US 11402417

I'll search for authoritative information on US 11402417, including any CAFC/litigation activity, before summarizing.

Let me run a few more targeted searches to check for litigation/docket activity and to confirm claim structure.

Let me retrieve the actual claim set for US 11,402,417.

US Patent 11,402,417 B2 — Analyst Summary

Sourcing note: The analysis below is grounded primarily in the authoritative full text of US 11,402,417 B2 supplied in the record (fetched from Google Patents), supplemented by search results for the patent family. Where I could not verify something directly from an authoritative source, I flag it explicitly.

1. Bibliographic data

Note on a potentially confusing near-neighbor: Search results repeatedly surface WO 2020/216671 A1 (Sagemcom, also 29 Oct 2020) — this is a different Sagemcom family (appearing here as prior art cited in an FR search report for another application). It should not be conflated with this patent's WO 2020/207922 A1.

2. Abstract (verbatim, authoritative text)

"A centralising meter is connected to a data concentrator via a first powerline communication network and to remote basic meters via a second powerline communication network. The centralising meter emulates a smart electricity meter application with respect to the data concentrator for each remote basic meter, in order in particular to construct a corresponding load curve. When the ratio of an energy consumption declared by all the remote basic meters to an energy consumption measured by the centralising meter is below a predefined threshold and when furthermore a plurality of load curves show a drop to zero, a fraud is detected. The centralising meter compares total energy consumptions estimated from extrapolations of certain load curves with the measured energy consumption, to identify which remote basic meter is subject to the fraud."

3. Litigation / CAFC docket check (as of the current date)

I searched for this patent number across USPTO/PTAB and Federal Circuit (CAFC) 2026 docket information.

• No CAFC appeal, PTAB proceeding (IPR/PGR), or district-court case asserting US 11,402,417 was found. I state this with moderate confidence: absence of a hit is not proof of absence, and my search was not an exhaustive PACER/PTAB scrub.
• Searches did surface Sagemcom-related litigations, but none involve this patent:
  • Entropic Communications, LLC v. Sagemcom Broadband SAS, No. 2:25-cv-00379 (E.D. Tex.) — asserts Entropic's patents (e.g., US 11,785,275), not the '417 patent; related PTAB matter IPR2026-00263.
  • Encryptawave Technologies LLC v. Sagemcom USA LLC, No. 2:26-cv-00639 (E.D. Tex.).
  • NetConnect Wireless LLC v. Sagemcom Broadband SAS et al., No. 2:26-cv-00696 (E.D. Tex.).
  • These are different Sagemcom entities (Sagemcom Broadband SAS / Sagemcom USA LLC) and different patents than the assignee/patent at issue here.

4. Plain-language overview of the invention

The patent addresses a proxy/centralising-meter architecture for smart metering. Expensive "smart" electricity meters each carry heavy application intelligence (DLMS/COSEM objects, load curves, tariffs, communications). The patent's architecture instead deploys cheap "remote basic meters" (RBMs) in each dwelling/room, which only do metrology and answer a small set of atomic commands. A single centralising meter (CM) in the building/property complex emulates a full smart meter toward the grid's data concentrator for each RBM, so the concentrator "sees" normal smart meters.

The twist: with the anti-fraud intelligence stripped out of the cheap RBMs, how do you detect energy-consumption fraud (i.e., an RBM declaring zero consumption while its installation is actually drawing power)? The answer: compare the total declared consumption (sum of all RBMs) against the total actually measured at the CM's own general metrology unit, then use load-curve extrapolation to pinpoint the culprit RBM.

5. Independent claims — plain-language overview

Based on the authoritative text and the retrieved claim text for the family (claim 9 and claim 10 are quoted in the search results for US 2022/0146558 A1; claim 1's substance is mirrored in the "Disclosure of the Invention" and WO abstract). I could not retrieve the verbatim, numbered claim set from the truncated full text, so the following describes the independent claims by their substance; exact wording should be verified against the printed claims.

Independent Claim 1 — Method in a centralising meter
A method performed by a "centralising meter" that has (a) a first powerline communication unit talking to a data concentrator over a first PLC network in a first frequency band, and (b) a second powerline communication unit talking to multiple "remote basic meters" over a second PLC network in a distinct second frequency band. The method:

1. Emulates a smart-electricity-meter application toward the data concentrator for each registered RBM, each emulated application building a load curve for its RBM;
2. Measures, with a general metrology unit in the CM, the total electrical consumption of all installations supervised by the RBMs;
3. Detects fraud when the ratio ΔEaD/ΔEaG (declared total energy ΔEaD over all RBMs during time T ÷ measured total energy ΔEaG from the CM's metrology unit during T) falls below a predefined threshold AND a plurality of load curves show a drop to zero of consumption;
4. Identifies which RBM is fraudulent by extrapolating the load curves of those plurality of RBMs and comparing the estimated total energy ΔEaE from the extrapolations against the measured ΔEaG.

Key dependent concepts in the description (likely recited in dependent claims): ΔEaE includes a predetermined intrinsic consumption ΔEC of the RBMs; the culprit is the RBM whose ΔEaE is closest to ΔEaG (e.g., minimum of |1 − ΔEaE/ΔEaG|, or maximum ΔEaE/ΔEaG ratio); state interrogation of dropped-to-zero RBMs (via a dedicated atomic command) and exclusion of RBMs whose own breaker BRK or a downstream breaker DBRK is open; and extrapolating over a window of T/2 up to the drop-to-zero moment.

Independent Claim 9 — Non-transitory storage medium
A non-transitory storage medium storing a computer program whose instructions cause a centralising meter to execute the method of claim 1. (Claim text confirmed via search result for US 2022/0146558 A1.)

Independent Claim 10 — Centralising meter (apparatus)
A centralising meter for an automated metering management system, comprising: the first PLC unit (first frequency band, to a data concentrator); the second PLC unit (second distinct frequency band, to multiple remote basic meters); a control unit configured to emulate a smart-meter application per RBM (each building a load curve); a general metrology unit reading total consumption of all supervised installations; and a metrology control module that — when ΔEaD/ΔEaG is below a predefined threshold AND a plurality of load curves show a drop to zero — identifies the fraudulent RBM by extrapolating those load curves and comparing total estimated consumption ΔEaE with the measured ΔEaG. (Claim text confirmed via search result for US 2022/0146558 A1.)

Possible additional independent claim: The specification also recites a computer program per se (distinct from the storage medium). It is plausible that the claim set includes a separate computer-program claim (e.g., claim 8) alongside the claim-9 storage medium and claim-10 apparatus, but I could not verify this from an authoritative claim listing — treat it as unconfirmed.

6. Uncertainty flags (explicit)

1. Verbatim claim set / independent-claim count. The full-text record supplied to me truncates before the printed claims. I confirmed the substance of claim 9 and claim 10 from the family publication text, and claim 1's substance from the disclosure and WO abstract, but I have not verified exact claim numbering or whether a separate computer-program claim exists. Recommend confirming against the printed patent or USPTO PatentCenter.
2. Litigation/docket absence. My "no CAFC 2026 or district-court docket found" conclusion reflects a web search, not an authoritative PACER/Docket Navigator sweep. Do not treat it as a certified clearance.
3. Assignee naming. Sources vary between "Sagemcom Energy and Telecom SAS" and "Sagemcom Energy & Telecom SAS" — same entity, different punctuation.
4. No reissue, reexamination certificate, or disclaimer was surfaced; absence is not independently verified.

I'll search for litigation involving US patent 11402417.

Let me try more targeted searches on litigation databases.

Let me try more specific litigation-database searches.

Litigation Search Results: US Patent No. 11402417

Summary: No litigation involving US Patent 11402417 was found in the sources searched.

I was unable to identify any district court, CAFC, or ITC proceedings asserting or otherwise involving US 11402417 ("Detection of energy-consumption fraud in an electrical distribution service," assigned to Sagemcom Energy and Telecom SAS) as of the current date.

Search details and what the searches actually returned

Searches were run against litigation-oriented queries (Unified Patents, PACER/CAFC-style query formulations, and general web coverage of patent suits) using the exact identifier "11402417." The results did not contain any case captions, docket numbers, or complaints referencing this patent. Instead, the hits returned were unrelated to this patent number:

• Unrelated Sagemcom litigation (different patents): An RPX litigation document from the District of Delaware involving Sagemcom as a defendant — but the asserted patent was the '581 Patent, and the accused product was the F@ST 5345 wireless gateway. This is a Sagemcom Broadband matter concerning Wi-Fi/OFDM technology, not US 11402417 and not Sagemcom Energy & Telecom. I want to be explicit that this is not the patent you asked about; I am not auto-mapping it onto US 11402417.
• Prosecution records: A USPTO PTAB/Office action document referencing application 13/357,549 and a Kuhlman reference — a different application entirely.
• Foreign registration bulletins: EPO/Belgian, Australian, Brazilian, and French INPI patent register entries listing various Sagemcom Energy & Telecom patents (e.g., flow meters, solenoid valves, electric-meter adjustment methods) — none referencing US 11402417 and none constituting litigation.

No result matched the exact number "11402417" in a litigation context.

Caveats on this finding

• Absence of evidence is not conclusive. My search coverage of PACER and the CAFC docket was not a direct, exhaustive paid-database query; free web sources frequently lag and under-index recent complaints. A patent granted in 2022 (filing date 2020-04-03, priority 2019-04-08) could plausibly be involved in recently filed or sealed cases that free sources do not surface.
• SEP/portfolio context: Sagemcom Energy & Telecom holds a substantial smart-metering portfolio and has been active in licensing/enforcement discussions in that sector, so future or ongoing assertions cannot be ruled out from this search alone.

Recommended next steps to confirm

To definitively confirm the absence of litigation, I would (subject to your authorization/access) query:

1. Unified Patents Portal – Litigation (https://portal.unifiedpatents.com/litigation) with the exact patent number.
2. MaxVal / Lex Machina / Docket Navigator (direct patent-number litigation search — most authoritative).
3. PACER docket search by party name ("Sagemcom Energy and Telecom") to catch any case where this patent is asserted alongside others.
4. USPTO Patent Center / Assignment and the ITC EDIS database for Section 337 investigations.

If you can provide or authorize access to one of these paid databases, I can compile the full case list (plaintiff, defendant, jurisdiction, case number, filing date, status). Based on the sources available to me now, I must report: no known litigation identified for US 11402417.

I'll start by verifying the structured ODP data point against live searches, since the instructions permit flagging anything ODP hasn't indexed yet.

The first two searches returned nothing relevant. Let me try a few more targeted queries covering PTAB docket sources, litigation, and any Unified Patents activity.

Let me run a final pair of targeted checks on PTAB dockets specific to the patent owner and any third-party challenger.

PTAB proceedings on US 11,402,417 — analyst report

Bottom line up front: no AIA trial activity on file. This is a genuine negative, not a data gap I could paper over with search results.

Proceedings overview

Total AIA trial proceedings on US 11,402,417: 0 (0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denied). The USPTO Open Data Portal returns no IPR, PGR, or CBM proceeding naming this patent, and independent web searches of PTAB docket sources surfaced nothing to contradict that. The defensive posture for a defendant is therefore "the patent is untested at the Board" — every claim is live, un-narrowed by any PTAB final written decision, and the patent owner has never had to defend these claims in an adversarial Patent Office forum. That cuts both ways: there is no cancellation to exploit, but there is also no prosecution-backed PTAB record, no claim-construction order, and no institution decision to learn from. Whoever attacks this patent first writes on a clean slate.

{No proceedings to enumerate}

Because the count is zero, the per-proceeding template (type / filed / status / panel / grounds / institution / FWD / settlement / appeal / defensive value) has nothing to populate. I am deliberately not inventing proceeding numbers, panels, or claim-level outcomes. The prompt's illustrative framing ("the patent has survived two IPRs and is hardened" / "claims 1–5 have been canceled") describes possibilities that are not supported by any record for this patent, and I decline to adopt either.

What I searched, and the negative result:

Near-miss hits I checked and rejected (do not cite these as proceedings on this patent):

• IPR2026-00263 — Sagemcom Broadband SAS v. Entropic Communications, LLC — different patent (US 11,785,275), and here a Sagemcom entity is the petitioner, not the patent owner. Unrelated.
• The family of Entropic IPRs (IPR2024-00435 / -00438 / -00441 / -00442 / -00446, IPR2025-00180/-00181/-00182) — all concern Entropic's set-top/network patents asserted against Sagemcom Broadband. Unrelated to the '417 patent.
• The "417" hits that appear in Netlist/Comcast petition excerpts (US 11,093,417, US 10,996,417) are different patents with coincidentally similar numbers. This is a real trap in this docket space — three separate patents in circulation end in "417," and search engines conflate them. Anyone citing "the 417 IPR" should confirm the number digit-by-digit.
• The European counterpart EP 3 953 717 B1 ("Détection de fraude à la consommation énergétique dans un service de distribution électrique") and EP 3 709 671 are EPO proceedings, not PTAB, and any EPO opposition activity would not appear on this US docket.

Strategic summary

Claim status. All claims of US 11,402,417 are UNTESTED before the PTAB. There are no CANCELED claims and no SUSTAINED claims in the AIA-trial sense. The claim set as granted on 2022-08-02 (priority 2019-04-08, filed 2020-04-03, applicant/drafter Sagemcom Energy & Telecom SAS, inventor Henri Teboulle) stands exactly as issued. Because the patent issued only in 2022, the § 315(b) one-year window runs from service of any complaint — meaning a first-wave defendant served in, say, 2025–2026 has had a short runway to date, which is consistent with the absence of filings.

Estoppel landscape. With no petitions on file, there is no § 315(e)(2) estoppel for anyone. No petitioner has a PTAB record to be estopped on; a fresh defendant retains the full universe of § 102/§ 103 art, including art that an earlier (hypothetical) petitioner would have had to disclose. Two practical consequences: (1) IPR is fully available — nothing is time-barred unless you have been served and sat on a complaint for more than a year, so calendar the one-year anniversary of service immediately if you've been sued; (2) because the patent has never been construed by the Board or by a district court on the merits (no litigation hits found), the claim-construction risk is symmetric and unsettled.

Pattern signals. No repeat petitioner exists on this patent — no defensive aggregator (Unified Patents, RPX, AST-type filer) has taken a run at it, and this patent owner shows up in PTAB only as a petitioner in unrelated Entropic campaigns through the separate Sagemcom Broadband entity. That is a notable asymmetry: Sagemcom knows the PTAB playbook well (it has filed competing IPRs and joined/settled in Entropic matters), so a defendant should expect a sophisticated patent owner on the other side of any IPR, not a novice. On the patent-owner side there is no history of aggressively appealing PTAB losses for this patent, simply because there are no losses to appeal.

The absence of IPR activity is itself the signal here: it suggests the patent has not been broadly asserted in US litigation to date. Well-asserted patents in this space eventually attract IPRs; this one has not. That either means assertion has been minimal/targeted, or that it is coming.

Recommended next steps

1. Do not rely on any supposed PTAB outcome. There is no FWD to link to, no institution decision, and no claim-level disposition. If opposing counsel or a demand letter references a "PTAB ruling" on this patent, that reference is wrong — ask for the proceeding number and verify it against ODP before responding.
2. Verify the patent number against the litigation complaint. Confirm the asserted patent is US 11,402,417 and not US 10,996,417 or US 11,093,417, both of which do have PTAB activity and both of which are different patents. Getting this wrong will corrupt your invalidity and estoppel analysis from day one.
3. If you have been served, file your § 315(b) deadline now. The one-year bar from service is the single most consequential date in this posture, precisely because no one else has burned the patent's IPR runway.
4. Budget a full-freedom IPR. With zero estoppel and zero prior PTAB construction, a § 103 challenge built on primary references that were never substantively applied during prosecution is the highest-leverage move available. The patent's own specification gives you a roadmap of the field to mine: G3-PLC (ITU-T G.9903), PRIME (ITU-T G.9904), DLMS/COSEM (IEC 62056-5-3), 6LoWPAN, LOADng, and EAP-PSK are all admitted prior art in the description.
5. Monitor the docket, not the aggregators. Set a watch on the USPTO ODP PTAB endpoint and on Docket Alarm / Docket Navigator for 11,402,417 specifically. If a first petition is filed, the institution decision (statutory deadline ~6 months from filing) and the FWD (~12 months from institution where the § 316(a)(11) clock is not extended for good cause) are the milestones to track. I flag explicitly that today's date is 2026-09-29, and that a first petition, if it lands, would likely be filed as an IPR2026- or IPR2027-series case.

Caveats and confidence. I am high-confidence that no AIA trial proceeding on US 11,402,417 exists as of the ODP ingest and as corroborated by web search. I am lower-confidence on the completeness of any litigation picture, because the prompt scoped this task to PTAB and I did not run a full district-court docket sweep — the absence of litigation hits in my searches is suggestive, not dispositive. If litigation status matters to your analysis, that should be confirmed against PACER/Docket Navigator directly rather than inferred from this report.

I'll research the assignment record for US 11402417 from the sources described.

Let me dig further into the assignment and legal-events records.

Let me verify the ownership/legal-events record and check for any assertion activity.

Assignment & Ownership Analysis — US 11402417 B2

Patent: "Detection of energy-consumption fraud in an electrical distribution service"
App. 17/601,629 (national stage) · Filed 2020-04-03 · Granted 2022-08-02 · Priority 2019-04-08 (FR 1903749)
Family: WO2020216671A1 · FR3095568A1 · EP3953717B1 (granted 2023-07-19)

Source-access note (read first): I could reach the patent record and several secondary registers, but I was not able to pull the live USPTO Assignment Center entry (reel/frame and correspondent of record) from the sources available to me. Where a field is not verified, I say so rather than guessing. The single recorded transfer below is corroborated by the Google Patents legal-events panel embedded in the authoritative patent text.

Inventors

Pattern check — no departure signal. Teboulle did not leave the original assignee. He remains a prolific Sagemcom inventor, with grant activity continuing well past filing — e.g. US 12,510,572 "Electricity meter arranged to detect fraudulent opening" (granted 2025-12-30) and published applications into 2026 (US 2026/0067721, US 2026/0067684). Third-party inventor profiles credit him with ~71 US grants assigned to Sagemcom. This is the opposite of the "inventor exits within 12 months → fire-sale precursor" tell. His co-inventor profile is normal (sole inventor here; co-inventors on other Sagemcom filings). Nothing anomalous.

Original assignee

Sagemcom Energy and Telecom SAS — a French société par actions simplifiée, a subsidiary within the Sagemcom group (itself part of Sagemcom Broadband/Energy corporate family). Address of record evolved from 250 Route de l'Empereur, 92500 Rueil-Malmaison, FR (as on the issued patent and 2021 assignment) to 4 Allée des Messageries, 92270 Bois-Colombes, FR on later documents — a registered-office move only, not a change of name or ownership.

• Primary business: design/manufacture of smart electricity, gas and water meters and the associated AMM/AMI (automated metering management) head-end, concentrator and PLC/RF communication equipment. This is a genuine, shipping operating company; the patent's claimed centralising-meter / dual-band PLC / emulated-smart-meter architecture maps directly onto Sagemcom's metering product line.
• Status: operating. Sagemcom is a privately held French group, so there are no SEC 10-K/8-K filings to cross-reference; I could not verify the current private ownership/control structure with high confidence, and I am not asserting one.
• Relationship to claims: the disputed subject matter is anti-fraud logic inside Sagemcom's own AMM devices — i.e., an operating-company implementation, not a licensing abstract.
• Current assignee of record: Sagemcom Energy and Telecom SAS (unchanged since grant).

Assignment timeline

The accessible record shows exactly one recorded assignment — the inventor-to-employer confirmatory assignment — and no post-issuance, third-party, security-interest or reversionary record.

• Executed: not stated on the sources available to me / recorded 2021-12-03 — Reel not retrievable from my sources (see note above)
  • Conveyance: Assignment
  • Assignor: TEBOULLE, Henri
  • Assignee: SAGEMCOM ENERGY & TELECOM SAS
  • Correspondent: not retrievable from the accessible record — cannot be assessed. (For context only, Sagemcom's European filings route through prosecution agencies such as Cabinet Le Guen Maillet, Cabinet Boettcher and Regimbeau; these are EPO agents, not the US recording correspondent, and they are not a finding here.)
  • Context: standard inventor-to-employer assignment perfecting Sagemcom's title (Sagemcom filed the application 2020-04-03 as applicant before this recordation). No NPE, aggregation or reorg character.

No further assignments exist on the record. There is no transfer to any LLC, no security agreement, no merger deed, no license recordation, and no change-of-name entry. The chain is one link long and terminates where it started.

Timeline diagram

timeline
    title Ownership of US 11402417
    2019 : French priority filed by Sagemcom
    2020 : PCT and US national stage filed
    2021 : Inventor assignment recorded to Sagemcom
    2022 : US 11402417 granted
    2023 : EP family member EP3953717 granted

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The only transfer is Henri Teboulle → Sagemcom Energy and Telecom SAS (recorded 2021-12-03). No "IP / Holdings / Ventures / Licensing" suffix, no single-purpose LLC, no registered-agent service address. Assignee is the same operating manufacturer that filed the application.

2. Known asserter in the chain — not present. Current and sole assignee (Sagemcom Energy and Telecom SAS) does not match any entity on the lists named — Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or Spangenberg vehicles. No RPX/Unified high-frequency-plaintiff hit surfaced for this patent or assignee.

3. Repeat correspondent across the chain — unclear. A single link cannot evidence recurrence, and I could not retrieve the correspondent of record for the 2021-12-03 entry. No finding. Absent the reel/frame data this signal is simply unevaluable.

4. Cascading transfers — not present. Zero chained LLC transfers; one assignment in ~8 years on record.

5. Pre-litigation transfer — not present. No assignment precedes any suit; my searches surfaced no US infringement litigation naming US 11402417 or app. 17/601,629. Sagemcom appears as a patent owner/prosecutor in these records, not as an asserting plaintiff on this patent.

6. Bankruptcy fire-sale — not present. No evidence of any Chapter 7/11 proceeding involving the assignee; the assignee is a going concern with continuous filing and grant activity through 2026.

7. Privateering — not present. No operating-company → NPE transfer of any kind; the patent never left the operating company.

8. Defensive aggregator — not present. Chain does not terminate at RPX, AST, LOT Network, Unified Patents or OIN. The patent has not been neutralized; it simply remains with its operating owner.

Verdict

Insufficient data — the ownership chain consists only of the original inventor-to-employer assignment recorded 2021-12-03 (Teboulle → Sagemcom Energy and Telecom SAS); there are no post-issuance or third-party transfers, so no NPE pattern can be built. The sole assignee is the original operating manufacturer, the sole inventor remains employed and actively filing for the same assignee through 2026, and no assertion activity on this patent was found — i.e., every hallmark of a fire-sale/privateering chain is absent. This is the "clean single-link chain, original owner retains" scenario, not a defect in the data; the only genuinely missing items are the reel/frame number and the correspondent of record.

Verify directly:

• USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search by patent number 11402417 / app. 17/601,629)
• Legacy Assignment Search — https://assignment.uspto.gov/patent/index.html
• Google Patents (legal events panel showing the 2021-12-03 "Assigned to SAGEMCOM ENERGY & TELECOM SAS" entry) — https://patents.google.com/patent/US11402417/en

Caveat: the reel/frame and correspondent fields should be confirmed against the live Assignment Center, which I could not query directly. Everything else above is either drawn from the authoritative patent text or corroborated by the cited registers.

I'll research this systematically. Let me start with searches for the patent's cited prior art and prosecution references.

I found the Justia "Referenced Cited" list. Let me gather details on each cited reference and check for additional citations.

Prior-Art Analysis — US Patent No. 11402417

Step-limit note: I hit the research-tool step limit partway through this task. The citation list itself is confirmed (Justia/Google Patents prosecution record); the full-text verification of several individual references is incomplete. I flag confidence on each entry below rather than fabricate descriptions.

1. Patent identified (confirmed)

This matches the number queried exactly; no similar-number substitution was used.

2. "References Cited" on the face of US 11402417 (prosecution prior art)

The patent's own bibliographic record (via Justia, "Referenced Cited") lists eight patent documents plus the ISR. These are the citations the Examiner/Applicant placed of record — i.e., the "patent citations for 11402417" the task asks for.

3. Claim-by-claim § 102 potential (based on what is verified)

Independent claim 1 is a method requiring, in combination: (a) a centralising meter with two distinct-band PLC units; (b) emulating a smart-meter application per remote basic meter at the data concentrator, each building a load curve; (c) a general metrology unit measuring the aggregate consumption of all supervised installations; and (d) a two-condition trigger — ΔEaD/ΔEaG below a threshold and a plurality of load curves showing a drop to zero — followed by identifying the fraudulent meter by extrapolating load curves and comparing estimated totals ΔEaE to measured ΔEaG.

• US 8461823 (de Buda) — does NOT anticipate claim 1. It teaches the generic "aggregate-measured vs. sum-of-declared" theft-detection comparison (relevant to the ΔEaD/ΔEaG ratio of step (d)), but it does not disclose the centralising-meter/emulated-application architecture, the load-curve drop-to-zero condition, or extrapolation-based localization of a single fraudulent meter. Best characterized as § 103 background art, not § 102 art.
• References 2–8 are cited electronic-metering / metering-data documents whose likely relevance is the general concept of comparing declared consumption against a master measurement or of anomaly/load-curve analysis. I could not verify their disclosures, so I decline to assert that any of them anticipates a specific claim. On the face of the claim language, none is expected to disclose the full combination of claim 1 (two-band PLC centralising meter + per-meter emulation + drop-to-zero + extrapolation), because that combination is the point of novelty.
• Dependent claims map to discrete features the art would need to show to anticipate:
  • ΔEaE includes intrinsic consumption ΔE_C;
  • select meter minimizing 1 − ΔEaE/ΔEaG;
  • select meter maximizing ΔEaE/ΔEaG;
  • recover state via an atomic command and exclude meters whose breaker or downstream breaker is open;
  • extrapolate over T/2 until the drop to zero.
    No cited reference verified to date discloses these specific selection formulas or the breaker-state exclusion; they are not anticipated on the present record.

Provisional conclusion on § 102: No reference on the face of US 11402417 that I verified anticipates any claim. The cited art is consistent with § 103 relevance (aggregate-vs-declared theft detection being a known field), while the asserted novelty (the centralising-meter emulation + extrapolation-based offender identification) is not shown in the verified references.

4. Related art surfaced from a sibling Sagemcom search report (flagged as different application)

A French preliminary search report for FR 2101219 (pub. FR 3119039) — a different Sagemcom application, not US 11402417 — cites the following as relevant ("Y"/"A"), which may nonetheless be useful comparators for US 11402417's field:

• WO 2020/216671 A1 (Sagemcom Energy & Telecom SAS), pub. 2020-10-29 — Sagemcom's own PCT; the search report cites it as "Y" for a later application.
• EP 2 660 612 A1 (Light Serviços de Eletricidade S/A et al.), pub. 2013-11-06.
• EP 3 734 984 A1 (Sagemcom Energy & Telecom SAS), pub. 2020-11-04.
• US 2020/288222 A1 (Mathems et al.), pub. 2020-09-10.

Important caveat: these are listed on a different application's search report, so I am not treating them as citations of US 11402417. I flag them only as potential § 103 comparators. Conversely, WO 2020/216671 may be a family member of the instant patent, which would make it a § 102(a)(1) self-publication rather than prior art.

5. What I could not complete (honest limitations)

1. I did not finish retrieving the full bibliographic details, abstracts, or § 102/§ 103 mapping for references 2–8 (Krishnan, Rhoads, Dietrich, Arya, Turner, Bandyopadhyay, FR 3025600) before the tool limit.
2. I did not retrieve the ISR (PCT/EP2020/059590) citation table, which would name the Examiner's actual citable art for the international phase — this is the single most probative document still outstanding.
3. I could not confirm the exact claim numbers of the issued US 11402417 claim set (the granted claims may be renumbered vs. the application). The mapping above is to the application/description feature set.

Recommended next queries (to finish rigorously): (a) Google Patents "Citations" + "Cited By" tables for US 11402417; (b) the WO 2020/216671 A1 search-report front page (ISA citations); (c) USPTO Patent Center file wrapper (IFW) for application 17/601,629 to capture the PTOL-892/1449 reference list and any examiner-cited art; (d) full text of references 2–8.

Bottom line: On the verified record, no cited reference anticipates a claim of US 11402417; the citations of record read as § 103 background art in the field of aggregate-vs-declared electricity-theft detection. The complete § 102 picture requires the outstanding full-text and ISR verification flagged above.

I'll research prior art relevant to the claims of US 11402417 to ground the obviousness analysis.

Obviousness Analysis — US 11402417 B2 under 35 U.S.C. § 103

Evidentiary basis and caveats (read first)

Two limitations on this analysis, stated up front:

1. The "Prior Art" section of the fetched page is the patent's own background narrative, not a citation list. The Google Patents text I have does not include the automated "Citations / Cited By / Similar Documents" panels. So I cannot tell you which references the examiner actually applied. What I can do is (a) treat the specification's admitted prior art as prior art for § 103 purposes, and (b) use search-surfaced references in the same field as secondary art. I have not verified the citations on the face of the patent, and I flag this rather than assume.
2. The claims section was not in the fetched text. I reconstructed claim scope from the "Disclosure of the Invention" paragraphs and from the published application US 2022/0146558 A1, where the search results confirm claim 9 = non-transitory storage medium and claim 10 = centralising meter (apparatus). The dependent claims are numbered below by inference; treat the numbering beyond claims 9–10 as unverified.

1. The claims at issue (in substance)

Claim 1 (method, centralising meter):

• first PLC unit ↔ data concentrator, first frequency band; second PLC unit ↔ remote basic meters (RBMs), second distinct frequency band;
• emulate a smart-meter application toward the concentrator for each registered RBM, each emulated application constructing a load curve;
• measure, via a general metrology unit, consumption of all installations the RBMs supervise;
• when (i) ΔEa_D/ΔEa_G falls below a threshold and (ii) the load curves of a plurality of RBMs show a drop to zero, identify the fraudulent RBM by extrapolating those load curves and comparing the estimated total ΔEa_E to the measured ΔEa_G.

Dependent claims (variously 2–8): ΔEa_E includes a predetermined intrinsic consumption ΔE_C; "closest" identification; min |1 − ΔEa_E/ΔEa_G|; max ΔEa_E/ΔEa_G; state-recovery via a dedicated atomic command frame and exclusion of RBMs whose breaker member and/or downstream breaker is open; and a T/2 extrapolation window.

Claim 9: CRM storing the program. Claim 10: apparatus (centralising meter) counterpart of claim 1.

2. Person having ordinary skill in the art (POSITA)

A smart-metering / powerline-communication engineer with a few years' experience in AMM systems, familiar with G3-PLC (ITU-T G.9903), PRIME (G.9904), DLMS/COSEM (IEC 62056-5-3 and the DLMS "Bluebook"), with knowledge of load-profile objects, atomic meter commands, and non-technical-loss (NTL) / theft detection by energy balancing. This is the skill level the patent itself assumes (it defines all these acronyms as known).

3. Scope and content of the prior art

3.1 Admitted prior art in the specification (the strongest § 103 footing)

The background (§§ [0002]–[0005]) expressly concedes:

• PLC-AMM networks, G3-PLC, PRIME, data concentrators, DLMS/COSEM, and smart meters whose load curves are read "typically … every 15 minutes" — all known;
• the centralising-meter-as-proxy architecture itself is known: "it is possible to concentrate the applicative aspects of a plurality of electricity meters in one and the same centralising meter … a role of proxy … remote basic meters … essentially responsible for performing metrological operations and for responding to a simple set of atomic commands."

That admission disposes of the entire architectural half of claim 1 (dual-band PLC, emulation, load-curve construction). What remains is the fraud-detection logic.

3.2 Secondary prior art in the same field

The common thread: compare an aggregate "measured/supplied" figure with the sum of individual meter declarations, model the non-theft components (technical losses, intrinsic consumption), and localize the residual to a meter. That is precisely the ΔEa_D/ΔEa_G comparison and the intrinsic-consumption ΔE_C handling.

4. Claim-by-claim obviousness

Claim 1

Combination A — Admitted centralising-meter architecture + aggregate energy-balance NTL detection (US 9,945,889 B2 in view of EP 3 032 269 A1):

• Dual-band PLC units, emulation toward the concentrator, and load-curve construction → admitted prior art (§ [0004]); no inventive contribution.
• The general metrology unit measuring all supervised installations → inherent in the admitted proxy topology ("the electrical supply distribution of these electrical installations passes through the centralising meter").
• Comparing declared vs. measured consumption against a threshold → US 9,945,889 B2 (loss-rate threshold) and EP 3 032 269 A1 (aggregate discrepancy vs. individual meters).

Motivation to combine: Both the centralising-meter art and the NTL art are in the same field (AMM / electricity distribution), address the same problem (energy theft), and the specification itself states the objective is "to detect situations of energy-consumption fraud, despite this structural simplification." A POSITA implementing the admitted proxy meter would naturally reuse the well-established head-end-vs-sum balance already standard in the metering world — a combination of known elements producing only a predictable result (KSR rationale (a): known elements, known method, predictable outcome).

Combination B — adding the "plurality of zero-dropping load curves" trigger and the extrapolation step:

The only arguably non-routine element is using (i) the simultaneity of multiple zero-consumption load curves as the fraud signature and (ii) extrapolation of pre-drop load curves + recombination against ΔEa_G to isolate the culprit.

• The load curves are already being built (admitted). Detecting a drop to zero in a load profile — and recognizing that a sudden, sustained zero is anomalous — is routine profile-monitoring.
• Localizing the culprit from an aggregate imbalance is expressly taught by EP 3 032 269 A1 (statistical ratio function identifies "that electric meter as being suspicious") and by ES 2705086 A1 (residual after modelling losses = fraudulent consumption).
• The extrapolation step is, in substance, a leave-one-in energy balance: for each candidate, re-inject its estimated would-be consumption and ask whether the total then reconciles with ΔEa_G. Estimating missing/zeroed consumption from the immediately preceding profile is the same modelling operation ES 2705086 and US 9,945,889 perform for unknown loads and losses. Applying a known estimation technique to a known data set is a KSR rationale (b)/(c) combination.

A POSITA with the fraud signal in hand would have a clear, concrete motivation to localize (a fraud report that does not name a meter is commercially useless), and would predictably reach for the only data available — the pre-drop load curves — to do it. No unexpected result is claimed.

Dependent claims

Claim 9 (CRM) and claim 10 (apparatus)

Obvious for the same reasons on which claim 1 falls; a programmed processor and a device configured to carry out the method add nothing patentable beyond the method steps.

5. Counter-considerations and honest limits

• The EPO granted the counterpart (EP 3 953 717 B1, active; LOT Network grant list, filed 2020-04-03). A granted European counterpart is a data point against obviousness, though the EPO's problem-and-solution approach differs from § 103 and European grants are not dispositive in the U.S.
• Family members exist (FR 3 094 795 A1, CN 113 728 239 A, US 2022/0146558 A1), meaning the disclosure was prosecuted across multiple offices; this cuts both ways.
• No secondary considerations are evidenced in the record before me: no unexpected results, no licensing-due-to-patentability, no long-felt unmet need tied to the anti-fraud logic specifically. The stated commercial advantage ("identify fraud without increasing RBM processing resources," § [0005]) flows from the admitted proxy architecture, not from the extrapolation step, so there is no nexus between that advantage and the asserted inventive contribution.
• I have not seen the face-of-patent citations, so I cannot confirm whether the examiner already considered and distinguished US 9,945,889 B2, EP 3 032 269 A1, ES 2705086 A1, or US 2008/0109387 A1. If they were never applied, they are plausible new § 103 grounds; if they were applied and overcome, the patentee's arguments would need to be evaluated.

6. Bottom line

Claim 1 is prima facie obvious. Its architectural elements are admitted prior art in the patent's own background. Its detection logic — aggregate declared-vs-measured energy balance with a threshold and modelling of non-theft contributions — is old in the metering art (US 9,945,889 B2; EP 3 032 269 A1; ES 2705086 A1; US 2008/0109387 A1). Its localization logic — identifying the specific meter responsible from an aggregate imbalance and estimating would-be consumption from existing profile data — is the predictable union of (a) known NTL localization techniques and (b) routine extrapolation of a load profile, combined with the admitted load-curve infrastructure. KSR rationales (a)–(c) all apply.

Dependent claims 2–8 add only routine accounting, selection criteria, a known atomic command used for anti-false-positive filtering, and an arbitrary extrapolation window — none of which supplies an unexpected result or a nexus to a non-obvious advance. Claims 9 and 10 rise and fall with claim 1.

Confidence: Moderate-high on the § 103 substance, given the breadth of the admission and the density of the NTL prior art. Lower on completeness, because I could not retrieve the face-of-patent citation list or the exact dependent-claim numbering — those should be verified against the patent's front page and the file history before this is used in any IPR or office action.

Sources consulted: the fetched Google Patents full text of US 11402417 B2; US 2022/0146558 A1 and the FR/CN/EP family pages; search-surfaced disclosures of US 9,945,889 B2, EP 3 032 269 A1, ES 2705086 A1, and US 2008/0109387 A1. No litigation reference to US 11402417 was located in the earlier search, consistent with the litigation section above.