US 11402443

I'll search for this specific patent number across USPTO and CAFC docket sources.

Let me probe specifically for litigation or PTAB proceedings tied to this patent number, and verify the CAFC "'243" reference isn't a different patent.

Let me verify whether the CAFC "'243 patent" reference (involving a "Tian" publication) is actually this patent, and check for any Fujitsu-related appeals.

Summary: US Patent No. 11,402,443 B2

Scope note on the search: I searched for the exact number 11402443 and for its litigation/PTAB/CAFC footprint. I found no CAFC 2026 docket, no district court case, and no PTAB (IPR/PGR) proceeding naming US 11,402,443. One important near-miss to flag so it isn't confused: the heavily reported CAFC 2026 decision The Nielsen Company (US), LLC v. TVision Insights, Inc., No. 2025‑1371 (Fed. Cir. Aug. 14, 2026), concerns a "’243 patent," but that patent is U.S. Patent No. 11,470,243 (audience measurement/image capture, Nielsen) — a different patent addressing an unrelated technology. It is not US 11,402,443.

Also flagged: one search hit (ptab.us, an Ex Parte Appeal listing) shows the bare string "11402443" in connection with "Ex Parte Fermin et al … Avery Dennison Corporation." That appears to be an application/appeal serial number (11/402,443), not the patent 11,402,443, and is unrelated. I did not auto-correct either identifier.

---

Bibliographic data (from the patent text and Justia)

Abstract (as issued): A magnetic characteristic measuring apparatus measures a magnetic characteristic of a sample having a core wound by a primary coil and a secondary coil with high accuracy at high frequency. The apparatus includes an air-core coil wound by a primary coil and a secondary coil, and a capacitor. The primary coil of the sample, the primary coil of the air-core coil and the capacitor are serially connected, and the capacitance C_L of the capacitor is selected as ⅓ω²L_a ≤ C_L ≤ 1/ω²L_a, wherein L_a is an inductance of the air-core coil at a frequency ω/2π.

---

Plain-language overview of the independent claims

Claim 1 — Apparatus (the core hardware claim).
A magnetic-characteristic measuring apparatus for testing a sample that itself has a primary coil and a secondary coil (e.g., a magnetic core wound with drive and sense windings). The apparatus adds:

• an air-core coil (a coil with no magnetic core) that also has its own primary and secondary windings;
• the sample's primary coil is serially connected to the air-core coil's primary coil; and
• a capacitor is serially connected in that same loop.

Two numeric conditions must hold:

1. Capacitance window: ⅓ω²L_a ≤ C_L ≤ 1/ω²L_a, where L_a is the air-core coil's inductance at frequency ω/2π.
2. Inductance matching: the air-core coil's inductance L_a must be 0.95 to 1.05 times the sample's inductance L_m.

Put simply: you measure core loss by putting a reference air-core coil in series with the device under test and resonating/cancelling the inductive component with a series capacitor, so that the current and voltage stay closer to phase and the loss can be extracted accurately at high frequency — while keeping the supply voltage (and thus the impedance seen by the source) from ballooning. The claim ties the concept down to a specific capacitor range and a ±5% inductance match between reference coil and sample. The inductance-match limitation (L_a = 0.95–1.05 × L_m) is what distinguishes the granted claim from the broader disclosure; it originated in the second-embodiment teachings (Table 1).

Claim 4 — Method (the corresponding process claim).
A magnetic-characteristic measuring method comprising, in substance, the steps of:

1. serially connecting the sample's primary coil (sample = primary + secondary coil) to the primary coil of an air-core coil (which has primary and secondary coils);
2. serially connecting a capacitor meeting the same ⅓ω²L_a ≤ C_L ≤ 1/ω²L_a range to the air-core coil's primary coil, where L_a is the air-core coil inductance at ω/2π and L_a is 0.95–1.05 times the sample inductance L_m;
3. applying AC voltage across the sample primary coil, air-core coil primary coil, and capacitor; and
4. measuring the sample's secondary voltage and the air-core coil's induced (secondary) voltage.

Dependent claims (context): Claim 2 adds an exciting power supply with an AC frequency ≥ 1 MHz. Claim 3 adds a measuring unit that reads the sample secondary voltage and the air-core coil induced voltage, plus a control unit that computes the core loss from those values (and the current-derived values).

---

Technological context and prosecution history (from the file)

• The specification explains that conventional B-H-analyzer core-loss measurement degrades at high frequency (1–10 MHz) because the phase angle θ between exciting current I_R and secondary voltage V₂ approaches 90°, collapsing effective power (I_R V₂ cos θ).
• Two reviewed alternatives are described: (a) a series capacitor tuned to resonance (FIG. 2), which works only at the resonant frequency and loses versatility; and (b) a series air-core coil without a capacitor (FIG. 3), which cancels inductive error but raises total impedance so much that a large, high-power exciting supply is needed. The claimed invention combines both: air-core coil plus a capacitance selected in the derived window, which brings θ near zero while suppressing the impedance rise.
• The capacitance bound is derived mathematically in the specification. Expression (6) is obtained from |Z₂|² ≤ |Z₁|² under L_m = L_a: ⅓ω²L_a ≤ C_L ≤ 1/ω²L_a.
• The inventor's empirical finding (Table 1) is that measurement error is minimized when L_a ≈ L_m; a ±3% error tolerance corresponds to L_a between about 4.77 µH and 5.22 µH for an L_m = 5.0 µH sample at 5 MHz, i.e., roughly 95%–105% — reflected in the claim as 0.95–1.05.
• Prior art cited on the face of the patent: JPH 03‑221886 A (Matsushita Electric Works, "Method for measuring core loss"), JPH 05‑172923 A (Fuji Electric, "System for measuring high-frequency magnetic characteristic"), US 5,175,476 A (General Electric), and JP 2005‑233818 A (Toyota Central R&D, inductance measuring device/probe/method).
• Non-patent literature cited: M. Mu et al., "New Core Loss Measurement Method for High-Frequency Magnetic Materials," IEEE Trans. Power Electron. 29(8):4374‑4381 (Aug. 2014); D. Hou et al., "New High-Frequency Core Loss Measurement Method With Partial Cancellation Concept," IEEE Trans. Power Electron. 32(4):2987‑2994 (Apr. 2017).
• Prosecution timeline: application filed June 25, 2019; assigned to Fujitsu Limited July 1, 2019; non-final office action Sept. 15, 2021; response Dec. 23, 2021; final rejection Jan. 25, 2022; response after final Apr. 29, 2022; notice of allowance May 16, 2022; patent granted Aug. 2, 2022. A maintenance-fee reminder was mailed March 23, 2026 (per the Google Patents legal-events record) — consistent with the patent being maintained but not indicative of any contest.
• Later documents citing this family include JP 2021‑196257 A (Fujitsu, magnetic loss estimation), JP 2022‑038455 A (Magnetic Device Laboratory), and US 11,307,266 B2 (Virginia Tech Intellectual Properties, "Core loss characterization and measurement").

---

Uncertainty / limitations of this report

• No litigation or post-grant proceeding found. Searches of PTAB/IPR docket aggregators and Federal Circuit 2026 reporting surfaced no case, IPR, PGR, or appeal involving US 11,402,443. I cannot rule out an action that simply is not indexed in the sources my searches reached (I could not directly query USPTO PatentCenter, PACER, or CourtListener for this exact patent). If litigation exists, it is not visible in the public results I retrieved.
• I expressly did not treat Nielsen v. TVision (U.S. 11,470,243) as relating to this patent; the numbers are different and the subject matter is unrelated.
• The quoted figures (issue date, expiration, family members, cited art) come from the Google Patents record for US11402443 and Justia's parallel page; the authoritative patent text supplied to me corroborates title, claims, and disclosure.
• Date note: my tool-collected results include material dated as late as August/September 2026 (e.g., the Aug. 14, 2026 Nielsen v. TVision opinion and a Sept. 1, 2026 blog post), which is later than the "April 26, 2026" date stated in the request. I have not reconciled those dates; I simply note the discrepancy rather than assume which is correct.

I'll search for litigation involving US patent 11402443 across the sources you specified.

Initial searches returned only unrelated matches (e.g., an unrelated application number in a 2014 PTAB decision, a UK company registration number). Let me search targeted litigation sources.

Summary of Findings

No litigation involving US Patent No. 11,402,443 is known.

I searched litigation-focused sources for the specific number 11402443 (and the equivalent "11,402,443"), and did not identify any district court case, ITC investigation, or Federal Circuit appeal asserting or adjudicating this patent. Per your instruction, I did not return results for similar numbers; I am reporting only what matched the exact identifier.

What the patent is (for context)

• Patent: US 11,402,443 B2 — "Magnetic characteristic measuring apparatus and method"
• Assignee: Fujitsu Limited
• Inventors: Hiroyasu Kawano, Yuji Uehara, Hirotaka Oshima
• Appl. No.: US 16/451,562; Filed: 2019-06-25; Priority: 2018-07-05 (JP 2018-128547)
• Granted: 2022-08-02; Adjusted expiration: 2040-08-24; status Active
• Claim scope (independent claim 1): a magnetic characteristic measuring apparatus with a sample coil, an air-core coil, and a series capacitor, where capacitance C_L satisfies ⅓ω²L_a ≤ C_L ≤ 1/ω²L_a, and where the air-core coil inductance L_a is 0.95–1.05 × the sample inductance L_m.

Sources checked

• Unified Patents portal (litigation/caselist and patent pages) — no litigation entry for US 11,402,443
• Google Patents page for US 11,402,443 (the authoritative record) — shows no "Litigation" section and no litigation-family entries; only the patent family (US 16/451,562; JP 7147299) and citing-family references
• CAFC and Docket Alarm style results — no matches tied to this patent number
• General web searches for the number plus litigation terms — matches returned were unrelated (see below)

False positives to note (explicitly NOT this patent)

• A 2014 PTAB entry "Ex Parte Fermin et al 11402443" (Tech Center 1712) — this is an application/decision number, entirely unrelated to the Fujitsu patent.
• A UK Companies House notice listing "11403887" etc. — a UK company registration number, unrelated.
• A "RE'543" reference in an IPR — this is RE45543 (a different, unrelated reissue patent).
• The "Fujitsu v. UI" materials relate to Fujitsu's LCD/plasma-display patent litigation (different patents entirely), not the '443 patent.

Caveats

• My coverage of PACER is indirect (I do not have live PACER access here); a complaint filed very recently (e.g., within the last few weeks) might not yet be reflected in the indexed sources I searched. The patent issued in August 2022 and the sources above are current as of the search, so any unnoticed case would have to be recent and low-profile.
• A patent may also be implicated indirectly via its Japanese counterpart JP 7147299 B2, but that is a separate patent and not US 11,402,443.

Conclusion

Based on the searches run, there are no known litigation proceedings involving US Patent No. 11,402,443. If you need a definitive negative, I'd recommend a direct PACER/RECAP docket search on the patent number and a Lex Machina or Westlaw Docket search cross-referenced to "11,402,443," since those proprietary databases provide the most complete coverage of recently filed complaints.

No PTAB hits so far. Let me run a few more targeted checks before concluding.

Proceedings overview

Total AIA trial proceedings on US 11,402,443: 0. The structured "PTAB proceedings on file" block (USPTO Open Data Portal) returns no IPR, PGR, or CBM for this patent, and targeted web searches (PTAB petition documents, PTAB E2E/PTABTS, Docket Alarm, and general web) surfaced no AIA trial proceeding, no Final Written Decision, and no Federal Circuit appeal involving the '443 patent as of 2026-09-29. That is the whole picture: 0 active / 0 claims invalidated by PTAB / 0 claims sustained / 0 settled / 0 institution denials. The bottom-line defensive posture is therefore not "the patent has survived IPRs and is hardened" and not "the claims are canceled" — it is simply "no one has tested these claims at the PTAB yet." All four issued claims (1–4) are untested in an AIA forum, and no § 315(e)(2) estoppel has attached to any party.

One framing correction worth flagging up front: the patent owner is Fujitsu Limited, an operating company, and the '443 patent (a core-loss measurement instrument) shows no sign of the mass-assertion/PAE pattern the prompt's "troll" language assumes. Treat this as a defensive-clearance / freedom-to-operate file, not a troll-defense file.

---

No proceedings to report

There is no proceeding entry to populate. I checked:

• The canonical ODP-derived block in the prompt → no AIA trials.
• Web search for IPR/PGR/CBM petitions naming "11402443," "11,402,443," or the title Magnetic characteristic measuring apparatus and method → no petition documents, no institution decisions, no FWDs.
• Federal Circuit / CourtListener-style docket queries for the patent number → no appeal found.

I did not find any proceeding number, so per your "do not invent proceeding numbers" constraint, I am listing none. Other patents appear in the same technical space and in PTAB filings (e.g., magnetic/core-loss art in IPR2023-00993, IPR2025-01098, and various Fujitsu litigation over unrelated patents), but none of those involve the '443 patent and none should be conflated with it.

---

Strategic summary

Claim status. All four claims — independent claim 1, dependent claims 2 and 3, and independent method claim 4 — are UNTESTED at the PTAB. None has been canceled, disclaimed, or held unpatentable in an AIA trial. Claim 1 recites the core structural combination (sample primary coil + air-core coil primary + series capacitor, with C_L bounded by ⅓ω²L_a ≤ C_L ≤ 1/ω²L_a) and the added limitation that the air-core coil inductance L_a is 0.95–1.05× the sample inductance L_m; claim 4 mirrors that onto a method. The patent is Active with an adjusted expiration of 2040-08-24, so there is roughly 14 years of runway — plenty of time for an IPR/PGR to be filed if commercial exposure appears.

Estoppel landscape. Because no IPR/PGR has been instituted, no § 315(e)(2) estoppel exists against anyone. For a defendant now served with an infringement complaint, the full toolbox remains open: § 102 anticipation, § 103 obviousness, and § 112 written-description/enablement challenges to L_a–L_m range limitations and the C_L formula are all available in an IPR (Note: § 112 is not available in an IPR; it is available in a PGR only. If you need § 112, the vehicle is a PGR filed within 9 months of grant or a district-court invalidity defense — the '443 patent granted 2022-08-02, so the PGR window is long closed at 2026-09-29.) Practically, that means IPR under §§ 102/103 is the live PTAB vehicle, and § 112 arguments go to the district court. Two constraints matter: the § 315(b) one-year bar runs from service of a complaint alleging infringement, and § 325(d) discretionary denial may bite if you rely on art already of record (the examiner considered the Mu 2014 and Hou 2017 papers and the JP references — see below).

Pattern signals. No petitioner has filed against this patent — not once, let alone multiple times. No defensive aggregator (Unified Patents, RPX, etc.) appears in the chain. The family is small: one US patent and one JP counterpart, JP7147299B2, plus the pre-grant publication US20200011944A1. The patent owner has pursued no PTAB appeal (there is nothing to appeal). The only third-party activity in the record is citing art — e.g., US11307266B2 (Virginia Tech, "Core loss characterization and measurement") and JP publications JP2021196257A and JP2022038455A — which are evidence of a live technical field, not of validity challenges.

Why the absence is itself informative. Well-asserted, commercially significant patents in litigious sectors almost always attract IPRs. That this one has not suggests either (a) Fujitsu has not asserted it against anyone, or (b) any assertion has been confined to contexts (licensing, standard-setting, internal design) that never triggered a § 315(b) clock. Either way, an accused infringer today would be the first to test these claims — meaning you would bear the cost of building the prior-art record from scratch, but you would also face zero adverse estoppel and zero Fintiv-driven "the patent has already been validated by the Board" narrative.

---

Recommended next steps

1. Treat the record clean. Confirm independently on PTAB E2E / the USPTO Patent Trial and Appeal Board End-to-End system and on the Federal Circuit's docket (and CourtListener) that no proceeding exists before relying on this memo — the ODP ingest can lag. A five-minute check of PTAB E2E and CourtListener for "11402443" is warranted before any formal opinion.

2. If you are a defendant and want PTAB relief, act within § 315(b). File an IPR within one year of service of the first complaint alleging infringement. There is no estoppel working against you, so you can raise the strongest §§ 102/103 grounds immediately. Because § 112 is unavailable in IPR, pair the IPR with a district-court § 112 defense targeting the numeric range limitations (L_a = 0.95–1.05×L_m and ⅓ω²L_a ≤ C_L ≤ 1/ω²L_a), which are classic written-description/enablement attack surfaces given the specification's relatively thin disclosure of how those ranges were derived (the spec reports only a single test series at 5 MHz with L_m = 5.0 μH).

3. Mine the prosecution and file histories before filing. The examiner expressly considered M. Mu et al., IEEE Trans. Power Electron. 29(8):4374-4381 (2014) and D. Hou et al., IEEE Trans. Power Electron. 32(4):2987-2994 (2017), along with JPH03221886A, JPH05172923A, and JP2005233818A. Anticipate a § 325(d) argument if you recycle those; better to lead with art not of record. Note the file history shows a non-final action (2021-09-15), a final rejection (2022-01-25), and an after-final response (2022-04-29) before allowance (2022-05-16) — worth reading to see exactly what overcame the rejection.

4. If no proceeding is pending and none is filed, there is nothing to track. Do not calendar institution deadlines, oral hearings, or FWD dates — there are none. The only dates that matter are the patent's 2040-08-24 expiration and any § 315(b) one-year window triggered by a future complaint.

I'll reconstruct the ownership chain for US 11402443 using the patent's own legal-events record plus live assignment-search and litigation sources.

Ownership Analysis — US 11,402,443 B2

Patent: US 11,402,443 B2 — "Magnetic characteristic measuring apparatus and method"
App. No.: 16/451,562 · Filed: 2019-06-25 · Granted: 2022-08-02 · Status: Active (adjusted expiration 2040-08-24)
Priority: JP 2018-128547 (2018-07-05)

Sourcing note (read first): My tool access was limited to web search; I was not able to open the USPTO Assignment Center record page directly. The assignment data below is taken from the patent's Google Patents legal-events feed, which mirrors the underlying USPTO assignment record (reel/frame and conveyance text are reproduced verbatim from that feed). Where a field — specifically the correspondent of record — was not surfaced by any source I could reach, I flag it as unverified rather than infer it. Verification links are listed under Verdict.

---

Inventors

Basis for employer attribution: all three appear under the "Fujitsu Limited" assignee profile on PatentLeaderboard, and all three executed the inventor-to-employer assignment recorded at Reel 049641/0213 (see timeline). The application itself is a Fujitsu-filed application with a Japanese priority filing (JP 2018-128547) from the same corporate applicant.

Unusual patterns: none detected. There is no evidence in any source I could reach that any of the three inventors left Fujitsu, and no co-inventor group departure signal. This is a three-inventor, single-employer, single-country team — the opposite of the "inventors scatter shortly after filing" pattern that often precedes a portfolio sale. I could not verify post-filing employment continuity from public records; if that matters to you, the JP corporate registry / Fujitsu 有価証券報告書 would be the next source, not the assignment record.

---

Original assignee

Fujitsu Limited (Kawasaki, Kanagawa, Japan) — named on the face of the issued patent and confirmed as original assignee. The application was also filed by Fujitsu Ltd per the Google Patents "Application filed by" event (2019-06-25).

• Primary line of business: operating ICT company — servers, supercomputers, network/telecom equipment, software and services, plus semiconductor and power-electronics R&D. Publicly traded (Tokyo Stock Exchange). Not a holding or licensing entity.
• Product embodying the claims: this is the weak link in any "operating company ships the invention" analysis. The patent claims a core-loss measurement instrument/bench setup (exciting supply + current-sense resistor + air-core coil + series capacitor), i.e. an R&D metrology rig for characterizing high-frequency magnetic cores. There is no indication Fujitsu sells a commercial magnetic-characteristic analyzer; the apparatus reads as an internal design-support tool for high-frequency (1–10 MHz) DC-DC converter magnetics. I found no evidence either way on commercialization, so I record this as unclear / probably internal use.
• Current status: operating, not acquired, not dissolved, not in bankruptcy. Litigation of note (Fujitsu v. Belkin / Netgear, N.D. Cal. 5:10-cv-03972) is a decade-old, unrelated Wi-Fi portfolio matter and does not touch this patent.

---

Assignment timeline

One recorded assignment exists. It is the filing-time inventor→employer transfer. There are no recorded post-issuance assignments — no transfer to any LLC, aggregator, or asserter.

• 2019-05-31 to 2019-06-05 (executed; per-record signing dates) / recorded 2019-07-01 — Reel 049641 / Frame 0213
  • Conveyance: Assignment — recorded text: "ASSIGNMENT OF ASSIGNORS' INTEREST"
  • Assignor: Kawano, Hiroyasu; Uehara, Yuji; Oshima, Hirotaka (each signing individually)
  • Assignee: Fujitsu Limited (Japan)
  • Correspondent: not verified. The Google Patents legal-events entry records only the reel/frame and the assignment text; it does not reproduce the correspondent of record, and my searches did not surface it. I will not guess a name. Pull the record at Reel 049641/0213 in the Assignment Center to capture the correspondent and address fields.
  • Context: routine inventor-to-employer assignment of rights, executed pre-filing (all signing dates precede the 2019-06-25 US filing date) and recorded six days after filing — the ordinary Fujitsu corporate patent-capture sequence, not a monetization event.

Subsequent legal events are administrative only: allowance 2022-05-16, patent grant 2022-07-13, maintenance-fee reminder mailed 2026-03-23. The "2040-08-24" adjusted expiration is a term adjustment, not a transfer.

---

Timeline diagram

timeline
    title Ownership of US 11402443
    2018 : JP priority application filed by Fujitsu
    2019 : Inventors assign rights to Fujitsu Limited
         : US application 16451562 filed
    2022 : US 11402443 granted to Fujitsu
    2026 : Maintenance fee reminder mailed

---

NPE / troll-pattern signals

1. Shell-entity transfer — not present. No assignment out of Fujitsu is recorded. The only recorded transfer runs into an operating company, at Reel 049641/0213 on 2019-07-01. No "IP / Holdings / Ventures / Licensing" entity appears anywhere in the chain.

2. Known asserter in the chain — not present. Neither the original assignee nor any recorded assignee matches Acacia, Marathon, IV, IPNav, Wi-LAN, Conversant/Mosaid, Vringo, Pendrell, Round Rock, or any Spangenberg-affiliated entity. The current owner of record is Fujitsu Limited.

3. Repeat correspondent across the chain — not present / unclear on the merits. There is only one link in the chain, so recurrence is impossible by construction. I could not retrieve the single correspondent name, so I cannot state whether it recurs on other Fujitsu recordings. This is the one field you should re-pull manually — it is the only missing data point of substance in this docket.

4. Cascading transfers — not present. One recorded assignment in a seven-year ownership history; no chained LLCs, no transfers within 24 months of each other, no shared-principal pattern to analyze.

5. Pre-litigation transfer — not present. No infringement suit naming this patent was found in any source I could reach (RPX litigation feed, CourtListener, Law360). The sole recorded assignment predates filing, not litigation.

6. Bankruptcy fire-sale — not present. Fujitsu has not filed Chapter 7/11, and no trustee/conveyance-of-estate recording exists at this reel block for this patent.

7. Privateering — not present. No transfer to a third party, let alone one asserting on Fujitsu's behalf.

8. Defensive aggregator — not present. Chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN.

---

Verdict

Insufficient data — the chain consists solely of the original inventor-to-employer assignment.

The complete ownership record is a single conveyance: the three inventors assigned to Fujitsu Limited, executed 2019-05-31 through 2019-06-05 and recorded 2019-07-01 at Reel 049641 / Frame 0213, with Fujitsu still the owner of record and the patent active through 2040-08-24. No NPE signal is present, and the subject matter — a bench-top core-loss measurement instrument for high-frequency magnetics — is low-value to assertion campaigns, which is consistent with a portfolio that has never been asserted. The correct classification under your rubric is "only the original assignment," not "operating-company assertion," because no assertion exists.

Verification links

• USPTO Assignment Center: https://assignmentcenter.uspto.gov/ (search 11402443 or reel/frame 049641/0213)
• Legacy Patent Assignment Search: https://assignment.uspto.gov/patent/index.html
• Patent text and legal events: https://patents.google.com/patent/US11402443/en

Open item: confirm the correspondent of record at Reel 049641/0213. It is the only field that would change the analysis if it turned out to be a repeat NPE-recording firm — though given the single-link chain and operating-company assignee, that outcome is unlikely.

I'll search for the patent and its cited prior art references.

Let me search for the specific prior-art references on patent databases.

US 11,402,443 B2 — Prior-Art Analysis of the Cited References

Note on method/limits: My live searches confirmed the patent's identity, ownership, and its own front-page citation lists, but the search tool hit its step limit before I could pull the full text of every cited reference. The identification, dates, and assignees below are taken from the authoritative Google Patents record you supplied (https://patents.google.com/patent/US11402443/en) and from the searches I did complete. Where I could not verify a reference's internal disclosure, I say so explicitly rather than guessing at its content.

---

1. The patent under analysis (bibliographic anchor)

Key limitations to be met for §102 anticipation:

• Claim 1: sample with first primary + first secondary coil; air-core coil with second primary + second secondary coil; first primary serially connected to second primary; capacitor serially connected to the second primary; and ⅓ω²L_a ≤ C_L ≤ 1/ω²L_a; and L_a = 0.95 to 1.05 × L_m.
• Claim 2: adds an exciting supply with AC frequency ≥ 1 MHz.
• Claim 3: adds a measuring unit (secondary voltage V₂ and induced voltage V_L) and a control unit computing core loss.
• Claim 4: method counterpart of claim 1 (with the same capacitor range and 0.95–1.05 L_a range).

---

2. References cited on the face of US 11,402,443 (the "prior art" of record)

A. Patent citations (4)

1. JPH03221886A — "Method for measuring core loss"

• Assignee: Matsushita Electric Works Ltd
• Filing/priority: 1990-01-26; Publication: 1991-09-30
• Description: A core-loss measurement method — a classic two-winding (exciting/primary + sensing/secondary) approach around a magnetic core, i.e., the same genus as the patent's own background FIG. 1.
• §102 relevance to claims 1/4: None apparent. It predates and does not appear to disclose the air-core coil, the series capacitor, or the ⅓ω²L_a–1/ω²L_a / 0.95–1.05 L_a ranges. It is background art for the core-loss measurement step only; any use would be §103, not §102.

2. US5175476A — "Magnetically tunable starting circuit for an electrodeless high intensity discharge lamp"

• Assignee: General Electric Company
• Filing: 1992-04-16; Grant: 1992-12-29
• Marked "cited by examiner" on the front page.
• Description: A starting circuit for an electrodeless HID lamp using a tuned (inductive–capacitive) resonant network — the kind of series L–C tuning arrangement used to control impedance/phase.
• §102 relevance: None. It is not a magnetic-characteristic or core-loss instrument. It is almost certainly cited only as evidence that series L–C tuning of a coil circuit was known. That is a §103-style "general knowledge" citation; it does not teach a sample coil, an air-core coil, or the claimed capacitance constraint.

3. JPH05172923A — "System for measuring high-frequency magnetic characteristic"

• Assignee: Fuji Electric Co Ltd
• Filing/priority: 1990-09-14; Publication: 1993-07-13
• Description: A high-frequency magnetic-characteristic measurement system. This is one of the three JP references the specification itself names as "related art," and is the closest subject-matter match among the patent citations (a high-frequency B–H/core-loss-type instrument).
• §102 relevance to claims 1–4: None apparent. High-frequency magnetic measurement as a field is disclosed, but the distinguishing elements of claim 1 — the air-core coil in series with the sample primary, the series capacitor, the ⅓ω²L_a–1/ω²L_a bound, and the 0.95–1.05 L_a relation — are not disclosed. Use would be §103 at most.

4. JP2005233818A — "Inductance measuring device, measuring probe and measuring method"

• Assignee: Toyota Central R&D Labs Inc
• Filing/priority: 2004-02-20; Publication: 2005-09-02
• Description: An inductance-measuring device/probe — relevant to the measurement of L_a and L_m that the patent's method steps (S102/S104/S202) rely on.
• §102 relevance to claims 1/4: None. A tool for measuring inductance is not the claimed measuring apparatus; it lacks the series capacitor, the air-core coil topology, and both numeric ranges.

B. Non-patent literature citations (2) — the most relevant art

5. M. Mu, Q. Li, D. J. Gilham, F. C. Lee, K. D. Ngo, "New Core Loss Measurement Method for High-Frequency Magnetic Materials," IEEE Trans. Power Electron., vol. 29, no. 8, pp. 4374–4381, Aug. 2014 (8 pages).

• Description: Introduces the modern "cancellation" approach to high-frequency core-loss measurement using a cancellation inductor in series with the device-under-test so that the inductive (non-loss) voltage component is subtracted, avoiding the near-90° phase problem.
• §102 relevance: This is the closest conceptual prior art to the air-core coil element of claims 1/4. It does not, on the record available, disclose the series capacitor or either numeric range (⅓ω²L_a–1/ω²L_a; L_a = 0.95–1.05 L_m). So it supports, at most, §103 against the "air-core coil" element — which is precisely why the granted claims carry the tuned-capacitor range and the 0.95–1.05 L_a limitation.

6. D. Hou, M. Mu, F. C. Lee, Q. Li, "New High-Frequency Core Loss Measurement Method With Partial Cancellation Concept," IEEE Trans. Power Electron., vol. 32, no. 4, pp. 2987–2994, Apr. 2017 (8 pages).

• Description: Extends Mu et al. with a "partial cancellation" idea to reduce the size/drive requirement of the cancellation coil.
• §102 relevance: Again the closest art on the air-core/cancellation coiI concept, and closer in time (2017, just before the 2018 priority date). Still does not teach the series capacitor with C_L within ⅓ω²L_a–1/ω²L_a, nor the L_a = 0.95–1.05 L_m relationship. Supports §103, not §102.

C. References cited by/against the patent family (3)

These are marked as examiner-cited against a family member (likely the JP counterpart):

7. US5420741A — "Passive tuned magnetic flux rate feedback sensing and control arrangement"

• Assignee: Allied-Signal Inc. | Filing 1993-11-15; Grant 1995-05-30
• Description: Magnetic-flux-rate feedback sensing using a passively tuned (L–C) network. Relevant only as evidence of tuned-LC magnetic sensing.

8. JPH0943328A — "Superconducting magnetic detector"

• Assignee: Sharp Corp | Filing 1995-07-28; Publication 1997-02-14
• Description: Magnetic detector (superconducting). Only tangentially related to magnetic sensing.

9. US8823370B2 — "High frequency loss measurement apparatus and methods for inductors and transformers"

• Assignee: Virginia Tech Intellectual Properties, Inc. | Filing 2011-08-31; Grant 2014-09-02
• Description: The most pertinent reference of this group — a high-frequency loss-measurement apparatus for inductors/transformers, i.e., the same technical problem (accurate high-frequency loss characterization, phase-error mitigation).
• §102 relevance to claims 1/4: None apparent on the available record. It is directed to loss measurement generally; it does not disclose the sample-coil + air-core-coil + series-capacitor topology with the ⅓ω²L_a–1/ω²L_a bound and the 0.95–1.05 L_a relation. Strongest candidate for §103 in this group.

---

3. Post-dating documents (NOT available as prior art)

For completeness, the front page also lists three documents that cite this family. All were filed in 2020, after the 2018-07-05 priority date, so they are not prior art under §102(a)(1)/(2) and cannot anticipate claims 1–4:

• JP2021196257A (Fujitsu, 2020-06-15 / 2021-12-27) — "Information processing device, magnetic loss estimation method…"
• JP2022038455A (磁気デバイス研究所, 2020-08-26 / 2022-03-10) — "Method for measuring loss of magnetic body"
• US11307266B2 (Virginia Tech, 2020-08-28 / 2022-04-19) — "Core loss characterization and measurement"

---

4. Bottom-line assessment

My conclusion: On the record available, none of the cited references (patent or non-patent) appears to anticipate any of claims 1–4 under 35 U.S.C. § 102. The individually closest references — Mu et al. 2014 and Hou et al. 2017 for the air-core "cancellation coil," and US8823370B2 for high-frequency loss measurement — supply the general coiI-cancellation/loss-measurement concepts, but none of them (per the material I could retrieve) discloses the combination that defines the claims: a series capacitor satisfying ⅓ω²L_a ≤ C_L ≤ 1/ω²L_a together with L_a = 0.95 to 1.05 × L_m. Those two limitations are the likely points of novelty over the art — consistent with the file history (non-final 2021-09-15, final rejection 2022-01-25, after-final response 2022-04-29, allowance 2022-05-16), which strongly suggests the claims were allowed only after amendment narrowing to the L_a/L_m range now recited in claims 1 and 4. These references are therefore best characterized as §103 obviousness candidates, not §102 anticipatory art.

Caveats: (a) I could not open the full specifications of JPH03221886A, JPH05172923A, JP2005233818A, JPH0943328A, US5175476A, US5420741A, or US8823370B2 before the search limit was reached, so my element-by-element statements for those depend on their titles/abstracts/assignees and the '443 specification's own characterizations. (b) The examiner cited only US5175476A with the "cited by examiner" marker in the four-patent-citation list; the others' examiner/third-party status should be confirmed on the face of the grant. (c) For a definitive anticipation/obviousness opinion, the full texts (and English translations of the JP references) must be reviewed — especially Mu 2014 and Hou 2017 — to verify whether either discloses a series capacitor or any inductance-matching relationship.

Sources: Google Patents record (authoritative, as supplied) https://patents.google.com/patent/US11402443/en; Justia patent page https://patents.justia.com/patent/[11402443](/patent/11402443); the IEEE citation details were confirmed in an independent bibliography (DTU "Advances in High-Frequency Inductor Design" reference list, which lists Mu et al. 2014 and Hou et al. 2017 exactly as cited on the '443 face).

I'll analyze obviousness of US 11,402,443 under § 103. First, let me verify the disclosures of the key prior-art references — especially the two IEEE papers and the cited Japanese applications, since those are the most likely primary references.

Now let me verify the disclosures of the cited Japanese applications and the remaining references, which the examiner relied upon.

Obviousness Analysis — U.S. Patent No. 11,402,443 B2

"Magnetic characteristic measuring apparatus and method" (Fujitsu Limited; inventors Kawano, Uehara, Oshima)
Effective filing / priority date: July 5, 2018 (JP 2018‑128547) — therefore all cited art dated 2017 and earlier qualifies as prior art.
Post‑AIA § 103 applies; the Graham/KSR framework governs.

---

0. Methodological note and evidence caveats

Per the strict rule, I have not auto‑corrected any identifiers. Two limits on my evidence must be stated up front:

• I could not retrieve the full text of JPH 03‑221886 A (Matsushita Electric Works, "Method for measuring core loss"), JPH 05‑172923 A (Fuji Electric, "System for measuring high‑frequency magnetic characteristic"), US 5,175,476 A (General Electric), or US 8,823,370 B2 (Virginia Tech) within this session. For those I rely on their titles, assignees, and the patentee's own characterization, and I flag where a conclusion depends on unverified content.
• I did verify the substance of the two IEEE papers that constitute the strongest prior art: Mu et al. 2014 (capacitive cancellation) and Hou et al. 2017 (partial cancellation using "a cancellation capacitor or inductor"). Sources: https://library.cnu.ac.kr/eds/detail/edseee_edseee.[6648460](/patent/6648460) (Mu abstract) and https://xplorestaging.ieee.org/document/[7479554](/patent/7479554) and https://www.semanticscholar.org/paper/.../f96edf63abc6769ff88a509e232f41c5b993a911 (Hou abstract).

The single most important observation for this analysis: the applicant's own specification supplies much of the prior art expressly. FIGS. 1–3 and their accompanying text are an admitted review of prior measurement apparatus (capacitor‑only cancellation at FIG. 2; air‑core‑coil‑only cancellation at FIG. 3). Applicant‑admitted prior art (AAPA) is fully available for § 103. Riverwood Int'l Corp. v. R.A. Jones & Co., 324 F.3d 1346 (Fed. Cir. 2003); MPEP 2129.

---

1. The claims to be tested

Note the drafting history: the Summary section reproduces the as‑filed claim language, which contains only the capacitance window — the L_a = 0.95–1.05·L_m limitation is not in the summary. That strongly suggests the inductance‑ratio limitation was added during prosecution (a non‑final action was mailed 2021‑09‑15 and a final rejection 2022‑01‑25 before allowance on 2022‑05‑16). The obviousness question therefore turns almost entirely on whether the 0.95–1.05 ratio would have been obvious.

---

2. Scope and content of the prior art

2.1 The two‑winding core‑loss method (background, undisputed)

The "two‑winding" method — excite primary, sense secondary, compute loss from ∫V₂·I_R — is the classical baseline. Hou states plainly: "the two‑winding method is widely used and considered a classical method… applicable to arbitrary wave excitation," but it "suffers from its sensitivity to phase discrepancy." Both the specification and Mu/Hou treat this as known. Elements "(sample with) primary + secondary coil," "measuring unit measuring secondary voltage," and "control unit computing core loss" are all standard.

2.2 Mu et al. 2014 — capacitive cancellation (primary reference)

Verified disclosure: a high‑frequency (1 MHz–50 MHz) core‑loss method that "utilizes capacitive cancellation," reducing phase‑induced error "from over 100% to <5%," demonstrated at 10 MHz. This is precisely the FIG. 2 concept — a series capacitor selected to cancel the reactive voltage of the device under test. Mu thus discloses or renders obvious every element of claim 1 except the air‑core coil, the air‑core‑coil secondary, and the 0.95–1.05 ratio; Mu supplies the claimed capacitance‑selection teaching and its rationale.

2.3 Hou et al. 2017 — partial cancellation (primary reference)

Verified disclosure: "[P]revious methods proposed to cancel the reactive voltage of the testing core by using a cancellation capacitor or inductor. However, the value of the cancellation component is critical, and a small variation can induce a big measurement error… [T]he paper presents a new measurement method with a partial cancellation concept that enables accurate core loss measurement for arbitrary wave excitation without the requirement to fine‑tune the cancellation component value," verified to 10 MHz.

Two points are decisive:

1. Hou expressly names an inductor as an alternative cancellation component — i.e., the air‑core coil of the claim — and places capacitor and inductor on equal footing as the same solution to the same problem.
2. Hou teaches that exact tuning of the cancellation component is undesirable, motivating operation anywhere inside a band of component values rather than at a single tuned point. That is exactly the claimed window ⅓ω²L_a ≤ C_L ≤ 1/ω²L_a (rather than a single value), and it directly rebuts any "critical range" argument.

2.4 Applicant‑admitted prior art (FIGS. 2 and 3)

The specification concedes:

• FIG. 2: series capacitor cancellation → phase ≈ 0, precise core loss, but "may require the measuring frequency to be a resonance frequency… lowers versatility."
• FIG. 3: series air‑core coil (primary + coaxially wound secondary) whose secondary induced voltage V_L is measured and used to "subtract[] a loss error generated due to the inductive component." Applicant concedes this yields a precise core loss, but "the inductance generated therein increases, resulting in a significantly increased impedance… the exciting power supply… may be required to be a large‑sized one with high power."

This is a textbook "known problem + two known partial solutions, each with a known deficiency" fact pattern.

2.5 Secondary references

• JPH 03‑221886 A (Matsushita, core‑loss measuring method) and JPH 05‑172923 A (Fuji Electric, high‑frequency magnetic‑characteristic measuring system) — cited on the face; both are in the same field (G01R 33/12) and are corroborative of the state of the art in high‑frequency core‑loss measurement. Content unverified; not load‑bearing below.
• US 5,175,476 A (GE) — a magnetically tunable starting circuit using an LC resonant network. Its likely evidentiary role is generic: capacitor‑inductor resonant tuning of a magnetic circuit is old. Content unverified.
• JP 2005‑233818 A (Toyota Central R&D) — verified as an inductance‑measuring device/probe/method (4‑terminal‑pair precision inductance/impedance measurement, including of parasitic inductance between terminals). This is the natural reference for the method step of measuring L_m (and L_a) to enable the ratio limitation.
• US 8,823,370 B2 (Virginia Tech, 2014) — "High frequency loss measurement apparatus and methods for inductors and transformers." Content unverified; same CPES lineage as Mu/Hou.

2.6 Level of ordinary skill (Graham factor 2)

A POSITA here is an electrical engineer (or physicist) with an M.S. or a B.S. plus 2–3 years in power‑magnetics/measurement instrumentation, familiar with impedance/admittance algebra, series LC resonance, B‑H analyzers, and phase‑discrepancy error in high‑frequency core‑loss measurement. Determining a series capacitance that makes total impedance no worse than the load's own impedance is routine algebra at this level.

---

3. The central obviousness ground: Mu in view of Hou

3.1 Claim chart for claim 1

3.2 Motivation to combine (KSR factor 3)

The motivation is explicit in the references themselves and in the applicant's own problem statement, satisfying KSR, 550 U.S. at 418–421, and MPEP 2144.04:

1. Same field, same problem, same solution mechanism. Mu, Hou, and the '443 patent all address "accurate high‑frequency core‑loss measurement." KSR: "if a technique has been used to improve one device, and a person of ordinary skill in the art would recognize that it would improve similar devices in the same way, using the technique is obvious." Both the capacitor and the inductor are described by Hou as the same means to the same end (cancelling reactive voltage).
2. Complementary, non‑overlapping benefits with known tradeoffs. The capacitor (Mu) nulls phase but is frequency‑locked; the air‑core coil (AAPA FIG. 3) permits the reactive‑voltage subtraction that yields precise loss but raises impedance and supply power. A POSITA seeking precision without a large supply has a finite, predictable set of options — one of which is to add the capacitor to the air‑core‑coil circuit. That is the classic "finite number of identified, predictable solutions" scenario, which KSR treats as obvious (id. at 421).
3. The combination is physically coherent, not a bodily incorporation. As the patent's own expression (3) shows, the net series reactance becomes (2ωL_a − 1/ωC_L); a capacitor lets the designer trim the total impedance back down. This is ordinary, predictable impedance engineering — the "predictable variation" of In re Keller / In re Merck.
4. No teaching away. Mu's capacitor method does not disparage inductor‑based cancellation; Hou affirmatively preserves it. AAPA FIG. 3 does not call the air‑core coil unworkable — it says it works accurately but is power‑hungry; a known disadvantage motivates a known cure, not abandonment. (A teaching away requires a "criticism, statement, or admonition" of the combination; none exists.)

3.3 The capacitance window is a routine, result‑effective optimization

The claimed window is not a magic number; it is the exact solution set of an inequality the patent derives itself. From expression (3)–(5), imposing |Z₂|² ≤ |Z₁|² with L_m = L_a yields, in one step of algebra, ⅓ω²L_a ≤ C_L ≤ 1/ω²L_a. The window's endpoints are simply the capacitances at which the total series impedance equals the sample's own impedance:

• C_L = 1/ω²L_a → net reactance (2ωL_a − ωL_a) = ωL_a = ωL_m (equal to the sample alone);
• C_L = 1/2ω²L_a → net reactance = 0 (full cancellation, the patent's own preferred value);
• C_L = 1/3ω²L_a → net reactance = −ωL_a (again equal in magnitude to the sample alone).

A POSITA who has decided to add a series capacitor so as not to increase the impedance seen by the supply will compute this same interval by routine arithmetic. "It is not inventive to discover the optimum or workable ranges by routine experimentation." In re Aller, 220 F.2d 454, 456 (CCPA 1955); In re Boesch, 617 F.2d 272 (CCPA 1980); MPEP 2144.05. Hou independently removes any contrary incentive to pin the value to a single point by teaching that fine‑tuning the cancellation component is unnecessary and that a partial‑cancellation band suffices.

3.4 The crux: L_a = 0.95–1.05·L_m

This is the only limitation not directly named in Mu or Hou, so it deserves the most scrutiny.

(a) The result is predictable, not unexpected. Applicant's own Table 1 shows error varying monotonically (with scatter) as a function of how close L_a is to L_m: −5.3% at 93.6%, −2.8% at 95.4%, +0.7% at 97.2%, −1.4% at 99.0%, +0.4% at 100.0%, +1.4% at 101.0%, +2.5% at 102.6%, +1.9% at 104.4%, +3.9% at 105.4%. Any skilled artisan would expect error to be minimized when the reference (air‑core) inductance equals the sample inductance — because the air‑core coil's induced voltage V_L is the quantity that is subtracted as the reactive‑component correction, and its fidelity improves with matching. The specification itself asserts precisely this as the inventor's finding of a natural design rule ("when the inductance of the air‑core coil 60 was equal to the inductance of the sample 10, a core loss might be obtained with significantly high precision"). A result that a POSITA would have predicted ex ante cannot support patentability. In re Geisler, 116 F.3d 1465 (Fed. Cir. 1997) (range must be shown critical/unexpected); In re Woodruff, 919 F.2d 1575 (Fed. Cir. 1990).

(b) The ±5% band is a "performance‑tolerance" range, not a criticality. The patentee selects ±5% only because a ±3.0% measurement error is "acceptable" — an arbitrary acceptance threshold ("If an error in a range of ±3.0% is acceptable, a core loss may be measured with a less error when the air‑core coil has an inductance L_a in a range equal to or higher than 4.77 μH and equal to or lower than 5.22 μH"). Choosing a tolerance band around a predictable optimum is the archetype of obvious‑to‑try optimization (KSR at 421; MPEP 2144.05). Note also the data are not clean: within the claimed band, 104.4% (+1.9%) and 102.6% (+2.5%) are worse than 97.2% (+0.7%), so the band boundary does not mark a sharp change in behavior — undermining any "critical range" theory.

(c) The "match the reference to the DUT" principle is itself routine measurement practice. Precision measurement by comparing a DUT against a reference element of nominally the same value is a generic technique; the application of it here (choose/select from a set of air‑core coils the one whose L_a is "close to" L_m, as the method claim's step 204 recites) is a matter of ordinary design choice. JP 2005‑233818 (precision inductance measurement, including of small parasitic inductance) supplies the enabling measurement step for L_m; if the examiner preferred a documentary anchor for the ratio, this is the reference to combine.

(d) KSR "design incentive." Once the air‑core coil is in the circuit, its inductance is a design parameter affecting both the magnitude of V_L and the total impedance. Tuning it toward L_m maximizes correction fidelity while (with the capacitor) holding impedance at or below the sample's own. MPEP 2144.04: a design incentive flowing from the intended use renders the variation obvious.

3.5 Claims 2 and 3 fall with claim 1

• Claim 2 (AC ≥ 1 MHz): Mu is directed to 1 MHz–50 MHz and demonstrates at 10 MHz; Hou is verified "up to 10 MHz." The claimed ≥1 MHz (spec's preferred 5 MHz) is squarely disclosed/obvious. No separate inventive contribution.
• Claim 3 (measuring unit + control unit computing core loss): the two‑winding method, the AAPA, Mu and Hou all measure the secondary voltage V₂ (and the reactive correction voltage V_L) and compute core loss by integration/multiplication; a personal‑computer control unit and an oscilloscope measuring unit are the applicant's own admitted off‑the‑shelf components ("The measuring unit 40 is an oscilloscope, for example… The control unit 80 is a personal computer, for example"). Implementing a known computation on a general‑purpose computer is obvious. In re Venner, 262 F.2d 91 (CCPA 1958).

3.6 Claim 4

Claim 4 is the process counterpart of claim 1, reciting the same serial‑connection topology, the same capacitance window, and the same 0.95–1.05·L_m ratio, plus the steps of applying AC voltage and measuring the two secondary voltages. All reasoning in §§ 3.1–3.4 applies mutatis mutandis; the method steps are the ordinary use of the Mu/Hou apparatus. In re Kuehl, 475 F.2d 658 (CCPA 1973) (apparatus/method obviousness track together).

---

4. Alternative and secondary obviousness grounds

1. AAPA (FIG. 2) + AAPA (FIG. 3), alone or with Mu/Hou. Both a series capacitor and a series air‑core coil are conceded prior art for cancelling the inductive error. Combining two known solutions that the specification itself says each cure part of the problem — FIG. 2 cures phase but is frequency‑locked; FIG. 3 is accurate but raises impedance — states a motivation verbatim in the four corners of the patent. No reference outside the patent is even required for the combination; only the numerical limits need supporting art (Mu/Hou/JP 2005‑233818).
2. Hou alone + JP 2005‑233818. Hou names a "cancellation… inductor," teaches a partial (band) cancellation needing no fine‑tuning, and is verified to 10 MHz; JP 2005‑233818 provides precision inductance measurement of the DUT. This ground is slightly weaker than § 3 because Hou does not itself emphasize matching the reference to the DUT, but it is defensible under KSR's "obvious to try" and design‑incentive rationales.
3. Any of the above further in view of US 5,175,476 (GE), if its disclosure confirms a tuned series‑LC magnetic‑circuit arrangement, as corroboration that capacitor‑plus‑inductor resonant reactance control in magnetics was old. Flagged: verification pending; do not rest the rejection on this reference without the full text.

---

5. Anticipated patentee rebuttals and their weaknesses

---

6. Conclusion

The claims of US 11,402,443 are, in my analysis, vulnerable to § 103 rejection. The strongest ground is:

Mu et al. (2014) in view of Hou et al. (2017), optionally further in view of JP 2005‑233818 A, and in view of Applicant's Admitted Prior Art (FIGS. 2–3) —

under which:

• Claim 1 would have been obvious: Mu supplies series‑capacitor reactive cancellation with a capacit‑ance chosen to null the reactive voltage; Hou supplies the series‑inductor cancellation alternative and expressly teaches that a band of cancellation values suffices without fine‑tuning; the capacitance window is ordinary minimization of total impedance; and the 0.95–1.05 inductance ratio is a predictable, evidence‑confirmed optimum whose ±5% tolerance derives from an arbitrary accuracy target — a routine optimization barring demonstrated criticality.
• Claims 2, 3, and 4 add only known elements (≥1 MHz operation verified in both papers; oscilloscope/P‑C measurement and computation; the corresponding method steps) and fall with claim 1.

If the Office were to maintain these claims, the only colorable distinction is the 0.95–1.05·L_m limitation, since the as‑filed disclosure/claim appears to have covered only the capacitance window. The obviousness of that ratio therefore reduces to a criticality/unexpected‑results question — and on the face of the specification and Table 1, the applicant has not made that showing.

Caveat on completeness: this conclusion should be confirmed against the full text of JPH 03‑221886 A, JPH 05‑172923 A, US 5,175,476 A and US 8,823,370 B2, which I could not retrieve here. If any of those expressly teaches (i) measuring the DUT inductance and (ii) selecting a reference/air‑core inductance equal to it, the case for obviousness becomes materially stronger and the 0.95–1.05 ratio would be squarely met by art of record.