent No. 5,640,002. Based on the searches I was able to run, there is no known infringement suit, declaratory judgment action, ITC investigation, or PTAB proceeding (IPR/PGR/CBM) asserting or challenging this patent. I want to be explicit that this is a "no results found" answer rather than a confirmed negative — I could not directly query PACER, Docket Navigator, or the Unified Patents case list (the latter returned only a patent-record page, not a case list), so I cannot rule out an unindexed or very old case.

What I searched and what came back

What the searches did surface (useful context, but not litigation)

• The patent itself: U.S. 5,640,002, "Portable RF ID tag and barcode reader," Ruppert et al., filed Aug. 15, 1995, issued June 17, 1997, Appl. No. 08/515,257, original assignee SCAN-EM CORP, later assigned to Jonathan P. Ruppert (2008).
• Status per Google Patents: Expired – Fee Related, anticipated expiration Aug. 15, 2015. A patent that lapsed for failure to pay maintenance fees and expired in 2015 is an unlikely candidate for recent assertion.
• Third-party citation only: The patent appears repeatedly as an [A] category reference in EPO/PCT search reports and as a cited prior-art reference in other patents (e.g., EP 0 919 971 A3, WO 99/045493 A3, WO 2004/059598 A1). These are prior-art citations, not litigation.

Important caveats about my search

1. No live PACER/Docket Navigator/CourtListener API access was used for a definitive docket query, so an old suit filed and terminated before broad docket digitization (the patent's relevant window is roughly 1997–2015) could exist without appearing in general search results.
2. The absence of hits is consistent with the patent's history: a shopper self-scanning patent from a small entity (SCAN-EM Corp. / individual inventors), expired for fee non-payment in 2015, with no large commercial licensor or NPE asserting it.
3. All identifiers were treated literally: I looked only for 5,640,002 / US5640002, and did not substitute near-neighbors such as US 5,640,020, US 5,640,024, or the unrelated JP5640002B2 or the various "564" patents (e.g., U.S. 10,298,564) that appeared in the search results.

Recommended verification steps (to convert this into a definitive answer)

• PACER Case Locator / CourtListener RECAP: search party names Ruppert, Fish, Yap, Ames, Scan-Em Corp., and patent number "5640002."
• USPTO Patent Assignment / litigation docket and USPTO PTAB (IPR/PGR/CBM) search for Patent No. 5,640,002.
• Docket Navigator or Unified Patents Portal → Litigation → Case List, searching the exact patent number.

If you'd like, I can draft the exact query strings for each of those databases, or — if you have access — you can run them and I can help interpret the results.

If no litigation exists (my working conclusion), the correct answer to your question is: no litigation is known involving U.S. Patent No. 5,640,002, and therefore no plaintiff, defendant, jurisdiction, case number, filing date, or outcome can be provided.

I'll start by verifying whether any AIA trial proceedings exist for this patent, since the structured ODP block reports none.

The first pass surfaced mostly noise (note: hits referring to "the '564 patent" in Philips ITC matters are U.S. 10,298,564 — an unrelated patent). Let me run targeted checks on this specific patent and its owner.

Let me check for district court enforcement activity, defensive-aggregator filings, and any USPTO reexamination history, since those are the places an AIA trial would surface.

Proceedings overview

Zero AIA trial proceedings on file. The USPTO Open Data Portal's structured "PTAB proceedings on file" block returns no IPR, PGR, or CBM petition against US 5,640,002, and my independent web checks surfaced no petition, institution decision, final written decision, or Federal Circuit appeal naming this patent. The breakdown is therefore 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denials — the defensive posture is not "hardened patent" but moot patent: US 5,640,002 issued 1997-06-17 from an application filed 1995-08-15, reached its 20-year term on 2015-08-15 (Google Patents lists status "Expired – Fee Related," anticipated expiration 2015-08-15), so there is nothing left to invalidate and no live damages window for a demand letter to leverage.

---

No proceedings to report individually

There is no {PROCEEDING_NUMBER} to populate. Rather than fabricate one, here is what I checked and what I found:

• Structured ODP data (canonical source): no AIA trial proceedings. This is the authoritative list for this task.
• Web search for IPR/PGR/CBM on "5,640,002" / "US5640002": no hits. Every appearance of this patent in the search results is as a cited prior-art reference in other parties' patents and prosecution files — e.g., KR 10-0885432 (Nokia), WO 2013/021233, WO 2016/189096, DE 19924161, and the EP 1 255 217 / EP 0919971 search reports. Being cited as prior art is not a PTAB proceeding.
• Web search for enforcement activity by SCAN-EM CORP / Ruppert / Fish / Yap / Ames: nothing. Google Patents shows a 2008-06-14 assignment from SCAN'EM CORPORATION to Jonathan P. Ruppert, and no litigation is surfaced. No identified petitioner, and no defensive aggregator (Unified Patents, RPX, etc.) in the chain. — flagging as not fully verified: absence of litigation hits in general web search is weaker evidence than absence of PTAB hits in ODP.
• Federal Circuit / CourtListener: no appeal found, which is expected given there is no FWD to appeal.

⚠️ False-positive trap — read this before docketing. Searching for "the '564 patent" returns a large volume of PTAB papers, ITC briefing (Inv. No. 337-TA-823 line and successor Philips matters), and district court briefing. Those references are to U.S. Patent No. 10,298,564 (Philips), an unrelated patent about HDCP round-trip-time authentication. Sample hits: ptacts.uspto.gov/.../petitions/1545103, ptacts.uspto.gov/.../petitions/1546667. None of that is US 5,640,002. If opposing counsel's chart or a vendor docket cites "'564 IPRs" against your client, verify the number — the '564 they mean is almost certainly the Philips patent, not Ruppert's barcode reader.

---

Strategic summary

Claim status. No claim of US 5,640,002 has been canceled, narrowed, or confirmed by the PTAB, because no claim was ever challenged there. Every claim is simultaneously "untested" (no PTAB record) and largely irrelevant (expired). I could not verify the exact claim set from the authoritative full text in a form I'd want to state with high confidence — the OCR of the claim set I retrieved shows at least apparatus claim 1 (hand-held housing, barcode scan engine, display, touchscreen, microcomputer, and a passive-RF-ID-tag reader/writer that writes "paid for" data into tags) and a second independent apparatus claim 6, with dependent claims addressing a thermal printer (claim 2), an infrared transceiver (claim 3), a rechargeable battery plus detachable base unit (claim 4), and a base-unit LAN interface (claim 5). I am flagging that claim listing as secondary-source and not relied on for any legal conclusion.

Estoppel landscape. § 315(e)(2) estoppel is a non-issue here — estoppel attaches only to petitioners who obtain an FWD, and there is no petitioner. That cuts against a defendant in one narrow sense: there is no free invalidity roadmap from a prior IPR to inherit. But it is irrelevant in practice, because the far stronger bars are (a) expiration and (b) the 6-year § 286 damages lookback, which cannot reach any period when the patent was in force if a demand letter arrived now.

Pattern signals. No repeat-petitioner pattern, no patent-owner PTAB appeals (there was nothing to appeal), no aggregator involvement. The notable pattern is institutional: a 1995-filed shopping-cart self-scanner patent that lapsed for fee reasons in 2015 and was never asserted in an AIA trial. The absence of PTAB activity here is not the usual "well-asserted patents attract IPRs" signal — it's the signal of a patent that was never worth asserting.

---

Recommended next steps

1. If you received a demand letter citing US 5,640,002 — the patent is expired, and that is the whole defense.

• Confirm status at the source: https://patents.google.com/patent/US5640002/en (status "Expired – Fee Related"; anticipated expiration 2015-08-15), and independently confirm the term calculation and any maintenance-fee lapse in USPTO Patent Center (https://patentcenter.uspto.gov/).
• An expired patent cannot be infringed. Damages under 35 U.S.C. § 286 reach back only six years from suit, and there is no unexpired term to enjoin. Any demand premised on this patent should be rejected outright, not negotiated.
• Because no PTAB activity exists, there is no FWD to quote and no PTAB E2E link to cite. Do not point to PTAB papers in your response — a citation to a "'564 IPR" will almost certainly be the Philips patent and will damage your credibility.

2. Verify the numbers before you respond. If the letter or a chart cites proceedings on "the '564 patent," confirm the patent number digit-for-digit (5,640,002 vs. 10,298,564). See the false-positive note above and the ITC/PTAB filings at https://ptacts.uspto.gov/ and the ITC EDIS docket for Inv. No. 337-TA-823-line matters.

3. Search the record yourself for the two things I could not fully close out. I found no evidence of, but cannot affirmatively rule out, (i) an ex parte reexamination or (ii) pre-AIA district court litigation asserting this patent in the 1997–2008 window. ODP's "PTAB proceedings" list does not cover reexaminations, so a clean ODP result does not resolve (i). Run a PAIR/Patent Center transaction-history check and a CourtListener docket search (https://www.courtlistener.com/?q=%225640002%22) to close the loop.

4. If you are defending on a different patent in this family. The Ruppert/Fish/Yap/Ames family around RF-ID-plus-barcode handhelds is prior art against others — it is cited in exactly that role by Nokia, by Samsung-adjacent filings, and in multiple PCT search reports. Its highest and best use to you is as a § 102/§ 103 reference against a later-issued patent, not as a target.

Confidence statement: I have high confidence that no AIA trial proceeding exists against US 5,640,002 — the canonical ODP list says none, the patent expired in 2015, and targeted searches produced no petition or decision. I have moderate confidence that no reexamination or pre-2008 litigation exists, and I have not been able to verify the complete official claim set. Do not treat the absence of PTAB activity as evidence of a strong patent; here it reflects a patent that lapsed and was never asserted in an AIA trial.

I'll verify the chain against live sources before writing up the record.

US 5640002 — Ownership Chain Reconstruction

Data-availability caveat up front: I was unable to retrieve the Assignment Center records directly (the USPTO Assignment Search interface is not retrievable through the tools available to me, and no reel/frame numbers appear in any indexed source I could reach). Everything below is grounded in the Google Patents legal-events record as fetched 2026-09-24 from https://patents.google.com/patent/US5640002/en. Reel/frame numbers and correspondents of record are therefore missing from this report — that is a data gap, not a finding, and I flag it rather than infer it. Verify at https://assignmentcenter.uspto.gov/ and https://assignment.uspto.gov/patent/index.html.

---

Inventors

Per the Google Patents bibliographic record and the 1995-08-15 assignment record (US08/515,257, filed 1995-08-15):

Unusual patterns:

• Employer allocation of relative rights is by inference, not by statement. I found no employment agreements or SEC-type disclosure; the only evidence that all four were Scan-Em personnel is that all four assigned to SCAN-EM CORP in a single instrument at filing.
• Geographic split of the team. Three inventors are in the Santa Clara / Morgan Hill, CA corridor; Ames is in Aurora, CO. That is a far-flung-design-team pattern, sometimes consistent with contract engineering, but I have no engagement documents and will not characterize it further.
• No "inventor exit" signal in the 12 months post-filing. All four assigned to the company at filing and no inventor-directed transfer appears until 2008 (see below), so the "inventors depart, then fire-sale" precursor is not observable here.

Original assignee

SCAN-EM CORP (also rendered in the record as "SCAN'EM CORPORATION"), original assignee of record and the assignee named on the issued patent.

• Line of business: handheld barcode / RF-ID "Personal Scanner™" and "Personal Information Device (PID)" hardware for self-scanning grocery checkout. The specification describes the physical product in detail (FIGS. 16–23; on-board Seiko-Epson MTP201-G128B thermal printer; HSDL-1000 IR transceiver; PCMCIA Type I/II/III slot; charging base unit) and refers at FIG. 16 to "a commercial embodiment of a Personal Information Device (PID) according to the teachings of the invention." That is the specification's own characterization of a commercial product, not independently corroborated market evidence.
• Did they ship a product embodying the claims? Unclear. The specification's "commercial embodiment" language plus the level of mechanical/electrical detail (down to vendor part numbers and a designed docking station) indicates a product program that was at minimum near-production. I found no independent evidence of retail deployment, and no revenue, catalog, or FCC ID record.
• Current status: No trace found. The determinative datum is the 2008 transfer of the patent out of SCAN'EM CORPORATION to a single named inventor (Google Patents legal event, 2008-06-14). A 15-year-old single-patent company conveying its crown asset to an individual co-founder is consistent with wind-down/dissolution, but I have no bankruptcy filing, no state dissolution record, and no SEC filing — so "dissolved/dormant" is an inference from the transfer, explicitly labeled as such.
• Note the patent itself reached end of life as "Expired – Fee Related" with legal-status "Anticipated expiration 2015-08-15" — i.e., it lapsed for non-payment of maintenance fees rather than being asserted to term.

Assignment timeline

Two recorded transfers exist in the Google Patents legal-events record. Reel/frame and correspondent are not exposed by that source and I could not obtain them; I am not going to invent them.

• 1995-08-15 (executed) / recorded date not exposed — Reel not retrievable

  • Conveyance: Assignment of Assignors' Interest (expressly, per record text: "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
  • Assignor: Ronald Craig Fish; Jonathan Paul Ruppert; Thomas Allan Yap; Ronald Merle Ames (all four, jointly)
  • Assignee: SCAN-EM CORP
  • Correspondent: not retrievable — cannot assess recurrence
  • Context: Founding-company assignment — all four inventors convey their rights to the employer-entity at filing, the standard new-company capitalization event.

• 2008-06-14 (date as recorded by Google Patents; whether this is execution or recording date is not determinable from the source) / recorded date not exposed — Reel not retrievable

  • Conveyance: Assignment of Assignors' Interest ("ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
  • Assignor: SCAN'EM CORPORATION (note the apostrophe variant in the assignment record vs. "SCAN-EM CORP" on the patent face — same entity, differing transcription)
  • Assignee: RUPPERT, JONATHAN P (the first-named inventor, taking as an individual, not through an entity)
  • Correspondent: not retrievable — cannot assess recurrence
  • Context: Corporate wind-down / asset return to founder. The patent was then ~11 years post-issuance and ~7 years from expiry, being conveyed back to a named inventor with no intervening licensee, aggregator, or holding entity. Google Patents currently lists "RUPPERT JONATHAN P" as current assignee.

No records of any kind were found for: any assignment to an "IP/Holdings/Licensing/Ventures"-suffixed entity; any assignment to Acacia, Marathon, Intellectual Ventures, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg-linked entity; any transfer to RPX, AST, LOT Network, Unified Patents, or OIN; any security agreement, license, merger, or release instrument.

⚠️ Verification request to close the gap: the two items above should be confirmed with reel/frame and correspondent pulled from Assignment Center. The correspondent is the single highest-value missing datum in this file — if a repeat-player NPE-recording attorney appears on the 2008 entry, the reading of that transaction changes materially.

Timeline diagram

timeline
    title Ownership of US 5640002
    1995 : Four inventors file application
         : All rights assigned to Scan Em Corp
    1997 : US patent issues
    2008 : Reassigned to inventor Jonathan P Ruppert
    2015 : Patent lapses for unpaid maintenance fee

NPE / troll-pattern signals

1. Shell-entity transfer — not present. No entity with an "IP / Patents / Licensing / Holdings / Ventures" suffix ever appears in the chain. The only post-issuance transfer (2008-06-14) moved the patent from a corporation to a natural person, the inverse of the shell-formation pattern. No single-member Delaware/Texas LLC, no registered-agent service address, and no commercial-product gap to exploit (the patent was 7 years from expiry). Cite: Google Patents legal event 2008-06-14.

2. Known asserter in the chain — not present. Neither assignee of record (SCAN-EM CORP; Jonathan P. Ruppert, individually) matches any entity on the Acacia / Marathon / IV / IPNav / Wi-LAN / Mosaid-Conversant / Vringo / Pendrell / Innovatio / MPHJ / Lumen View / Round Rock / Document Generation Corp lists, nor did any entity surfaced by Unified Patents or RPX appear. Neither name appeared in any of the assertion-oriented sources I searched.

3. Repeat correspondent across the chain — unclear, and unassessable on current data. Correspondent of record is not exposed by Google Patents legal events and I could not reach Assignment Center. With only two links in the chain and neither party being an entity, the a priori likelihood of an NPE-recording repeat player is low — but I will not call this "not present" when I simply could not read the field.

4. Cascading transfers — not present. There are two recorded transfers, 13 years apart (1995-08-15 → 2008-06-14). The chained-LLC-in-under-24-months pattern requires shared correspondent addresses, shared correspondents, or common principals — none visible, and the second link terminates at an individual rather than feeding another entity.

5. Pre-litigation transfer — not present. I located no infringement litigation ever naming US 5640002. The 2008 transfer precedes expiry by seven years and precedes nothing litigable; the patent's late term ran unasserted. (Note for the record: I did surface an E.D. Tex. RFID/retail-inventory case from 2006 against Wal-Mart, Gillette, and Target — that suit involved US 5,865,563, a 2005 patent, not this patent. Do not conflate them.)

6. Bankruptcy fire-sale — unclear. No Chapter 7/11 proceeding involving SCAN-EM CORP was located, and there is no Kodak/Nortel/Polaroid-style sale. The 2008 conveyance of a near-expiry patent by a dormant-looking corporation to its own named inventor is consistent with a quiet wind-down but is not documented as a bankruptcy sale. Because the asset moved to an insider rather than to a buyer, this is the weakest possible version of the signal even if a proceeding were later found.

7. Privateering — not present. Privateering requires an operating company conveying rights to an NPE that asserts on its behalf. Here nothing was conveyed to any NPE, and nothing was ever asserted. There is no SEC 10-K/8-K (Scan-Em was not public as far as I can determine) and no Patent Progress/EFF coverage involving this patent number.

8. Defensive aggregator — not present. RPX, AST, LOT Network, Unified Patents, and OIN appear nowhere in the chain. The chain terminates at an individual. However, note the practical inverse: the patent is recorded as "Expired – Fee Related" with legal-status entry 2015-08-15 Anticipated expiration. Whatever offensive value it had has been neutralized by its own abandonment, not by a defensive purchase.

Additional observation, clearly labeled as inference: the reassignment to Ruppert in June 2008 fell within roughly six months of the opening of the 11.5-year maintenance-fee window (due 2008-12-17 on a 1997-06-17 grant). That proximity is suggestive of a controlled lapse — the founder taking a patent the company had decided not to maintain — but I found no fee-payment record confirming the lapse date, so treat this as a lead to check, not as evidence.

Verdict

Insufficient data.

The record contains exactly two ownership events, both of which I can date but neither of which I can fully document: the 1995-08-15 founder assignment to SCAN-EM CORP and a 2008-06-14 assignment from SCAN'EM CORPORATION to the inventor Jonathan P. Ruppert individually. With no reel/frame and no correspondent of record obtainable, signal 3 — the one signal most likely to change the picture — cannot be evaluated, which is precisely the "no records / incomplete records" condition this verdict is reserved for.

Bearing the caveat in mind, the affirmative evidence points away from an NPE: a two-link chain spanning 13 years, terminating at a natural person rather than a licensing entity, with the patent recorded "Expired – Fee Related" at the 2015-08-15 term end and no infringement suit found naming US 5640002 in any source I searched. If the reel/frame and correspondent pulled from Assignment Center confirm only these two entries with no aggregator or repeat-player filer, the file should be re-scored as an expired single-patent operating-company asset with no NPE involvement — a clean negative rather than a troll finding.

Verify the chain directly: https://assignmentcenter.uspto.gov/ (search by patent number 5640002) or the legacy index at https://assignment.uspto.gov/patent/index.html.

I'll search for the cited references on US 5,640,002 specifically.

I found the front-page "References Cited" list. Let me verify it against primary sources and pull details on each reference.

Prior Art of U.S. Patent No. 5,640,002 — Reference-by-Reference §102 Analysis

A. Search scope, source, and ground-truth statement

Patent searched (literal): US 5,640,002 — "Portable RF ID tag and barcode reader," Appl. No. 08/515,257, filed 1995-08-15, granted 1997-06-17. The number was taken literally; no substitution of near numbers.

Sources actually returned by my searches: Google Patents (patents.google.com/patent/US5640002/en), FreePatentsOnline (freepatentsonline.com/5640002.html), EveryPatent (everypatent.com/comp/pat5640002.html), uspto.report (uspto.report/patent/grant/5640002), and Espacenet (ie.espacenet.com). Caveat: I did not obtain a direct hit from a USPTO-hosted full-text page (Patent Center / PatFT) in this session. The "References Cited" list reproduced below comes from EveryPatent's mirror of the printed patent front page, and should be confirmed against the USPTO-hosted full text before being relied on for a filing or an invalidity contention.

Near-number results I filtered out as different patents (per the literal-ID rule):

• US 5,660,002 (Lashinger) — appeared in the search report of EP1678997A1; different patent.
• US 10,298,564 (Philips) — the "'564 patent" in ITC/PTAB papers flagged in the earlier PTAB section; unrelated.
• US 5,640,020 / 5,640,024 / JP5640002B2 / KR 10-0885432 — different documents.

---

B. Record update — two uncertainty flags from earlier sections are now resolved

1. Abstract — now retrieved. EveryPatent reproduces the abstract verbatim: "A portable barcode and RF ID tag reader that gathers information about items to be purchased etc by reading barcodes or RF ID tags. A store host computer gathers information about items to be purchased from the portable barcode/ID Tag readers and then the items are bagged by the customer at the checkout stand or by employees of the store… or in a separate warehouse… The portable barcode/RF ID tag reader can also be used in authenticating articles by accessing a factory computer using a serial number for the article scanned from an RF ID tag on the article. The portable barcode/RF ID tag reader is comprised of a microprocessor coupled to a bar code reader, an RF ID tag reader, a spread spectrum RF transceiver, a communication port, an audible feedback device, a touchscreen or light pen and display, a thermal printer and a magnetic stripe card reader and a smart card reader." (Verify against the printed patent; EveryPatent is a secondary mirror.)

2. Claim count — resolved: 8 claims, not 6. The earlier sections correctly flagged the claim set as incomplete. The full set is claims 1–8: claim 1 (independent, barcode + touchscreen + RFID reader that writes "paid for" data), 2 (thermal printer, dep. on 1), 3 (IR transceiver, dep. on 1), 4 (rechargeable battery + detachable base unit with AC→DC converter, dep. on 3), 5 (base unit with LAN interface + line-of-sight IR, dep. on 4), 6 (independent — IR + battery + base + rotatable charging mount), 7 (independent — as 6, plus RF transceiver in the base unit), and 8 (a method claim — using the device with an RFID tag reader and a PCMCIA/credit-card/smart-card slot). Field of Search as printed: 235/472, 380, 462, 383, 382, 435, 492, 493. Primary Examiner: Hajec, Donald T.; Assistant Examiner: Lee, Michael G.

Contradiction to flag in the earlier "Obviousness" section: that section referred to "U.S. 5,250,789 (Kipp), 'Shopping cart.'" The patent is correctly US 5,250,789 to Edward L. Johnsen, "Shopping cart," filed 1991-10-31, granted 1993-10-05 (verified this session from the full text). The "Kipp" attribution is an error and should be corrected wherever it appears.

---

C. The complete "References Cited" list — U.S. Patent Documents

As printed on the face of US 5,640,002 (27 U.S. references). "US Class" is the class printed on the patent face; §102 category is my classification under pre-AIA §102 (the patent was filed 1995-08-15, so pre-AIA governs).

Note on #13 (5,340,973, "Aug. 1994"): if it granted on or before 1994-08-14 it is §102(b); if after 1994-08-15 it is §102(a). The front page gives only the month; verify.

---

D. The complete "References Cited" list — Foreign Patent Documents

Honesty flag: I could not retrieve the disclosure content of any of these five foreign documents in this session. I will not assign them to specific claim limitations on the basis of their numbers alone; they are listed with dates and §102 status only. Verify each against Espacenet/Global Dossier.

---

E. Reference-by-reference analysis

Tier 1 — Most relevant (in-specimen verified this session)

1. US 5,424,524 — Ruppert, Fish? et al., "Personal scanner/computer for displaying shopping lists and scanning barcodes to aid shoppers."

• Full citation: US 5,424,524 A; Appl. filed 1993-06-24; granted 1995-06-13; US Class 235/462.
• Description (verified): Hand-held/palmtop "Personal Scanner™" with microprocessor, CCD barcode reader with built-in decoder outputting ASCII, display, RAM/ROM, touchscreen or light pen, modem or infrared transceiver LAN interface, download of a store price list over a comm port or IR link, shopping-list display and modification, fuzzy-logic matching of scanned items to list entries, running total, budget entry, coupon-list download and automatic coupon matching, credit-card information entry and download to the store computer for payment, security-strip magnets at the scan window. The specification text of '524 and of '002 is near-identical — the "Definitions" block on the Google Patents pages of the two documents matches word for word.
• §102 posture: This is the single most dangerous reference on the list, and the most important one to pull in full. It is the applicants' own earlier patent, but the inventive entity differs: '524 names Ruppert; '002 names Ruppert, Fish, Yap and Ames. Under pre-AIA §102 a reference need only be "by another," and a different inventive entity qualifies for the differing inventors, so '524 is available both under §102(a) (granted 1995-06-13, two months before the '002 filing) and under §102(e) as of its 1993-06-24 filing date — well before any plausible invention date for '002.
• Claims it potentially anticipates: On the retrieved text, '524 discloses in substance the subcombination of claim 1 limitations (1a)–(1f) and (1h) and the claim 2/3 subject matter (printer; IR transceiver), and much of the claim 8 method context (credit-card magnetic-stripe reading, PCMCIA-type expansion memory). It does not disclose claim 1's (1g) passive-RF-ID-tag reader with RF powering/commanding, or (1i) writing "paid for" data into tags. Therefore '524 does not anticipate claim 1 (or claims 2–3, which incorporate it) in full. It is, however, the reference most likely to be invoked for §102(e) or for obviousness-type double patenting against the '002, and it is the strongest §103 starting point.
• ⚠ Flag: I could not determine in this session whether '524 contains the FIGS. 16–44 "PID" embodiment (charging base unit, thermal printer, spread-spectrum transceiver, RFID reader) that appears in '002. If it does, it reaches claims 6–8 as well. This is the key follow-up.

2. US 5,250,789 — Johnsen, "Shopping cart."

• Full citation: US 5,250,789 A; filed 1991-10-31; granted 1993-10-05; US Class 235/383; 22 claims; Non-Patent Literature also cited.
• Description (verified in full text): A shopping-cart-mounted display system with a microprocessor 38, memory 50, display screen 14 (expressly "may use touch screen technology"), product code scanner 22 mounted on the cart, rechargeable power supply 44, thermal printer 32, data port 34, wireless communications (antenna 40 / receiver 42 / transmitter 48) to the store's central computer or POS register, a UPC database with product description and price per bar code, a running total that adds each scanned item's price, an "ability to remove a product by pressing an appropriate button… then scanning the product being removed" (unscanning), coupon-list matching, shopping-list organization, price comparison, and a magnetic-stripe-less identification-card scan.
• §102 posture: Unambiguous §102(b) art — granted more than one year before the 1995-08-15 filing.
• Claims it potentially anticipates: This reference discloses every element of the portable-terminal core (1a–1f), including the touchscreen-over-display element (1e) and the microcomputer storage/display/input control (1f), plus the printer of claim 2 and the running-total/coupon/unscan behaviors underlying the claim 8 method. It does not disclose claim 1's (1g)/(1i) RF-ID-tag reader or write-paid function, and it is cart-mounted rather than "small enough to be carried by hand," so it does not anticipate claim 1 as a whole and therefore cannot anticipate dependent claims 2–5. It is nonetheless the best single §102(b)/§103 reference for the non-RFID half of claim 1.

3. US 5,151,684 — Johnsen.

• Full citation: US 5,151,684; granted Sep 1992; US Class 340/572 (the electronic article surveillance / anti-theft class).
• Description (moderate confidence, from class + inventor/title pattern): An electronic article surveillance / security-tag system — i.e., the alarm-at-the-exit side of the problem. Same inventor (Johnsen) as the '789 "Shopping cart" and cited alongside it, suggesting the examiner grouped the shopping-cart and the EAS/security aspects.
• §102 posture: §102(b) (granted >1 yr before filing).
• Claims it potentially anticipates: Relevant to the "do not cause activation of store alarms… at store exits" functional language of claim 1 (limitation 1i). It cannot anticipate claim 1 because it lacks the barcode/touchscreen/microcomputer core. ⚠ Unverified disclosure — do not treat as a §102 mapping until the text is pulled; my recollection of the title/scope is not certain enough to quote.

4. US 5,381,137 — Ghaem et al., and US 5,528,222 — Moskowitz et al. (both US Class 340/572)

• Citations: US 5,381,137, granted Jan 1995 (§102(a)); US 5,528,222, granted Jun 1996 (§102(e) only — its U.S. filing must predate the '002 invention date to be available at all).
• Description (moderate confidence): Electronic article surveillance / RFID-type security tag and deactivation systems.
• Claims: Relevant to claim 1 (1i) — the security-alarm-at-exit requirement — and to the '002 specification's own EAS discussion (permanent magnets 120/122, switched DC electromagnets). Neither discloses the write-data-into-a-functional-tag step, so neither anticipates claim 1.

Tier 2 — RFID / RF-transponder art (post-filing grants; §102(e) only)

5. US 5,434,572 (Smith) and US 5,479,172 (Smith et al.) — US Classes 342/44 and 342/51; granted Jul 1995 (§102(a)/§102(e)) and Dec 1995 (§102(e) only).
6. US 5,517,194 (Carroll et al.) — US Class 342/50; granted May 1996 (§102(e) only).
7. US 5,550,547 (Chan et al.) — US Class 342/42; granted Aug 1996 (§102(e) only).
8. US 5,317,330 (Everett et al.) — US Class 343/867; granted May 1994 (§102(b)).

• Description (moderate confidence, class-derived): Radio-frequency interrogation/transponder art — interrogators, tag antennas, and reader-to-tag communication. The 342/4x and 343/867 classifications are the RFID/transponder subclasses.
• Claims: These are the references potentially supplying claim 1 limitation (1g) (a transmitter that supplies power and commands to a passive tag, and a receiver that gets product data back). Caveat: limitation 1g as drafted is specific — the RF transmissions must both supply power and commands to an integrated-circuit tag that has no power supply of its own. Whether any of these four shows that exact combination must be checked against the reference text. None of them alone anticipates claim 1.

Tier 3 — Hand-held code-reader / portable-terminal art (the §102(b) baseline)

9. US 4,857,716 (Gombrich et al.) — Aug 1989, 235/472.
10. US 4,970,379 (Danstrom) — Nov 1990, 235/472.
11. US 5,047,615 (Fukumoto et al.) — Sep 1991, 235/472.
12. US 5,221,838 (Gutman et al.) — Jun 1993, 235/472.
13. US 5,227,614 (Danielson et al.) — Jul 1993, 235/462.
14. US 5,324,922 (Roberts) — Jun 1994, 235/472.
15. US 5,340,973 (Knowles et al.) — Aug 1994, 235/472.
16. US 5,406,063 (Jelen) — Apr 1995, 235/472.
17. US 5,444,223 (Blama) — Aug 1995, 235/435.
18. US 5,468,947 (Danielson et al.) — Nov 1995, 235/472 (§102(e) only).

• Description (class-derived; individual disclosures NOT verified in this session): Class 235/472 is the hand-held/portable optical code-reader subclass; 235/462 is the optical-code-reader-combination subclass; 235/435 is optical. These are the references the examiner used for limitations (1a) hand-carried housing and (1b) barcode scan engine that decodes and outputs alphanumeric characters.
• Claims: They can supply (1a) and (1b). None anticipates claim 1, because each lacks the touchscreen (1e), the RFID reader (1g), and the write-paid function (1i). The '002 specification itself concedes this art: it expressly incorporates U.S. Pat. No. 4,204,636 (Hayman) by reference for bar code scanning and names Densei's BCH5532 CCD reader as off-the-shelf.

Tier 4 — POS / checkout / card art

19. US 5,059,951 (Kaltner) — Oct 1991, 235/383.
20. US 5,408,077 (Campo et al.) — Apr 1995, 235/383.
21. US 5,280,159 (Schultz et al.) — Jan 1994, 235/382.
22. US 5,293,424 (Holtey et al.) — Mar 1994, 235/380.
23. US 5,412,192 (Hoss) — May 1995, 235/380.
24. US 5,442,704 (Holtey) — Aug 1995, 235/380.
25. US 5,513,261 (Maher) — Apr 1996, 235/380 (§102(e) only).

• Description (class-derived; NOT verified): Class 235/383 is the cash-register/POS-checkout subclass (~"checkout procedures with a code reader," which is exactly where the '002 itself is classified: G07G1/0045); 235/380 is the credit/smart-card-verification subclass; 235/382 is the check-operated/POS subclass.
• Claims: These map to claim 8's PCMCIA/credit-card/smart-card-slot limitation and to the specification's credit-card-payment and self-checkout embodiments. They do not touch claim 1's RFID limitations.

---

F. Bottom line — what actually anticipates what

No single reference cited on the face of US 5,640,002 anticipates any of claims 1–8 in full. The reason is structural: every claim 1–5 traces back to independent claim 1, which requires (1g) a transmitter that powers and commands a passive, self-powered-less IC tag and (1i) the writing of "paid for" data into tags so exit alarms do not trigger. None of the 32 cited documents describes either limitation, let alone both together with the touchscreen/display/microcomputer core.

The cited art therefore divides cleanly, and this is the finding that matters for any invalidity work:

The two findings I would carry forward into any invalidity position:

1. US 5,424,524 (Ruppert et al., granted 1995-06-13, filed 1993-06-24) is the reference to pull first. It is the applicants' own earlier disclosure with a different inventive entity, making it §102(a) and §102(e) art, and its specification text is materially the same as '002's. If it also contains the PID/base-unit embodiment, it reaches claims 6–8. It also raises obviousness-type double patenting.
2. U.S. 5,382,784 (Eberhardt, 1995-01-17) is conspicuously NOT on the '002 "References Cited" list. The earlier Obviousness section flagged it as a possible hand-held dual-technology (barcode + RF-ID) reading head and named it "the single most important item to pull." The front-page citation record confirms the examiner did not cite it. That makes it third-party art to be developed independently, not "cited art" of this patent — and if it does disclose a dual-technology head, it bears directly on limitation (1b)+(1g).

---

G. Forward citations (context, not "cited art")

For completeness, US 5,640,002 is itself cited as prior art by later filings, e.g. KR 10-0885432 (Nokia — RFID reader power management in a mobile terminal), US 6,314,406 ("Customer information network," Symbol/Telxon), US 9,794,353, and numerous PCT/EP search reports (EP 0 919 971, WO 99/045493, WO 2004/059598). These are forward citations by others and are not part of the §102 analysis of the '002's own claims.

---

H. Confidence and verification steps

• High confidence: the patent identification, dates, assignee, classification, the 8-claim structure, and the abstract (cross-checked against the Google Patents, FreePatentsOnline and EveryPatent records).
• Moderate-to-high confidence: the 27 U.S. and 5 foreign references in the "References Cited" list, and their printed grant dates and U.S. classes (single secondary source — EveryPatent).
• Low confidence / explicitly unverified: the disclosure content of nearly all 32 references. I verified the full text of only US 5,250,789 and the bibliographic/definition material of US 5,424,524 in this session. Class-derived descriptions are labeled as such and must not be quoted as holdings.

To close the gaps:

1. Pull the USPTO-hosted full text and the file wrapper (Patent Center) for 5,640,002 to confirm the printed "References Cited" list and see which references the examiner actually applied (Form PTO-892) versus which the applicant submitted in an IDS.
2. Pull US 5,424,524 in full and compare its disclosure against FIGS. 16–44 of '002 — determinative for claims 6–8 and for double patenting.
3. Pull US 5,382,784 (Eberhardt) and the four 342/4x Smith/Carroll/Chan references to test limitation 1g.
4. Pull US 5,151,684, US 5,381,137, US 5,528,222 to test limitation 1i.
5. Verify the five foreign documents on Espacenet — and note that the Google Patents family field for US 5,640,002 shows "NONE," which conflicts with the shared-specification relationship to 5,424,524; treat that family record as incomplete.

I'll start by pulling the claim text and the prior-art citation records for US 5,640,002 so the analysis is grounded in specific references.

I have partial claim text. Let me retrieve the full claim set and the cited prior-art reference list.

Let me retrieve the front-page U.S. references cited in the patent itself, plus key related pre-1995 art.

Obviousness Analysis — U.S. Pat. No. 5,640,002 ("Portable RF ID Tag and Barcode Reader")

1. Record limitations and confidence flags (read first)

• The document I was given (fetched from https://patents.google.com/patent/US5640002/en on 2026‑09‑24) is truncated. It ends mid‑description at FIG. 26M and does not contain the patent's own "References Cited" list, the claims, or the prosecution history. I therefore cannot reproduce the examiner's cited art or the applicant's remarks.
• The only "Prior Art" metadata present on the supplied page is the auto‑generated Prior art keywords: data, portable information, information gathering, block, tag. These are Google‑generated search terms, not references; they do, however, map onto the claim preamble ("portable information gathering apparatus") and the RFID limitation.
• Full claim text was obtained from FreePatentsOnline (https://FreePatentsOnline.com/5640002.html) via search; claims 1, 2, 3, 4, 5 and a fragment of 6 were retrieved. I could not confirm the total claim count.
• Where I rely on a reference I could not open directly in this session, I say so explicitly. Nothing below should be treated as a legal conclusion; legal status shown is "Expired – Fee Related" (anticipated expiration 2015‑08‑15).

---

2. Claim 1 deconstructed

Claim 1 (as retrieved) is a single independent apparatus claim. Its limitations:

Dependent/related claims: 2 (thermal printer), 3 (IR transceiver, bi‑directional), 4 (rechargeable battery + detachable base unit with AC→DC power converter/charging contacts), 5 (base unit with LAN interface + IR transceiver in line‑of‑sight with the portable unit's IR transceiver, download-on‑command), 6 (independent apparatus claim emphasizing the IR‑transceiver + rechargeable‑battery + detachable‑base combination).

Key observation: the specification itself concedes that essentially every hardware block is off‑the‑shelf. It names Densei (CCD scan engine BCH5532), Hewlett‑Packard (HSDL‑1000 IR module), Disys Inc. (90 Series RF ID tags and "90 Series RF/ID System Applications Manual for CRM‑90 Readers and 90 Series Tags"), Hughes ID Corp., Atmel, Eurosil/Daimler‑Benz, Indala (Motorola), AEG Telefunken, Photonics (IR transceiver), Seiko‑Epson (MTP201‑G128B thermal printer), Inforite/Touch Tech (touchscreens), Proxim and Wireless Logic (RF transceivers), plus the PCMCIA standards. It also expressly incorporates U.S. Pat. No. 4,204,636 to Hayman (bar code scanner) by reference. Under §103 these admissions are prior art against the claims whether or not they were cited on the front page.

---

3. Candidate prior art (with dates and grounding)

Solidly grounded, pre‑1995:

Retrieved but unverified as to content:

• U.S. 5,382,784 (Eberhardt), 17 Jan 1995 — cited as X art in EPO search report EP 02 25 2460. My recollection is that this is a hand‑held dual‑technology reading head for both bar codes and RF identification tags (Indala). If that is correct, it is the single most damaging reference against the barcode‑plus‑RFID combination in claim 1. I could not verify the disclosure in this session; treat as a lead, not a finding.
• U.S. 5,543,607 (Akimoto), 6 Aug 1996 — cited as A art in EP0903683A3. Issued after the 15 Aug 1995 filing date; potentially §102(e)/§103 art only if its filing predates the '002 filing. Content unverified.
• EP 0 918 085‑family EAS references (US 4,686,513 Farrar; US 5,264,829 Paul; EP 0 798 681 Sensormatic; EP 0 020 062 Parmeko; EP 0 449 173 Braehler) — surfaced as A art in EP0919971A3; relevant to alarm‑at‑exit/EAS deactivation, but I did not verify each.

---

4. Obviousness combinations

Combination A — Claims 1 (1a–1h), 2, 3

Primary: IBM "Grocery Buggy" (1988) or Kipp U.S. 5,250,789, in view of the admitted CCD scan engine art (Densei BCH5532 / Hayman '636) and a passive RF‑ID reader/writer system (Disys CRM‑90 or Hughes ID).

• "Grocery Buggy"/Kipp disclose the portable, hand‑carried/cart‑mounted information‑gathering terminal with scan engine, display, memory and microcomputer — limitations 1a–1f and 1h's storage function, and motivation (faster checkout, shopper‑assisted item entry).
• The admitted commercial systems supply 1g in full: the specification states that the RF ID tag "is powered by power derived from rectification of incoming RF signals" and that "the RF ID reader/writer unit 314 knows the unlock sequence … and transmits that sequence plus interrogation or other commands," with "Read Word n / Write Word n" commands — i.e., a reader whose transmitter both powers and commands a passive IC tag and receives product data back.
• Motivation to combine: both Kipp/IBM and the applicant frame the problem identically — get item identity into a portable terminal without a clerk. A POSA facing grocery items that carry either bar codes (low‑value, ubiquitous) or RF tags (higher‑value, EAS‑relevant) would combine two known reading technologies into one hand‑held housing because it (i) eliminates a second device, (ii) allows mixed inventories, and (iii) yields nothing more than the predictable aggregation of two known readers with a known microcomputer bus. KSR Int'l v. Teleflex, 550 U.S. 398 (2007): combination of familiar elements according to known methods, yielding predictable results, is obvious.
• Dependent claim 2 (thermal printer): the specification itself names a commercially available Seiko‑Epson MTP201‑G128B and analogizes to the TI‑59 print cradle. Portable data terminals with integral printers were a recognized class (see, e.g., the "PORTABLE WORK STATION‑TYPE DATA COLLECTION SYSTEM," U.S. 5,710,728, retrieved as an exhibit; date of its underlying filing not verified). Adding a printer to print receipts/coupons is a predictable, market‑driven addition.
• Dependent claim 3 (IR transceiver): admitted — Photonics IR transceiver, HP HSDL‑1000 (1 cm–1 m, 30° cone, 2.4–115.2 KBd, IrDA‑compatible), and the applicant's citation of the IEEE 802.11 working group. Point‑to‑point IR data transfer between a portable terminal and a fixed host was a standard, well‑understood substitution for a cable.

Combination B — Claim 1 (the RFID subcombination specifically)

Eberhardt U.S. 5,382,784 (if confirmed as a dual‑technology bar‑code/RF‑ID hand‑held reading head) in view of Kipp/IBM "Grocery Buggy" and the admitted Disys/Hughes reader‑writer systems.

If Eberhardt '784 discloses a hand‑held head that reads both symbologies, claim 1's 1b + 1g combination is met by a single reference, and only 1e (touchscreen) and 1i remain — the touchscreen being admitted commercial art (Inforite/Touch Tech) and 1i addressed below.

Combination C — Claim 1 limitation 1i ("write paid data so exit readers don't alarm")

Grocery Buggy/Kipp + admitted RFID reader/writer commands (Write Word n) + electronic article surveillance (EAS) deactivation art (e.g., Farrar U.S. 4,686,513; Sensormatic EP 0 798 681; and the '002 specification's own discussion of magnetic security‑strip deactivation by permanent magnets 120/122 or switched DC electromagnets).

• EAS tags were already deactivated at the point of sale; the '002 specification devotes several paragraphs to exactly this (magnets mounted at the scan window, switched only when the scan button is pressed to prevent abuse), which is an admission that "neutralize the alarm at exit upon payment" was a known requirement.
• Given a writable passive tag (the admitted Disys tags accept "Write Word n"), writing a paid‑status datum instead of magnetically killing the strip is a substitution of one known deactivation mechanism for another, producing the same predictable result (no alarm at exit). The functional language "such that said RF ID tags do not cause activation of store alarms" recites the intended result of that substitution, not a new structure.

Combination D — Claims 4, 5 and 6 (base unit, charging, LAN, IR docking)

Portable terminal of Combination A/B + conventional docking/charging cradle art + LAN drop.

• Claim 4 is a charging cradle with an AC→DC converter and mating contacts — ubiquitous (the specification itself refers to a TI‑59 print cradle; the retrieved U.S. 5,710,728 lists U.S. 4,845,419 (Hacker), a battery‑charging reference).
• Claim 5 adds an IR transceiver aligned with the portable unit's IR window and connected to a LAN interface: the specification describes precisely such a base unit (base unit 312 with an IR transceiver behind a window in charging unit 316, plus an RF module/LAN interface card). Combining a known cradle with a known IrDA link and a known Ethernet/LAN interface to a host is a routine, predictable aggregation; the only added feature is a "download command" flag, i.e., software.
• Claim 6, as retrieved, is essentially the claim‑1 housing/display/memory/touchscreen/microcomputer core plus the IR transceiver and the rechargeable‑battery/detachable‑base subcombination — so it falls with Combination A plus the admitted IR art, and does not require the RFID limitations at all.

---

5. Rebuttal considerations (arguments against obviousness)

A rigorous §103 analysis should flag the counters a patent owner would raise:

1. "The references teach killing (deactivating) a tag, not writing a persistent 'paid' semantic datum into it." This is the strongest non‑obviousness argument. Sensormatic‑type EAS art disables a resonant element; the claim requires the RF ID tag to remain functional and carry state. Counter: the admitted Disys tag supports "Write Word n," and the advantage (a tamper‑evident, auditable paid record) is a predictable benefit of a writable memory the artisan already had in hand.
2. Cost/availability of RF tags on grocery items in Aug. 1995. Tags were then economic only for higher‑value goods; applying a tag reader to a grocery self‑scan terminal could be argued non‑obvious. Counter: the '002 specification itself uses inventory/factory and authentication applications, which undercuts an argument that the inventors recognized a grocery‑only obstacle.
3. Separate‑accommodation argument — that the '002 device integrates barcode, RFID, spread‑spectrum RF, IR, printer, magnetic stripe and smart card in one hand‑held unit. But each peripheral is individually admitted as off‑the‑shelf, and the claim does not recite any unforeseen interaction among them.
4. No secondary‑considerations record was available to me (no evidence of commercial success, copying, licensing, or long‑felt need in the retrieved material). Absent such evidence, KSR's "predictable results" standard is likely controlling.

---

6. Bottom line

• Claims 1–5 are, on the available record, prime §103 candidates. The overwhelming majority of limitations are the applicant's own admitted prior art (off‑the‑shelf CCD scan engines, IR modules, PCMCIA cards, thermal printers, touchscreens, and — critically — commercially available passive RF‑ID tag reader/writer systems with "Read/Write Word n" command sets). Adding "Grocery Buggy" (Nov. 1988) or Kipp U.S. 5,250,789 for the self‑scanning shopping terminal, plus EAS deactivation art for limitation 1i, supplies every remaining element with a strong, problem‑driven motivation to combine.
• If U.S. 5,382,784 (Eberhardt, 17 Jan 1995) is indeed a hand‑held bar‑code/RF‑ID dual‑technology reading head, claim 1 is very likely obvious as a two‑ or three‑reference combination. I could not verify that reference's disclosure in this session — it is the single most important item to pull before finalizing any invalidity contention.
• The only genuinely contestable feature is limitation 1i (writing paid‑status data into tags rather than deactivating them). Expect the obviousness case to turn on whether the EAS‑deactivation art plus the admitted writable tags is accepted as a predictable substitution under KSR.