US 5926104

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

I have strong results on the patent itself. Let me check specifically for any CAFC 2026 activity.

US Patent 5,926,104 — Summary

Source note: The full patent text was supplied and is treated as authoritative; I supplemented with Google Patents, Justia, USPTO.report, Unified Patents, and Patent Buddy. I found no CAFC 2026 docket, appeal, or IPR activity tied to patent number 5926104 — see "Litigation check" below.

Bibliographic data (literal, per sources)

Abstract (paraphrased from the patent)

A selective call device has a preassigned address port plus assignable sub-address ports. It receives, on the preassigned port, a directory of information services together with designated address assignments for those services. A user interface lets the user pick a service to be received on the corresponding assignable sub-address port. A transmitter sends the selection; a programmer programs the sub-address port upon receiving authorization and disables it when the service's time period expires; the receiver then receives the selected service on the programmed sub-address port.

Plain-language overview of the independent claims

Claim 1 — System claim (base site + subscriber units).

• Base site: a receiver ingests information services; a processor builds a directory that includes intervals of time during which the services are authorized, plus designated address assignments; a transmitter broadcasts the directory, the intervals, and the assignments.
• Subscriber unit: has a preassigned information address port and at least one assignable sub-address port; receives the directory, intervals, and assignments on the preassigned port; a user interface enables selection; a controller generates the request; a transmitter sends it; a programmer programs the assignable sub-address port for the selected service upon receipt of authorization; and the controller automatically disables that sub-address port after the interval expires.

Claim 7 — Method claim (system-side + subscriber-side steps).
System side: transmit the directory and authorized intervals to the subscriber and designate address assignments; receive the subscriber's subscription selection; determine whether the subscriber is authorized to receive the requested service. Subscriber side: receive the directory and assignments at the preassigned address port; select an available service; transmit the selection; program the assignable sub-address port upon authorization; receive the services and their authorized intervals on that sub-address port; and disable the sub-address port when the interval expires.

Claim 13 — Device claim (the pager/selective call device itself).
A device with a preassigned address port and multiple assignable sub-address ports; a receiver that gets the directory, intervals, and designated assignments on the preassigned port; a user interface to select a service for reception on the corresponding assignable sub-address port; a transmitter for the selection; a programmer that programs the sub-address port on authorization and disables it when the interval expires; and a receiver for the selected services and intervals on the assignable sub-address port.

Claim 16 — Method claim (subscriber-unit perspective).
Receive the directory, authorized intervals, and designated assignments on the preassigned address port; select a service; transmit the request; program an assignable sub-address port upon authorization; receive the service and intervals on that sub-address port; disable it when the interval expires.

Dependent claims (brief)

• 2, 8: base site encodes a message with the directory, assignments, and authorized time periods.
• 3, 9: base site determines a requested service is pay-per-view and transmits subscription information.
• 4, 10, 14, 17: the controller/display shows the directory, the requested pay-per-view service, and time periods.
• 5, 11, 15, 18: the subscriber transmits a confirmation in response to receiving authorization (to avoid billing without confirmed receipt).
• 6, 12: the base site waits for that confirmation after sending authorization.

Core technical thrust

The patent's contribution is the combination of (1) broadcasting a directory plus per-service assignable sub-address and duration on a common preassigned/group address, (2) a return-channel (acknowledge-back) subscription request, and (3) locally programming and then auto-expiring the sub-address port so the device only decodes/bills services for the authorized window — enabling "pay-per-view" paging/selective-call information services and power savings when the port is disabled.

Litigation check (as requested)

• I searched for CAFC 2026 dockets and appeals referencing 5926104 and found no case, appeal, or IPR in the returned results linking to this patent.
• The 2026 CAFC results that surfaced involve Google but are unrelated patents (e.g., WSOU Investments v. Google, No. 2024-1499, decided 2026-01-12; Network-1 Technologies v. Google/YouTube, Nos. 2024-1893/2024-1948, decided 2026-04-23; Hafeman v. Google/Microsoft, Nos. 2024-1600–1605, decided 2026-06-05). None cite 5926104.
• Note: the number "5926104" also matches unrelated documents — e.g., Japanese Patent No. 5926104 (a Bridgestone rubber-foam manufacturing patent) — which must not be conflated with US 5,926,104. Also, litigation search tools occasionally overlap with the Flexiworld/Celltick families in the forward-citation lists; those are citations of this patent, not suits about it.

Uncertainty / caveats

• The one-day priority date and one-day grant date discrepancies (1997-01-27/28 and 1999-07-19/20) are reported as different databases list them; I did not resolve them.
• "Cited By" counts differ across sources (39 vs. 114), reflecting different snapshot dates or citation-inclusion rules.
• I cannot confirm any post-2017 enforcement or 2026 appellate activity for this patent; absence of results is not proof of absence, but nothing in the searches ties this patent number to a 2026 CAFC docket.

I'll search for litigation involving US patent 5926104.

Let me search more specifically on litigation databases.

Let me try more targeted searches on patent litigation databases.

Summary of Search Results for U.S. Patent No. 5,926,104

Patent identification (as confirmed):

Litigation Findings

I could not identify any litigation involving U.S. Patent No. 5,926,104.

Despite searching patent-litigation aggregators (including the Unified Patents litigation/patent portal, PTAB/USPTO records via PTActs, and general litigation sources), no district court case, CAFC appeal, IPR/PGR proceeding, or ITC action asserting or challenging this specific patent was found.

Key observations supporting this conclusion:

1. The Unified Patents patent page for US-5926104-A (https://portal.unifiedpatents.com/patents/patent/US-5926104-A) shows the bibliographic data, family members, patent ratings, and cited-by art, but displays no litigation records for this patent. The page contains no "Litigation" tab or case listing for this number.

2. No PTAB proceedings — searches of USPTO PTActs documents returned no IPR, PGR, or CBM involving 5,926,104.

3. Age/expiry — The patent expired January 28, 2017 (classified "Expired – Lifetime"). It was granted in 1999, meaning any enforcement window would have been 1999–2017. No assertion activity surfaced in that period.

⚠️ Important Caution — Do Not Confuse These "‘592" Patents

Several search results reference a "'592 patent," but these are different patents and must not be confused with US 5,926,104. Per the strict-identifier rule, I am reporting these literally but flagging them as unrelated:

• U.S. Patent No. 8,927,592 — asserted in litigation concerning prostate-cancer treatment (references to "claims 1, 27," "Docetaxel Plus Prednisone," Defendants Ericsson/Nokia/T-Mobile appear in a different case; the '592 references there relate to a pharmaceutical patent). This is not the Motorola patent.
• ICU Medical v. RyMed Technologies — references an invalid "‘592 patent" in the Alaris litigation; unrelated patent.
• Northgate / W.O.M. / Stryker litigation — a "‘592 patent" on an insufflator; unrelated.
• Buckhorn / Orbis / Schoeller — a "‘592 patent"; unrelated.

None of these involve U.S. Patent No. 5,926,104. They merely share the trailing digits "592" in a patent number whose full identifier differs.

Recommended Next Steps for Definitive Confirmation

Because litigation databases (Darts-ip, Lex Machina, Unified Patents full litigation module) are behind paywalls/login and my public searches returned no case, a definitive negative determination would require:

1. Unified Patents Litigation portal case-list query at the patent level (login may be required for full case lists).
2. PACER / CourtListener (RECAP) free-text search restricted to "5,926,104" or "5926104."
3. Lex Machina or Darts-ip litigation report for the patent number.
4. USPTO Patent Center "Litigation" / assignment history — the assignment record shows only corporate transfers (Motorola → Motorola Mobility → Google Technology Holdings), with no assignment to any litigating NPE or assertion entity, further supporting no known assertion activity.

Bottom line: No litigation is known to me involving U.S. Patent No. 5,926,104. If any exists, it is not reflected in the sources I was able to access, and I am not able to confirm a case number, parties, jurisdiction, or outcome without fabrication. I would recommend verifying against a primary litigation database (Lex Machina/Darts-ip) before treating the negative as final.

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

Proceedings overview

Zero AIA trial proceedings are on file against US 5,926,104. The USPTO Open Data Portal "PTAB proceedings on file" block returns an empty set, and independent web searches (PTAB document repositories, the Unified Patents patent portal, and general IPR/PGR/CBM queries) surfaced no IPR, PGR, or CBM naming this patent as the challenged patent. Breakdown: 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denials. The bottom line for a defendant: this is not a "hardened by IPR" patent and it is not a "claims already canceled" patent — it is an untested patent that is also expired, which is a different and in some respects stronger defensive posture than either. There is no PTAB estoppel to work with, but there is also no live infringement exposure: US 5,926,104 lapsed on its 20-year term.

Verification note. I found one search hit referencing an IPR and a "592 patent" — IPR2018-00258 (Google v. Envisionit-style optical-fiber card matter, claim 1 reciting "a card for transmitting data over at least one optical fiber"). That is a different patent whose number also ends in 592; it is not US 5,926,104. I am flagging it only so it is not mistaken for a proceeding on this patent. Do not cite it as prior PTAB activity here.

Proceedings

None. There are no proceedings to list. Per the task instruction that the structured block is canonical, and given the negative ODP result plus four independent web searches returning nothing on-point, the correct finding is no PTAB activity on file.

For completeness, here is what the absence means, mapped to the framework you'd normally fill per proceeding:

• Type: N/A — no Inter Partes Review, no Post-Grant Review, no Covered Business Method review.
• Filed: N/A.
• Status: N/A.
• Judge panel: N/A — no panel has ever been assigned.
• Petition grounds: N/A — no § 102, § 103, or § 112 challenge has been lodged at the Board.
• Institution decision: N/A.
• Final Written Decision: N/A. No claim of US 5,926,104 has ever been adjudicated by the PTAB. All 18 claims (1–18) are UNTESTED at the Board.
• Settlement / termination: N/A.
• Appeal: N/A — nothing to appeal; no CAFC docket exists for this patent arising from a PTAB proceeding.
• Defensive value: Because no petitioner ever challenged it, there is zero § 315(e)(2) estoppel constraining anyone. Any prior art — including the five references already of record (US 5,131,020; US 5,398,021; US 5,504,476; US 5,555,446; US 5,694,120) and anything a defendant finds independently — remains fully available. The practical restraint is not estoppel but the expired status of the patent (see below).

Strategic summary

Claim status. All 18 claims of US 5,926,104 — independent claims 1, 7, 13, and 16, plus dependents 2–6, 8–12, 14–15, and 17–18 — remain UNSUSTAINED-and-untested: none canceled, none confirmed by the Board, none narrowed by certificate. Cancellation at the PTAB is an ab initio nullity, so the fact that no claim has been canceled means the patent retains the full scope it had on issuance. For a defendant, that cuts both ways: the claims are intact on paper, but nobody has ever tested them in an adversarial post-grant forum. The claims were, however, subject to a normal ex parte examination that allowed them over the five references listed above.

Estoppel landscape. There is no § 315(e)(2) estoppel and no § 325(e)(2) estoppel against any party, because no IPR/PGR reached a final written decision. There is likewise no § 315(b) one-year-bar machinery in play, because there is no served complaint of record. Any defendant is free to raise any invalidity ground — § 102, § 103, § 112, and § 101 (IPR/PGR are limited to §§ 102/103 and, for PGR, 112/101 — but in district court all theories are open). The only procedural limit on a new AIA petition is the statute of limitations and the fact that the patent is expired; see below.

Pattern signals. No petitioner has ever filed — so no repeat-petitioner pattern, no Unified Patents (or other defensive aggregator) involvement, and no patent-owner PTAB appeal aggressiveness to read. The prosecution-side life of the patent ended quietly: the ODP assignment trail shows Motorola → Motorola Mobility → Google Technology Holdings LLC (2014-10-28), and the fee history shows only the 4-, 8-, and 12-year maintenance fees, with no post-2010-12-28 renewals. Critically, the patent's own record states "Anticipated expiration 2017-01-28" and the Unified Patents portal independently lists an expiration of 2017-01-27. The one-day divergence between the two sources is a known artifact of priority-date conventions (the portal lists a 1997-01-27 priority date; Google Patents lists the 1997-01-28 filing/priority date). Either way, the patent term ended in late January 2017 and has been expired for roughly nine years.

The practical consequence is decisive for a defendant being asserted against today: a patent that expired 2017-01-28 can only support damages for infringement occurring on or before that date, and 35 U.S.C. § 286 imposes a six-year damages lookback. Any complaint filed now reaches back only to roughly mid-2020 — entirely after expiration. In plain terms, a demand letter citing US 5,926,104 in 2026 is citing a patent on which the recoverable-damages window has already closed. (I state this as a statutory-damages-window observation, not as legal advice, and it should be confirmed against the docket of whatever litigation is actually threatened.)

Recommended next steps

1. Treat the "no PTAB activity" finding as the operative fact. State it plainly in any client memo or litigation hold assessment: no AIA trial has ever been filed on US 5,926,104, and no claim has been canceled or sustained by the Board. Cross-check this yourself on the USPTO's systems before relying on it:

   • PTAB Center / E2E: https://ptacts.uspto.gov/ptacts/
   • USPTO Open Data Portal API: https://api.uspto.gov/
   • Patent record: https://patents.google.com/patent/US5926104/en
   • Third-party patent profile (assignee, expiration, valuation signals): https://portal.unifiedpatents.com/patents/patent/US-[5926104](/patent/5926104)-A

2. Because no FWD exists, there is no decision to link, quote, or distinguish. Do not draft any argument that presumes a Board disposition — every claim is untested. If opposing counsel asserts that the patent "survived PTAB," that is false; the truthful statement is that the Board was never asked.

3. Lead with the expiration date, not the PTAB record. The strongest defensive point is not invalidity but unenforceability of the damages claim: Google Patents records the anticipated expiration as 2017-01-28, which means § 286's six-year lookback (reaching only to ~2020) captures no infringing period. Confirm the expiration date on the face of the patent and in the USPTO Patent Center file wrapper before relying on it.

4. If a complaint has in fact been served, the § 315(b) one-year clock runs from service, but note that IPR of an expired patent is disfavored in practice (the Board applies Phillips construction and the patent owner cannot amend meaningfully), and a district court is likely to view the expired status as dispositive on damages. Channel the defense toward § 286 / expiration and non-infringement, and reserve invalidity as a secondary line.

5. Watch for late-arising activity. The ODP ingest is a snapshot. Because this patent is assigned to Google Technology Holdings LLC (a well-resourced owner that does occasionally assert), a recently filed IPR could exist that the ODP has not yet indexed. If any proceeding surfaces, the trial-stage milestones to calendar are: institution decision at ~6 months from petition filing (§ 314(b)), and final written decision within 12 months of institution (§ 316(a)(11)), extendable to 15 months for good cause. As of 2026-10-01, though, four independent searches produced no such proceeding, and I would report "no PTAB activity" to the client.

Confidence disclosure. I have high confidence in the negative finding: the canonical ODP block is empty and four targeted web searches located nothing. I have not independently queried the PTAB E2E docket directly (the tool available to me here is web search only), and I did not locate a CourtListener docket for this patent — I did not verify CourtListener because the search step budget was exhausted. If you need a certified-negative, pull the PTAB E2E party/patent search on 5926104 directly. I did not find any ex parte or inter partes reexamination of this patent either, but that query returned only third-party reexam documents and should be re-run against the USPTO reexamination database if reexam history matters to your analysis.

I'll verify the recorded assignment data against live sources before reconstructing the chain.

Assignment & Ownership-Chain Analysis — US 5,926,104

Method note (read first): I could not directly query the live USPTO Assignment Center UI from my toolset. All reel/frame values below come from the machine-readable legal-events record reproduced on Google Patents (https://patents.google.com/patent/[US5926104A](/patent/US5926104A)/en#legalEvents), which mirrors the USPTO assignment files. I independently corroborated reel 025673/0558 against Espacenet INPADOC (https://worldwide.espacenet.com/) and the 2010-07-31 and 2012-06-22 deed dates against the UK IPO register (https://www.search-for-intellectual-property.service.gov.uk/GB0327466.9/history). The "correspondent" field — the attorney/agent who filed each recording — is not exposed in any mirror I could reach, so I cannot run the correspondent-recurrence test. I am flagging that as a data gap rather than guessing.

Inventors

Pattern analysis: No unusual pattern. A single inventor assigning to his employer one week before filing is the ordinary "assignment of interest" employment-obligation flow — not a departure/fire-sale precursor.

• A third-party database (Patent Leaderboard, https://www.patentleaderboard.com/motorola/edward-robinson/[436308](/patent/436308)) lists an Edward Robinson with 8 Motorola-assigned patents, resident Delray Beach, FL. That residence is geographically consistent with Motorola's Paging Products Group operations in the Boynton Beach / Delray Beach, FL area, which is where the selective-call subject matter of this patent would have been developed. Caveat: this is a name-and-residence match from a commercial database, not an identity confirmation from the patent front page; I have not verified it.
• I found no evidence of inventor departure within 12 months of filing, no co-inventor fragmentation, and no re-assignment of the inventor's interest to any third party.

Original assignee

Motorola, Inc. (Illinois), Schaumburg, IL — per the issued patent and reel 008421/0374.

• Line of business at the time: diversified electronics; the relevant unit was its paging / selective-call infrastructure and subscriber-device business (FLEX/ReFLEX paging networks and pagers). The patent's claimed subject matter — broadcasting a directory of information services to pagers on a group/mail-drop address, taking a return-channel subscription, and self-programming a sub-address port for a paid window — sits squarely in that paging line.
• Did they ship a product embodying the claims? Motorola unquestionably manufactured the selective call devices and operated the networks in which the claimed system would be practiced. However, I found no evidence of a commercial "pay-per-view" paging information-service product matching the full claim set (directory + assignable sub-address + auto-expiring subscription). Treat "shipped product embodying the claims" as unconfirmed.
• Current status: Motorola, Inc. no longer exists as such. It split effective January 4, 2011 into Motorola Solutions, Inc. (public safety/networks — still operating) and Motorola Mobility Holdings, Inc. (mobile devices). Motorola Mobility was acquired by Google in May 2012 (~$12.5B) and the device business was sold to Lenovo on 2014-10-30 (~$2.91B), with Google retaining the "vast majority" of the Motorola patent portfolio — see Alphabet/Google Form 8-K filed 2014-01-29 (https://www.sec.gov / mirrored at https://app.quotemedia.com/data/downloadFiling?webmasterId=[101533](/patent/101533)&ref=[9347825](/patent/9347825)&type=DOC&symbol=GOOG). This patent was in the retained portfolio, not the Lenovo tranche (see reel 034485/0449 below).
• Bankruptcy: None relevant. The 2011 Motorola event was a corporate split, not a Chapter 7/11 proceeding.

⚠️ Cross-source contradiction to flag: Unified Patents (https://portal.unifiedpatents.com/patents/patent/US-[5926104](/patent/5926104)-A) lists the "Original Assignee" as Motorola Solutions Inc and the parent company as Google LLC. The recorded assignment chain says Motorola, Inc. → Motorola Mobility, Inc. I cannot reconcile the Unified Patents "Motorola Solutions" label with the reel records; the assignment chain is the better evidence because it cites reel/frame numbers. (Unified also repeats the one-day discrepancies already flagged in the Patent Summary: priority 1997-01-27, grant 1999-07-19, expiration 2017-01-27.)

Assignment timeline

Four recorded events. The chain is short, corporate-structural, and terminates at an operating company's IP-holding subsidiary.

1. 1997-01-21 (executed) / recorded 1997-01-28 — Reel 008421/0374

• Conveyance: Assignment (Assignment of interest; "ASSIGNMENT OF INTEREST (SEE DOCUMENT FOR DETAILS)")
• Assignor: Robinson, Edward Herbert
• Assignee: Motorola, Inc. (Illinois)
• Correspondent: Not exposed in the accessible record — data gap.
• Context: Standard inventor-to-employer assignment of entire interest, executed seven days before the application was filed.

2. 2010-07-31 (executed) / recorded 2010-12-13 — Reel 025673/0558

• Conveyance: Assignment
• Assignor: Motorola, Inc.
• Assignee: Motorola Mobility, Inc. (Illinois)
• Correspondent: Not exposed in the accessible record — data gap.
• Context: Internal corporate separation — the patent was swept into the entity that became Motorola Mobility as part of the Motorola split. Not an arm's-length sale.

3. 2012-06-22 (executed) / recorded 2012-10-02 — Reel 029216/0282

• Conveyance: Change of Name (explicitly so captioned: "CHANGE OF NAME (SEE DOCUMENT FOR DETAILS)")
• Assignor: Motorola Mobility, Inc.
• Assignee: Motorola Mobility LLC
• Correspondent: Not exposed in the accessible record — data gap.
• Context: Change of name only — Delaware corporation-to-LLC conversion. No change in beneficial ownership.

4. 2014-10-28 (executed) / recorded 2014-11-27 — Reel 034485/0449

• Conveyance: Assignment ("ASSIGNMENT OF INTEREST"; "ASSIGNMENT OF ASSIGNOR'S INTEREST")
• Assignor: Motorola Mobility LLC (600 North US Highway 45, Libertyville, IL 60048)
• Assignee: Google Technology Holdings LLC (1600 Amphitheatre Parkway, Mountain View, CA 94043)
• Correspondent: Not exposed in the accessible record — data gap.
• Context: Internal reorganization within the Google group — executed two days before the Lenovo/Motorola device-business closing (2014-10-30) and consistent with the 8-K disclosure that Google retained the Motorola patent portfolio. The UK IPO register independently records this same deed of assignment dated 28.10.2014 for a sibling Motorola patent (https://www.search-for-intellectual-property.service.gov.uk/GB0327466.9/history).

No subsequent assignments are recorded. The chain has been static since 2014-11-27.

Timeline diagram

timeline
    title Ownership of US 5926104
    1997 : Inventor Robinson assigns to Motorola Inc
         : Application filed 28 Jan 1997
    1999 : Patent issued 20 Jul 1999
    2010 : Motorola Inc to Motorola Mobility Inc
         : Internal split of the Motorola businesses
    2012 : Change of name to Motorola Mobility LLC
    2014 : Motorola Mobility LLC to Google Technology Holdings LLC
         : Google retains portfolio in Lenovo device sale
    2017 : Patent expires 28 Jan 2017

NPE / troll-pattern signals

Verdict

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

Justification: The ownership chain is fully reconstructable across four recorded events (reels 008421/0374 → 025673/0558 → 029216/0282 → 034485/0449) and every hop is either an inventor's employment assignment, an internal corporate split, a bare change of name, or an intra-group portfolio retention. Zero of the eight enumerated NPE signals is affirmatively present; signals 5 and 6 are affirmatively negated by the "no litigation" finding and the absence of any bankruptcy proceeding, and signal 4 fails on timing (51-month span, no LLC chain). The patent expired at end of term on 2017-01-28 with no recorded assertion.

Caveat, stated plainly: the literal trigger for this verdict category is "chain terminates at a defensive aggregator," and Google Technology Holdings LLC is not one of the named aggregators. It is an operating company's holding subsidiary. I am selecting "Defensive / non-asserting" because it is the only category in the forced-choice list that fits a chain ending at a large operating parent with no NPE signals and no assertion — but I want it on the record that the fit is functional, not literal. If your rubric requires the literal aggregator trigger, the honest alternative is that no category in the list maps cleanly to a clean chain ending at Google.

Data gap that could change this call: signal 3 (repeat correspondent). If the four recordings share a single attorney or filing firm — particularly if it recurs across the wider Motorola→Google assignment family — that would be worth a second look, though on this record it would not, by itself, move the verdict. Verify at: https://assignmentcenter.uspto.gov/ (search patent number 5926104; open each reel/frame and read the "Correspondent" block) and the mirror at https://assignment.uspto.gov/patent/index.html. Google Patents legal events for cross-check: https://patents.google.com/patent/US5926104A/en#legalEvents

Bottom line: US 5,926,104 shows a clean corporate chain from a Motorola inventor to a Google holding subsidiary, with no shell entities, no known asserters, no cascading transfers, no bankruptcy, and no enforcement activity. It reads as a defensive-portfolio asset that simply aged out, not as an NPE instrument.

I'll research each cited prior-art reference for US 5,926,104 to give you a claim-by-claim § 102 analysis. Let me pull authoritative details on each.

I have the patent text and all five examiner-cited references. Let me confirm a few bibliographic details and the terminal-disclaimer relationship before finalizing.

US 5,926,104 — Prior-Art Analysis of the Examiner-Cited References

Scoping note. The authoritative citation list for US 5,926,104 is the five U.S. patents on its own front page ("References Cited"), which matches Google Patents' "Citations (5)" exactly. There are no foreign patent documents among the examiner citations. The family members (EP0973795B1, WO1998032764A1, DE69837323T2, JP2001509346A, KR100311887B1, CN1124048C, AU713966C, HK1026557B) are the same invention, not prior art. Below I treat the five citations in turn, then state which claims each could reach under § 102.

USPTO / bibliographic confirmation (literal): US 5,926,104; App. No. 08/791,043; filed 1997-01-28; granted 1999-07-20; inventor Edward Herbert Robinson; assignee Motorola, Inc. (now Google Technology Holdings LLC); 18 claims (independent 1, 7, 13, 16); status Expired – Lifetime (anticipated expiration 2017-01-28). Because the effective filing date is 1997-01-28, pre-AIA § 102 governs.

The five cited references (§ 102 prior-art table)

Why US 5,694,120 is § 102(e), not § 102(b): it granted 1997-12-02, after the '104 filing date (1997-01-28), so its printed publication cannot be § 102(a)/(b) art. Its application was filed 1996-02-26, before the '104 filing date, so it is available only as a § 102(e) reference. This is the key reason the '104 was subjected to a terminal disclaimer (see the "[*] Notice" on the '104 front page) — i.e., the same-assignee (Motorola) overlap was resolved by disclaimer rather than a pure § 102 rejection.

Reference-by-reference analysis

1. US 5,131,020 (Liebesny / SmartRoutes) — telephone traffic-report system.
Discloses a central station that collects/updates information, transmits a report to a caller in response to a keyboard-entered zone code, and — critically — lets the caller request automatic updates within a caller-selected, variable time period, with the station storing the caller's number and calling back on significant changes. It also mentions "validating pay-per-caller and subscriber caller calls."

• Overlap with '104: the notion of an information subscription valid for a predetermined time period, and subscriber/pay validation (cf. '104 claim 3/9 "pay-per-view" concept).
• Missing: any selective-call device with a preassigned address port and assignable sub-address port; broadcast of a directory with designated address assignments; an MCR return-channel; programming a sub-address port upon authorization; time-based auto-disable of an address port. It is a telephony/facsimile system, not an over-the-air paging system.
• Claim reach: anticipates none of claims 1, 7, 13, 16 (each requires the sub-address-port programming/expiry machinery). At most background for the "interval of time" limitation and dependent claims 3/9.

2. US 5,398,021 (Moore / Motorola) — reliable information-service message delivery.
Discloses information-service messages carried on a group address, receiver-side storage of address/group tables, and disable-code messages that cause a receiver to selectively enable/disable groups.

• Overlap with '104: the group-address broadcast concept (cf. '104 "group or mail drop address") and the enable/disable of address(es) at the receiver (cf. '104's enabled/disabled status information in memory 222).
• Missing: the directory of services with designated sub-address assignments; the return-channel subscription request; programming a sub-address upon authorization; time-based auto-disable.
• Claim reach: anticipates none of the independent claims; usable as secondary art for the "group/mail-drop address" and "enable/disable address" elements of claims 1/7/13/16.

3. US 5,504,476 (Marrs / Motorola) — content-based alerts.
Discloses a common message on one paging address containing multiple recipient IDs and multiple alert codes; each receiver locates its own ID (a form of sub-address) within the common message and extracts the data/alert meant for it, so different receivers pull different data from the same transmission.

• Overlap with '104: the "sub-address" extracted from a shared/common message concept, and efficient use of a single group transmission.
• Missing: any subscription/billing flow; assignable sub-address port; acknowledge-back request; interval-of-time and auto-disable.
• Claim reach: anticipates none of the independent claims; evidence only for the "sub-address within a broadcast" idea.

4. US 5,555,446 (Jasinski / Motorola) — selective call receiver that requests information.
Discloses a base station storing information in a plurality of files, receiving an information service request from a receiver, and transmitting the requested information back; the receiver generates and transmits the request; the base station determines whether the receiver is authorized/has access to the requested service and denies if not. Includes acknowledge-back teaching.

• Overlap with '104 — substantial: this is the closest art for the return-channel subscription/request and base-site authorization determination limitations that appear in claims 1, 7, 13, and 16. It also shows the subscriber device (preassigned address) acting as a transceiver.
• Missing: broadcast of a directory of services with designated address assignments on a preassigned port; the assignable sub-address port that is programmed upon authorization; and especially the "intervals of times during which corresponding information services are authorized" plus automatic disable on expiry.
• Claim reach: could be argued to reach the subscriber-side "transmit a request / receive information in response" and the "determine whether authorized" steps, but does not anticipate any independent claim as a whole because of the sub-address-programming + interval + auto-disable combination.

5. US 5,694,120 (Indekeu et al. / Motorola) — selecting services from a menu. ← MOST RELEVANT.
Discloses: a base site generates a directory of information services; broadcasts it on a mail-drop/group ("ubiquitous") address; the receiver decodes the directory, builds an interactive menu, and the user selects a service; the controller assigns/programs one of a plurality of assignable address ports to the selected service; the controller encodes the assigned address port in a request transmitted to the base site; the base site determines authorization (and may return a password/access code); the receiver thereafter receives the service on the assigned address port; status information enables power conservation when a port is disabled.

• Overlap with '104: nearly the entire apparatus/method of claims 13 and 16 — preassigned port + multiple assignable sub-address ports, directory broadcast on a preassigned/group port, user selection, transmit request, program the assigned port, receive the service on it. This is essentially the '104 invention minus the time-limited pay-per-view auto-expiry.
• Missing: the "interval of times during which corresponding information services are authorized" and the automatic disabling of the sub-address port when that interval expires — the very limitations that distinguish '104's independent claims.
• Claim reach: the closest reference to claims 13 and 16 and to the subscriber-side of claim 7, but it does not literally anticipate any independent claim because it lacks the interval/auto-disable limitation. Available as § 102(e) art (pre-'104 filing), and the likely double-patenting/terminal-disclaimer driver.

Bottom line

• Most relevant prior art: US 5,694,120 (closest overall — directory + assignable address port + return-channel request, missing only the time-limited auto-disable), followed by US 5,555,446 (the request/authorization flow). The remaining three (US 5,131,020, US 5,398,021, US 5,504,476) are secondary/background art supplying isolated elements (time-period updates, group-address enable/disable, sub-address extraction from a common message).
• No single cited reference anticipates any of independent claims 1, 7, 13, or 16 on its face. Each independent claim requires the combination of (a) a broadcast directory carrying designated sub-address assignments and authorized intervals, (b) a return-channel subscription request, and (c) a sub-address port that is programmed on authorization and automatically disabled on interval expiry. No cited reference discloses all three; the novelty resides in the combination, particularly the auto-expiring "pay-per-view" sub-address port.
• The relationships among these references are best characterized as § 103 (obviousness) combinations over US 5,694,120 / US 5,555,446, with the terminal disclaimer resolving the '120 double-patenting overlap — not clean § 102 anticipation.

Caveats / uncertainty. I traced the terminal disclaimer on '104 to US 5,694,120 as the likely basis (same assignee, earlier-filed, near-identical disclosure); the front page does not name the disclaimed-against patent, so this is inference, not confirmed fact. I also relied on the front-page citation list and Google Patents' "Citations (5)" for the set of examiner references; the "Cited By" counts differ across mirrors (39 vs. 114) and reflect forward citations, not the prior art cited against '104. If a definitive § 102/§ 103 record is needed, the prosecution history (USPTO Patent Center / Global Dossier) would show the examiner's actual rejections and the specific double-patenting rejection underlying the terminal disclaimer.

I'll ground this obviousness analysis in the actual examiner-cited references. Let me pull details on each.

Obviousness Analysis — U.S. Patent No. 5,926,104 Under 35 U.S.C. § 103

Source note. The full text of US 5,926,104 (supplied in this thread) is authoritative and is quoted below. For the five examiner-cited references I relied on their own patents/specifications retrieved live: US 5,555,446 (Google Patents, uspto.report), US 5,694,120 (Google Patents, uspto.report), US 5,398,021 (Google Patents), US 5,504,476 (Google Patents), and US 5,131,020 (Google Patents). This section builds on, and does not repeat, the prior "Patent summary" and "Litigation summary."

⚠️ Identifier caution (strict-rule flag). Two of the searches surfaced US 5,542,120 (Smith et al.), an Ericsson-family reference appearing in an unrelated PCT search report (PCT/SE98/01643). That is not examiner-cited art here and must not be confused with the Motorola reference US 5,694,120 (Indekeu et al.), which is cited on the face of 5,926,104. The difference is one digit (554 vs 569); I am reporting them literally and treating only 5,694,120 as relevant.

1. Governing framework

• Statute: The application was filed 1997-01-28, so pre-AIA § 103(a) applies (not AIA § 103). Prior art is measured against the § 102 categories in force at that time.
• Test: Graham v. John Deere — scope/content of prior art, differences, PHOSITA level, secondary considerations; coupled with KSR Int'l v. Teleflex — combinations of known elements are obvious where there was a known problem, a finite number of identified solutions, and a reasonable expectation of success; "obvious to try" is enough where the solution is one of a small set of predictable options.
• POSITA: A person with a bachelor's degree in electrical engineering or computer science and 2–3 years' experience designing two-way selective-call/paging systems and paging terminals, or equivalent. All five references are analogous art — same field of endeavor (selective call / message delivery to portable receivers) or reasonably pertinent to the problem of controlling which broadcast information a subscriber receives (Liebesny, a telephony information service, is reasonably pertinent to time-limited delivery of updated information to mobile subscribers).

2. The cited art, as it actually reads

3. The distinguishing features of '104

Comparing claim 1 of '104 to the art, the only elements not squarely met by 5,694,120 (+5,555,446) are:

1. the directory includes "interval[s] of times during which corresponding information services are authorized" (a time field, transmitted to the device); and
2. the controller "automatically disables the assignable sub-address port … subsequent to an expiration of the interval of time."

Everything else in independent claims 1, 7, 13 and 16 — preassigned/group address port carrying a directory; plural assignable sub-address ports; menu/user selection; transmit-back request; base-side authorization; programming the selected sub-address port upon authorization; reception of the service on that port — is disclosed in 5,694,120 and/or 5,555,446.

Also note two objective indicia that the feature set was not far from the art: (a) the '104 background section is verbatim the background of 5,694,120 (compare "hundreds of information services … exceed the amount of information a user … can easily receive based upon the limitation of the number of … address ports"), and (b) the '104 front page carries the notation "[*] Notice: This patent is subject to a terminal disclaimer" — the hallmark of an obviousness-type double-patenting rejection over a commonly-owned Motorola case, consistent with the examiner having found a mere variation over the same family of disclosures.

4. The obviousness combinations

Combination A (primary): 5,694,120 + 5,131,020 → claims 1, 7, 13, 16

• 5,694,120 supplies the complete directory → menu → select → request → authorize → program assignable port architecture.
• 5,131,020 supplies the missing temporal limitation: information is authorized and delivered only for a "limited predetermined time period," after which the service stops.

Motivation to combine (KSR rationales):

1. Same problem, same field. Both address delivering recurring updates to a mobile user to whom only current, relevant information should flow.
2. Billing/market force. The '104 spec itself concedes the driver: users are charged for services, and pay-per-view requires a defined, metered interval ("the duration … of the subscription to that service"). A time-limited subscription is the natural commercial unit; 5,131,020's "predetermined time period" is a ready-made teaching of that unit.
3. Battery/link economy. Both systems seek to avoid receiving information the user does not want; auto-terminating a subscription at expiry conserves receiver power — the same goal 5,694,120 already pursues with status information that lets the controller skip decoding disabled ports.
4. Predictable result. Adding an expiry timer to a reprogrammable address port is a mechanical, predictable combination (start a timer when the port is programmed; clear the port when it expires), squarely within KSR's "known technique to improve similar devices in the same way."

Combination B: 5,555,446 + 5,694,120 + 5,131,020 → claims 1, 7, 13, 16

5,555,446 independently supplies the base-side "plurality of information files / receive service request / store per-subscriber subscription access info / transmit requested file" half of claim 1 and the acknowledge-back half of claims 5/6/11/12/15/18. 5,694,120 supplies the device-side assignable-port programming. 5,131,020 supplies the interval. This is a straightforward three-reference combination among commonly-owned Motorola pagers addressing the same paging-channel-capacity problem.

Combination C: 5,694,120 + 5,398,021 → the "disable the port" step

Where 5,694,120 programs a port to enable a service, 5,398,021 teaches the complementary act — a disable code turning off reception of a group's information at the receiver. A POSITA seeking to end a subscription (or an expired subscription) would readily apply group/enable-disable control to the programmable port of 5,694,120. In '104 the disable is timer-driven rather than code-driven, but the function (cease decoding that service) and the mechanism (a status bit/port in the receiver) are the same; substituting a local timer for a downlink disable code is an obvious design choice where, as here, the interval is already transmitted with the directory (and it conserves downlink bandwidth).

Combination D: + 5,504,476 / 5,555,446 → "designated address assignments" and sub-address mapping

Claim 1/13 require the directory itself to carry "designated address assignments." 5,694,120 leaves the port assignment at the transceiver; 5,504,476 teaches the base associating different data sets with sub-addresses within a common message, and 5,555,446 teaches the base storing per-receiver access/subscription information. A POSITA would see immediately that if the base must transmit a service on a specific sub-address, the base must tell the device which sub-address — i.e., put the assignment in the directory. That is a design choice with only two options (device-assigned vs. base-designated), both known, yielding a predictable result — obvious under KSR.

5. Element-by-element mapping — claim 1 (representative)

Claims 7/16 (methods) map identically; claim 13 (device) differs only in claiming the device standalone with plural assignable sub-address ports — expressly disclosed by 5,694,120's "plurality of assignable address ports (246)."

6. Dependent claims

• Claims 3 and 9 (pay-per-view + subscription info): essentially anticipated in substance by 5,694,120 claim 3 ("determines when the information service requested is a pay-to-receive service; and … transmits subscription information"). Language is near-verbatim; obvious a fortiori.
• Claims 4, 10, 14, 17 (display directory / pay-per-view service / time periods): 5,694,120 discloses displaying the menu (display 228, memory 226); adding the time periods already received (per 5,131,020) to the display is routine.
• Claims 2, 8 (encode message with directory, assignments, time periods): message encoding of these fields is conventional; 5,694,120 encodes the directory and the assigned address, 5,131,020 supplies the period.
• Claims 5, 11, 15, 18 (subscriber transmits a confirmation upon receipt of authorization): 5,555,446 expressly teaches acknowledge-back ("transmit an acknowledge-back response to the central station"); 5,694,120 teaches the transceiver "waits for service confirmation from the communication system." Combining the two gives the claimed confirmation, with the motivation stated in '104 itself (avoid billing when authorization was not received) — a routine reliability measure.
• Claims 6, 12 (base waits for confirmation): same references; the base "wait[ing] for" a confirmation is inherent in an acknowledge-back protocol and disclosed by 5,555,446.

7. Anticipated counterarguments

1. "Time-based auto-disable was non-obvious — nothing in paging taught self-expiring subscriptions." Counter: 5,131,020 teaches exactly time-delimited information delivery to a mobile subscriber, and 5,694,120 already uses per-port status information to disable functions. Combining a timer with a programmable port is a predictable arrangement of known elements (KSR).
2. Common ownership / § 103(c). All cited references are Motorola-assigned, like '104. Important caveat for pre-AIA cases: because the '104 application was filed 1997-01-28, before the AIPA amendment effective 1999-11-29, the pre-amendment § 103(c) excluded only § 102(f)/(g) art — not § 102(e) art. 5,694,120 (different inventive entity — Indekeu/Henz vs. Robinson, "by another," filed 1996-02-26, granted 1997-12-02) therefore remains available as § 102(e) prior art for § 103, notwithstanding common ownership. Secondary consideration evidence (commercial success, long-felt need, industry praise) is not in the record and, on a "pay-per-view pager" commercial model, would face a strong nexus problem.
3. Teaching away? None identified in any reference; all five point toward more, not less, subscriber control over which broadcast services are received.

8. Bottom line

• Strongest single rejection: Claims 3, 9 (and the display/confirmation dependents) as obvious over US 5,694,120 alone, given its near-verbatim disclosure of pay-to-receive subscription and its menu/assignable-port/authorization architecture.
• Strongest combined rejection for the independents (1, 7, 13, 16): US 5,694,120 in view of US 5,131,020, optionally further in view of US 5,555,446 (base-side files/request/acknowledge-back) and US 5,398,021 or US 5,504,476 (enable/disable and base-designated sub-addresses).
• The feature that would most likely have defeated a clean prima facie case is the specific "interval of times … authorized" carried in the directory plus automatic, timer-driven disabling of the assignable sub-address port. I judge these to be obvious over 5,131,020's time-limited auto-update teaching combined with either 5,398,021's disable control or the ordinary expedient of a subscription timer, but a reasonable examiner could also find a secondary-consideration/packaging argument here, since no single reference shows a pager that self-expires a programmable information-service port. That is the strongest (and narrow) non-obviousness position for '104.

Caveats: (1) I have not seen the actual file wrapper/office actions for 5,926,104, so I cannot confirm which reference(s) the examiner applied or which claim feature drove allowance; the analysis above is reconstructed from the cited-art list and the issued claims. (2) The 5,694,120 content is taken from its Google Patents/uspto.report text; its post-grant Certificate of Correction (noted on uspto.report) could change specific wording. (3) As a legal conclusion, this is an analytical assessment, not legal advice.