US 7064692

I'll search for authoritative information on US patent 7064692, including any litigation activity.

Let me check for any litigation or CAFC docket activity involving this patent.

Summary: US Patent 7,064,692

Verification note: My searches of the open web (Google Patents, Justia, EPO Global Patent Index, FreshPatents/docket aggregators) surfaced consistent bibliographic data for US 7,064,692. I did not retrieve a direct USPTO PatentCenter/Patent Application Locator (PAL) page or a CAFC docket record in these searches. Below, I flag what is authoritative (from the patent document itself) versus what is aggregated.

Bibliographic data

Abstract (verbatim)

"Methods and systems for simply and inexpensive converting the signals of solid-state sensors for use by analog systems and indicators. An embodiment of the system receives a DC voltage value from at least one sensor, converts the DC voltage value into one or more analog signals based on a reference AC voltage signal, and performs at least one of outputting or storing the generated analog signals. The conversion is performed digitally then converted to analog or is performed using an analog trigonometric converter."

Plain-language overview of the independent claims

Claim 1 — Apparatus (system). A sensor puts out a DC voltage value (e.g., a MEMS or other solid-state aircraft sensor). A "trigonometric converter" turns that DC value into one or more analog outputs based on a reference AC voltage signal. A downstream component either outputs or stores the resulting analog signals. The claim then recites the digital implementation of the converter, comprising:

• an analog-to-digital (A/D) converter that converts the DC value to a digital value;
• a non-linear transfer function component that converts the digital value into a digital trigonometric value (i.e., computing sine-like values by formula rather than via a look-up table);
• a digital offset component that adds an offset value to the digital trigonometric value;
• a first multiplier that multiplies the reference AC voltage signal with the digital trigonometric value (yielding something like A·sin(X)); and
• a "second digital analog multiplier" that multiplies the reference AC voltage signal with the offset output (yielding something like A·sin(X + offset)).

Claim 8 — Method. The method counterpart: receive a DC voltage value from at least one sensor; convert it into one or more analog signals based on a reference AC voltage signal; output or store the analog signals. The converting step specifically comprises: converting the DC value to a digital value; converting that digital value to a digital trigonometric value using a non-linear transfer function; adding an offset value; multiplying the reference AC signal by the digital trigonometric value; and multiplying the reference AC signal by the offset-added trigonometric value.

Dependent claims (context): Claim 2–3 cover use of an off-the-shelf trigonometric chip (claim 3 specifically names the AD639); claim 4 adds an analog multiplier for the reference AC signal; claim 5 adds a phase-shift voltage generator; claim 7 implements the non-linear transfer function in an FPGA. The patent's Equation (1) gives an example non-linear approximation: sin(X) ≈ X(1 − X²)(2.83 − X²) / [0.9(1 + 0.3X²)].

Drafting anomalies worth flagging (do not auto-correct)

• Claim 6 reads "The method of claim 4" even though claim 4 is a system claim ("The system of claim 2…"). This appears to be an internal claim-dependency/typo issue in the printed patent.
• The specification's Figure 2 embodiment is analog (trigonometric chips + analog multipliers), while the claims as issued are directed to the digital/FPGA embodiment of Figure 3. The analog embodiment is largely described but not claimed in independent form.
• Claims 1 and 8 mix "a first multiplier" with "a second digital analog multiplier," an imprecise label in the original text.

Litigation / CAFC 2026 docket check

I found no CAFC 2026 docket, ITC action, or district-court case referencing US 7,064,692 in any of my searches. The Honeywell-related CAFC activity I did surface (e.g., Honeywell International Inc. v. US, Fed. Cir. No. 26-1377, docketed Jan. 26, 2026) is a tariff-classification appeal from the Court of International Trade, unrelated to this patent. I cannot rule out a case that is not publicly indexed, but I have no authoritative evidence of litigation involving this patent. This is a reasonable outcome given the patent lapsed for nonpayment of maintenance fees in 2010 (and again recorded as lapsed in 2018).

Prior art cited on the face of the patent

US 4,375,636 (United Technologies, AC synchro signal to digital converter); US 4,935,692 (Hewlett-Packard); US 5,034,743 (Flight Visions, AC encoded signal to digital converter); US 5,481,451 (Arex Electronics); US 5,880,947 (Mitsubishi Denki); US 2002/0019710 A1 (Foxboro, digital flowmeter).

Numeric-identity caution (per your "do not auto-correct" rule)

My search for the bare string "7064692" also returned unrelated records that share the numeric string but not the patent: JP 7064692 B2 (Japanese patent, GS Yuasa, "power storage device, management device, and method for determining state of power storage device," filed Sept. 26, 2017, registered Apr. 27, 2022) and a Lithuanian corporate-registry notice number. These are not US 7,064,692 and should not be conflated with it.

Uncertainty statement

Bibliographic fields (title, inventors, assignee, filing/issue dates, claims, abstract) are taken verbatim from the patent document and are high confidence. The maintenance-fee lapse dates and family/EP/WO details come from aggregated sources (Google Patents, EPO GPI) and are consistent with one another but were not independently confirmed against a raw USPTO PatentCenter record. The absence of CAFC/litigation activity reflects the absence of results in my searches, not a positive certification that none exists.

I'll search for litigation involving US patent 7064692 specifically.

Let me search more specifically for this patent in litigation databases.

Let me try the Unified Patents litigation portal and other litigation-specific sources.

Litigation Search: U.S. Patent No. 7,064,692

Bottom line: I found no record of any litigation (district court, ITC, or PTAB/AIA proceeding) involving U.S. Patent No. 7,064,692.

I searched for the specific number "7064692" / "7,064,692" across litigation-oriented sources (Unified Patents litigation portal, CAFC opinions/orders, Docket Alarm, PACER-referencing results, and general web). Every hit returned by those queries was either:

• An unrelated patent or case with a similar or coincidentally overlapping number string — e.g., U.S. Pat. No. 6,703,469 (the air-mattress patent litigated in Intex Recreation Corp. v. Team Worldwide Corp., D.D.C. No. 1:04-cv-01785); U.S. Pat. No. 10,311,073 or similar in Unified Patents IPRs; "Patent No. 7064692" in a Japanese filing (JP 7064692 B2, GS Yuasa, a battery device — a different jurisdiction and unrelated subject matter); a Lithuanian registry notice number; or
• The patent's own Google Patents/JP family-citation page (JP2007248479A citing US 7,064,692 B1).

None of these is litigation against U.S. 7,064,692.

Why no litigation result is expected here

The patent's own record (Google Patents, authoritative text supplied) is consistent with a patent that was never asserted:

The patent went abandoned early (first maintenance fee unpaid in 2010) and the family shows no counterpart granted in Europe — EP 1 861 806 A2 is listed as "Withdrawn" and WO 2006/101687 A2 as "Ceased." There is no "Cited By" or "Litigation" entry in the source record indicating assertion, and the 9 claims (a DC-voltage sensor feeding a trigonometric converter producing A·SIN(X) and A·SIN(X+offset)) describe a niche aviation retrofit component rather than a high-value assertion target.

Caveats on completeness

• I could not directly query PACER (paywall) or run the number through the Unified Patents case-list its interactive search; my searches were via web/CAFC/aggregator results.
• A case that was filed and dismissed/settled without a published opinion, filed in state court, or filed outside the U.S., could in principle be invisible to these searches.
• Very recent filings (2025–2026) may not yet be indexed.

If you need an exhaustive negative clearance, the reliable next steps are: (1) PACER "Find a Case" / Docket Alarm search on patent number 7,064,692; (2) the Unified Patents litigation case-list search; and (3) a Lex Machina / Darts-IP patent-number query, which index complaints by patent number.

No plaintiff, defendant, jurisdiction, case number, filing date, or outcome is reported here because none was found. I will not supply placeholder case numbers for litigation that does not appear to exist.

I'll verify the canonical "no proceedings on file" result against public PTAB/CAFC sources before writing.

Let me run a few targeted checks on the PTAB/CAFC side to confirm the negative.

Proceedings overview

Zero AIA trial proceedings are on file for US 7,064,692. The canonical USPTO Open Data Portal block returns no IPR, PGR, or CBM for this patent, and my independent web checks surfaced no petition, institution decision, Final Written Decision, or appeal naming it. The breakdown is therefore: 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denials. For a defendant, that means the patent is untested at the PTAB and un-hardened by any adverse or favorable Board ruling — but "untested" here is a weak signal, because the reason no one filed is almost certainly that the patent was expired for failure to pay maintenance fees since 2010-06-20 and offers no leverage worth an IPR filing fee. There is no FWD to point to, no cancellation to lean on, and no estoppel record; the defensive posture is that you must litigate or design around on your own, not inherit a PTAB win.

The requested per-proceeding sections are omitted deliberately. I will not generate proceeding numbers, petitioner names, panels, or dispositions for proceedings that do not exist — inventing them would be worse than useless, since a fabricated IPR number in a defendant's invalidity or clearance memo is a disciplinary problem.

What I actually checked, and what came back

Numeric collisions encountered (flagging, not auto-correcting): hits for the bare string "7064692" resolve to JP 7064692 B2 (GS Yuasa, 蓄電装置 / power-storage device, filed 2017-09-26, registered 2022-04-27 — a different jurisdiction, different subject matter) and a Lithuanian corporate-registry notice (Pranešimo Nr. 7064692, UAB "Bilubilis"). Neither is US 7,064,692. The only genuine references to US 7,064,692 outside its own document are forward-citation entries in Google Patents tables (e.g., US 4,933,674 "Method and apparatus for correcting resolver errors" lists it under "Families Citing this family," and JP 2007248479A cites it). Citations are not proceedings.

Honeywell PTAB/CAFC activity that is not this patent (to close off false matches a keyword search may surface): Honeywell Int'l v. Arkema Inc., PGR2016-00011 and related, Fed. Cir. Nos. 2018-1151, 2018-1153 (decided 2019-10-01, involving US 9,157,017 — refrigerant/catalyst patents); Honeywell Int'l v. Mexichem Amanco, Fed. Cir. No. 2016-1996 (2017-08-01, inter partes reexamination on refrigerant compositions); Honeywell Int'l Inc. v. 3G Licensing, S.A. (Fed. Cir., 2025-01-02). None involves US 7,064,692.

Strategic summary

Claim status — all nine claims are UNTESTED, not sustained. No claim of US 7,064,692 has been canceled, confirmed, or even construed by the Board. The full original claim set remains as issued on the face of the patent: independent claim 1 (system) and claim 8 (method), with dependents 2–7 and 9. Any statement that a claim "survived" an IPR would be false — survival requires a proceeding, and there wasn't one.

Estoppel landscape — § 315(e)(2) is a non-issue. Because no IPR/PGR was ever instituted, there is no petitioner-side estoppel barring anyone, and no patent-owner-side record to build on either. A defendant today faces a completely clean slate: every § 102/§ 103 ground, every reference in the art, and every § 112 theory remain fully available, constrained only by the ordinary district-court rules (IPR estoppel under § 315(e)(2), which applies only to grounds actually raised in an instituted IPR, does not attach). Conversely, there is no Board claim construction to borrow, so claim terms — including the imprecise "second digital analog multiplier" in claim 1 and the anomalous "The method of claim 4" dependency in claim 6 flagged in the summary section above — are unconstrued and open to argument.

Pattern signals — none to read. No serial petitioner, no defensive aggregator (Unified Patents, RPX, etc.) in the chain, and no PTAB-to-CAFC appellate trajectory for this patent. The real pattern is economic abandonment: the first maintenance fee came due and was not paid (reminder mailed 2010-01-25; lapse recorded 2010-06-20 under 37 CFR 1.362), with a further lapse event recorded 2018-08-21 effective 2018-06-20. The family is equally dead — EP 1 861 806 A2 is "Withdrawn" and WO 2006/101687 A2 is "Ceased." Nothing here was ever valuable enough to assert, which is precisely why nothing was ever IPR'd.

Recommended next steps

• If you are a defendant and a demand letter cites US 7,064,692: check the maintenance-fee status before anything else. The patent is listed Expired – Fee Related in the source record, with lapse events dated 2010-06-20 and 2018-06-20 and an anticipated expiration of 2025-03-18. A patent whose claims lapsed for nonpayment cannot support a live infringement claim for post-lapse conduct — that is a Rule 11 / § 285 conversation, not an IPR conversation. Confirm the current fee status in USPTO PatentCenter (https://patentcenter.uspto.gov) before relying on the aggregated status.
• Because no proceeding exists, there is no FWD to link or quote. Do not represent to a court or client that claims were canceled or sustained — I found no basis for either. If you need an exhaustive negative on PTAB activity, the authoritative direct queries are the PTAB P-TACTS portal (https://ptacts.uspto.gov/ptacts/), the USPTO Open Data Portal (https://api.uspto.gov), and CourtListener (https://www.courtlistener.com/?q=%227%2C064%2C692%22) for any related appeal; my results are from indexed web search, not a raw API pull.
• If you nonetheless need a PTAB filing (e.g., a client insists on an IPR track): an IPR remains technically available on an expired patent, but the Board's practice is to weigh the patent's expired status and the absence of any live infringement exposure heavily against institution. Given the 2010 lapse, this is very likely to be an exercise with no practical benefit — spend the budget on a PatentCenter status confirmation and a § 112 challenge to the claim-6 dependency defect instead.

No proceeding numbers, parties, panels, dates, or outcomes are reported because none was found. I have not supplied placeholders.

I'll verify the assignment record against authoritative sources before writing up.

Assignment & Ownership Reconstruction — US 7,064,692

Source note (important limitation): The USPTO Assignment Center (assignmentcenter.uspto.gov) and the legacy Assignment Search (assignment.uspto.gov/patent/index.html) are interactive/search-gated; I could not pull the raw record page for this patent in my searches. The assignment data below is taken from the authoritative patent document supplied in the analysis (Google Patents legal events, which mirror the Assignment Center record), specifically the Legal Events entry and the Assignment entry:

"2005-03-18 — AS — Assignment — Owner name: HONEYWELL INTERNATIONAL INC., NEW JERSEY — Free format text: ASSIGNMENT OF ASSIGNORS INTEREST; ASSIGNORS: KOSHT, DANIAL L.; TERPSTRA, LARRY J.; REEL/FRAME: 015796/0912 — Effective date: 20050307"

Because I could not open the Assignment Center record directly, the correspondent-of-record field was not retrieved and is flagged as such rather than guessed. This does not affect the substance of the finding: the chain consists of a single assignment, from the inventors to their employer, at filing.

Numeric-identity caution (per the "do not auto-correct" rule): searches for the bare string "7064692" also return JP 7064692 B2 (GS Yuasa, "power storage device / management device / method for determining state of power storage device," filed 2017-09-26, registered 2022-04-27). That is a different patent in a different jurisdiction and is not part of this chain.

Inventors

Pattern assessment — no adverse pattern. The two inventors are a stable, repeat co-inventor pair at Honeywell, appearing together on multiple Honeywell applications filed in the same 2005 window (e.g., US 2006/0164052, assigned to Honeywell International Inc.). There is no evidence of inventors departing the assignee within 12 months of filing, and no re-assignment activity by them. This is the profile of an internal corporate R&D team filing a cluster of related applications — the opposite of the "inventors bail out before a portfolio fire-sale" pattern.

Uncertainty: I did not retrieve inventor residence/citizenship details or the specific Honeywell business-unit affiliation (e.g., Honeywell Aerospace vs. Honeywell Sensing). The subject matter (aircraft sensors, cockpit indicators, flight data recorders) strongly implies Honeywell's aerospace/avionics business, but I state that as inference, not record fact.

Original assignee

Honeywell International Inc. (recorded as "Honeywell International Inc., New Jersey"; corporate address of record in related Honeywell filings: 101 Columbia Road, P.O. Box 2245, Morristown, NJ 07960).

• Primary line of business: Diversified industrial/technology conglomerate; the relevant operating unit here is aerospace/avionics (sensors, cockpit instruments, flight data recorders, air-data and position sensing) — exactly the field the specification addresses ("control surface position, pressure sensors, temperature sensors," "analog indicators or gauges, such as those used in an aircraft cockpit").
• Did they ship a product embodying the claims? Not established from the record. The specification frames the invention as a retrofit converter to let modern solid-state/MEMS sensors drive legacy analog cockpit indicators and data acquisition units. That is a plausible Honeywell product-line use (avionics retrofit/obsolescence management), but I found no product datasheet, catalog entry, or marketing material tying this specific patented converter to a shipped Honeywell part. I decline to assert product practice without that evidence.
• Current status: Honeywell International Inc. is an operating company (not dissolved, not in bankruptcy). No SEC-filed divestiture, spinoff, or asset sale of this patent appears in the record. The patent itself, however, was abandoned for fee non-payment in 2010 and its EP/WO family members were Withdrawn/Ceased — consistent with Honeywell internally deprioritizing this particular filing rather than monetizing it.

Assignment timeline

Exactly one assignment is recorded against US 7,064,692. There are no post-issuance assignments, no security agreements, no change-of-name filings, no licenses, and no releases in the record.

• 2005-03-07 (executed / effective) / recorded 2005-03-18 — Reel 015796 / 0912
  • Conveyance: Assignment (ASSIGNMENT OF ASSIGNORS' INTEREST)
  • Assignors: Danial L. Kosht; Larry J. Terpstra
  • Assignee: Honeywell International Inc. (New Jersey)
  • Correspondent: Not retrieved — the Assignment Center correspondent field could not be opened in my searches. Flagged as a gap, not a null result. (Note: Honeywell filings of this era commonly list Honeywell's own in-house docketing address as correspondent, but I will not state a name I did not verify.)
  • Context: Initial inventor-to-employer assignment at filing — standard employee invention assignment, not an acquisition, fire-sale, securitization, or transfer-to-asserter.

Post-issuance ownership: none recorded. Per the Assignment Center's own structure, the absence of any conveyance after the original inventor assignment means Honeywell International Inc. remained the owner of record through the patent's entire life, up to and past the 2010 lapse for non-payment of the first maintenance fee.

Timeline diagram

timeline
    title Ownership of US 7064692
    2005 : Inventors execute assignment to Honeywell
         : Recorded Reel 015796 frame 0912
    2006 : US 7064692 B1 issued to Honeywell
    2010 : First maintenance fee unpaid
         : Patent lapses for fee nonpayment
    2025 : Anticipated expiration date

NPE / troll-pattern signals

Cross-references checked (per task instructions):

• Google Patents legal events — single Assignment event (2005-03-18, Reel 015796/0912); maintenance-fee reminder 2010-01-25; lapse 2010-06-20; further lapse event 2018-06-20; status Expired – Fee Related.
• USPTO PEDS/ODP & SEC filings — no divestiture/spinoff of this patent surfaced in Honeywell filings.
• RPX / Unified Patents asserter directories — no listing for US 7,064,692 or for any entity owning it other than Honeywell.
• Litigation — consistent with the earlier-generated litigation section, no district-court, ITC, PTAB/AIA, or CAFC matter naming US 7,064,692 was found. (The unrelated Honeywell Int'l v. US, Fed. Cir. No. 26-1377, is a tariff-classification appeal and does not involve this patent.)

Verdict

Insufficient data — specifically, only the original assignment is on record, and it terminates at the original operating assignee.

Justification: The complete ownership chain is a single recorded conveyance — inventors Kosht and Terpstra to Honeywell International Inc., executed 2005-03-07 and recorded 2005-03-18 at Reel 015796/0912 — followed by issuance (2006-06-20) and lapse for non-payment of maintenance fees (2010-06-20). There is no second link, no LLC transferee, no known asserter, no cascading transfer, and no pre-suit assignment, so none of the 8 NPE signals can be marked "present" on the record. This is the textbook "single-employer chain" outcome: the patent never left the company that filed it, was never asserted, and was simply abandoned — hence Insufficient data rather than any NPE or defensive-aggregator characterization.

Verification link: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search by patent number 7064692; the legacy interface is at https://assignment.uspto.gov/patent/index.html). To close the one open gap, pull the Reel/Frame 015796/0912 record directly to read the correspondent of record for that single entry.

Confidence & gaps

• High confidence: one and only one recorded assignment; assignor/assignee names; reel/frame 015796/0912; effective date 2005-03-07; recording date 2005-03-18; no post-issuance assignments; lapse history. These come from the patent document's own legal-events data.
• Not established / open: (a) the correspondent of record on Reel 015796/0912 (interactive field not retrieved); (b) whether Honeywell shipped a product practicing the claims; (c) inventor residence/business-unit details. None of these would change the verdict, since the chain has only one link.

I'll research the cited prior art references for US 7,064,692. Let me search for details on each.

Let me research the remaining cited references.

Prior-Art Analysis for US 7,064,692 B1

Scope note / search limitations. I was not able to reach an interactive USPTO PatentCenter or Patent Application Locator (PAL) session within these searches; the authoritative full-text record used here is the Google Patents record for US 7,064,692 B1 supplied in this task (https://patents.google.com/patent/US7064692/en), including its face-citation list. Where a reference's own text was retrievable, I quote/paraphrase it and cite the URL. I flag below the one reference (Foxboro) whose full text I could not retrieve. Per your rule, I treat "7064692" literally — and I re-flag that JP 7064692 B2 (GS Yuasa, battery device) is a different, unrelated patent sharing the number string and is not analyzed as prior art here.

The two citation lists that exist for this patent

The Google Patents record distinguishes two sets:

• "Patent Citations (6)" — six references cited against the application on the US face.
• "Family Cites Families (4)" — four references cited in the family (the WO 2006/101687 and EP 1 861 806 counterparts), i.e., extended family citations rather than US-face art.

I address both, since the task says "each patent citation."

All six face references and all four family references predate the 2005-03-18 filing date, so each is available as prior art under § 102(a)/(b) (subject to the usual one-year/§ 102(e) mechanics). The decisive analytical issue is directionality of conversion, addressed after the individual entries.

A. The six face citations ("Patent Citations")

1. US 4,375,636 A — "AC Synchro signal to digital signal converter"

• Assignee: United Technologies Corporation
• Filed: 1980-06-12 | Published/Issued: 1983-03-01
• URL: https://patents.google.com/patent/US4375636
• Description (retrieved): A synchro-signal conditioning interface for a Data Acquisition Unit (DAU). It converts sensed AC synchro stator voltages (per ARINC-717) into ratiometric analog values and phase "discretes," which a DAU signal processor reads to compute synchro angle. It explicitly avoids a Scott-T network; the analog interface (multiplexer, differential receiver, programmable-gain amps) feeds a signal processor that ultimately yields a digital representation of angle.
• § 102 assessment vs. claims 1 / 8: No anticipation. The reference converts in the opposite direction (AC synchro → digital angle), whereas claims 1 and 8 require DC sensor value → analog sine-type signals. It discloses an A/D-style conversion path and a signal processor, so it maps only to isolated claim-1 elements (an A/D converter stage, a processor). It contains no non-linear-transfer-function sine generator, no digital offset component, and no pair of multipliers combining a reference AC signal with sine values (A·sin(X), A·sin(X+offset)). It cannot anticipate claims 1 or 8 as a whole; at most it is background art.

2. US 4,935,692 A — "Vector ammeter having digital signal processing unit"

• Assignee: Hewlett-Packard Company (inventor Wakasugi, JP priority 1987-11-30)
• Filed: 1988-11-30 | Published/Issued: 1990-06-19
• URL: https://patents.google.com/patent/US4935692
• Description (retrieved): A vector impedance/ammeter instrument. An AC reference-source signal drives the device under test; an amplifier output is digitized by an A/D converter and fed to a digital signal processing unit that "orthogonally decomposes the AC measurement signal into real and imaginary components." The DSP output feeds a digital-to-analog converter within a cascaded forward-amplification chain with feedback. It computes a vector ratio (impedance/admittance).
• § 102 assessment vs. claims 1 / 8: No anticipation. It does disclose several individual building blocks named in claim 1 — an AC reference signal, an A/D converter, a DSP block, and a D/A converter. But it is a measurement instrument, not a sensor-to-synchro signal generator; there is no DC sensor input, no sine-generation non-linear transfer function, no angular-offset stage, and no dual multiplier producing A·sin(X) and A·sin(X+offset). Element-by-element, it does not disclose the claim as arranged. It is potentially § 103 art against the generic "A/D + DSP + D/A" architecture, not § 102 art against claims 1/8.

3. US 5,034,743 A — "AC encoded signal to digital converter"

• Assignee: Flight Visions Incorporated
• Filed: 1989-08-07 | Published/Issued: 1991-07-23
• URL: https://patents.google.com/patent/US5034743
• Description (retrieved): A synchro/resolver → digital converter. A synchronous demodulator uses the reference excitation signal to convert each AC synchro/resolver signal to a DC value (negative if 180° out of phase with the reference), which is then digitized and "digitally processed based on their trigonometric relationship to extract the encoded angular position value." A specific embodiment reuses an existing A/D converter + microprocessor (e.g., Motorola 68000; A/D model AD574) in a HUD data-acquisition system.
• § 102 assessment vs. claims 1 / 8: No anticipation. Same directional mismatch: it goes AC-encoded → digital angle. It does highlight a reference-AC/reference-excitation signal, synchronous demodulation, A/D conversion, and digital trigonometric processing — the closest conceptual cousin among the six to claim 1's "digital trigonometric value." But it does not teach generating a sine value from a DC sensor input via a non-linear transfer function, does not teach a digital offset adder, and does not teach the two multiplying D/A converters yielding A·sin(X) and A·sin(X+offset). No single-reference anticipation of claims 1/8.

4. US 5,481,451 A — "AC-to-AC power inverter apparatus functioning without smoothing capacitor, and control method thereof"

• Assignee: Arex Electronics Corporation
• Filed: 1992-10-30 | Published/Issued: 1996-01-02
• URL: https://patents.google.com/patent/US5481451
• Description (retrieved): A power inverter control. Input AC is converted to a DC voltage (with ripple) and then converted back to a 3-phase AC output by a switch-mode inverter under PWM control. A microcomputer reads a reference vector memory (stored unit-length/zero-length vectors corresponding to a reference output waveform) and modulates switching so the output tracks the target, compensating for DC ripple.
• § 102 assessment vs. claims 1 / 8: No anticipation. This is the only face reference that is directionally akin to the patent (DC intermediate → AC output generation). However, its waveform generation uses a stored reference-vector look-up memory (a look-up-table approach) — precisely the technique the patent's Background criticizes — not a non-linear transfer function computed from a sensor DC value. There is no sensor-angle input, no digital angular-offset adder, and no pair of multipliers producing A·sin(X)/A·sin(X+offset). It does not anticipate claims 1/8; it is at most general § 103 art and is relevant to the "look-up table vs. computed sine" distinction.

5. US 5,880,947 A — "Control device for a PWM controlled converter having a voltage controller"

• Assignee: Mitsubishi Denki Kabushiki Kaisha
• Filed: 1995-09-27 | Published/Issued: 1999-03-09
• URL: https://patents.google.com/patent/US5880947
• Description (retrieved): A PWM converter control. A unit sine wave generator (11) produces R-phase and T-phase unit sine waves synchronized to the AC source; multipliers (12, 13) multiply a peak-value instruction by those unit sine waves to form input-current instruction signals; PI controllers generate PWM gating.
• § 102 assessment vs. claims 1 / 8: No anticipation. It discloses multipliers that multiply a reference sine signal — superficially parallel to claim 1's "multiplier … multipl[ying] the reference AC voltage signal." But the sine source is a synchronized oscillator/unit-sine generator (again, the oscillator approach the patent disparages), not a non-linear-transfer-function computation from a sensor DC value; and there is no digital angular-offset adder, no DC-sensor-to-synchro conversion, and no A·sin(X+offset) output. It cannot anticipate claims 1/8.

6. US 2002/0019710 A1 — "Digital flowmeter"

• Assignee: The Foxboro Company (Massachusetts)
• Filed: 1997-11-26 | Published: 2002-02-14
• Description: A digital (presumed digital-signal-processing / Coriolis-type) flowmeter. ⚠️ I was unable to retrieve the full specification text of this publication in these searches, so the description is based on its title and assignment only, and my § 102 assessment is correspondingly lower-confidence.
• § 102 assessment vs. claims 1 / 8: On the face of it, a digital flowmeter is in a different field (mass/volumetric flow measurement via a vibrating-tube sensor) and is not directed to synchro/resolver signal synthesis. There is no apparent disclosure of a DC-sensor-to-AC-sine converter with a non-linear sine transfer function, a digital angular offset, or dual multiplying D/A converters. I would expect no anticipation of claims 1/8, but I flag this as unverified because I could not read the reference. If a full § 102 opinion is needed, this text should be pulled and read.

B. The four family citations ("Family Cites Families")

These were cited in the WO/EP family counterparts. All are resolver/synchro-angle-processing references.

C. Bottom-line § 102 conclusion

No single one of the ten cited references anticipates independent claim 1 or independent claim 8.

The controlling reason is a direction-of-conversion mismatch:

• The patent's claims are directed at DC-sensor-value → analog synchro/resolver-style output — i.e., a digital-to-synchro/resolver (D/S, D/R) function — via (a) an A/D stage, (b) a non-linear transfer function that computes a sine value, (c) a digital offset adder, and (d) two multipliers yielding A·sin(X) and A·sin(X+offset).
• The closest cited references (US 4,375,636; US 5,034,743; US 4,933,674; US 4,857,926; US 5,162,798; JP 2000-337924) go the opposite way (AC synchro/resolver → digital angle).
• The references that do generate AC (US 5,481,451; US 5,880,947) do so for power-converter control, using a stored look-up table or a synchronized oscillator/unit-sine generator — not the claimed non-linear transfer function computed from a sensor DC value, and without the digital angular offset or the A·sin(X)/A·sin(X+offset) paired output.
• The HP vector ammeter (US 4,935,692) supplies generic building blocks (AC reference, A/D, DSP, D/A), and the Foxboro digital flowmeter is a different field entirely — neither is a full-claim disclosure.

Consequently, the substantive exposure from these references is under 35 U.S.C. § 103 (obviousness as background/combinations), not § 102. This is consistent with the patent having issued over them in 2006.

Where partial element overlap does exist (useful for an invalidity/§ 103 mapping rather than a § 102 mapping):

• Element "A/D converter converts … into a digital value" (claim 1): disclosed by US 4,375,636; US 4,935,692; US 5,034,743.
• Element "digital trigonometric processing" (claim 1): conceptually approached by US 5,034,743 (digital trigonometric extraction) and US 4,935,692 (DSP orthogonality decomposition).
• Element "multiplying a reference signal with a sine signal" (claim 1): approached by US 5,880,947 (unit-sine multipliers) and US 4,935,692 (phase-sensitive detectors/modulators).
• Element "generating AC from a DC intermediate" (claim 1/8 preamble): approached by US 5,481,451 (DC intermediate → 3-phase AC) and US 5,880,947.
• Closest single reference overall: US 5,034,743 (Flight Visions) for the reference-AC/demodulate/A-D/digital-trigonometry concept, and US 5,880,947 (Mitsubishi) for the sine-multiplier concept — but neither directionally or completely meets claims 1/8.

Uncertainty / next steps. Two caveats for a formal opinion: (1) I could not retrieve or analyze the full text of US 2002/0019710 A1 (Foxboro) — its § 102 analysis above is title-based and should be confirmed; (2) I did not independently re-pull the USPTO face-citation list from PatentCenter, relying instead on the supplied Google Patents record. If you want a strict § 102 "all-elements-in-one-reference" chart mapped limitation-by-limitation, the references to read in full are US 5,034,743 and US 5,880,947 (the two with the most element overlap) and the Foxboro publication (unread).

I'll pull details on the cited prior-art references so the § 103 analysis is grounded in what those references actually disclose.

Now let me pull the remaining cited references and the family-cited Honeywell/HP references.

Obviousness Analysis — U.S. Patent No. 7,064,692 under 35 U.S.C. § 103

Patent: US 7,064,692 B1, "Solid-state synchro/resolver converter" (Honeywell International Inc.; Kosht & Terpstra; filed 2005‑03‑18; issued 2006‑06‑20; claims 1–9; independent claims 1 and 8).

Scope note / not legal advice. This is a technical prior-art analysis, not a validity opinion or freedom-to-operate conclusion. I have not performed claim construction, and I have not reviewed prosecution history (no office actions, no IDS, no applicant remarks were available to me). Ambiguities in claim language are flagged rather than resolved.

Source discipline (per your rule). Element mappings below are grounded in the text I actually retrieved. I retrieved full disclosure text for US 4,375,636, US 5,034,743, US 4,933,674, US 5,162,798, and US 4,935,692. I did not retrieve disclosure text for US 5,481,451 (Arex), US 5,880,947 (Mitsubishi), US 2002/0019710 A1 (Foxboro), US 4,857,926 (Honeywell), or JP 2000‑337924 A (Minebea), so I do not assert what they disclose beyond their titles. Statements about general knowledge (e.g., polynomial approximation of sine, CORDIC, FPGA implementation) are flagged as such and are not tied to a specific retrieved reference.

Timing note. The application was filed 2005‑03‑18, pre‑KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007). Any contemporaneous obviousness determination would have applied the pre‑KSR Federal Circuit "motivation/suggestion/teaching" framework. A modern re-analysis (IPR, district court) applies KSR's broader "reason to combine" and "predictable results" rubric. I analyze under KSR and note where the older framework would be more favorable to the patentee.

1. The inventive subject matter, reduced to elements

Independent claim 1 is a hybrid: an analog-facing preamble (DC-voltage sensor → trigonometric converter producing analog output based on a reference AC voltage signal → output/store component) wrapped around a wholesale replacement of that converter with a digital signal chain:

Dependent claims add: trigonometric chip (2), AD639 specifically (3), analog multiplier for the reference AC signal (4), phase-shift voltage generator (5), FPGA implementation (7). Claim 6 is mis-drafted ("The method of claim 4," while claim 4 is a system claim) — a § 112 defect I flagged in the summary section, not a § 103 point, but it means claim 6's scope is ambiguous and its validity is separately assailable.

2. What the cited references actually teach

• US 5,034,743 — Flight Visions, "AC encoded signal to digital converter." A synchronous demodulator uses the associated reference excitation signal to turn AC synchro/resolver signals into DC values; those DC values are digitized (A/D converter 14 / 40, e.g., Analog Devices AD574) and then "digitally processed based on their trigonometric relationship to extract the encoded angular position value." Critically, the reference states: "Virtually any acquisition system comprising an A/D converter and processing capability sufficient to process digital data using trigonometric functions can provide the conversion and processing capability suitable to practice the invention," and it expressly motivates reuse of existing aircraft A/D and processing capability in a head-up display system to reduce cost, weight, volume, and power. It also handles a DC sensor among "other analog inputs 36." (https://patents.google.com/patent/US5034743)

• US 5,162,798 — Pacific Scientific, "Resolver to digital converter." A resolver excitation source provides a sinusoidal AC carrier reference. The "control transformer" contains a pair of multiplying digital-to-analog converters (MDACs) that each take a digital input and an analog multiplying reference and produce the product of the resolver winding voltage and the sine or cosine of the digital input. A digital position estimate θ_E is fed back to the MDACs. (https://patents.google.com/patent/US5162798)

• US 4,933,674 — Allen-Bradley, "Method and apparatus for correcting resolver errors." Discloses a synchro-to-digital tracking converter that uses a digitized reference waveform stored in a read-only memory — i.e., a digitally generated sine reference used in the conversion loop. (https://patents.google.com/patent/US4933674)

• US 4,375,636 — United Technologies, "AC synchro signal to digital signal converter." An ARINC-717 data acquisition unit receiving AC synchro sensors 15A and a DC sensor 15B; a ratiometric A/D converter digitizes; a digital computation device implements a trigonometric expression yielding sine, cosine, and angle information; output goes to data utilization circuits / a flight data recorder. (https://patents.google.com/patent/US4375636)

• US 4,935,692 — Hewlett-Packard, "Vector ammeter having digital signal processing unit." A/D converter (21) feeding a digital signal processing unit (23) which performs digital phase-sensitive detection and digital integration, with a D/A converter on the output of the DSP chain and microcode-based "transfer functions, phase compensation, and corrections." (https://patents.google.com/patent/US4935692)

Family-cited (listed under "Family Cites Families"). US 4,933,674 (above), US 4,857,926 (Honeywell, Phase-responsive angular tracking device), US 5,162,798 (above), and JP 2000‑337924 A (Minebea, R/D converter) — I did not retrieve disclosure text for US 4,857,926 or JP 2000‑337924 A and do not rely on them substantively.

3. Primary ground of rejection: US 5,034,743 in view of US 5,162,798 (optionally with US 4,933,674)

Mapping.

Motivation to combine — specific and articulated.

1. Same field, same problem, interoperable art. All three references are in synchro/resolver signal conversion. US 5,034,743's stated objective — "reduced cost, weight, volume, and power… while retaining high reliability and making use of existing processing capability" — is verbatim the motivation recited in the '692 background (retrofit legacy analog cockpits/recorders, avoid replacing indicators). This is not hindsight; it is the reference's own stated purpose.
2. US 5,034,743 itself supplies the bridge. It instructs that any A/D-plus-processor system "sufficient to process digital data using trigonometric functions" is suitable, and it expressly contemplates a DC sensor input on the same multiplexed interface. A POSITA reading that reference is directly led to implement trig extraction in the digital domain.
3. US 5,162,798 supplies the output stage the patent claims. The patent's elements G/H require multiplying a reference AC voltage signal by a digital trigonometric value. US 5,162,798's multiplying D/A converters do exactly that operation (digital word × analog carrier reference). Substituting a multiplying D/A converter for a discrete analog multiplier is a substitution of one known element for another to obtain the predictable result — a KSR rationale.
4. Predictable, finite solution set. Given the patent's own framing ("perform the trig conversion digitally then convert to analog, or using an analog trig converter"), the field had a small, enumerated set of implementations (analog trig chip; digital computation + multiplying D/A; DDS/LUT; CORDIC; polynomial approximation). KSR holds that where a finite number of identified, predictable solutions exists and a POSITA would have pursued the known options, the resulting combination is obvious.

4. Secondary ground: US 4,375,636 in view of US 5,034,743 and US 5,162,798

This ground is useful because it supplies the avionics retrofit context and the flight-data-recorder/output component that the '692 specification recites ("An example data acquisition unit is the processor associated with a flight data recorder"):

• '636 Fig. 1 groups an AC synchro sensor with a DC sensor 15B on the same ARINC-717 interface — the patent's element A.
• '636's ratiometric A/D + digital computation device implementing a trigonometric expression — elements D/E.
• '636's data utilization circuits 36 — element C.
• Combining '636 (avionics DAU architecture) with '743 (digital trig extraction) and '798 (multiplying-D/A output stage with an AC reference) yields every element of claim 1.

Motivation: all three are avionics/flight-data-acquisition art; the '692 background expressly characterizes the problem as retrofitting aircraft sensors into existing aircraft indicators and recorders. Combining avionics references that address the same aircraft signal-format problem is the paradigm case of an articulated reason to combine.

5. The dependent claims are weaker than the independent claims

Note also that claims 2/4/5 describe the analog embodiment of Figure 2 but depend from claim 1, whose converter is defined digitally. That internal inconsistency (§ 112) does not help the patentee on § 103 and, if the claims were read to import the analog chips into the digital claim, it merely collapses claim 1 into the Figure 2 embodiment, which is the more squarely anticipated/disclosed subject matter.

6. The one limitation with genuine argument space: "non-linear transfer function"

I want to be candid about where the patent's best non-obviousness footing lies, and what it would take to knock it down.

The patentee's argument would be: the claims require producing a digital trigonometric value by a non-linear transfer function — i.e., computing sin(X) by formula, not by indexing a table. The '692 background expressly disparages the alternatives: "Current translators or converters utilize look‑up tables or oscillators… However, these methods are costly and difficult to implement. Traditional methods require memory (for look‑up tables) and/or oscillators that lower system reliability… Implementation using look‑up tables also has finite resolutions bounded by their digital size." Under that reading:

• US 4,933,674 is arguably away from the claim, because it teaches the very PROM look‑up table waveform generation the patent criticizes.
• US 5,034,743 and US 4,375,636 describe trig processing at a level of generality that does not specify how sine is computed.

Why this argument is nonetheless vulnerable, in my assessment:

1. Claim 1 is not limited to a polynomial or rational approximant. The specific Equation (1) identity (a rational approximation of sin X) appears only in the specification, not in any claim. Construed literally, "non-linear transfer function" covers any non-linear mapping from the digital input to a digital trig value — including a memory-based mapping. If so construed, US 4,933,674's digitized sine reference waveform (and the general DDS/LUT art) reads directly on element E.
2. Disparagement of a reference is not automatically teaching away. KSR and its progeny require that the criticism amount to a teaching that the reference's approach would not work or could not be adapted — not merely that it has drawbacks (memory cost, finite resolution). The patent's criticisms are cost/quality trade-offs, and its own specification concedes the alternatives are usable ("The trigonometric conversion is performed either digitally then converted to analog, or is performed using an analog trigonometric converter").
3. Polynomial/rational approximation of sine was standard general knowledge. Minimax and Padé-type rational approximations of sin were textbook material well before 2005, as was CORDIC (Volder, 1959). I am asserting this as general knowledge, not from a retrieved reference — if this ground is to be relied on formally, a specific pre‑2005 reference (a DSP text, a math-handbook approximation chapter, or a CORDIC/DDS reference) must be put in the record.
4. The performance advantage is asserted, not evidenced. Even if the patentee ties the advantage to formula-based sine generation, secondary considerations require a nexus plus evidence. There is none in the record I retrieved, and the patent lapsed for non‑payment of maintenance fees in 2010 (and again recorded lapsed in 2018), which tends to undermine any commercial-success narrative.

Net: the "non-linear transfer function" limitation is the only element with meaningful non-obviousness potential. It is a claim-construction–dependent limitation: broad construction → obvious (LUT art reads on it); narrow construction (polynomial/rational only) → the patent likely survives, but only because the claim would then cover a narrow slice of the disclosed embodiments, and its validity would hinge entirely on the Equation (1)-type approximant not being in the prior art.

7. The strongest counterargument the patentee has on the independent claims: direction of conversion

Most of the cited art runs the opposite direction from the claimed invention:

The '692 specification is explicit that this is the point: legacy indicators and recorders "still need to receive the data in the same format that was outputted by the old style sensors," so the converter must manufacture the A·SIN(X) / A·SIN(X + offset) pair that a synchro or resolver would have produced. This is functionally a digital-to-synchro (D/S) converter, and the cited art is almost entirely S/D. A patentee can legitimately argue that references directed to extracting an angle from a real synchro say little about synthesizing a quadrature-modulated carrier pair from a memoryless DC input.

My assessment of this counterargument:

• It is credible but not decisive. The claimed digital core (D/A conversion, non-linear mapping to a trig value, offset addition, and multiplication by a reference carrier) is reciprocal to the S/D function: the operations and the required precision/linearity concerns are the same, and US 5,162,798's multiplying D/A converters driven by a digital angle word and an AC carrier reference are structurally the claimed output stage, merely vectoring the other way. A POSITA asked to build a D/S converter would naturally start from the MDAC-based architectures already used in R/D converters to drive a resolver (e.g., the resolver excitation/excitation-synthesis art, and US 4,857,926's phase-responsive tracking device, both of which I could not fully retrieve).
• However, a well-constructed § 103 challenge should add a reference squarely in the D/S or synchro/resolver synthesis art — i.e., a reference that generates sin/cos-modulated outputs from a digital input for driving a synchro indicator or resolver. The six references cited on the face of the '692 patent do not obviously include one, and that gap is the patentee's best structural defense. If someone were to mount an IPR on the two independent claims, this is where I would look first for a missing element in the primary reference set.

8. Bottom-line assessment

Confidence and limits. Element-to-reference mappings for US 5,034,743, US 5,162,798, US 4,933,674, US 4,375,636, and US 4,935,692 are grounded in the disclosure text I retrieved (URLs above) and are high confidence. My characterizations of US 5,481,451, US 5,880,947, US 2002/0019710 A1, US 4,857,926, and JP 2000‑337924 A are based on their titles only — I did not retrieve their disclosures within this session and I make no claim about what they teach. Assertions that CORDIC, polynomial/rational sine approximation, and FPGA implementation were general knowledge as of 2005 are my own assessment, not retrieved evidence, and would need documentary support before being relied on. Finally, this analysis is only as good as the claim constructions of "trigonometric converter," "non-linear transfer function component," "digital offset component," and "digital analog multiplier" — none of which I have construed, and the last of which is an undefined term appearing only in claims 1 and 8.