US 11402354

I'll run several targeted searches to verify the patent's bibliographic data and check for any 2026 litigation activity.

The initial searches found the patent but no litigation. Let me dig for the actual claim text and any docket activity.

US 11,402,354 B2 — Verification Summary

Scope note: My searches returned the patent's own documents and secondary aggregators, but I could not confirm any USPTO assignment-record updates beyond the original, and I found no district-court or CAFC (2026 or otherwise) docket referencing this patent number. Details below, including one literal-identifier collision you should be aware of.

---

1. Bibliographic data (confirmed)

Primary CPC classifications include B06B 1/0215 (driving circuits for generating pulses/bursts), G01N 29/348 (excitation circuits with frequency characteristics), G01N 29/346, A61B 8/4488 (phased array transducers), and G01S 7/5202 / G10K 11/346.

---

2. Abstract (verbatim)

"A method is provided for generating ultrasound transmission waves comprising: a) providing an array of electroacoustic transducer elements each one being connected to an electric excitation signal generator by a dedicated feeding channel; b) feeding at least a part of the electroacoustic transducer elements with a pulsed electric signal having a predetermined frequency, a predetermined amplitude, a predetermined length or duration and a predetermined phase or a predefined delay with respect to the pulsed electric signals fed to the adjacent transducer elements, the said pulsed electric excitation signals comprising a sequence of pulses; c) modulating each or at least part of the pulsed electric signals by feeding to the transducer elements only a predetermined portion of the pulses of the said pulsed electric signals. Modulating is carried out by cutting the duration of each pulse of the pulsed electric excitation signal according to a predetermined time period."

---

3. Independent claims — plain language

Claim 1 (the only independent claim I could verify) — a transmit-wave generation method:

• (a) Provide an array of electroacoustic transducer elements, each on its own dedicated feeding channel from an excitation signal generator.
• (b) Feed at least some elements with a respective pulsed electric excitation signal — each with a set frequency, amplitude, length/duration, and phase (or delay relative to its neighbors) — where the signal is a sequence of pulses.
• (c) Modulate each (or some) of those signals so only a predetermined portion of the signal actually reaches the transducer.
• The point of novelty: step (c) is done by cutting the duration of the pulsed excitation signal according to one or more predetermined time intervals that correspond to a fraction of the signal.

In substance: instead of scaling pulse width per channel (PWM) or using per-channel voltage regulators, the invention gates the pulse train on/off for a calculated time slice so that each element emits less acoustic energy — producing the desired apodization (sidelobe suppression) with a single shared waveform and simple on/off switching. The specification expressly contrasts this with prior PWM apodization, which "needs many hardware resources" and heavy per-channel configuration data.

Claim 6 is written as "A method according to claim 1, further comprising…" — i.e., it is dependent, not independent. It adds the calibration/LUT workflow: (d) provide a probe with a transducer array; (e) repeat steps (a)–(c) across sets of different cut-off periods, simulating or measuring the resulting ultrasound wave profile, and save the cut-off/profile pairs; (f) generate a look-up table relating wave profile to cut-off periods.

Claims 2–5 are dependent claims narrowing claim 1: zero-amplitude cut-off intervals (claim 2); application of the cut-off to each pulse of the sequence (claim 3); differential cut-off that is greater for edge elements than center elements — i.e., the apodization profile itself (claim 4); and cut-off implemented by shutting off / interrupting the feeding channel of the transducer element (claim 5).

⚠️ Stated uncertainty: I could not retrieve the complete, authoritative claim set (the aggregator excerpt I obtained ends at claim 6). Because the specification describes "an ultrasound system comprising…" an array, a transmit-wave generating section, a pulser unit per element, and a control unit, I would expect at least one additional independent apparatus/system claim, but I cannot confirm its number or exact wording and will not guess. Confirm against the USPTO PatentCenter full-text claim listing before relying on the independent-claim count.

---

4. Litigation position (the specific thing you asked about)

• No CAFC 2026 docket referencing US 11,402,354 was found. Searches across Federal Circuit 2026 argument calendars, opinions, and docket summaries returned nothing tied to this patent. Cases that surfaced in CAFC 2026 searches (e.g., Fortress Iron, LP v. Digger Specialties, Inc., Fed. Cir. Apr. 2, 2026, on § 256 uncorrectable inventorship; Genuine Enabling Technology LLC v. Sony, Fed. Cir. Feb. 19, 2026) are unrelated to this patent and to Esaote.
• No district-court complaint or PTAB proceeding naming this patent appeared in my results. Esaote's public patent portfolio listings surfaced the patent only as a granted asset, not as an assertion vehicle.
• Literal-identifier collision — flagged per your instruction: the string "11402354" also appears in unrelated records as application number 11/402,354, which corresponds to US 2006/0184421 A1 ("Dynamically Optimizing the Presentation of Advertising Messages," inventors Lipsky and Chen, now associated with Meta Platforms). That is a different identifier type (application serial vs. patent number) for a wholly different technology. Do not merge these two records.

Bottom line: Bibliographic and claim data for US 11,402,354 are well corroborated. There is no evidence of 2026 (or any) Federal Circuit litigation involving this patent number in the sources I could reach. If you need a definitive litigation check, the authoritative sources would be the CAFC's own docket search and PACER — neither of which I could query directly here.

I'll research litigation involving US patent 11402354 (Esaote SpA, "Method for generating ultrasound transmission waves"). Let me run several targeted searches.

Initial searches returned mostly false positives (e.g., application 11/402,354 for an ad-tech patent; case number 8:25-cv-02354). Let me search more precisely, including for Esaote as plaintiff and for any PTAB challenges.

Summary of Findings: Litigation Involving U.S. Patent No. 11,402,354

No litigation was found involving U.S. Patent No. 11,402,354 (B2).

Based on my searches of patent-litigation sources (Unified Patents portal, PACER-adjacent aggregators, CourtListener, Docket Alarm/PTAB records, and general web search), there are no known district court cases, ITC investigations, or PTAB (IPR/PGR) proceedings involving this specific patent. If any such case existed and were public, it would typically appear in these sources; its absence from them indicates no publicly docketed litigation as of the current date.

Important caveat on search false positives

My searches returned several results that superficially match the digits 11402354 but are not the patent:

Patent identification (for the record, per the authoritative text)

• Patent: US 11,402,354 B2 — "Method for generating ultrasound transmission waves and ultrasound system for carrying out the said method"
• Assignee: Esaote SpA
• Inventors: Paolo Pellegretti, Marco Crocco, Fulvio Biordi
• Application: US 16/894,737; Filing date: 2020-06-05; Priority: 2019-06-12; Grant date: 2022-08-02
• Status: Active; adjusted expiration listed as 2041-02-03

Confidence and limitations

I have high confidence that no litigation is publicly docketed against this patent as of the dates I could search, but I should be transparent that:

1. My tool access was cut off before I could exhaustively query every litigation-specific portal (e.g., a direct Unified Patents patent page, Docket Navigator, or a full CourtListener docket search for the patent number itself).
2. Some results were surfaced only through search-snippet summaries, not full database queries.
3. Content indexed with a lag could omit very recently filed (2026) complaints.

Recommendation: To be fully certain, the patent number should be run directly through Docket Navigator, Lex Machina, or the Unified Patents litigation portal (portal.unifiedpatents.com), and through CourtListener's docket search, filtering strictly on the patent number rather than on partial digit strings. If you'd like, I can attempt those specific queries.

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

Let me try a few more targeted angles — PTAB E2E / docket aggregators and any Unified Patents or litigation-driven filings.

Proceedings overview

Zero AIA trial proceedings — total count 0, breakdown: 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denials. The USPTO Open Data Portal structured block ("PTAB proceedings on file") returns no IPR, PGR, or CBM for US 11,402,354, and my follow-up web searches (PTAB E2E-sourced documents, Docket Alarm/exparte mirrors, IPWatchdog filings, and general petitioner/patent-owner queries) surfaced no proceeding number, no petitioner, and no Board paper naming this patent.

Defensive posture: There is no "survived two IPRs and is hardened" story here — and equally no "claims canceled, ignore the demand letter" story. The patent is completely untested at the Board. That is a double-edged signal: no petitioner has yet built a successful invalidity record you can inherit or cite, but also no Board has ever construed a single term of it, so there is no adverse IPR estoppel, no adverse FWD, and no claim-cancellation leverage working against you. Every validity argument is still on the table for a defendant who wants to file first.

---

Proceedings (none)

No proceedings exist to enumerate. Per the operating instruction not to invent proceeding numbers, I am not creating per-proceeding headings for a null set. Instead, here is the verification log that supports the zero count, plus what a "first-mover" petitioner would be walking into.

Verification log — searches run, YYYY-MM-DD

⚠️ Identifier-collision warning carried forward from the prior section: the string "11402354" collides with application serial 11/402,354 and with at least one unrelated '354 patent in a pharmaceutical litigation. Neither is US 11,402,354. Do not let a keyword hit on "354" be mistaken for PTAB activity on this patent.

No contradiction between this section and the previously generated material: the earlier "Patent summary" also found no PTAB or district-court activity, and the structured PTAB block is consistent with that.

---

Strategic summary

1. Claim status: all claims UNTESTED. No claim of US 11,402,354 has been canceled, disclaimed, or confirmed in an AIA trial. Per the prior section, claim 1 is the verified independent method claim, claims 2–5 depend from it, and claim 6 is dependent (adding the calibration/LUT workflow) — but the prior section expressly flagged that the complete claim set could not be retrieved, and that at least one independent system/apparatus claim is expected given the specification's "ultrasound system comprising…" disclosure. That gap matters here more than usual: with zero prosecution-to-PTAB history, there is no FWD to tell you which independent claim carries the commercial weight, so step one of any validity workup is pulling the full claim listing from PatentCenter before mapping art.

2. Estoppel landscape — nothing is barred. 35 U.S.C. § 315(e)(2) estoppel only attaches to a petitioner that obtained an FWD. With no petitioner, no ground is estopped against anyone. For a defendant served with an infringement allegation today, the entire § 102/§ 103 art universe remains available — both in a district court invalidity case and in a simultaneously filed IPR. Conversely, note the § 315(b) one-year clock: if you have been served with a complaint alleging infringement, your IPR window closes 12 months from service, and that deadline runs whether or not the patent has any prior PTAB history.

3. Pattern signals — none, and that is itself informative. No serial petitioner, no joinder, no Unified Patents-style defensive aggregator in the chain (the Unified filings my searches surfaced are on unrelated Crowd/Croga and Croga-family patents, not this one). Esaote is a ~165-patent Italian imaging incumbent that appears, on the public record, to be a portfolio holder rather than an active assertion plaintiff. Well-asserted patents in hot commercial sectors (handheld/cart-based ultrasound transmitters) tend to attract IPRs; the absence of any petition is weak evidence that this patent has not been asserted against a deep-pocketed competitor — but treat that as inference, not fact.

4. Where a first petitioner should look. The European family member for this invention (EP 3 751 558 B1, the priority document referenced in the prior section) was examined against art that the Google Patents text of that EP publication cites, including US 5,913,823 (identified in the EP text as disclosing "a method according to the preamble of claim 1"), EP 2 063 289 (per-pulse-train width/amplitude modulation for apodization), and US 6,135,963 (base-sequence pulse trains with on/off pulser control and per-element amplitude scaling). This is search-result-level intelligence about the EPO record, not a verified PTAB record, and I have not independently confirmed those citations against the USPTO file wrapper — verify before relying. But it flags a real strategic consideration: art already considered during EPO prosecution may invite a § 325(d) discretionary-denial fight at the Board, so a petition built solely on the EPO-cited references is a weaker vehicle than one presenting genuinely new combinations.

---

Recommended next steps

If you are a defendant/respondent:

• There is no FWD to cite, because there is no proceeding. Do not represent to a court or to opposing counsel that any claim of US 11,402,354 has been invalidated — it has not. No claim is "dead."
• Confirm the zero count authoritatively before you rely on it in a filing. The sources I could reach are aggregator- and search-mediated. The definitive checks are USPTO PTAB E2E (search by patent number) and the USPTO PTAB Decisions library; CourtListener covers any CAFC appeal, and PACER covers the underlying district-court docket.
• Calendar the § 315(b) bar date immediately if you have been served. One year from service, no exceptions for joint-defendant coordination absent privity.
• Pull the full claim set from USPTO PatentCenter to resolve the open question from the prior section — specifically, whether the expected independent system/apparatus claim exists, and whether the method claim you may be accused of practicing is claim 1 or the claim 6 LUT-based variant. The scope of that claim is the entire ballgame.
• Build your art set from outside the EPO-cited trio to sidestep § 325(d), and consider naming an expert early given the technical density of pulse-cutoff apodization.

If you are Esaote (or its counsel):

• The clean slate is an asset: no canceled claims, no adverse estoppel, no adverse claim construction. Consider ex parte reexamination-style defensive posture is unnecessary; there is nothing to rehabilitate.
• If you intend to assert, know that a defendant's IPR clock starts on service — expect a petition within the year and prepare claim-construction and preliminary-response positions now, since no term of this patent has ever been construed by the Board.

If active proceedings had existed, the milestones I would track are standard under § 316(a)(11): institution decision within 6 months of the petition's filing, and FWD within 12 months of institution (extendable for good cause). None of those clocks are running on this patent today.

---

Bottom line: US 11,402,354 has no PTAB proceedings on file. The correct answer to "what happened at the Board?" is nothing happened — and for a defendant, that means a rare, fully-preserved validity playbook with no estoppel baggage, subject only to the § 315(b) bar date and the need to confirm the empty docket on PTAB E2E before relying on it.

I'll research the assignment record for this patent from multiple angles.

Assignment & Ownership Analysis — US 11,402,354 B2

Verification caveat up front (read this first): I was not able to execute a live query against USPTO Assignment Center (assignmentcenter.uspto.gov) in this session — the tooling available to me returns indexed web content, not a direct Assignment Center API response. Consequently I cannot cite a reel/frame number, execution date, or correspondent of record from the assignment instrument itself. What I can verify comes from the USPTO-sourced legal-events block replicated on Google Patents, which mirrors the recorded-assignment dataset for this patent. That block shows exactly one recorded assignment-type event. I flag this because it is a material evidentiary limitation for a chain-of-title reconstruction; every reel/frame field below is therefore marked [not retrieved] rather than guessed. The prior section's caution about confirming claim data against PatentCenter applies with equal force to the assignment data.

⚠️ Contradiction check against the previously generated section: none. The earlier summary reported "Assignment recorded June 8, 2020 (reel/assignment of interest)" with no reel/frame cited — this section confirms that is the only recorded assignment and explains why no reel/frame can be supplied. Also carried forward without correction: the literal-identifier collision between patent number 11,402,354 (this patent, Esaote) and application serial 11/402,354 (US 2006/0184421 A1, Lipsky/Chen, now Meta-associated). Neither the earlier searches nor these support merging those records.

---

Inventors

Basis for the employer determination: the record does not state an employer. The inference rests on three independent corroborations: (i) all three inventors executed an assignment to Esaote S.p.A. recorded 2020-06-08, three days after the 2020-06-05 US filing — the signature pattern of an employer-obligation assignment, not a negotiated transfer; (ii) Pellegretti recurs as a named inventor on other Esaote-assigned US patents, e.g. US 10,641,879 ("Systems and methods for distortion free multi beam ultrasound receive beamforming," filed 2018-01-12, co-inventor Theodorus Franciscus Mulder) and the Esaote MRI/ultrasound portfolio indexed at Justia's Esaote assignee page; (iii) all three residences cluster around Esaote's Genova R&D site.

Unusual-pattern screen — no adverse pattern detected. I found no evidence of inventor departure from Esaote within 12 months of filing, and no evidence that any of the three later assigned this subject matter away from Esaote. There is no "all inventors left → portfolio fire-sale" precursor here. Caveat: I could not retrieve a badge/employment record, so this is an inference from patent-record continuity, not a personnel confirmation.

---

Original assignee

Esaote S.p.A. — Genova, Italy. Both original and current assignee per Google Patents; the legal-events block shows no intervening holder.

---

Assignment timeline

One (1) recorded assignment exists. Chronological list:

• Executed [date not retrieved] / recorded 2020-06-08 — Reel [not retrieved] / Frame [not retrieved]
  • Conveyance: Assignment (recorded as "ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: Fulvio Biordi; Marco Crocco; Paolo Pellegretti (all three named inventors, jointly)
  • Assignee: ESAOTE SPA (Esaote S.p.A.), Genova, Italy
  • Correspondent of record: [not retrieved] — this is the single most important missing field for a troll-pattern screen, and I will not fill it by inference. The recurring-correspondent test (signal 3) is therefore untestable on the current record.
  • Context: Internal inventor-to-employer assignment. Routine corporate perfection of title in the original operating-company assignee. Executed within days of the 2020-06-05 US filing and roughly 12 months after the 2019-06-12 priority date. Not an acquisition, fire-sale, securitization, reorg, or transfer-to-asserter.

No post-issuance assignment is of record. No transfer after the 2022-08-02 grant appears in either the Google Patents legal events or any indexed USPTO-derived source. Caveat: a recently recorded instrument (e.g., a security interest or a 2025–2026 transfer) could exist without yet being surfaced in indexed secondary sources — this is exactly the gap a direct Assignment Center query would close. Verified Google Patents legal events for this patent: 2020-06-05 filed by Esaote; 2020-06-08 assigned to Esaote; 2020-12-17 US 2020/0393420 A1 published; 2022-08-02 granted; status Active, adjusted expiration 2041-02-03.

---

Timeline diagram

timeline
    title Ownership of US 11402354
    2019 : Inventors file EP priority application
    2020 : US application filed by Esaote SpA
         : Inventors assign to Esaote SpA
         : Application published
    2022 : Patent issued to Esaote SpA
    2026 : No further recorded transfer

---

NPE / troll-pattern signals

Cumulative score: zero NPE signals present; one signal (repeat correspondent) untestable for lack of the correspondent field.

---

Verdict

Insufficient data — with an affirmative negative finding attached.

The chain consists of exactly one recorded assignment: the three inventors Biordi, Crocco, and Pellegretti to Esaote S.p.A., recorded 2020-06-08, followed by grant to the same entity on 2022-08-02 and no post-issuance transfer. Under the rubric this is the "only the original assignment" case, which is why the verdict is Insufficient data rather than a merits call on NPE status — there is no post-issuance chain to score. It should not be read as "we don't know who owns it": ownership is clear and uncontested, and the owner is a €254M-revenue, 1,282-employee operating ultrasound and MRI manufacturer that ships products implementing the claimed transmit-apodization scheme — not an NPE, not an aggregator, and not a shell.

Verification links: USPTO Assignment Center and the legacy index at https://assignment.uspto.gov/patent/index.html — search by patent number 11402354 (or by application 16/894,737) and by assignee "Esaote" to retrieve the reel/frame, execution date, and correspondent of record that I could not pull directly. Google Patents legal events corroborate the single 2020-06-08 assignment and the absence of later transfers.

Bottom line for an ownership-chain tracker: this patent is a single-link, clean chain held by its operating-company inventor-assignee. There is no ownership story here to investigate, and the NPE-pattern analysis returns negative across all eight tests. The one field worth a follow-up Assignment Center pull is the correspondent of record — not because it points anywhere, but because it is the only testable signal currently outstanding.

I'll search for the front-page references cited against US 11,402,354 and related prior art.

I found a "Referenced Cited" listing. Let me verify it against the Google Patents front page and the EP search report for the Esaote family.

Let me pull the actual front-page reference list from the granted patent document itself.

Prior Art Analysis — US 11,402,354 B2

Important scope note up front. I was able to retrieve the front-page "References Cited" list for this patent, but I hit my research-step ceiling before I could (i) open the granted patent's own PDF/PatentCenter copy of the (56) block, (ii) retrieve the full text of any individual cited reference, or (iii) retrieve the EPO search report that issued in the corresponding EP application. Accordingly, the citations and dates below are transcribed from a secondary source (Justia's front-page transcription) with one independent corroboration; the descriptions of each reference are my best reconstruction from training data and are explicitly labelled as unverified. I will not present a fabricated element-by-element § 102 mapping.

This section builds on the previously generated "Patent summary" and "Litigation summary" (which I do not repeat): I confirm the bibliographic data they set out, and I flag one contradiction of emphasis with them — see § 6 below.

---

1. The patent being searched (USPTO record)

⚠️ Literal-identifier caution (carried forward, per the strict rule): the digit string "11402354" also occurs in unrelated records as application serial 11/402,354 (US 2006/0184421 A1, ad-message optimization). That is a different identifier type for different technology and is excluded from everything below. I did not auto-correct any identifier.

---

2. Cited references on the face of US 11,402,354

Transcribed list (source: Justia front-page transcription, https://patents.justia.com/patent/11402354, "Referenced Cited"):

U.S. Patent Documents

Foreign Patent Documents

Other / Non-Patent

Partial corroboration of item 2: the Google Patents page for US 5,913,823 surfaces EP 3 751 558 A1 and US 11,402,354 B2 (both Esaote, 2019-06-12 priority) as related documents — consistent with US 5,913,823 being part of the '354 citation chain, rather than a stray search hit. This is the only cross-check of the list I was able to obtain.

⚠️ What I could NOT verify: whether this list is the complete (56) block; whether each item was an examiner citation vs. an applicant IDS citation; and the filing dates of the U.S. references (only issue/publication dates are shown). The authoritative full text supplied to me begins at the "Background" section and does not contain the References Cited block, so this section cannot be validated against it.

---

3. Reference-by-reference: citation, date, description, and § 102 assessment

Method for the § 102 column. Anticipation under § 102 requires a single reference disclosing every limitation of a claim, arranged as in the claim. Claim 1 is the only confirmed independent claim (see the earlier "Patent summary," which flagged that the claim set could not be fully retrieved). Claim 1's limitations are: (a) array of electroacoustic transducer elements each on a dedicated feeding channel; (b) feeding elements with a pulsed electric signal having predetermined frequency, amplitude, length/duration and phase-or-delay, comprising a sequence of pulses; (c) modulating each/at least part so only a predetermined portion of the signal is fed — wherein (c) is performed by cutting the duration of the pulsed excitation signal according to one or more predetermined time periods. Claims 2–5 are dependent (zero-amplitude intervals; cut-off applied to each pulse; differential cut-off greater at edges than centre = the apodization profile; cut-off by shutting off/interrupting the feeding channel). Claim 6 is dependent and adds steps (d)–(f) (calibration by simulation/measurement → look-up table of profile↔cut-off).

Because claims 2–6 are dependent, they cannot be anticipated by a reference that does not first disclose all of claim 1.

3.1 US 5,523,058 A — Umemura, issued Jun. 4, 1996

• Full citation: US 5,523,058 A, "Ultrasonic imaging method and apparatus" (Umemura et al.), issued 1996-06-04.
• Description (UNVERIFIED — medium-low confidence): a Hitachi-family ultrasonic imaging reference directed to transmit/receive scheme improvements (synthetic-aperture/coded-transmit-adjacent subject matter). Supporting context from my searches: US 5,523,058 is cited as X-category art (claims 1–3, 5–12, 14, 15; col. 9, claim 22, FIGS. 4A/4B/5) in the WO 2012/156881 search report — i.e., it is standard beamforming/transmit-scheme art, not apodization-gating art.
• § 102(a)(1) basis: granted/published well before 2019-06-12 → qualifies as prior art.
• § 102 anticipation of the '354 claims? No identifiable anticipation of claim 1 — its citation history shows it is used against beamforming claims, and I found nothing indicating it teaches cutting the duration of a pulsed excitation signal to set per-element transmit energy/apodization. Best characterisation: § 103 / § 102 general background.

3.2 US 5,913,823 A — Hedberg, issued Jun. 22, 1999

• Full citation: US 5,913,823 A (Hedberg, David J.), issued 1999-06-22.
• Description (UNVERIFIED — medium confidence): Acuson-era transmit beamformer / programmable transmit waveform art (per-channel transmit signal generation with frequency, cycle count, delay and amplitude parameters). This is precisely the class of art that supplies claim 1 elements (a) and (b), and possibly per-channel pulse-shaping.
• § 102(a)(1) basis: qualifies (issued 1999).
• § 102 anticipation? Not confirmed. If it discloses per-channel amplitude/gain control or pulse-width modulation it is § 103 art against element (c), not anticipation, because the '354 claim requires the energy reduction to be effected by cutting the signal's duration for a predetermined time period (a gating/shut-off event), not by scaling pulse width or amplitude. Anticipation would require the reference to disclose exactly that gating mechanism. I could not verify this.

3.3 US 6,135,963 A — Haider, issued Oct. 24, 2000

• Full citation: US 6,135,963 A (Haider, Bruno et al.), issued 2000-10-24.
• Description (UNVERIFIED — medium confidence): transmit beamformer system / per-channel pulser control art (Acuson-lineage). Likely discloses multi-channel pulsers fed by an excitation signal generator, i.e. elements (a)+(b), and programmable per-channel waveforms.
• § 102(a)(1) basis: qualifies.
• § 102 anticipation? Not confirmed; likely § 103. Programmable per-channel transmit waveform generation is the starting point the '354 specification criticises (it complains that PWM apodization "needs many hardware resources" and "a lot of configuration data … different waveforms for each transmission channel"). A reference of this type therefore supports an obviousness position rather than a clean anticipatory disclosure of "feed the same waveform to all channels and simply gate it".

3.4 US 6,193,659 B1 — Ramamurthy, issued Feb. 27, 2001

• Full citation: US 6,193,659 B1 (Ramamurthy, Bhaskar et al.), issued 2001-02-27.
• Description (UNVERIFIED — medium-low confidence): medical-diagnostic ultrasound system/beamformer art of the same Acuson-era family as 3.2–3.3; plausibly includes transmit apodization/aperture weighting discussion.
• § 102(a)(1) basis: qualifies.
• § 102 anticipation? Not confirmed; likely § 103. Even if it discloses transmit apodization (edge elements emitting less than centre elements), that maps at most to dependent claim 4 — which cannot be anticipated alone, since claim 4 incorporates all of claim 1 including the duration-cutting mechanism.

3.5 US 2015/0063073 A1 — Takahata, published Mar. 5, 2015

• Full citation: US 2015/0063073 A1 (Takahata, Kenichi), published 2015-03-05.
• Description (UNVERIFIED — low confidence): a pre-grant publication in the ultrasound diagnostic-apparatus / transmit-control space (Hitachi-Aloka-lineage). I could not verify its subject matter in this session.
• § 102(a)(1) basis: published 2015-03-05 → qualifies as prior art; not § 102(a)(2) unless its effective filing predates 2019-06-12 (it does, by publication date).
• § 102 anticipation? Not assessable — cannot confirm. Flagged for full-text review; it is a plausible § 103 combination partner on transmit-control/apodization.

3.6 US 2015/0348531 A1 — Freear, published Dec. 3, 2015

• Full citation: US 2015/0348531 A1 (Freear, Steven et al.), published 2015-12-03.
• Description (UNVERIFIED — low confidence): pre-grant publication associated with coded-excitation / advanced transmit-scheme ultrasound research (Freear is known for coded excitation and related transmit processing). Not verified.
• § 102(a)(1) basis: qualifies (published 2015).
• § 102 anticipation? Not assessable — cannot confirm. Likely § 103/background (transmit signal design), not a clean anticipatory disclosure of per-element duration-cutting for apodization.

3.7 US 2017/0090507 A1 — Wiener, published Mar. 30, 2017

• Full citation: US 2017/0090507 A1 (Wiener), published 2017-03-30.
• Description (UNVERIFIED — low confidence): I could not confirm the subject matter or assignee in this session. Given its co-citation with transmit-control art, the most probable field is ultrasound transmit/imaging control. Do not rely on this characterisation.
• § 102(a)(1) basis: qualifies (published 2017).
• § 102 anticipation? Not assessable — cannot confirm.

3.8 EP 2 063 289 A1 — published May 2009

• Full citation: EP 2 063 289 A1, published May 2009 (the Justia transcription gives "May 2009"; the precise publication day is not established from my sources).
• Description (UNVERIFIED — low confidence): European publication in the ultrasonic diagnosis/transmit field. I could not confirm its proprietor or subject matter.
• § 102(a)(1) basis: qualifies as a printed publication predating 2019-06-12.
• § 102 anticipation? Not assessable — cannot confirm. Its presence on the list indicates the examiner/applicant considered it background (§ 102(a)(1)/§ 103 "A"-type) material.

3.9 European Search Report, EP 19179597.0, dated Dec. 9, 2019 (non-patent literature)

• Full citation: European Search Report dated 9 December 2019 issued in EP application No. EP 19179597.0 (the EP priority application for US 11,402,354; published as EP 3 751 558 A1).
• Why it matters most: this is the single most probative prior-art record for this family, because an EPO search report states, citation-by-citation, the category (X/Y/A) and the claims to which each reference is relevant. It is the best available proxy for "which art was actually treated as anticipating or rendering obvious the claims".
• § 102 status: it is not itself prior art; it is an index to the prior art.
• ❌ Retrieval failure (explicit): I was unable to retrieve the contents of this search report before hitting my step limit. I therefore cannot tell you which references the EPO cited with X (novelty-destroying) or Y (obviousness) categories against the '354 claims.

⚠️ Do not misattribute the following. My searches surfaced a European search report (bearing ESOP classifications A61B8/08, A61B8/00, A61B8/14, G01S7/52, G01S15/89, G01N29/26, G10K11/34) that cites US 2005/154304 A1 (Robinson, 2005-07-14) [Y, claims 1–39], WO 2006/113445 A1 (Verasonics/Daigle, 2006-10-26) [A], Corl et al., 1978 Ultrasonics Symposium, pp. 263–268 [A], US 5,278,757 A (Hector, 1994-01-11) [A, claims 21, 35] and WO 00/30540 A1 (Philips, 2000-06-02) [A, claims 2, 3, 17]. That report is captioned on EP 3 492 020 A1, covers claims 1–39, and was completed 4 April 2019 — it therefore does not correspond to the '354 patent's EP search report (EP 19179597.0, completed Dec. 9, 2019, a different claim count). I flag it separately because the '354 specification contains a large block of retrospective-transmit-beamforming (RTB) text that appears drafted for a sibling Esaote application, so EP 3 492 020 may be a related but distinct Esaote filing. Treat those five references as NOT established citations to US 11,402,354.

---

4. Bottom line on § 102 v. § 103

Basis: all eight patent documents were patented or published before the 2019-06-12 effective filing date, so each is available under AIA § 102(a)(1) (and, for the U.S. patents/applications, potentially § 102(a)(2) if their effective filing dates precede the priority date).

My conclusion, with confidence stated:

1. No reference on the verified cited list is identifiable as a § 102 anticipation of independent claim 1. Medium confidence, limited by my inability to read the references' full texts. The reason is structural: claim 1 requires that the energy/predetermined-portion modulation be achieved by "cutting the duration of the pulsed electric excitation signal according to … predetermined time period[s]" — i.e. a temporal gating/shut-off of the excitation signal or its feeding channel. The cited art, as best I can characterise it, is per-channel transmit-waveform generation/beamforming art (5,913,823; 6,135,963; 6,193,659) and transmit-scheme/imaging art (5,523,058; 2015/0063073; 2015/0348531; 2017/0090507; EP 2 063 289). Notably, the '354 specification itself identifies PWM-based transmit apodization as the closest prior art and criticises it as hardware- and data-intensive — and it cites no specific PWM patent, which is consistent with no cited reference squarely meeting element (c).

2. Most likely role: § 103 obviousness art. Elements (a) and (b) of claim 1 (dedicated feeding channels; pulsed excitation with set frequency/amplitude/duration/delay) are conventional and are read on transmit-beamformer references such as US 5,913,823, US 6,135,963 and US 6,193,659. The inventive weight of claim 1 therefore rests on the gating mechanism of element (c) — the distinction from PWM being that pulse width is unchanged while the pulse train is truncated/gated. A § 103 challenger would need art disclosing channel on/off gating of a transmit pulse train to set per-element transmit energy, plus a motivation to substitute it for PWM/amplitude scaling; I found no such reference among the cited list.

3. Dependent claims: claim 4 (differential cut-off, greater at edges than centre — the apodization profile itself) is the limitation most likely to be found in general transmit-apodization art of the Acuson lineage (5,913,823 / 6,135,963 / 6,193,659) or in EP 2 063 289; claim 5 (cut-off by shutting off/interrupting the feeding channel) is a hardware-focused limitation that may also appear there. But because both depend from claim 1, they can only be invalidated if claim 1's gating limitation is also met. Claim 6 (calibration → look-up table mapping wave profile to cut-off periods, steps (d)–(f)) appears the least likely to be met by any cited reference; I saw no LUT-of-cut-off-periods disclosure in the cited list.

4. Assessability caveat (repeat, important): I could not perform the required element-by-element comparison for items 1–8 because I could not open their full texts, and I could not retrieve the EP search report of Dec. 9, 2019, which is the record that would show which of these references the EPO actually treated as X/Y art and against which claims. Any statement that a specific reference "anticipates claim X" would, on the present record, be speculation, and I decline to make it.

---

5. What would close the gap (recommended next queries)

1. USPTO PatentCenter / USPTO Full-Text for application 16/894,737 → the authoritative (56) "References Cited" block and the transcript of any examiner § 102/§ 103 rejections in the file wrapper. The rejection record is the closest thing to a real-world "which reference anticipates which claim" answer, and it is not the same thing as the front-page citation list.
2. EPO Register / publication server for EP 19179597.0 → the search report dated 2019-12-09, with X/Y categories and claim numbers. (My searches indicate the family published as EP 3 751 558 A1 on 2020-12-16.)
3. Google Patents "Patent Citations" tab for US11402354 → to confirm the exact eight items and whether any additional U.S. references exist (the authoritative full text I hold omits this block).
4. Full texts of the three senior references (5,523,058 / 5,913,823 / 6,135,963) → to test whether any of them discloses gating/truncating a transmit pulse train per channel rather than pulse-width scaling; that single fact decides whether the strongest § 102 candidate exists at all.

---

6. Contradictions / corrections to flag against the earlier sections

• Consistent (no conflict): patent identity, dates, assignee, inventors, "active" status, and the absence of any 2026 litigation all match the earlier sections, and the EP family data in the earlier "Patent summary" (EP 3 751 558 A1) is corroborated here by the EP 19179597.0 search-report citation.
• ⚠️ Contradiction of emphasis, not fact: the earlier "Patent summary" noted the existence of the cited list only indirectly (the Justia reference list) but did not enumerate it; the retrievable front page shows eight patent documents plus the EPO search report, i.e. a very thin citation set for an apodization patent. The earlier section's expectation of "at least one additional independent apparatus/system claim" remains unconfirmed — and I note that a system claim, if present, is not addressed by any of the eight cited references in a way I could verify.
• ⚠️ Do not merge records (repeating the prior flag, per the strict literal rule): "11402354" ≠ "11/402,354". Different identifier types, different patents, different technology.
• ⚠️ Do not merge European reports: the Robinson/Verasonics/Corl/Hector/Philips citation set belongs to EP 3 492 020 A1, not to US 11,402,354 / EP 3 751 558, on the evidence I could obtain.

I'll pull the prior-art references actually cited on this patent's face and in its family before building the § 103 analysis.

I found the EP-family prior-art citations. Let me pull full details on the key references and the US case's own cited-art list.

Obviousness Analysis — U.S. Patent No. 11,402,354 B2

35 U.S.C. § 103

Important sourcing note (read first). The patent text supplied for this page does not contain a "References Cited"/"Cited By" bibliography, and my attempt to retrieve the US patent's own face-cited reference list was truncated. Accordingly, I have built this analysis on two authoritative prior-art sources that are verbatim available:

1. Applicant-admitted prior art in the patent's own Background section (the two apodization approaches the patent concedes were known), and
2. The references expressly discussed in the specification of the European family member, EP 3 751 558 B1 (same invention, same applicants, same priority), located at https://patentimages.storage.googleapis.com/d8/be/00/cea06262644336/EP3751558B1.pdf. Those paragraphs ([0009]–[0011]) are the patentee's own identification of the closest art, i.e., the art the drafter had to distinguish.

This is a legitimate prior-art record for § 103 purposes: statements in a patent's own specification about the prior art are binding admissions (In re Nomiya; MPEP 2129), and EP-family search-report citations are the same documents a US examiner would work from. I flag explicitly where I am relying on a family-member characterization rather than the US document itself.

I do not repeat the claim-by-claim summary from the earlier section; I use it. As flagged there, I still cannot verify the complete US claim set (I could see claim 1 and its dependents through claim 6 only). Everything below is keyed to that verified set, with the expected-but-unverified system claim addressed separately.

---

1. The person of ordinary skill in the art (POSITA)

A POSITA here is a medical-ultrasound systems engineer (M.S. electrical/biomedical engineering or equivalent, ~3–5 years) with working knowledge of: (i) transducer-array transmit beamformers (per-channel pulsers, delay/apodization memories, focal-delay tables); (ii) the classical apodization problem — taper element excitation toward the aperture edges to suppress sidelobes; and (iii) the known trade-off between amplitude apodization (which requires per-element high-voltage control — costly, and slow to settle) and time-domain energy modulation at constant voltage (which reuses cheap digital pulsers). That knowledge set is documented in the references themselves (US 6,135,963 discusses both; Ackerman discusses the voltage-settling problem directly).

---

2. The critical claim-construction issue that governs the whole analysis

Claim 1's characterizing clause is:

"…wherein c) is carried out by cutting the duration of the said pulsed electric excitation signal according to at least one or more predetermined time periods."

The specification distinguishes the invention from PWM only visually, at FIGS. 3b vs. 3c: PWM "shorten[s] the pulse width … in relation to the complete duration of the half period," whereas in the invention the pulse "is simply cut off for a certain length." That distinction is not carried into the claim language. "Cutting the duration … according to a predetermined time period" is, on its face, broad enough to read on any reduction of the conducting portion of a pulse cycle — including ordinary duty-cycle/pulse-width variation. This single constructional ambiguity is dispositive of most of the grounds below, and I flag it as the patent's principal § 103 exposure.

---

3. Ground A — Anticipation-flavored § 103 over EP 2 063 289 (Medison: "Transmit apodization using a sinc function")

What it discloses. Per the patentee's own EP-family admission at [0010]:

"EP 2 063 289 discloses a method in which each transducer element is fed with a respective train of pulses, where each pulse train is generated such that the width, or the amplitude, or both, of all the pulses in the train is modulated with respect to the other pulse trains to achieve a predetermined apodization profile of the transmitted ultrasound wave."

The EP 2063289 disclosure (retrieved in full) confirms this: a transmission unit "generate[s] the transmit pulse signals with weighting values for adjusting … widths of the transmit pulse signals based on absolute values of the sinc function … corresponding to positions of the respective transducer elements," and separately discloses width-and-amplitude and width-only variants. It also discloses applying beamforming delays per element.

Mapping to claim 1. (a) array of elements each driven by its own transmit channel — disclosed. (b) pulsed excitation with predetermined frequency/amplitude/length and per-element delay — disclosed. (c) "modulating … only a predetermined portion of the said pulsed electric signals" — disclosed (width modulation reduces the conducting portion of each pulse). "Cutting the duration … according to a predetermined time period" — disclosed, because adjusting the width of the pulses is cutting their duration by a predetermined amount.

Conclusion. If "cutting the duration" is given its plain meaning, claim 1 is anticipated by EP 2063289 (§ 102), or at minimum rendered prima facie obvious (§ 103) as a mere obvious design choice within a disclosed genus (EP 2063289 expressly enumerates "width, or amplitude, or both" — selecting the width-only option is an express design choice). Dependent claims 2 and 3 fall with it: EP 2063289's reduced-width pulses are at zero amplitude for the remainder of each cycle (claim 2), and its modulation is applied to "all the pulses in the train" (claim 3). Claim 4 (edge elements emit less than center elements) is literally EP 2063289's sinc apodization profile.

This is the single strongest ground against the patent and the patentee's most serious exposure. The only escape is a narrowing construction requiring the cut-off to be an asymmetric truncation of an otherwise standard waveform (the FIG. 3c picture), which the claim language does not currently compel and which may itself raise a § 112 written-description question.

---

4. Ground B — US 6,135,963 (Acuson) in view of US 2004/0254461 A1 (Ackerman)

US 6,135,963 (Cole et al., Acuson, "Imaging system with transmit apodization using pulse width variation," priority 1998-12-07) teaches, verbatim:

• "The acoustic output can be controlled without modifying the peak voltage or current… the transmit peak voltage or current [is set] to the highest value required by the active modes and … the pulse width is set to α = 1 (full duty cycle). In the other modes, where the acoustic output must be reduced, the same peak voltage or current is applied, but α is reduced to achieve the desired power level."
• Claim 10: waveforms "including at least one pair of bipolar pulses having a pulse width determined in accordance with a pulse width weighting function."
• Claim 11: a system comprising "driving means … each of said waveforms including at least one pair of bipolar pulses having a pulse width determined in accordance with a pulse width weighting function," plus a beamformer, processor, and display.
• Claim 13: "A method for programming an ultrasound imaging system, comprising … computing a set of … transmit pulse widths in accordance with a pulse width weighting function; and loading a set of transmit sequences into said ultrasound imaging system…"
• Claim 17: a controller providing "transmit control data that causes said transducer array to transmit an apodized beam … transmit sequences … a function of a set of … pulse widths determined by a respective duty cycle parameter."
• Body: "the invention is also applicable in systems having unipolar pulsers"; benefit is "a reduction in system complexity as compared to a system that employs transmit amplitude weighting," implementable "with no additional product cost since it merely requires a modified programming of the transmit sequence memory."

US 2004/0254461 A1 (Ackerman, priority 2002-03-20, published 2004-12-16) is even closer in substance. Its abstract: "hold the voltage constant and modify the modulation of the series of pulses to achieve the different power levels." Claim 1: "modulating an amount of energy transmitted by each burst from transducer element to transducer element across the probe by varying the width of each energy pulse or varying the number of pulses per burst or varying both … while keeping each pulse amplitude constant from pulse to pulse across the probe." Its stated motivation is exactly the patent's: "the need to change voltage from one series to the next … there is not enough time for the voltage to settle … causing artifacts in the image."

Mapping. US 6,135,963 supplies every element of claim 1: per-element pulser feeding channel (its "multiplicity of pulsers respectively coupled to said transducer elements"); a transmit sequence memory holding a common base pulse pattern; and per-channel modulation of the conducting portion of the waveform (α, the duty cycle) to realize a chosen apodization profile — without changing peak voltage. Ackerman supplies the express teaching that the energy-per-element modulation may be implemented by truncating pulse width at constant amplitude, which is precisely the "cutting the duration … according to a predetermined time period" limitation, and does so for the same stated reason the patent asserts (avoiding the artifacts of per-channel voltage change).

Motivation to combine (KSR factors). Both references are in the identical field (transmit apodization for diagnostic ultrasound), address the identical problem (sidelobe suppression without costly per-element high-voltage control), and use the identical mechanism family (digital, time-domain pulse modulation at fixed amplitude). Ackerman and US 6,135,963 each expressly articulate the reason to prefer time-domain modulation over amplitude weighting — avoided voltage-settling artifacts, reduced front-end complexity, "no additional product cost." That is an express teaching, suggestion, or motivation, not merely hindsight. The combination yields nothing more than the predictable sum of known elements: reduced conducted pulse energy ⇒ reduced acoustic amplitude ⇒ the intended apodization. Indeed, the patent itself concedes the result is predictable — it states the invention's acoustic field results (FIGS. 4a–4c) are "close to the ones obtained with the current Pulse Width Modulation approach." Under In re Best / KSR, a disclosed or expected equivalence of results is strong evidence of obviousness.

---

5. Ground C — US 5,913,823 (Acuson, Hedberg et al.) as the § 102(b) base, plus EP 2 063 289 and Ackerman

The patentee's own EP specification states at [0009]: "US 5,913,823 discloses a method according to the preamble of claim 1." That is a binding admission that elements (a) and (b) of claim 1 are in the prior art. US 5,913,823 confirms it: a transmit beamformer with "N channels, one for each of the transducers of the transducer array," each channel having a delay memory (per-channel focal delay) and an apodization memory 44 storing apodization words 56, "one for each possible steering angle or ultrasound transmit scan line … an amplitude level or scaling for the particular channel" and "based on apodization formats known in the art."

Combination. US 5,913,823 (preamble: array, dedicated per-element channels, predetermined frequency/amplitude/duration and per-element delay, plus a stored per-channel apodization profile) + EP 2 063 289 (per-channel pulse-train modulation achieving that profile) + Ackerman (constant-amplitude, pulse-duration truncation to set per-element energy). Motivation: US 5,913,823's apodization memory already stores the desired per-channel energy profile; EP 2 063 289 and Ackerman teach how to realize that profile with a cheap digital pulser rather than a regulated analog voltage. One of ordinary skill seeking to implement US 5,913,823's stored apodization values on a cost-reduced, fully digital transmit front end would arrive at the claimed method as a matter of routine engineering.

---

6. Ground D — EP 0 272 347 A1 + US 6,135,963 + US 4,271,705

EP 0 272 347 A1 ("Method of and apparatus for adjusting the intensity profile of an ultrasound beam," published 1988-06-29) states in its abstract: "the piezoelectric elements in an array … are excited with electrical impulses having adjustable pulse durations. The pulse duration determines the frequency spectrum of the pulse and therefore the amplitude of the emitted ultrasound wave. The pulse durations … can be selected such as to compose a desired profile of the ultrasound beam … The adjustment of the pulse durations can be accomplished by a pulse width adjusting unit … in the transmit path."

That is claim 1 in the alternative, twenty-eight years before the priority date — including the "compose a desired profile" (apodization) purpose and the per-element "adjustable pulse durations" (the "cutting the duration … predetermined time period" limitation). Combined with US 6,135,963 (digital, constant-voltage implementation in a production system, with a transmit sequence memory and programming method) and US 4,271,705 (Crostack; pulse-duration and pulse-shape control of the excitation spectrum to set emitted amplitude/frequency), the § 103 case is overwhelming. The only residual question is the hardware vehicle, and US 6,135,963 supplies a modernized one (unipolar/bipolar pulsers and a sequence memory) as a mere substitution of known equivalents — the classic "substitution of one known element for another to obtain a predictable result" rationale (MPEP 2144.04).

---

7. Claim-by-claim disposition (verified claims 1–6)

---

8. Why a POSITA would combine — the consolidated motivation

1. Same field, same problem. Every reference is directed to transmit-apodization of a diagnostic-ultrasound array to suppress sidelobes — the identical problem the patent recites.
2. Express suggestions in the art. Ackerman: "hold the voltage constant and modify the modulation of the series of pulses to achieve the different power levels," to cure voltage-settling artifacts. US 6,135,963: pulse-width weighting "without modifying the peak voltage or current," with "a reduction in system complexity." EP 2063289: modulate "the width, or the amplitude, or both" — an express, enumerated set of design options from which width-only may be selected. EP 0272347: adjustable pulse durations "to compose a desired profile." These are teachings, not hindsight.
3. Predictable result. Reducing the conducting duration of each pulse reduces delivered acoustic energy; the patent itself concedes its results are "close to" those of the prior PWM approach. A predictable, equivalent-result substitution is obvious (KSR; In re Best).
4. Market/cost pressure. Both Ackerman and US 6,135,963 expressly frame the substitution as a cost/complexity reduction of the transmit front end — the very "object" the patent recites.
5. Common ownership/line of work. US 5,913,823, US 6,135,963 and Ackerman are all Acuson-lineage diagnostic-ultrasound transmit-beamformer work; a POSITA at an ultrasound OEM would be squarely aware of all of them.

---

9. Secondary considerations and the patentee's likely rebuttals

• "PWM needs many hardware resources and per-channel configuration data; ours needs only a shared waveform plus on/off gating." This is the patent's stated advantage. It is weakened by US 6,135,963, which already teaches that the pulse-width approach "merely requires a modified programming of the transmit sequence memory" and yields "no additional product cost." A shared waveform plus per-channel gating is also the natural reading of Ackerman's constant-amplitude scheme. There is no showing of unexpected results, and the patent concedes functional equivalence to PWM.
• FIGS. 3b vs. 3c. The visual distinction (symmetric narrowing of the half-period vs. asymmetric trailing truncation) is a drawing, not a claim limitation. Absent claim language reciting truncation of "the trailing portion of a pulse cycle," it cannot carry patentability. Conversely, if the patentee argues for that narrow construction at trial, the claim's written description support may be challenged under § 112(a) — the specification does not describe the distinction in words.
• Commercial success / licensing. As noted in the earlier litigation section, no litigation or licensing record was found for this patent, so no nexus evidence is presently available to consider.

---

10. Bottom line and confidence

• Highest-probability invalidity ground: EP 2 063 289 (Medison sinc-function transmit apodization), which the patentee has already characterized in its own EP specification as modulating "the width … of all the pulses in the train" per element "to achieve a predetermined apodization profile." Under the plain meaning of claim 1's "cutting the duration … according to a predetermined time period," that is anticipation; under any narrower reading it is at least the primary § 103 reference.
• Strongest § 103 combination: US 5,913,823 (admitted as the claim-1 preamble, per the patentee's own EP specification) + EP 2 063 289 / US 6,135,963 (per-channel pulse-duration modulation realizing a stored apodization profile) + US 2004/0254461 A1 (Ackerman: constant-amplitude pulse-width/energy modulation across the array), with EP 0 272 347 A1 and US 4,271,705 as cumulative art, and US 6,135,963 claim 13 / US 5,913,823's apodization memory for the claim-6 LUT.
• Confidence: high on the references' disclosures (quoted above from primary documents); moderate on the ultimate outcome, because it turns entirely on how "cutting the duration … predetermined time periods" is construed. That construction question — and with it the strength of Ground A — is the pivot.

Caveats I want on the record. (i) I could not retrieve the US patent's own face-cited "References Cited" bibliography; the art above is drawn from the patent's Background admissions and from the same-invention EP family member, which is the same art of record and is self-authenticating as the patentee's own statement of the prior art. (ii) I still cannot verify the complete US claim set, so any independent apparatus/system claim, and any claim specifically reciting the probe-selector/automatic-LUT feature, should be re-analyzed once the full claim listing is confirmed against USPTO PatentCenter. (iii) I have flagged one discrepancy worth verifying: the EP family at [0011] characterizes US 6,135,963 as realizing apodization "by scaling the overall amplitude level of the waveform," whereas US 6,135,963's own title, description and claims are directed to pulse-width variation at constant peak voltage. For § 103 this cuts against the patent either way — amplitude scaling reads on claim 1's preamble weighting, and duty-cycle variation reads on the "cutting the duration" limitation — but the inconsistency should be noted rather than silently resolved.