US 5874896

I'll verify this patent against live sources, then summarize.

US 5,874,896 — Verification and Summary

Search scope caveat up front: I do not have direct API access to USPTO PatentCenter/assignment records or PACER/CAFC dockets; my available tool is open web search. The bibliographic data below is corroborated across Google Patents, uspto.report, and secondary patent databases, but a formal PACER/PTAB docket check was not performed. Statements about litigation are therefore "not found," not "confirmed nonexistent."

Number-identity check. The number resolves to exactly one live document: US 5,874,896. I explicitly did not conflate it with near-identical identifiers that surfaced during searching and are unrelated documents:

• US 5,587,896 (Hansen et al., issued 1996‑12‑24) — a different patent, cited as prior art in WO1998041907.
• USD392,939 — a design patent whose Google Patents internal application string reads "US5874896F"; coincidental string overlap, not this patent.
• Various later "'896" patents (e.g., 11,905,896 fuel selector; D959,896 silicone lid; 9,896,xxx family) — unrelated.

Bibliographic data (US 5,874,896)

Discrepancy flagged (not corrected): at least one aggregator (a Unified Patents portal page) lists the priority date for US‑5874896‑A as 1996‑08‑25 and assignee as "HID Corp," whereas Google Patents/USPTO‑derived data give 1996‑08‑26. I am reporting both rather than harmonizing them; the filing date is consistently 1996‑08‑26.

Abstract (source text)

The system monitors merchandise in a sales outlet to deter shoplifting. Transponder tags are attached to articles. A tag exciter at the exit emits an RF surveillance excitation signal. If an article passes without removal authorization, the tag is powered by that signal and emits an RF surveillance response signal that triggers an alarm. If the customer has paid, the tag is reprogrammed to change stored operational data — typically the response-signal frequency, the excitation frequency the tag responds to, and/or the modulation type — so it no longer generates the alarm-triggering signal. Sales data (purchase price, date of purchase) is also written into the tag and can later be read back by a clerk.

Plain-language overview of the independent claims

Claim 1 — Method (sales outlet). A four-step loop: (1) attach a transponder tag to merchandise; (2) put a tag exciter at the exit; (3) when an unauthorized article passes, the exciter's RF surveillance signal powers the tag, which replies with an RF surveillance response signal that trips an alarm; (4) when removal is authorized (i.e., paid for), reprogram the tag and store sales data in it, such that the reprogrammed tag no longer emits the alarm-triggering response signal as it passes the exciter. The commercial point: deactivation is achieved by rewriting the tag's operational parameters rather than by mechanically removing or physically disabling the tag.

Claim 12 — System. The apparatus counterpart: a tag exciter at the exit; an exciter/reader/writer (ERW) circuit that emits both an RF writing excitation signal carrying operational + sales data and an RF reading excitation signal; many transponder tags, each with a memory and a controller having three modes — surveillance (emit RF surveillance response signal), programming (store operational and sales data in memory), and reading (emit an RF reading response signal containing the stored sales data); an alarm responsive to the surveillance response signal; and the ERW receiving the reading response and outputting the sales data to an output device (display/printer/network/computer). This is the "EAS + RFID record-keeping" combination: the same tag both guards the exit and later serves as the proof-of-purchase/refund-control record.

Claim 15 — Method (broader "controlled area" variant). Substantively parallel to claim 1, but generalized: "article" instead of "article of merchandise," "controlled area" instead of "sales outlet," and "transaction data" instead of "sales data." The independent claim set thus spans retail-specific and generic-controlled-area framings.

Claim 26 — Method (signal-frequency variant). Parallel to claims 1/15 but with an added signal characteristic: the RF surveillance response signal must have a frequency at least about one quarter the frequency of the RF surveillance excitation signal. The specification explains the purpose — returning the response at a single high-frequency (bitrate) maximizes signal power and hence the effective detection range (dependent claim 28 narrows this to "about one half," and claim 11 mirrors it in the claim 1 family).

Dependent claim structure: claims 2–11 depend from claim 1; 13–14 from claim 12; 16–25 from claim 15; 27–28 from claim 26. Recurring themes: sales/transaction data content (price, date, customer name, merchandise ID); the three reprogramming mechanisms (response frequency, excitation frequency, modulation type); password-protected writing; the response being "substantially free of data" (presence-of-signal alarm logic); and use of an ERW to reprogram.

Observations worth noting

1. Antecedent-basis quirk in the claim set. Claims 6–8 (depending from claim 1) recite "wherein said operational data includes…", but "operational data" is first introduced only in claim 2. The same pattern appears at claims 21–22 (depending from claim 15, which introduces it only in claim 16). Whether this was ever litigated or corrected is not something I can confirm from available sources.
2. Legal status and 2026 docket expectation. The patent is recorded as expired (anticipated expiration 2016‑08‑26, status "Expired – Fee Related"). A 2026 CAFC appeal involving US 5,874,896 would be unusual given expiry, and I found no IPR, PTAB, or CAFC proceeding naming this patent. The HID/ASSA ABLOY‑related PTAB activity I did find (e.g., IPR2017‑01938/‑01939/‑01940/‑01941, HID Global v. Idemia; IPR2022‑01045, Assa Abloy v. CPC Patent Technologies) concerns entirely different patents (7,207,494 / 7,661,600 / 8,083,152 / 9,269,208) and should not be attributed to this patent.
3. Prior art cited on the face of the patent: US 4,399,437 (Falck et al.), US 4,656,463 (Intelli-Tech), US 4,730,188 (Milheiser), US 5,218,343 (Stobbe), US 5,214,409 and US 5,499,017 (Avid), plus the Temic e5550 IDIC non-patent literature (1994/1995). The preferred tag is the Temic e5550 read/write identification IC (Temic Eurosil, Eching, Germany).
4. Family-member-as-prior-art note. The Australian family member AU 712,986 B2 was later cited (category X/Y) in the international search report for WO2007/006085, so this disclosure family has downstream prior-art significance in the RFID/EAS space.
5. Uncertainty statement. I have high confidence in the title, inventors, assignees, dates, abstract, and claim scope (they come from the full patent text). I have low/uncertain confidence in any statement about enforcement history for US 5,874,896 specifically — I found none, but that is a negative search result from a web tool, not an authoritative docket clearance.

Primary sources used: Google Patents full text (https://patents.google.com/patent/US5874896/en), uspto.report/patent/grant/5874896, plus corroborating citation listings on justia/FPO patent pages.

I'll search for litigation involving US Patent 5874896.

Let me run more targeted searches on litigation databases.

Litigation Search Report — U.S. Patent No. 5,874,896

Patent identified literally as US 5,874,896 (Application No. 08/703,045; "Electronic anti-shoplifting system employing an RFID tag"; filed 1996‑08‑26; granted 1999‑02‑23; inventors Peter R. Lowe and Donald G. Small, Jr.; original assignee Palomar Technologies Corporation; reassigned to HID Corporation 1998‑03‑06; Google Patents lists current assignee as HID Global Corp; status "Expired – Fee Related," anticipated expiration 2016‑08‑26).

Bottom line

No litigation involving U.S. Patent No. 5,874,896 was identified. I found no district court complaint, CAFC appeal, ITC action, or PTAB proceeding in which the '896 patent is named as an asserted patent. I am stating this as "not found," not as a certified negative — my searches returned no hit tying this exact number to any case, and I could not complete the additional database-specific queries I had queued. Treat this as a preliminary clearance-style result pending direct docket confirmation.

What was searched

• Google Patents entry for US5874896A (full text, family, citations, "Cited By," "Families Citing," status/reassignment history).
• FreePatentsOnline and Justia Patents listings citing 5874896 (all appear as prior art citations in later RFID patents, e.g., Rockwell Automation US 7,510,110; not as litigation).
• General web queries for "5874896" combined with litigation/case/lawsuit terms, plus assignee-based queries (HID, Palomar Technologies, anti-shoplifting, RFID tag).
• Unified Patents litigation portal and Docket Alarm/CourtListener material surfaced in results.

Nothing found in the following name/assignee-adjacent matters (none assert the '896 patent)

To be precise about what the searches did return, so you can see I checked the right neighborhood and these are all different patents/parties:

Two identifier traps — do not conflate

1. "The '587 patent" in Metricolor LLC v. L'Oréal S.A. (C.D. Cal. No. 2:18‑cv‑00364‑R‑E, aff'd in part/vacated in part, Fed. Cir. 2018‑2397, Oct. 30, 2019) is U.S. Patent No. 9,301,587 (hair-dye dispensing), a completely unrelated patent. It is not US 5,874,896 and must not be recorded as litigation for this patent.
2. The string "US5874896F" appears in USD392939S (a battery design patent) as part of that application's internal document identifier (application 29/058,748). That is a different identifier, not US 5,874,896. Likewise, unrelated commercial/trademark records show the bare digit string "5874896" (e.g., a Lithuanian company code, an Indian trademark notice) — irrelevant.

Related family (for completeness; no litigation found on these either)

• EP 0827123 A1 (A System and Method for Monitoring an Article) — status "Ceased"
• CA 2213716 C — "Expired – Fee Related"
• AU 712986 B2 — "Ceased"
• JP H10124764 A — listed as active/pending on Google Patents

Note that EP 0827123 was cited as an X-category reference in WO 2005/015516 A1 and WO 1999/059112 A1 — that is patent prosecution prior art, not litigation.

Recommended verification before relying on this

Because I could not exhaustively query every docket system, confirm with:

• PACER / CourtListener RECAP — full-text search for "5,874,896" and "5874896" across all district and appellate dockets.
• Unified Patents Portal litigation and PTAB caselists (https://portal.unifiedpatents.com/litigation/caselist).
• Docket Alarm and RPX Insight by patent number.
• USPTO Patent Center / Assignment database for the '896 patent — the assignment records I retrieved show only corporate assignments (Palomar → HID, and the 1999 Chase Manhattan security agreement later terminated in 2001), with no indication of a lawsuit-driven transfer.
• Note the expiration date of 2016‑08‑26; even had a case existed, enforcement would have ceased by then, so litigation risk is now historical only.

I'll verify the structured "no proceedings" result against live sources before writing this up.

Now let me triangulate with targeted checks on the patent's owners and any assertion history.

Let me run a few final targeted checks on the PTAB dockets and any assertion history.

Proceedings overview

Total AIA trial proceedings on US 5,874,896: zero. The structured "PTAB proceedings on file" block (USPTO Open Data Portal, most recent ingest) returns no IPR, PGR, or CBM naming this patent: 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution-denied. My independent web searches corroborate that negative result. The bottom-line defensive posture is therefore neither "hardened by surviving IPRs" nor "claims already canceled" — it is "never stress-tested at the Board at all, and expired on 2016-08-26." That combination is unusually favorable to an accused infringer, because there is no adverse FWD to work around, no claim-construction record that binds anyone, and no estoppel running against any challenger.

Verification trail (what I actually checked)

I searched for PTAB/IPR/CAFC material keyed to the patent number and to its ownership chain (Palomar Technologies Corporation → HID Corporation → HID Global Corp). Every hit resolving to a "'896" proceeding was a different patent. I did not auto-correct any identifier:

I also looked for the Unified Patents portal page for US-5874896-A specifically (a defensive-aggregator filing would show up there). My search returned the Unified Patents page for US-10706412-B2 instead and I could not re-verify the US-5874896-A portal page in this session — so I am not asserting that page exists or that it records anything. The prior generated section referenced such a page (with a 1996-08-25 priority date and "HID Corp" assignee); I could not independently reconfirm it here. That is a gap, not a contradiction.

Search limitation, stated plainly: two of my planned targeted queries — one keyed to IPR2015/IPR2016/IPR2017/CBM petition numbers paired with this patent number, and one on the patent's district-court assertion history — did not execute (I hit a tool step limit). Those are the two remaining places a proceeding or suit could hide. The ODP structure block remains the authoritative source, and it says none.

(No proceeding entries — the per-proceeding template is inapplicable)

Because there are no proceedings, there are no panels, institution decisions, FWDs, settlements, or appeals to report for US 5,874,896. I am not going to populate the {PROCEEDING_NUMBER} template with placeholders, and I will not attribute the HID Global / ASSA ABLOY-related PTAB activity (IPR2017-01938/‑01939/‑01940/‑01941, HID Global v. Idemia; IPR2022-01045, Assa Abloy v. CPC Patent Technologies) to this patent — as the prior summary correctly noted, those concern US 7,207,494 / 7,661,600 / 8,083,152 / 9,269,208, not this one. Likewise, the MRSI Systems v. Palomar Technologies litigation (D. Mass., D.S. Cal.) involves a different Palomar entity (die/wire bonders) and completely different patents — the shared "Palomar Technologies" name is a trap and I am flagging it rather than treating it as evidence of assertion of this patent.

Strategic summary

Claim status: all 28 claims are UNTESTED — and expired. No claim of US 5,874,896 has been canceled, and none has been confirmed by the Board; there is simply no PTAB record. The independent claims at issue would be 1, 12, 15, and 26 (method/sales-outlet, system, method/controlled-area, and method/response-frequency variants per the prior section's claim map), with dependents 2–11, 13–14, 16–25, and 27–28 hanging beneath them. The critical fact for anyone being asserted against today is the anticipated expiration of 2016-08-26 — the patent term is over. Under 35 U.S.C. § 286, a complaint filed on or after 2020-08-26 cannot reach six years back to any date on which the patent was still in force (27 Sep 2026 minus six years = 2020-09-27, which postdates expiration). In practical terms: there is no live injunctive relief and essentially no recoverable damages window left for conduct after expiry. Any demand letter citing this patent is, at best, a historical past-damages play for pre-2016 conduct; at worst it is a bluff.

Estoppel landscape: none, which is pure upside for a defendant. Because no IPR was ever instituted — let alone carried to an FWD — 35 U.S.C. § 315(e)(2) estoppel never attached to anyone. There is no petitioner whose raised-or-reasonably-could-have-raised grounds are now locked out, and no privy chain to worry about. A defendant today (or any future challenger) may raise any § 102/§ 103/§ 112 ground, on any reference, in any forum, without runway restrictions from a prior Board proceeding. There is also no prior Board claim construction to distinguish — any district court or Board would construe the claims fresh (and, for an expired patent, the Board would still apply Phillips, though the patent owner could not meaningfully amend).

Pattern signals: a flat line. No petitioner has filed even one IPR against this patent, so there is no serial-filer or follow-on-petitioner pattern to model. No Unified Patents (or similar defensive aggregator) proceeding was found in the chain. No Federal Circuit appeal naming this patent was found — consistent with the prior section's "no IPR, PTAB, or CAFC proceeding naming this patent" conclusion, so there is no contradiction between the earlier work product and this analysis. The prior summary's strongest caveat — that a 2026 CAFC appeal would be anomalous given 2016 expiry — is reinforced here: the absence of PTAB activity is unsurprising for a patent that expired a decade ago. Note also the current discretionary-denial climate: the PTO's practice of declining institution based on a patent's "settled expectations" (the policy at issue in the Kahoot! AS v. Interstellar cert petition, where a denial keyed to "over six years" of patent life was held unreviewable) is squarely relevant to a 1996-priority patent — though this is a general trend observation, not a statement about any proceeding involving this patent.

Recommended next steps

If you are a defendant and someone asserts this patent: there is no FWD to point to and no canceled claims to quote — I will not invent one. Your first move is not an invalidity brief but a date check: confirm the 2016-08-26 expiration against the USPTO PatentCenter maintenance-fee record, then run the § 286 six-year lookback against the alleged conduct. If the demands relate to post-2016 conduct, the § 286 math above should end the conversation. Secondary moves, in order: (1) re-pull the full USPTO PTAB E2E docket and the ODP AIA-trial endpoint for application 08/703,045 to confirm the zero count at the moment you respond (I could not run these as live API queries — my tooling is open web search only); (2) check CourtListener for any district-court or CAFC docket naming 5,874,896, which is the one enforcement-history question I could not close in this session; (3) treat the prior section's antecedent-basis quirks (claims 6–8 referring to "operational data" first introduced in claim 2; claims 21–22 same pattern relative to claim 16) as § 112(b) arguments you can raise without any estoppel exposure, since no IPR ever fixed or foreclosed them.

If a proceeding were somehow pending: the milestones to have tracked are the § 314(b) three-month institution deadline, the § 316(a)(10) oral-hearing request, and the § 316(a)(11) one-year FWD deadline from institution. None of those clocks is running here.

The plain answer to "is there PTAB activity on US 5,874,896?": no. The ODP record and my searches agree, and the two independent web queries that could have overtaken that default (historic IPR2015–2017/CBM petitions; assertion-history search) failed to execute, so I am flagging them as open rather than confirming them. The absence is itself a signal: patents that are actually asserted in the modern era attract IPRs, and this one — expired since 2016-08-26, never asserted in any litigation or PTAB filing I could locate — never did.

I'll research the assignment chain for US 5,874,896, starting with the recorded conveyances and the entities involved.

Verification note (read first)

My only tool in this session is open web search — I have no direct API or authenticated session to USPTO Assignment Center (https://assignmentcenter.uspto.gov/) or to its legacy index (https://assignment.uspto.gov/patent/index.html). What I could retrieve are the legal-event records surfaced on the patent's Google Patents page (which are derived from USPTO assignment data) and corporate-history sources. Those records confirm that four conveyances were recorded and give their dates and party names, but they do not expose reel/frame numbers, correspondent names, or recorded addresses. I have therefore left those fields explicitly marked "not retrieved" rather than inventing them. Anyone relying on this for a clearance or standing opinion must pull the underlying reel/frame images from Assignment Center by searching patent number 5874896.

Inventors

Basis: The assignment-of-interest recorded at the filing date names Lowe, Peter R. and Small, Donald G., Jr. as assignors and PALOMAR TECHNOLOGIES CORPORATION as assignee. Inventor-to-applicant assignment executed on the filing date is the standard employee/employer pattern; combined with the corporate history (Palomar was the Hughes Identification Devices management buyout vehicle), it is strong circumstantial evidence both men were Palomar/HID personnel at filing.

Unusual-pattern check: No evidence of inventor departure within 12 months of filing. I found no re-assignment, no separate inventor-to-third-party conveyance, and no record of either inventor assigning away rights outside the employer chain. I also could not confirm (and do not assert) that the "Lowe" name appearing in later RFID patent citation lists (e.g., Lowe‑named publications cited in US 10,742,630) is the same Peter R. Lowe — treat that as unverified.

Original assignee

Palomar Technologies Corporation (name on the issued patent; Google Patents presents it as "Palomar Technologies Corp").

• Corporate origin: HID began in 1991 as Hughes Identification Devices, a Hughes Aircraft subsidiary. In October 1995 Hughes management, backed by Citicorp Venture Capital, combined the military communications/display groups, the AML Wireless Systems organization, and Hughes Identification Devices into Palomar Technologies Corporation. The security business within it was renamed HID Corporation.
• Primary line of business: RF identification for physical access control — 125 kHz "prox" cards and readers. Per ASSA ABLOY's acquisition press release (Nov. 6, 2000), HID held roughly 40% world market share in contactless access control with ~USD 100 M expected 2000 sales and ~350 employees (HQ Irvine, CA; R&D Denver, CO; manufacturing North Haven, CT).
• Did they ship a product embodying the claims? Qualified yes on the platform, unverified on the claimed system. HID commercially shipped RFID reader/tag hardware that matches the patent's exciter/reader writer architecture, and the patent's preferred tag is a purchased part (Temic e5550, Temic Eurosil, Eching, Germany). I found no evidence the specific anti‑shoplifting EAS system of claims 1/12/15/26 was commercialized as a product — the disclosure reads as an application of HID's RFID platform to the EAS market.
• Current status: Acquired. Palomar/HID's access-control business was acquired by ASSA ABLOY AB in November 2000 for USD 250 M plus a limited earn-out. The entity is now HID Global (Irvine, CA), still an operating manufacturer. The original "Palomar" corporate shell no longer operates in this space.
• Name-collision warning: a separate, still-operating Palomar Technologies, Inc. (semiconductor die-attach/assembly equipment, Carlsbad, CA) exists and is not the patent assignee — do not conflate. Note also the related parent name "Palomar Technological Companies (Delaware Corp)," which appears in UK Companies House records as corporate secretary of HID Corporation Limited (UK) from 1998‑05‑22 to 2003‑03‑01.

Assignment timeline

All four conveyances below are confirmed by the patent's recorded legal events. Reel/frame and correspondent fields could not be retrieved with the tools available (see verification note).

1. 1996‑08‑26 (executed) / recorded 1996‑08‑26 — Reel not retrieved / Frame not retrieved

   • Conveyance: Assignment of Interest
   • Assignor: Lowe, Peter R. and Small, Donald G., Jr.
   • Assignee: Palomar Technologies Corporation
   • Correspondent: not retrieved — flag: unknown; cannot be assessed for repeat-player status.
   • Context: standard inventor-to-employer assignment executed coincident with filing (original vesting, not a transfer).

2. 1998‑03‑06 (executed/recorded) — Reel not retrieved / Frame not retrieved

   • Conveyance: Assignment of Interest
   • Assignor: Palomar Technologies Corporation
   • Assignee: HID Corporation
   • Correspondent: not retrieved — flag: unknown.
   • Context: internal reorg / push-down of the RFID access-control IP into the HID operating subsidiary ahead of the company's eventual sale. Timing corroborates an internal restructuring rather than a third-party sale: Palomar Technological Companies was recorded as secretary of the UK HID subsidiary from 1998‑05‑22, and HID plc/HID Corporation was acquired by ASSA ABLOY in Nov. 2000. This is the single most important link in the chain because it moves title from the parent/holding name to the branding entity that survives today as HID Global.

3. 1999‑10‑13 (executed/recorded) — Reel not retrieved / Frame not retrieved

   • Conveyance: Security Agreement (not an ownership transfer)
   • Assignor: HID Corporation
   • Assignee: The Chase Manhattan Bank, as Agent
   • Correspondent: not retrieved — flag: unknown; security-agreement recordings of this era are typically filed by outside lender counsel, so a bank-side correspondent here would be expected and is not by itself an NPE tell.
   • Context: securitization — patent collateral pledged under a corporate credit facility (consistent with the Citicorp/venture-debt-era capital structure; note the agent of record is Chase Manhattan, not Citibank). No title passes.

4. 2001‑04‑03 (executed/recorded) — Reel not retrieved / Frame not retrieved

   • Conveyance: Termination of Security Agreement (release)
   • Assignor: The Chase Manhattan Bank, The, as Agent
   • Assignee: HID Corporation
   • Correspondent: not retrieved — flag: unknown.
   • Context: release of the 1999 lien — the facility was repaid/refinanced (HID had been acquired by ASSA ABLOY in Nov. 2000, whose press release states the purchase was financed through ASSA ABLOY's existing credit facility). Title is clean in HID from this date forward.

Post-2001: Google Patents lists the current assignee as HID Global Corp. This is consistent with a change of name / rebranding of HID Corporation to HID Global Corporation, not a new arm's-length assignment — no separate conveyance record was retrieved and I do not assert one exists.

Timeline diagram

timeline
    title Ownership of US 5874896
    1996 : Application filed
         : Inventors assign to Palomar Technologies Corp
    1998 : Palomar assigns to HID Corporation
    1999 : Patent issues Feb 1999
         : Security agreement to Chase Manhattan Bank
    2000 : ASSA ABLOY acquires HID
    2001 : Security agreement terminated
    2016 : Patent expires Aug 2016

NPE / troll-pattern signals

1. Shell-entity transfer — not present. No "IP/Holdings/Licensing/Ventures"-suffixed assignee appears anywhere in the chain. Every recorded party is either an operating manufacturer (Palomar Technologies Corporation, HID Corporation) or a chartered bank acting as collateral agent (The Chase Manhattan Bank, as Agent). The terminal record assignee, HID Global Corp, is a 2,000-plus-employee operating business (per HID/ASSA ABLOY sources).

2. Known asserter in the chain — not present. No assignee in the chain matches Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Round Rock, Erich Spangenberg entities, or any Unified Patents / RPX high-frequency plaintiff list that surfaced in searching. The only NPE-adjacent name in the immediate corporate neighborhood is CPC Patent Technologies Pty. Ltd. (Charter Pacific), which sued HID — but as plaintiff against HID, and over entirely different patents (US 9,269,208 / 9,665,705 / 8,620,039). It is not in this patent's chain.

3. Repeat correspondent across the chain — unclear / cannot be assessed. This is the one signal the available tooling defeats: none of the four recordings' correspondent fields could be retrieved, so I cannot test recurrence. Given only three ownership-relevant filings (1996 inventor assignment, 1998 corporate assignment, and the 1999/2001 security-agreement pair) spread across a bank agent and a corporate law department, recurrence would in any event be unlikely — but I am recording this as unclear, not as a negative, and it should be the first thing checked in Assignment Center.

4. Cascading transfers — not present. The chain contains zero successive third-party-to-third-party LLC hops. Between the 1996 filing and the 2016 expiry there is exactly one ownership transfer (Palomar → HID in 1998), plus a lien and its release. No two transfers occur within 24 months of each other, and none share a suspicious common address.

5. Pre-litigation transfer — not present. No infringement suit naming US 5,874,896 was identified in the previously generated litigation analysis, so there is no trigger event against which to measure a 6-month pre-suit assignment. The last ownership event is 1998; the patent expired 2016‑08‑26.

6. Bankruptcy fire-sale — not present. Palomar/HID did not go through Chapter 7/11. It was sold as a solvent, going-concern acquisition for USD 250 M cash (ASSA ABLOY press release, Nov. 6, 2000), with the IP transferring as part of the corporate share purchase rather than through an estate sale.

7. Privateering — not present. The 1998 Palomar → HID transfer is a downward corporate reorganization within one controlled group (Palomar parent/HID branding entity), not a transfer to a third party asserting on Palomar's behalf. HID continued to make and sell access-control products; no indication of an assertion mandate.

8. Defensive aggregator — not present. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. It terminates at an operating manufacturer, HID Global Corp, and then at legal expiration (predicted 2016‑08‑26), which neutralizes the patent by operation of law rather than by defensive acquisition.

Verdict

Defensive / non-asserting — with an explicit definitional caveat.

Substantively: the chain terminates at an operating manufacturer (HID Global Corp, the ASSA ABLOY subsidiary), contains one internal 1998 reorg assignment and one 1999 lien released in 2001, shows no shell entity, no known asserter, no cascading transfers, no bankruptcy sale, and no pre-suit transfer, and the patent expired 2016‑08‑26 with no litigation found in the prior analysis. Caveat: the literal menu definition of this verdict requires the chain to end at a defensive aggregator (RPX/AST/LOT/Unified), which it does not — the correct characterization is "held-to-expiry by an operating company, never asserted." The verdict should not be recorded as "Operating-company assertion" (no suit exists) nor as "NPE — high/moderate confidence" (zero qualifying signals were found). Note also that the 1998‑03‑06 date on the recorded event is the event/record date as published; the underlying executed instrument should be confirmed at reel/frame before this is relied on.

Verification link: https://assignment.uspto.gov/patent/index.html (search patent number 5874896) and https://assignmentcenter.uspto.gov/ — to pull the four reel/frame images and, critically, the correspondent of record for the 1996 and 1998 recordings, which is the single field this analysis could not populate.

I'll verify the patent and each of its cited references against live sources.

Prior Art Analysis — U.S. Patent No. 5,874,896

Verified patent: US 5,874,896 A, "Electronic anti-shoplifting system employing an RFID tag," App. No. 08/703,045, filed 1996‑08‑26, granted 1999‑02‑23; inventors Lowe & Small; original assignee Palomar Technologies Corp. (Google Patents: https://patents.google.com/patent/US5874896/en; uspto.report: https://uspto.report/patent/grant/5874896).

This section builds on the previously generated summary and litigation sections and does not repeat their bibliographic/claim-scope content. It does not re-run the litigation check.

Scope caveat: my available tool is open web search, not a USPTO API. The reference list below is taken from the authoritative full text of the patent (the "Citations (6)" and "Non-Patent Citations (4)" tables at the Google Patents page), cross-checked where the search returned hits. Where I could not independently verify a reference's disclosure, I say so rather than characterize it.

1. How the prior art was identified

US 5,874,896's prior-art universe has three tiers:

Critical procedural point that governs the whole §102 analysis: all six Tier‑A references are marked with an asterisk in Google's citation table, i.e., "cited by examiner." A patent issues over its examiner-cited art only after the examiner has concluded those references do not anticipate (and generally do not render obvious) the claims as filed. That creates a strong presumption-in-practice that none of the six alone reads on independent claims 1, 12, 15, or 26 — as confirmed by the element-by-element mapping below.

Second critical point — incorporation by reference. The specification expressly incorporates two of the six cited references (US 4,399,437 to Falck et al. and US 4,730,188 to Milheiser, both identified as teaching contactless/direct-contact programming) plus three commonly assigned pending applications (Ser. Nos. 08/540,631, 08/514,712, 08/316,653, the last now abandoned). Practitioner caution: subject matter a patent expressly incorporates into its own specification is generally treated as part of that specification, which materially undercuts (and in many cases bars) using that same material as anticipatory §102 art for the incorporated subject matter. I flag this as a strong argument rather than a certainty — it is not something I can confirm as applied to this patent from any docket or PTAB record (none exists, per the earlier section).

2. The six examiner-cited U.S. patents

2.1 US 4,399,437 — "Coded information arrangement"

• Full citation: US 4,399,437 A (Falck et al.), Tag Radionics Limited; filed 1980‑05‑19; issued 1983‑08‑16.
• URL: https://patents.google.com/patent/US4399437
• Description (verified from text): A programmable active transponder ("coded tag") on a CMOS IC operating up to 150 kHz. A receiver accepts a 132 kHz interrogation signal; a frequency translator divides it by two to produce a 66 kHz re-transmission carrier, which is modulated by a stored code word read from a code memory register. A second RF receiver is provided solely for programming, feeding an inhibit/memory-protection circuit that prevents accidental re-clocking of the code memory. A companion interrogator (master 132 kHz oscillator + receiver + half-bit format detector) detects the returned signal.
• Relevance / potential §102 mapping:
  • Claims 10, 11, 24, 25, 28 (frequency-of-response limitations): Falck's response carrier is exactly half the interrogation frequency (66 kHz from 132 kHz). Those claims recite a response frequency "at least about one quarter" / "about equal to one half" the excitation frequency. Falck is the most direct single-reference teaching of that numeric relationship in this art.
  • Claims 2 / 16 (write excitation signal + modifying tag operation + non-volatile memory): Falck discloses write-by-RF into tag memory plus a memory-protection/inhibit scheme.
  • Anticipation assessment: Falck cannot anticipate claims 1, 12, 15, or 26 — it discloses no exit-mounted EAS exciter, no alarm trigger, and no sales/transaction data. Because dependent claims incorporate their parent limitations, it cannot anticipate the dependent claims either. Falck is properly an element-reference (a §103 building block and a §112/claim-construction aid), not a §102 reference for this patent. It is additionally subject to the incorporation-by-reference caveat above.

2.2 US 4,656,463 — "LIMIS systems, devices and methods"

• Full citation: US 4,656,463 A (Anders et al.), Intelli‑Tech Corporation; priority/filing 1983‑04‑21; issued 1987‑04‑07.
• URL: https://patents.google.com/patent/US4656463
• Description: Cited in the same EAS/RF-identification field; classified alongside the other front-page references. My search confirmed the citation metadata (multiple later patents list it as "LIMIS systems, devices and methods | 1987‑04‑07 | Anders et al. | 340/572") but the search results did not return its descriptive text, so I am not characterizing its specific disclosure beyond its title and classification. I will not fabricate an element-by-element mapping for it.
• Relevance / potential §102 mapping: On its face it is an article-identification/surveillance system reference relevant to the claim 1 preamble elements (a)–(f) (tag on an article, interrogating field, tag response, detection). It contains no showing of post-sale reprogramming of a tag to suppress an exit alarm, nor of storing sales/transaction data. No independent claim is anticipated; treat as a general-field reference pending verification of its full text.

2.3 US 4,730,188 — "Identification system"

• Full citation: US 4,730,188 A (Thomas A. Milheiser), Identification Devices, Inc. (Boulder, CO); filed 1984‑02‑15 (continuation of Ser. No. 580,401); issued 1988‑03‑08.
• URL: https://patents.google.com/patent/US4730188
• Description (verified): The classic Destron/IDI Passive Integrated Transponder (PIT) system. The reader/exciter has three functional units — exciter 4, signal conditioner 6, demodulation/detection 8 — i.e., precisely the ERW architecture reproduced in the '896 specification (exciter/writer 200, signal conditioner 202, demodulation and detection 204). The tag is powered inductively by a 400 kHz excitation, internally divided down, and returns an FSK signal shifting between 40 kHz and 50 kHz, detected by a PLL tone-decoder and decoded by a microcomputer.
• Relevance / potential §102 mapping:
  • Claim 1 steps (b)–(e) and claim 12's "exciter/reader/writer circuit… generating an RF reading excitation signal": Milheiser discloses the exciter/reader signal chain and the inductively-powered tag that returns a modulated response to the reader.
  • Claim 8 (modulation type — FSK): Milheiser expressly teaches FSK response encoding. (That limitation is in the original spec's modulation list.)
  • Specification cross-reference: the '896 specification expressly cites and incorporates this patent as teaching alternative (direct-contact) programming.
  • Anticipation assessment: Milheiser discloses no surveillance/exit alarm mode, no reprogramming-to-disable, and no sales data. It cannot anticipate any claim of the '896 patent; it is an architecture/§103 reference and is subject to the incorporation-by-reference caveat, since the '896 patent relies on it for programming disclosure.

2.4 US 5,218,343 — "Portable field-programmable detection microchip"

• Full citation: US 5,218,343 A (Anatoli Stobbe); filed 1990‑02‑05; issued 1993‑06‑08.
• URL: https://uspto.report/patent/grant/[5218343](/patent/5218343)
• Description (verified): A battery-free read/write transponder ("detection microchip") with a resonant LC circuit, rectifier, oscillator, counter/decoder, memory circuit, and code generator. The same RF field carries energy, commands, and data to the chip. A decoder + counter + oscillator evaluates the modulation from the read/write device and programs data into the memory, including a write-protected region; the code generator emits FSK/BIPHASE formats. The patent expressly contemplates re-programmable chips storing personal data on customers or patients where the data has to be updated periodically, and contact-free programming without physical contact of a sealed enclosure.
• Relevance / potential §102 mapping — this is the most dangerous single reference:
  • Claims 2 and 16 (reprogram step: generate an RF write excitation signal including operational data and data; activate the tag; modify tag operation; store data in non-volatile memory): Stobbe discloses each of these mechanical steps — field-powered writing, decoder evaluation of the write signal, and storage in (non-volatile) memory.
  • Claim 8 (modulation type): Stobbe expressly discloses FSK and BIPHASE code formats.
  • Claims 3/4/13/14 (storing purchase price/date) and 17/18 (name/removal date): Stobbe's express example of storing customer personal data that is periodically updated is close to the "store data about the transaction/person in the tag" concept, though it is not retail point-of-sale data.
  • Anticipation assessment: Stobbe does not anticipate claims 1, 12, 15, or 26: it has no exit exciter, no RF surveillance excitation/response alarm loop, and no "reprogrammed tag does not generate the alarm-triggering response signal" teaching. It is, however, the strongest §103 reference for the reprogramming-and-data-storage core of the invention when combined with an EAS reference such as Falck, Milheiser, or US 4,656,463.

2.5 US 5,214,409 — "Multi-memory electronic identification tag"

• Full citation: US 5,214,409 A (Beigel), Avid Corporation; filed 1991‑12‑03; issued 1993‑05‑25.
• URL: https://patents.google.com/patent/US5214409 (corroborated in the reference lists of later patents, e.g. https://patents.justia.com/patent/[6646554](/patent/6646554))
• Description: A multi-memory electronic identification tag — the tag is partitioned into multiple memory areas with differing access/write characteristics (e.g., read-only vs. read/write). (I confirmed the citation, dates, assignee, and classification; I was not able to retrieve full descriptive text within this session, so I am limiting the characterization to the title's clear subject matter — a tag with a partitioned memory — and not asserting details I did not see.)
• Relevance / potential §102 mapping:
  • Claim 12's "each transponder tag including a memory… a programming mode wherein said controller stores said operational and sales data in said memory, and a reading mode wherein said controller generates an RF reading response signal containing said sales data": the multi-memory structure (protected vs. rewritable fields) is directly on point for the memory architecture that lets the '896 tag carry reprogrammable operational data alongside sales data.
  • Claims 2/16 (non-volatile memory): memory-partition references.
  • Anticipation assessment: no exit alarm, no surveillance mode, no sales/transaction data → cannot anticipate any claim; a §103 element-reference for the tag-memory limitations of claim 12.

2.6 US 5,499,017 — "Multi-memory electronic identification tag"

• Full citation: US 5,499,017 A (Beigel), Avid; filed 1992‑12‑02; issued 1996‑03‑12.
• URL: https://patents.google.com/patent/US5499017 (corroborated at https://patents.justia.com/patent/6646554)
• Description: Same subject matter as 5,214,409 — an electronic identification tag with multiple memory areas. Notably, its 12 March 1996 issue date is only ~5.5 months before the '896 filing (26 Aug 1996), making it a §102(a)/(e)-qualified reference. (Same verification limitation as 2.5 — citation confirmed, full text not retrieved this session.)
• Relevance / potential §102 mapping: same as 2.5 — relevant to the multi-memory/programming-mode limitations of claim 12 and to claims 2/16. No independent claim anticipated; §103 element-reference.

3. The non-patent literature (Tier B)

• Description (verified): datasheets for the Temic e5550 Read/Write Identification IC (IDIC), the chip the '896 specification designates as the preferred transponder tag. The '896 description attributes to it: the analog front end (rectifier, clock extraction, load-switching, field-gap detection); the bitrate generator (RF/8 through RF/128, with RF/2 and RF/4 added "with slight modification"); the mode register; the write decoder; the password-controlled read/write access; the EEPROM memory; and selectable FSK/PSK/Manchester/biphase modulation.
• §102 mapping: these datasheets are the technical source for the bitrate/frequency limitations of claims 10, 11, 24, 25, 28 and the modulation limitation of claims 8/22, and for the password concept referenced in the specification.
• Critical caveat: the '896 patent expressly incorporates these two documents by reference ("both of which are incorporated herein by reference"). Because the patent's own disclosure of the preferred tag depends on them, using them as anticipatory art for that same subject matter runs into the incorporation doctrine discussed in §1 — the reference has effectively been made part of the patent's specification. They remain usable as general art for anything the patent did not actually incorporate, but they are poor §102 references against the claims that read on the e5550 architecture.

4. References cited inside the '896 specification but not on the front page (Tier C)

• U.S. Ser. No. 08/540,631, "Contactless Programmable Radio Frequency Transponder," filed 1995‑10‑11.
• U.S. Ser. No. 08/514,712, "RF Identification Tag and Contactless Method of Programming the Same," filed 1995‑08‑14.
• U.S. Ser. No. 08/316,653, "High Field Programmable Transponder System and Method," filed 1994‑09‑30, now abandoned.

All three are commonly assigned applications expressly incorporated by reference. Because they were unpublished pending applications as of the '896 filing date, they are not §102(a)/(b) printed publications, but they are §102(e) candidates to the extent they later published or issued (the "High Field Programmable Transponder System and Method" subject matter later appears as EP 0 704 815 A2, and later RFID patents in the assignee family cite it). Pre-AIA §102(e) requires the reference application to have been filed before the '896 invention date and to have issued/been published; the earlier filing dates here (1994–1995) satisfy the temporal element, so 08/316/653 (Sept. 1994) is the most plausible §102(e) reference of the three.

5. Overall §102 / §103 assessment and claim mapping

Bottom-line answer to the task's specific question ("which claim(s) does each reference potentially anticipate under §102"):

No single cited reference anticipates any claim of US 5,874,896 — independent or dependent. Every independent claim (1, 12, 15, 26) requires the combination of (i) an exit-located tag exciter generating an RF surveillance excitation signal, (ii) an inductively-powered tag returning an RF surveillance response signal, (iii) an alarm triggered by that response, and (iv) reprogramming the tag and storing sales/transaction data upon removal authorization such that the tag no longer triggers the alarm. No cited reference contains all four elements arranged as claimed:

Where the real invalidity exposure lies (practice note, not a §102 conclusion): the genuine threat to this family is §103, not §102 — most plausibly Stobbe (US 5,218,343) + one EAS reference (Falck, Milheiser, or Anders), with Beigel (US 5,214,409 / 5,499,017) supplying the partitioned-memory architecture of claim 12. The most vulnerable specific limitations are the numeric response-frequency limitations (claims 10/11/24/25/28, taught squarely by Falck's 132 kHz → 66 kHz ÷2 response) and the FSK/modulation limitations (claims 8/22, taught by Milheiser and Stobbe). That is presumably exactly why these six references were cited and distinguished during the 1996–1999 prosecution.

6. Confidence and verification limits

• High confidence: citation identities, dates, assignees, titles, examiner-citation status, and the express-incorporation language — all read directly from the patent's own full text and corroborated at Google Patents and uspto.report.
• Medium confidence: the descriptive content of US 4,399,437 and US 4,730,188 (retrieved and read from search results) and US 5,218,343 (retrieved from uspto.report).
• Low confidence / not verified: the specific disclosures of US 4,656,463 (Anders), US 5,214,409, and US 5,499,017. My searches confirmed their bibliographic data but did not return their specification text, so I have deliberately not asserted element-by-element mappings for them beyond what the titles and claim-mapping logic support. If a formal claim-chart is needed, those three should be pulled in full text before any §102/§103 position is taken on them.
• The patent is expired (anticipated expiration 2016‑08‑26), so this prior-art analysis is of historical/validity-interest value only; per the earlier litigation section, no IPR/PTAB/CAFC proceeding names US 5,874,896.

Sources: Google Patents US5874896A (https://patents.google.com/patent/US5874896/en); uspto.report/patent/grant/5874896; Google Patents US4399437 (https://patents.google.com/patent/US4399437); Google Patents US4730188 (https://patents.google.com/patent/US4730188); uspto.report/patent/grant/5218343; Justia reference list at https://patents.justia.com/patent/6646554 (corroborating the Beigel and Stobbe citations).

I'll verify the cited prior-art references against live sources before building the § 103 analysis.

Let me fill in the remaining gaps on Milheiser, the Intelli-Tech LIMIS patent, and the Temic e5550 NPL.

Obviousness Analysis — U.S. Patent No. 5,874,896 under 35 U.S.C. § 103

Patent: US 5,874,896 A — "Electronic anti-shoplifting system employing an RFID tag" (App. No. 08/703,045; filed and priority 1996‑08‑26; granted 1999‑02‑23; inventors Lowe & Small; original assignee Palomar Technologies Corp.; status Expired – Fee Related, anticipated expiration 2016‑08‑26).

Scope note on sources. I am building on the verification, claim-summary, and litigation sections already generated (which I do not repeat). Per the task instruction I use the Prior Art section of the Google Patents page — i.e., the References Cited / Patent Citations (6), Non‑Patent Citations (4), and Family Cites Families (GB9202831D0) entries. I explicitly do not treat the "Cited By (112/212)" entries as prior art, because nearly all post‑date the 1996‑08‑26 critical date; a handful are pre‑1997 but their priority dates still fall on/after the critical date or they are cited-by-third-party only. That distinction is flagged again in § VIII.

Bottom line up front. On the art cited on the face of this patent, claims 1–9 and 15–25 in the two "sales outlet / controlled area" method families are, in my assessment, strongly vulnerable to § 103 challenges; claim 12–14 (system) is vulnerable; and the frequency‑defined family (claims 10–11, 24–25, 26–28) is the most defensible, though it is seriously weakened by the patent's own specification, which concedes the key parameter values are reached by "slight modification" of a commercial part whose datasheet is cited as prior art.

I. Legal framework applied

1. Graham v. John Deere Co. v. Laitram, 383 U.S. 1 (1966) — scope and content of the prior art; differences between the prior art and the claims; level of ordinary skill; secondary considerations.
2. KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) — a combination of known elements is obvious where the improvement is a predictable use of prior‑art elements according to their established functions; "if a technique has been used to improve one device, and a person of ordinary skill in the art would recognize that it would improve similar devices in the same way, using the technique is obvious." A design need, market pressure, or a finite number of identified, predictable solutions supplies the motivation. A mere change in a known parameter, with a recognized benefit, is generally obvious.
3. In re Kahn / KSR "articulated reasoning" requirement — motivation must be something more than the hindsight teaching of the claims.
4. § 102(b) admissions in the specification are prior art to the patent's own claims.

Each reference below on the face of the '896 patent issued well before 1996‑08‑26, so each is § 102(b) prior art; the pre‑1996 dates also mean no § 102(a)/(e) complexity for those references.

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

As of August 26, 1996, a person of ordinary skill in the relevant art would have had: a bachelor's degree in electrical engineering (or equivalent), plus roughly 2–4 years of experience designing inductively coupled, field‑powered RF identification transponders and/or electronic article surveillance (EAS) systems — i.e., familiarity with load/field modulation, field‑synchronous data rates, EEPROM‑based read/write tag ICs, and exciter/reader architectures. The '896 specification itself is written at that level (it treats the tag IC, the exciter, the reader front end, and the bandpass/bandstop conditioning as known building blocks). None of the claims recites software or algorithmic subject matter requiring a different skill level.

This matters because a substantial portion of the '896 disclosure is a system‑integration exercise over known tag ICs (see the Temic e5550 NPL citation), which is precisely the kind of predictable combination KSR makes obvious.

III. Scope and content of the prior art on the face of the '896 patent

§ 102(b) admissions inside the '896 specification itself. The '896 background and "Description of Preferred Embodiments" expressly describe as known: "electronic article surveillance (EAS) systems that include transponder tags attached to each article of merchandise … one or more electronic readers positioned at exits … the reader generates an excitation signal that powers the transponder tag [and] the transponder tag generates a response signal that triggers an alarm"; and "If a customer purchases the merchandise, the transponder tag is disabled or removed." These admissions place the entire EAS detect‑at‑the‑exit/alarm function (claims 1, 12, 15, 26 preamble elements) in the prior art.

IV. Claim‑by‑claim mapping and the proposed combinations

Combination 1 (primary) — Stobbe '343 + Falck '437 + Avid '409/'017 + admitted EAS art

This combination targets claims 1–9 and 15–25.

Motivation to combine (articulated, non‑hindsight):

1. Same field, same problem class. Stobbe '343, Falck '437, Milheiser '188 and Avid '409/'017 are all inductively coupled, field‑powered RF transponder systems in the same classification area (Avid's family is classified in G08B 13/24, the same EAS subclass as the '896 patent). Combination of references within a single field of endeavor is presumptively motivated.
2. The re‑write technique was recognized as generally applicable. Stobbe states the object is a field‑programmable chip "for use as an identification device," and expressly names applications beyond simple ID — periodic data updating, cash‑free payment transactions, object‑identifying data. Falck states re‑coding is "of particular advantage where such data is required for a short time only or where the data contains some element which needs to be updated periodically." Avid states the user "may utilize the reprogrammable memory option to recode the data assigned to particular objects." Each reference therefore supplies the reason to apply re‑writing to a merchandise tag.
3. The data content is already in the references, not in the claim. Falck already names price and date as appropriate tag data. The '896 claim element "storing sales data … purchase price … purchase date" is therefore not a new data‑type invention; it is the use of a known data field for a newly named purpose.
4. The "does not respond after reprogramming" function is an inherent consequence of Stobbe's mode change. Stobbe's mode register can change FSK frequencies, baud rate, and read/write vs. read‑only mode; any such change causes the chip to respond differently (or not at all) to a fixed exciter field. Reaching the '896's "reprogrammed tag does not generate the RF surveillance response signal" requires only the routine choice of a mode the fixed exciter cannot detect — a predictable result of a known parameter change (KSR).
5. A finite, predictable set of solutions. Stobbe enumerates exactly which operational parameters can be changed (baud rate, modulation format, FSK frequencies, read/write vs. read‑only). The '896 claims recite a subset of that list (response frequency, excitation frequency, modulation type). Where the prior art discloses a small, enumerated set of parameters and the claim selects among them, KSR's "finite number of identified, predictable solutions" rationale applies.

Combination 2 — Avid '409/'017 (primary) + Stobbe '343 + Milheiser '188 + admitted EAS art

This combination targets claim 12–14 (system) and, secondarily, claims 1/15.

Claim 12 requires a single tag with three modes: surveillance (emit RF surveillance response signal), programming (store operational + sales data in memory), and reading (emit RF reading response signal containing the sales data), with an ERW that receives the reading response and outputs sales data to an output device.

Motivation: Avid '409/'017 already supplies the single‑tag, dual‑source (programmer/reader) architecture that claim 12 recites as "modes"; the '896 system claim adds nothing structurally beyond combining that known tag with the admitted EAS exit‑detection loop. The motivation is the market/design pressure on retail EAS recited verbatim in the '896 background (return of sale‑priced goods at full price; expired price guarantees; cross‑store return fraud), coupled with Avid's own teaching that alterable tag memory can carry object‑specific data that the user may recode and later read to verify. Under KSR, the '896 background's own recitation of these business problems is admissible evidence of the design need driving the combination.

Combination 3 — Stobbe '343 + Milheiser '188 + admitted EAS art (alternative for claims 1–9 / 15–25)

If a tribunal does not accept Falck '437 and Avid '409/'017 as teaching the data content, this narrower three‑reference combination still reaches most method claims: Stobbe supplies field re‑programming of operational parameters and memory storage; Milheiser supplies the passive, field‑powered, load‑modulated tag attached to an article and responding at a selected frequency/modulation; the '896 specification's own EAS admissions supply the exciter, alarm, and the "deactivate or remove on purchase" model. The only element left untaught is the specific "consumer‑transaction" semantics of the stored data — a limitation the claims recite at a high level of generality ("sales data … purchase price"), which Falck '437 would fill.

Combination 4 — Frequency‑defined claims 10–11, 24–25, 26, 27, 28

Claim 26 (independent) requires the RF surveillance response signal to have a frequency "at least about one quarter" the excitation frequency; claim 28 narrows to "about one half"; claims 10/11 and 24/25 mirror these in the claim‑1 and claim‑15 families.

The most damaging material here is the '896 specification itself, which states:

"Typically, the bitrate generator allows selection of the following bitrates: RF/8, RF/16, RF/32, RF/40, RF/50, RF/64, RF/100, and RF/128 … With slight modification of the bitrate generator 128, additional bitrates of RF/2 and RF/4 are provided to maximize signal power of the RF surveillance response signal…"

and the specification elsewhere equates the "frequency" of the response signal with a "selected bitrate."

Three consequences:

1. The claimed values (≥ RF/4; ≈ RF/2) are, on the applicant's own characterization, reached by a "slight modification" of the bitrate generator of the Temic e5550 — a commercial part whose datasheet is cited as prior art in this patent. That is close to a self‑inflicted obviousness admission for a mere parameter change (KSR: change in a known parameter with a recognized benefit is obvious).
2. The stated benefit — maximizing signal power and hence detection range — is a functional result that the '896 specification asserts as the reason for the choice. A substitution justified by a known, predictable engineering trade‑off (higher bitrate ⇒ more power ⇒ greater read range) is the paradigm KSR case.
3. Supporting art: Stobbe '343 teaches that the data pulse rate is derived by dividing the RF field frequency and calls that derivation "obvious," and provides a mode register that selects baud rate; Milheiser '188 teaches FSK response frequencies and bit cells derived as integer fractions of a 400 kHz interrogation signal.

Candor about the strength of this combination: Milheiser's disclosed response carrier (40/50 kHz from a 400 kHz excitation, i.e., ≈ RF/8–RF/10) does not by itself meet "at least RF/4," and I did not locate a reference in the cited set that fixes RF/2 as a response frequency for this purpose. This limitation is therefore the patent's best non‑obviousness argument, and its presence as an independent claim (claim 26) suggests the applicant treated it as the point of novelty. A challenger would have to lean heavily on the specification's "slight modification" concession plus the KSR parameter‑change rationale; a patent owner would argue that the prior art taught away from high response frequencies (Milheiser's relatively low FSK tones) and that a new and useful detection‑range result flows from the choice.

V. Consolidated motivation‑to‑combine summary (KSR rationales invoked)

VI. Differences between the prior art and the claims — honest accounting

1. The refund‑control application — using the tag's stored purchase record to adjudicate a return at the point of sale — is, as a business method framing, not squarely disclosed in any single cited reference. But the '896 claims do not recite the refund adjudication; they recite only "storing sales data … purchase price … purchase date" (claims 3, 4, 13, 14) and "transaction data … name … removal date" (claims 17, 18), all of which Falck '437 already contemplates as tag data. The commercial narrative in the specification is broader than the claims, and only the claims govern § 103.
2. "Operational data" is used in claims 6–8 and 20–22 without antecedent introduction in their independent claims — this is a § 112(b) problem, not a § 103 save. A challenger would also note it shows the claim set was drafted by reference to the specification's FIG. 2 rather than to a distinct inventive concept.
3. Claim 12's three named "modes" are label‑level characterizations of functions Avid's tag performs; the claim adds no structural limitation beyond "a memory and a controller." A tribunal applying claim construction would likely read "surveillance mode"/"programming mode"/"reading mode" as functional descriptions rather than distinct circuits, which materially weakens the claim against the Avid combination.
4. No secondary‑considerations record that I can verify. I found no evidence (in the sources I retrieved) of an established nexus between any commercial product and this patent — the assignment chain is a pure corporate transfer (Palomar → HID Corp. 1998‑03‑06) plus a 1999 security agreement terminated in 2001. A patent owner would need product‑level evidence (e.g., HID/Palomar EAS‑RFID products practicing the claims), unexpected results, or industry praise attributable to the claimed combination to rebut the prima facie case. Long‑felt need is arguable from the '896 background's own recitation of the return‑fraud problem, but the "need" was satisfied in the art by 1983–1993 re‑writable tag technology.

VII. Overall assessment by claim group

VIII. Flags, contradictions, and uncertainties (per the operating rules)

1. Contradiction with a previously generated section — password. The earlier summary lists "password-protected writing" as among the "recurring themes" of the dependent claims. Checking the actual claim set (claims 1–28 as they appear on the Google Patents page): no claim recites a password. Password protection appears only in the specification ("Preferably, the write circuit of the transponder tag 58 requires a password… The ERW circuit 60 outputs the password as an initial portion of the write excitation signal 186"). I flag this because a password limitation, if claimed, would have been a meaningful obviousness obstacle (Falck '437's program‑protection circuit and Avid's keyed write mode would still likely neutralize it) — but it is not in the claims.
2. Contradiction (minor) in claim‑dependency identification. The earlier summary refers to the antecedent‑basis quirk at "claims 21–22"; the actual affected claims are 20–22 (claim 20 = response frequency; 21 = excitation frequency; 22 = modulation type), depending from claim 15, with "operational data" first appearing in dependent claim 16. The claim‑1‑family equivalent is claims 6–8 (with "operational data" first appearing in claim 2).
3. NPL date discrepancy (do not harmonize). The '896 specification cites the Temic Preliminary Information as dated Dec. 12, 1995; the Google Patents Non‑Patent Citations table lists it as Dec. 8, 1995, issued by "Temic Telefunken Semiconductors, pp. 93–106." I report both rather than correcting either. Note that either date makes the document a § 102(a) reference (within one year of the 1996‑08‑26 critical date), whereas the Oct. 13, 1994 Preliminary Product Features is § 102(b) art.
4. "Prior Art section" scope caveat. The Google Patents page's "Cited By (112)/(212)" and "Families Citing this family" tables are not prior art to the '896 claims — nearly all entries post‑date 1996‑08‑26 (e.g., the Checkpoint US 6,025,780 "RFID tags which are virtually activated and/or deactivated," 1997‑07‑25 priority, and the Sensormatic US 6,169,483 self‑checkout/EAS system, 1999‑05‑04). They are useful only as secondary/objective evidence of what the field regarded as the natural next step after '896. I have not used them as § 102/103 references. Similarly, the "References Cited" beyond the six patent citations on the '896 face (e.g., US 4,514,731 to Falck, a continuation‑in‑part of the '437 disclosure) are genuine pre‑1996 art but are not among the six references the Examiner cited on this patent — I mention US 4,514,731 only as a related Falck disclosure, not as a face‑of‑patent reference.
5. Reference I could not fully retrieve. US 4,656,463 (Intelli‑Tech "LIMIS systems, devices and methods") — my searches returned the citation record and its use as a cited reference in many later RFID patents, but not the specification text. I therefore do not rely on it for any affirmative teaching in the combinations above. If full text is needed, retrieve from https://patents.google.com/patent/US4656463/en or the USPTO full‑text database.
6. No litigation/PTAB history to reconcile. Consistent with the previously generated litigation section, I found no case naming US 5,874,896. The obviousness analysis above is therefore academic/retrospective, and I have not attempted to determine whether these references were ever the subject of a validity challenge. The patent expired 2016‑08‑26.
7. Confidence levels. High confidence in the bibliographic data, claim text, and the content of Stobbe '343, Falck '437, Milheiser '188, and the Avid '409/'017 family (I retrieved specification text for each). Low/uncertain confidence in the substance of US 4,656,463 and GB 9202831 D0, which I could not retrieve. Medium confidence in the ultimate legal conclusion — § 103 outcomes are fact‑intensive and dependent on claim construction ("operational data," "modes," "sales data") and on any secondary‑considerations evidence not present in the sources I consulted.