US 6650875

I'll search for litigation and docket information on this specific patent number before summarizing.

Let me do one more targeted check for any litigation or PTAB proceeding naming this patent.

Search result: no litigation/docket activity found for this patent

I searched for USPTO proceedings and Federal Circuit 2026 dockets naming US 6,650,875. I found no CAFC 2026 appeal, ITC investigation, district court action, or PTAB proceeding that specifically names US 6,650,875. What the searches did surface were similarly-numbered or same-party matters that must not be conflated with this patent:

• JP6650875B2 / Japanese Patent No. 6650875 — a Koninklijke Philips beverage-dispenser/consumable-recognition patent (issued 2020-02-19). This is a different jurisdiction and a different invention, matched only by numeric coincidence.
• Skyworks v. Kangxi (ITC Inv. No. 337-TA-1413; C.D. Cal. 8:24-cv-00974/00976) — asserted US 8,717,101; 9,917,563; 9,450,579; 9,148,194; 7,409,200. US 6,650,875 is not among them.
• Skyworks v. Kinetic Technologies (N.D. Cal. 3:14-cv-00010) — asserted US 7,921,320 and 8,539,275. Not this patent.
• Unrelated results (Netlist/Samsung PGR, Woodway/LifeCORE treadmill appeal) were false hits.

Important status point: per the patent record, US 6,650,875 is "Expired – Lifetime," with an adjusted expiration of 2021‑12‑27. That places it well outside any live 2026 enforcement posture, which is consistent with finding no current docket. (Uncertainty note: I cannot rule out a historical suit or a PTAB filing from the 2003–2021 window; my searches were targeted at 2026 activity and did not return an authoritative litigation history for this number.)

US 6,650,875 B1 — Summary

Source: https://patents.google.com/patent/[US6650875B1](/patent/US6650875B1)/en

Assignment-chain detail (interpreted literally): the record shows an initial 2000‑08‑30 assignment to Conexant Systems, Inc., with assignors listed as "DAMGAARD, MORTON; DOMINO, WILLIAM J.; ROZENBLIT, DMITRIY; VAKILIAN, NOOSHIN D." — note that the assignment entry spells the second inventor's given name "MORTON DAMGAARD" while the front-page inventor field reads "Morten Damgaard." I flag this discrepancy rather than harmonizing it. Later entries: Washington Sub, Inc. (2002‑08‑20) → Alpha Industries, Inc. (2002‑08‑26, merger) → Skyworks Solutions, Inc. (2002‑08‑26, merger), plus security-interest and release/reconveyance filings with Conexant (2002‑09‑16 / 2003‑10‑06).

Abstract (as published)

A system for using a translation loop upconverter and power amplifier including a secondary phase-error correction loop uses the output of the upconverter to lock a feedback loop during a time in which the output of the power amplifier is insufficient to provide feedback to the upconverter. After the power amplifier has developed sufficient power to provide feedback to the upconverter, the feedback to the upconverter is taken from the output of the power amplifier. By placing a phase detector and phase shifter in the secondary feedback path taken from the output of the power amplifier, any phase distortion present in the system can be detected and corrected.

Field and technical context

The patent addresses a polar (PM + AM) transmitter for burst-mode systems such as GSM/EDGE, where a linear amplifier would be too inefficient but a non-linear power amplifier introduces AM-to-PM conversion and phase error. The invention keeps the amplifier non-linear (efficient) by feeding the AM component into the power amplifier's control/bias port rather than the RF input, while an outer phase-correction loop removes the PA's phase distortion. The "amplitude reconstruction" in the title refers to circuit blocks that strip the AM component out of the feedback path (e.g., variable gain element 312 driven by an inverse-AM signal on line 311; in FIG. 7, IF variable gain element 402) so that no further AM-to-PM conversion is added in the feedback path, leaving a constant-envelope signal carrying only phase distortion.

Architecture in plain language

A translation loop (mixer 212 + synthesizer 142 + TX VCO 207 + phase detector 201) upconverts an IF modulated signal into the transmit frequency. During the first part of a burst (PA output too low to be usable), the loop feeds back only the TX VCO output (connection 215). Once the PA produces enough power, a second feedback path from the PA output (through an attenuator/variable gain element 312 that removes AM, then mixer 214) is added. A phase detector 209 compares that downconverted PA feedback against the original modulated input signal and drives a phase shifter 220 in the input path, so the PA's (and the rest of the chain's) phase error is cancelled. Figures 3–7 show alternatives: dividing the mixer output (FIG. 3), a single variable gain element serving both AM injection and phase feedback (FIG. 4), I/Q modulators inside the PLL (FIGS. 5–6), and an IF-domain amplitude-reconstruction variant (FIG. 7).

Independent claims — plain-language overview

The patent has five independent claims: 1 and 23 (methods), 12 and 28 (systems), and 33 (a means-plus-function system claim). All 33 claims are present; dependent claims are 2–11 (on 1), 13–22 (on 12), 24–27 (on 23), and 29–32 (on 28).

Claim 1 — Method (translation-loop transmitter with secondary phase-error correction).
Steps: (a) supply a first modulated signal to a translation loop that includes a PLL and outputs a frequency-specific modulated signal; (b) feed that to a power amplifier; (c) supply a second modulated signal to the power amplifier via a variable gain element; (d) detect a portion of the translation loop output and a portion of the PA output power; (e) apply the translation-loop detection to the loop input during a first time period; (f) apply the PA-output-power detection to the loop input during a second time period; (g) detect the phase difference between the first modulated signal and the detected PA output; (h) derive an error signal; and (i) feed that error signal to a phase shifter that adjusts the phase of the first modulated signal to compensate for phase variation in the PA output.

Claim 12 — System (counterpart to claim 1).
A translation loop with a PLL that receives the first modulated signal and produces the frequency-specific modulated signal; a power amplifier receiving that signal; a variable gain element that injects a second modulated signal into the PA; a first mixer taking a portion of the translation-loop output during the first time period; a second mixer taking a portion of the PA output power during the second time period; a phase detector receiving the second mixer output and the first modulated signal and producing an error signal; and a phase shifter adjusting the first modulated signal's phase to compensate for PA phase variation.

Claim 23 — Method (phase-error correction, generalized).
Same closed-loop scheme as claim 1 but recited over a single "a modulated signal" rather than a first/second pair, and without the variable-gain-element/AM-injection step: provide the modulated signal to a translation loop; feed the frequency-specific signal to the PA; detect portions of the translation-loop output and the PA output power; apply the detection to the loop input during first and second time periods; detect the phase difference between the modulated signal and the detected PA output; develop an error signal; supply it to a phase shifter to correct phase variation in the PA output.

Claim 28 — System (counterpart to claim 23).
The system analogue of claim 23: translation loop receiving the modulated signal; PA; first mixer (translation-loop output, first period); second mixer (PA output, second period); phase detector; and phase shifter. (Drafting note, not auto-corrected: claim 28 recites the phase detector receiving "the output of the second mixer and the first modulated signal," although claim 28's antecedent is "a modulated signal" — a literal antecedent-basis quirk in the issued text.)

Claim 33 — Means-plus-function system (counterpart to claim 23).
Recites the same functional sequence as claim 23 purely as "means for" elements: means for providing the modulated signal to a translation loop; means for providing the frequency-specific signal to the PA; means for detecting the translation-loop output portion and the PA output-power portion; means for applying each detection to the loop input during the first and second time periods respectively; means for detecting the phase difference; means for developing the error signal; and means for supplying the error signal to a phase shifter. This is the classic § 112 ¶ 6 style claim whose scope would be tied to the corresponding structures disclosed (phase detectors 201/209, mixers 212/214, phase shifter 220, TX VCO 207, etc.) and equivalents.

Notable dependent claims (scope anchors)

• Claims 2/3: variable gain element is a variable gain amplifier or an attenuator.
• Claim 4: first modulated signal is phase modulated; Claim 5: second modulated signal is amplitude modulated (the polar split).
• Claims 6/7: the variable gain element operates on an RF signal or on a baseband signal representing the PA's RF output power.
• Claim 8 (and 19, 24, 29): the first and second time periods overlap — i.e., the VCO-derived and PA-derived feedbacks are simultaneously applied, which is the crux of the burst ramp-up behavior described in the spec.
• Claims 9/25/30: a second variable gain element applies the PA output-power portion back into the translation loop.
• Claims 10/11 (and 21/22, 26/27, 31/32): the translation-loop feedback is a first feedback loop and the PA feedback is a second feedback loop.

Prosecution/prior-art context (from the record)

Fifteen references are cited, including US 4,600,892 (Hughes, RF amplifier frequency spectrum control), US 5,420,536 (Victoria University, linearized power amplifier), US 5,570,063 (Spectrian, predistortion), US 5,820,527 (TRL, GMSK PLL modulator), US 5,880,633 (Motorola, high-efficiency power amplifier), US 6,295,442 (Ericsson, AM-to-PM cancellation), US 6,427,900 (Philips, phase modulator + PLL), US 6,480,704 (Nokia) and US 6,526,265 (Skyworks). The patent is cited by 19 publications, including Skyworks follow-ons such as US 6,952,569 and US 6,792,282 and Trinh's US 7,218,951 (continuous closed-loop power control with modulation injection) — indicating this filing sits in a family of related Skyworks polar-transmitter work from the same 2000–2001 period.

Confidence and caveats

• Title, dates, inventors, assignee, abstract, claim text, and classifications are taken verbatim from the authoritative full patent text supplied and the Google Patents record above — high confidence.
• The absence of a 2026 CAFC docket or live proceeding is based on targeted searches; absence of results is not proof of absence, and I flag that I did not retrieve an authoritative complete litigation history for this number. Given the 2021‑12‑27 adjusted expiration, active 2026 litigation would be unusual.
• I have not auto-corrected the inventor-name spelling variance ("Morten Damgaard" vs. "MORTON DAMGAARD") or the antecedent quirk in claim 28; both are reported literally as they appear.

I'll research litigation involving US patent 6650875. Let me search several sources.

Let me search more targeted litigation databases and the patent's own litigation history.

Let me query litigation-specific databases directly.

Litigation Search Report — U.S. Patent No. 6,650,875

Bottom line: I found no litigation involving U.S. Patent No. 6,650,875. Based on the sources I was able to reach, there is no known district-court, ITC, PTAB, or Federal Circuit proceeding in which this patent was asserted, challenged, or at issue. I state this as a negative finding with the verification caveats noted below — not as an affirmative representation that no case has ever existed.

1. The patent as literally identified

• Patent number: US 6,650,875 B1
• Title: "Transmitter architecture having a secondary phase-error correction loop including an amplitude reconstruction system"
• Application no.: US 09/650,831
• Filing date / priority date: 2000-08-30
• Grant date: 2003-11-18
• Inventors: Dimitriy Rozenblit; William J. Domino; Morten Damgaard; Nooshin D. Vakilian
• Original assignee: Conexant Systems, Inc. → Washington Sub, Inc. → Alpha Industries, Inc. → Skyworks Solutions, Inc. (current assignee)
• Legal status: Expired – Lifetime (adjusted expiration 2021-12-27)

I confirmed this identity against the Google Patents record at https://patents.google.com/patent/US6650875/en. Note the important literal-reading caution: much of the web noise returned for the string "6650875" belongs to JP 6650875 B2, an unrelated Japanese patent (Koninklijke Philips, beverage-dispenser consumable recognition), and to CN 102356543 B, which merely cites US 6,650,875 B1 as prior art. Neither of those is litigation of the U.S. patent.

2. Sources consulted and results

The only Skyworks-versus-third-party results my searches produced involved patents other than 6,650,875, so I have not listed them as litigation of this patent.

3. Why litigation is unlikely (context, not proof)

• The patent expired 2021-12-27 (adjusted expiration), so it cannot be asserted prospectively.
• Its owner, Skyworks, acquired the portfolio via the Conexant → Washington Sub → Alpha Industries → Skyworks chain (2002–2003), and the patents I saw litigated in that portfolio are later, different patents (e.g., 7,921,320; 8,539,275; 7,758,979).
• The '875 patent's closest relatives surfaced in the record — Skyworks' U.S. 6,792,282 ("Multiple step switched translation loop…"), U.S. 6,754,471, and U.S. 2003/0139148 A1 (Damgaard, "Wireless transmitter having a modified translation loop architecture") — are family/technical neighbors, but I found no case naming 6,650,875 itself.

4. Verification caveats (please read)

• I was unable to directly reach Unified Patents' caselist, PACER, CourtListener, Docket Alarm, or RPX Insight for an authoritative hit-list on this docket; the available search tool returned generic and unrelated content and then stopped responding. My negative finding therefore rests on the absence of the patent number in the reachable records, including Google Patents' own litigation panel.
• To confirm with the level of certainty a litigation opinion requires, I recommend a paid/authoritative check of: Lex Machina, RPX Insight (patent-level "Litigation" tab), Docket Alarm (patent view for US6650875), PACER full-text and party search, and USPTO PTAB (IPR/PGR/CBM) plus the ITC EDIS docket system.

Conclusion: On the record available to me, no litigation is known involving U.S. Patent No. 6,650,875. If you need this certified for a diligence or freedom-to-operate opinion, run the paid databases listed in §4 before relying on it.

I'll verify the structured block's default with targeted web searches before writing anything.

Let me run additional checks for any PTAB filing, litigation history, or Federal Circuit involvement.

Two more targeted checks — litigation/reexam history and any Unified Patents or aggregator activity.

Proceedings overview

No AIA trial proceedings are on file for US 6650875. The USPTO Open Data Portal structured block for this patent returns zero IPR, PGR, or CBM proceedings, and my independent web checks of PTAB decision indexes and secondary trackers surfaced nothing contradicting that. So the count is 0 proceedings — 0 active, 0 with claims invalidated, 0 with claims sustained, 0 settled, 0 institution denials. Bottom line for a defendant: there is no IPR estoppel, no cancellation order, and no adversarial claim construction to hide behind — but the patent is also expired as of 2021-12-27, which is the single most useful defensive fact in the file. All 33 claims remain standing and untested, so validity must be attacked from scratch (district court or a fresh IPR) rather than piggybacked on an existing PTAB record.

Nothing to enumerate

The instruction to order proceedings "most-impactful first" has nothing to operate on. Rather than fabricate a docket, here is the verification trail.

• Structured source (canonical): the "PTAB proceedings on file" block states the ODP API returns no AIA trial proceedings for this patent as of the most recent ingest.
• Independent checks: keyword searches on the patent number in combination with IPR/PGR/PTAB terminology returned only (a) the patent's own Google Patents page, (b) third-party patents that cite 6650875 as prior art (e.g., its use as a reference during prosecution of Qualcomm-family CN102356543B), and (c) Skyworks PTAB activity directed at entirely different patents — IPR2014-00530 (US 7,921,320, Kinetic Technologies v. Skyworks), IPR2014-00690 (US 8,539,275, same parties), and the 2024–2025 Kangxi cluster (US 8,717,101, 9,450,579, 9,914,563, 9,148,194, 7,409,200). None of those dockets involves US 6650875.
• Related patent: the sibling application US 09/650,831-family patent US 6,792,282 ("Multiple step switched translation loop for power amplifier feedback control," also Skyworks) likewise shows no IPR/PGR in the material surfaced.

Confidence statement: I am highly confident there is no indexed AIA trial on this patent, because the ODP block is dispositive on that point and search corroborated it. I cannot rule out a very recently filed, not-yet-indexed petition or an ex parte reexamination request filed in the last few weeks; I found no evidence of either, and I am flagging the limit rather than asserting a negative I cannot fully prove.

Strategic summary

Claim status. All of claims 1–33 are UNTESTED and INTACT. There is no PTAB certificate, no final written decision, no statutory disclaimer on the record, and no reexamination certificate narrowing any claim. The claim set as issued on 2003-11-18 is the claim set in force today: independent claims 1 (method), 12 (system), 23 (method), 28 (system), and 33 (means-plus-function system), plus the dependent claims (2–11, 13–22, 24–27, 29–32) that track them. Because no claim has been canceled or confirmed, there is also no claim-level PTAB precedent — no adopted construction of "translation loop," "frequency specific modulated signal," "first time period / second time period," or "secondary phase-error correction loop" that you could cite or distinguish. That cuts both ways: you get no free invalidity, but the patent owner gets no favorable construction either.

Termination / expiration — the decisive fact. The structured record shows Adjusted expiration 2021-12-27, with legal status "Expired - Lifetime." The 20-year term from the 2000-08-30 filing date would have run to 2020-08-30; the extra ~16 months reflects PTA. Practically, this means: (i) the patent cannot be asserted for any conduct occurring after 2021-12-27, so any accused product that launched after that date is a non-issue for utility infringement; (ii) for pre-expiration conduct, damages are capped by the six-year lookback of 35 U.S.C. § 286, which reaches back only to roughly 2015-09 / 2020-09-21 (depending on the filing date of the complaint) — the recoverable window is at most a few years and shrinks every day; (iii) no injunction is available (eBay and simple mootness) and an ITC exclusion order under § 337 is off the table because there is no domestic industry founded on an expired patent; and (iv) enhanced damages/willfulness theories lose their bite. Any demand letter citing US 6650875 today is, in substance, a request to negotiate over expired-window back royalties.

Estoppel landscape. No petitioner has ever filed against this patent, so § 315(e)(2) estoppel does not exist for anyone, and the § 315(b) one-year bar has never started running for anyone (it is triggered only by service of a complaint alleging infringement of this patent — check your own file; a complaint served on a customer or distributor can start the clock for a privy). A defendant facing assertion therefore has the full prior-art universe available: the 15 references cited on the face of the patent during prosecution (including Hughes Aircraft's US 4,600,892, Victoria University's US 5,420,536, Spectrian's US 5,570,063, Motorola's US 5,880,633, Ericsson's US 6,295,442 for AM-to-PM cancellation, Nokia's US 6,480,704, and Tropian's US 6,377,784), plus the 19 documents citing it — several of which are highly relevant Skyworks-family continuations (US 6,792,282; US 6,952,569; US 7,218,951; US 7,787,570) and competitor polar-transmitter art (Infineon's US 7,427,903 "Polar modulator and a use thereof"; Quintic's US 8,301,086 "Low-power polar transmitter"). Because so much of the field's post-2003 art describes the same translation-loop-with-power-amplifier-feedback architecture, obviousness-type art under § 103 is likely abundant. The § 112 angle is also untouched — the specification's dependence on a "variable gain element" that "unmodulates" the AM component (col. discussion of variable gain elements 306/312 operating on an RF or baseband signal) invites a written-description/enablement attack on the breadth of "compensate for any phase variation."

Pattern signals. There is no multi-petition petitioner, no defensive aggregator, and no PTAB appeal history for this patent. Skyworks (as patent owner) has defended IPRs on its more recent RF front-end patents (Kinetic 2014; Kangxi 2024–2025), which shows the company litigates its portfolio aggressively and will defend claims rather than fold — but 6650875 is a 2003-issued, now-expired patent outside that active campaign. No Unified Patents, RPX, or other aggregator challenge was found. On the other side, Skyworks as plaintiff is currently on the back foot in the Kangxi matter (ITC Administrative Law Judge initial determination of non-infringement on US 9,917,563 and US 8,717,101, reported 2026-01-23/24, with the investigation terminated on joint withdrawal / settlement on 2026-04-24 and the parallel C.D. Cal. action dismissed) — evidence that even the patent owner's recent patents are vulnerable at the ITC, though it says nothing directly about 6650875.

Recommended next steps

1. Confirm expiration and the § 286 window first. Pull the patent's PTAB E2E / Patent Center file wrapper and the PTA calculation to verify the 2021-12-27 adjusted expiration independently. If it holds, your first response to any assertion should be a dated letter: no post-2021-12-27 acts are actionable, no injunction, no ITC relief, and damages limited to pre-expiration sales in the six-year lookback. That framing usually converts a "cancellation" posture into a nuisance-value posture.
2. Because there is no FWD to cite, build the invalidity record yourself. There is no PTAB decision to link here — the honest answer is that the record is empty. If you need an IPR, note that the Board can institute on an expired patent (claims cannot be amended, so it proceeds like ordinary litigation under Phillips), but weigh that against the near-zero commercial value of the remaining damages window. A § 282 district-court invalidity case may be the cheaper path.
3. Mine the citation corpus before drafting. Start with the 15 prosecution-cited references (particularly Ericsson US 6,295,442 for AM-to-PM cancellation and Motorola US 5,880,633 on high-efficiency amplification) plus the 19 citing documents, and look especially at the Skyworks continuations US 6,792,282 and US 6,952,569 — same inventors, same architecture, and the intervening filings may themselves supply § 102/§ 103 art or admissions about the scope of the "translation loop" teachings.
4. Check your § 315(b) clock and privity. If you or an indemnitee/customer were served with a complaint asserting 6650875, the one-year IPR bar is running for privies. No prior petitioner means no estoppel, but the bar is independent of estoppel.
5. Document the negative result. Save the ODP "no proceedings" output and the search record to the file. In a motion to stay or in settlement negotiations, "this patent has never been challenged at the PTAB and has now expired" is a materially stronger posture than it first appears — but be candid that the absence of IPRs here is explained by the patent's age and expiration, not by any demonstrated PTAB-proven strength.

I'll research the assignment chain for US 6650875 before writing up the analysis.

Let me dig into the specific assignment records and correspondents of record.

Let me try to pin down actual reel/frame numbers and the recorded correspondents.

I ran the available lookups (Google Patents legal-events mirror, Plainsite's USPTO assignment index, Skyworks SEC filings, and general web search). One important limitation up front, because it affects how you should read the tables below:

Reel/frame numbers for this patent are not retrievable from the sources available to me in this session. Google Patents' legal-events mirror (which is what the patent text you supplied carries) reproduces the conveyance type and parties of each USPTO recording but not the reel/frame. I could not open the Assignment Center record detail directly. I therefore do not print invented reel/frame numbers. Everything below is grounded in the Google Patents legal-events block you supplied plus the corresponding USPTO-sourced assignment records surfaced through Plainsite. Verify each entry at https://assignmentcenter.uspto.gov/ (search "6650875" or app. 09/650,831) before citing reel/frame in any filing.

Inventors

Pattern note — no fire-sale precursor. All four inventors assigned to Conexant on the filing date (2000-08-30), and there is no evidence of the inventors departing the original assignee en masse. To the contrary, the same inventors show up on later Skyworks-family filings (e.g. Damgaard on US 6,526,265 / US 20030139148A1, priority 1999-09-14; Rozenblit on US 20070014381A1, priority 2005-07-13). The inventive team rode the Conexant → Washington Sub → Alpha → Skyworks chain rather than exiting it — the opposite of the "all inventors gone within 12 months" tell.

The USPTO record spells one assignor "MORTON DAMGAARD" on the 2000-08-30 assignment while the patent names "Morten Damgaard." Per your instruction not to auto-correct, I flag the discrepancy as recorded rather than treat it as the same identifier.

Original assignee

Two different "original assignees" depending on which you mean, and the distinction matters here:

• At filing (2000-08-30): Conexant Systems, Inc. (Newport Beach, CA). Conexant was the Rockwell International semiconductor spin-off — an operating semiconductor company, not a holding vehicle. The claimed subject matter (a translation-loop upconverter with a secondary phase-error correction loop feeding a saturated power amplifier) sits squarely in Conexant's wireless/RF product line, which shipped GSM/EDGE-era handset RF front ends. Current status: the wireless business was divested in 2002 (below); Conexant itself later restructured through a Chapter 11 proceeding in 2013 and its remaining business was acquired by Synaptics in 2017. (Confidence: high on the 2002 divestiture, moderate on the post-2013 corporate history — treat the latter as context, not a court-grade citation.)
• Named on the issued patent (2003-11-18): Skyworks Solutions, Inc. Google Patents records the original assignee of the issued patent as Skyworks Solutions, Inc. That is consistent with the timeline: Alpha Industries, Inc. was the surviving corporation in the 2002-06-25 merger and was renamed Skyworks Solutions, Inc. roughly five months before this patent issued. Skyworks is an operating RF semiconductor company (Nasdaq: SWKS) that ships power amplifiers, front-end modules and transceivers embodying exactly this class of circuitry. Current status: operating.

The intermediate entities were not products companies: Washington Sub, Inc. was a Delaware wholly-owned Conexant subsidiary created to hold the spun-off wireless business (its separate existence terminated at the merger), and Alpha Industries, Inc. (Woburn, MA) was an operating RF/microwave semiconductor maker that became the surviving, renamed entity.

Assignment timeline

Recorded chain, as surfaced in the patent's legal-events block. Dates are the dates recorded in that block (Google Patents does not distinguish execution vs. recordation for these entries; the 2000-08-30 entry coincides with the filing date, which is typical of a filing-date assignment).

• 2000-08-30 / recorded 2000-08-30 — Reel/frame not retrieved

  • Conveyance: Assignment of assignors' interest
  • Assignor: Rozenblit, Domino, Damgaard (rec. "Morton"), Vakilian
  • Assignee: Conexant Systems, Inc.
  • Correspondent: not retrieved
  • Context: Ordinary inventor-to-employer assignment at filing — acquisition of rights by the operating employer.

• 2002-08-20 / recorded 2002-08-20 — Reel/frame not retrieved

  • Conveyance: Assignment of assignors' interest (internal transfer)
  • Assignor: Conexant Systems, Inc.
  • Assignee: Washington Sub, Inc.
  • Correspondent: not retrieved
  • Context: Internal corporate reorg — pre-merger contribution of the Conexant wireless business into a wholly-owned Delaware subsidiary ahead of the spin-off and merger.

• 2002-08-26 / recorded 2002-08-26 — Reel/frame not retrieved

  • Conveyance: Merger
  • Assignor: Washington Sub, Inc.
  • Assignee: Alpha Industries, Inc.
  • Correspondent: see note below (Horstemeyer / Thomas Horstemeyer LLP appears as correspondent on the related bulk Washington Sub→Alpha recording; not confirmed for this patent)
  • Context: Internal reorg / business combination — Washington Sub merged into Alpha, Alpha surviving (effective 2002-06-25 per the SEC-filed Agreement and Plan of Reorganization dated 2001-12-16).

• 2002-08-26 / recorded 2002-08-26 — Reel/frame not retrieved

  • Conveyance: Merger (the rename component)
  • Assignor: Alpha Industries, Inc.
  • Assignee: Skyworks Solutions, Inc.
  • Correspondent: not retrieved
  • Context: Change of name only, in substance — Alpha was renamed Skyworks Solutions, Inc. following the merger; the "assignee" is the same legal entity.

• 2002-09-16 / recorded 2002-09-16 — Reel/frame not retrieved

  • Conveyance: Security Interest (granted, not a transfer of title)
  • Assignor: Alpha Industries, Inc.
  • Assignee / secured party: Conexant Systems, Inc.
  • Correspondent: not retrieved
  • Context: Securitization / financing — corresponds to the $150M note Conexant took for the Mexicali operations and related loan facility disclosed in Skyworks' FY2003 10-K. A lien, not an ownership change.

• 2003-10-06 / recorded 2003-10-06 — Reel/frame not retrieved

  • Conveyance: Release and reconveyance of security interest
  • Assignor: Conexant Systems, Inc.
  • Assignee: Alpha Industries, Inc. (i.e. the secured party releases back to the obligor, Skyworks)
  • Correspondent: not retrieved
  • Context: Securitization payoff — lien released after the note was retired from the Junior Notes proceeds.

Correspondent-of-record note (signal 3). The only correspondent I could actually identify anywhere in this chain is Scott A. Horstemeyer, Thomas Horstemeyer, LLP, 100 Galleria Parkway, Suite 1750, Atlanta, GA 30339-5948, who appears as correspondent on the USPTO recording described as "Patent Assignment from Washington Sub, Inc. to Alpha Industries, Inc." (Plainsite's mirror of that assignment record: https://www.plainsite.org/patents/assignment.html?id=[2400591](/patent/2400591)). Two cautions: (i) the property list on that mirror does not visibly include US 6,650,875 (it is a large bulk recording covering applications such as 09/823,314, 09/704,930, 09/971,523), so I cannot confirm this correspondent covers this patent; and (ii) a single appearance is not a finding under your rule. Thomas Horstemeyer LLP is a general patent prosecution firm, and Alpha/Skyworks used Woburn- and Atlanta-area counsel routinely. There is no recurrence of a single correspondent across multiple links in this chain, and I found no Horstemeyer appearance on any RPX / Unified Patents NPE-assertion list. Do not treat this as a troll-lawyer tell.

Timeline diagram

timeline
    title Ownership of US 6650875
    2000 : Filed by Conexant Systems Inc
         : Inventors assign to Conexant
    2002 : Wireless unit moved to Washington Sub
         : Washington Sub merges into Alpha
         : Alpha renamed Skyworks Solutions
    2002 : Conexant security interest recorded
    2003 : Security interest released
         : Patent issues to Skyworks
    2021 : Patent expires

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The only intermediate entity that even resembles a special-purpose vehicle is Washington Sub, Inc., and it is not a licensing shell: it was a Delaware wholly-owned operating subsidiary formed for a taxable Section 355 spin-off of Conexant's wireless business, and its separate corporate existence terminated in the merger (2002-08-26 entries). No "IP / Patents / Licensing / Holdings / Ventures" suffix appears anywhere in the chain; every named entity had commercial products.

2. Known asserter in the chain — not present. None of the recorded assignees (Conexant Systems, Inc.; Washington Sub, Inc.; Alpha Industries, Inc.; Skyworks Solutions, Inc.) appears on the Acacia / Marathon / IV / IPNav / Wi-LAN / Mosaid-Conversant / Vringo / Pendrell / Innovatio / MPHJ / Lumen View / Round Rock / DGC / Spangenberg lists. The two most recent post-issuance events on the patent are Skyworks' own follow-on filings (e.g. US 10,826,570, 2020), i.e. the operating owner continuing to prosecute in the same family.

3. Repeat correspondent across the chain — unclear (single appearance only). See the correspondent note above: one identifiable correspondent, Scott A. Horstemeyer / Thomas Horstemeyer LLP, on an adjacent bulk recording, not confirmed for this patent and not recurring. Marking unclear because the correspondents for the six entries above were not retrievable, not because any recurrence was observed.

4. Cascading transfers — not present (form without substance). Three recordings land within ~6 days of each other in August 2002, which superficially looks like a cascade. It is not: they are the constituent steps of one previously-disclosed reorganization (contribution to Washington Sub → merger into Alpha → Alpha renamed Skyworks) executed under a single Agreement and Plan of Reorganization dated 2001-12-16 and disclosed in Skyworks' SEC filings. The endpoints are operating companies; there is no chained-LLC sequence, no shared correspondent address across unrelated assignees, and no common-principal shell structure.

5. Pre-litigation transfer — not present. The last title-affecting recording is 2002-08-26 (with only a lien and its release in 2002-09-16 and 2003-10-06), years before any date I could associate with assertion. I found no infringement suit naming US 6,650,875.

6. Bankruptcy fire-sale — not present for the original assignee. Conexant's later financial distress post-dates this patent's 2002 divestiture by more than a decade; this patent was not sold in any Conexant estate proceeding that I could identify.

7. Privateering — not present. The Conexant→Washington Sub→Alpha→Skyworks sequence is an operating-company merger, not a transfer to an assertion vehicle acting on a sponsor's behalf. No SEC filing, Patent Progress or EFF coverage links any chain participant to a privateering arrangement on this patent.

8. Defensive aggregator — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents or OIN. It terminates at Skyworks Solutions, Inc., an operating semiconductor company.

Verdict

Operating-company assertion — with the express caveat that the "assertion" prong is unverified: I found no litigation naming this patent, and the category is being used here as the non-NPE bucket rather than as a claim of suit activity.

Justification: every link in the recorded chain (2000-08-30 assignment to Conexant; 2002-08-20 to Washington Sub; 2002-08-26 merger into Alpha; 2002-08-26 rename to Skyworks) moves the patent between operating companies, and the only two subsequent recordings are a security interest and its release (2002-09-16; 2003-10-06) — securitization mechanics, not a title transfer to an asserter. The current owner, Skyworks Solutions, Inc., is an operating RF semiconductor maker whose products practice this subject matter, and zero of the eight NPE signals is present (one is only unclear for want of correspondent data). Because the reel/frame numbers could not be retrieved in this session, treat the ownership conclusion as well-supported but the citations as needing verification at https://assignmentcenter.uspto.gov/ (search 6650875) — do not carry the "not retrieved" placeholders into any filing.

I'll verify the cited references and their content before building the anticipation analysis. Let me run several targeted searches.

Let me verify the remaining cited references.

Prior-Art Analysis — U.S. Patent No. 6,650,875 B1

Scope note (literal-ID compliance): This report analyzes U.S. Patent No. 6,650,875 B1 only. I have not substituted, harmonized, or auto-corrected any identifier. Numbers that resemble but are not this patent (e.g., JP 6650875 B2, a Koninklijke Philips consumable-recognition patent; CN 102356543 B) are excluded, as established in the previously generated sections.

USPTO-check caveat: I could not complete a fresh live USPTO PatentCenter pull within this session's step budget. The reference list below is taken from the authoritative full patent text supplied (the "Patent Citations (15)" set from the Google Patents record at https://patents.google.com/patent/[US6650875B1](/patent/US6650875B1)/en). I verified the content of four of the fifteen (US 6,295,442; US 6,420,940; US 6,480,704; US 6,526,265) plus US 5,825,257 via web retrieval; the remaining descriptions rest on the citation titles, the '875 specification's own characterization of the art, and my knowledge of these references. I flag where my confidence is lower.

1. The 15 cited references — complete inventory

All fifteen have effective prior-art dates before the '875 priority date of 2000-08-30, so each is available under at least 35 U.S.C. § 102(a)/(b)/(e) as of the 2000 filing. Dates below are as recorded in the Google Patents citation data.

(Note: items 11, 13, 14, 15 are declared above as "verified." Note also that items 13–15 are 2002–2003 grants but with pre-2000 priority — they are prior art by their priority/PCT-publication dates, not merely their U.S. grant dates.)

Overarching § 102 conclusion up front: None of these fifteen references, taken individually, discloses every element of any of the five independent claims (1, 12, 23, 28, 33). The reason is structural: every independent claim requires the concurrence of (i) a translation loop (mixer + local oscillator + phase detector + VCO) and (ii) a second, PA-derived feedback path applied to that same loop, with the loop output and the PA output detected and applied during first and second time periods, plus (iii) a phase detector → phase shifter correction of the input modulated signal. No single cited reference teaches all three. The references therefore function primarily as § 103 material, which is consistent with the patent's own primary classification, H03F 1/3241 — "reducing non-linear distortion using predistortion circuits with feedback acting on the predistortion circuits."

Below, each reference is analyzed for what it can reach under § 102, with the missing-claim-element gaps stated explicitly.

2. Reference-by-reference § 102 analysis

2.1 US 6,480,704 B1 — Nokia — the closest reference (highest § 103 relevance)

• Full citation: U.S. Patent No. 6,480,704 B1, "Method and arrangement for forming a transmission signal," appl. (FI priority) 1998-10-27, issued 2002-11-12, Nokia Mobile Phones Ltd.
• Brief description (verified): Discloses forming an amplitude+phase modulated signal in which baseband A and φ control signals are created; the carrier frequency signal is amplified; and, per claim 1, "the phase of the carrier frequency signal also is controlled on the basis of the carrier frequency signal after the last amplification step in order to correct the transmission signal phase," with amplitude likewise corrected from the post-amplification sample. In the FIG. 2 embodiment, a directional coupler (6) samples the RF-amplifier output; the sample is divided (11) and compared (12) against a reference f_ref (13) to control the VCO; the description states "the RF amplifier thus constitutes a part of the phase-locked loop, and thus the phase nonlinearity of the amplifier is corrected." Claim 4's alternative forms a phase modulation signal from the reference and the preliminary phase control signal φ, then compares phases.
• Which claims it potentially anticipates under § 102: It is the only cited reference that reaches the core "correct phase from the PA output" idea, and it is the most plausible § 102 candidate against claims 23 and 28 (and their dependent claims 24–27, 29–32) and claim 33 (means-plus-function). It reaches claims 12/1 only weakly.
• Why it does not fully anticipate (missing elements):
  1. No "translation loop" as claimed. '704 compares the divided PA sample to a fixed reference oscillator (f_ref 13), not to a modulated IF carrier that is mixed with an LO to set the transmit frequency. The claimed "translation loop" (the '875 mixer 212 + synthesizer 142 + TX VCO 207 + phase detector 201 structure) is absent. In the '875, the phase comparison is made against the modulated signal itself (claim 23: "detect a phase difference between the modulated signal and the detected output power portion"); '704 compares against a reference, not the modulated signal.
  2. No first/second time periods and no dual feedback source. '704 contemplates a single PA-derived feedback path; the '875 claim requires applying the translation-loop's own output during a first time period and the PA output during a second time period (the burst-ramp-up behavior, dependent claims 8/19/24/29).
  3. No phase shifter in the input path. '704 corrects phase by steering the VCO frequency (summing unit 9 / phase comparator 21 → LPF 22 → VCO 10), not by means of a phase shifter (e.g., '875 phase shifter 220) inserted in the modulated-signal input path.
• Verdict: § 102 anticipation is not established for the independent claims; the reference is the primary § 103 reference, and it plausibly anticipates only a hypothetical broadened claim. In an IPR the '875 would distinguish on the translation-loop + input-phase-shifter + dual-time-period combination.

2.2 US 6,295,442 B1 — Ericsson (highly material; AM-to-PM cancellation + feedback-switching concept)

• Full citation: U.S. Patent No. 6,295,442 B1, "Amplitude modulation to phase modulation cancellation method in an RF amplifier," appl. 1998-12-07 (U.S. Ser. 09/207,167), issued 2001-09-25, Ericsson Inc. (inventors Camp, Schlang, Gore, Mannerstråle, per the DE 699 16 886 T2 counterpart record).
• Brief description (verified): Addresses unintended phase modulation caused by amplitude-modulating a power amplifier. It develops a phase-modulation command and an amplitude-modulation command, stores phase-correction information correlating the AM command to a phase error (correction table 34), and varies the phase-modulation command based on that phase error. Notably, its background states that the PA-within-the-PLL problem "has been addressed by including the amplitude and modulated power amplifier within the total phase locked loop," and that "the feedback point from the voltage-controlled oscillator can be shifted to the output of the power amplifier only after the loop has locked," citing US 3,900,823 (Sokal) and US 5,430,416 (Black) for that technique. Amplitude modulation is applied to the PA supply (FIG. 8 Δ-Σ modulator + Class-D stage).
• Which claims it potentially anticipates under § 102: Primarily relevant to the AM-to-PM / amplitude-reconstruction aspect and to the second-feedback-path concept. It arguably touches the "second modulated signal to the power amplifier" element (claims 1(c)/12) and the general phase-correction-from-AM idea. It does not reach an independent claim as a whole.
• Why it does not anticipate: It corrects phase open-loop via a stored correction table keyed on the AM command, rather than by detecting a phase difference between the modulated signal and the detected PA output and feeding a phase shifter (claim elements 1(g)–(i), 23). There is no variable-gain element in a PA feedback path as claimed, and no translation-loop with the claimed first-mixer/second-mixer detection.
• Verdict: No § 102 anticipation; strong § 103 secondary reference that directly addresses the "AM-to-PM conversion in the feedback path" problem the '875 solves. Important lead: the two references it cites by name — US 3,900,823 (Sokal) and US 5,430,416 (Black) — teach switching the PLL feedback from the VCO to the PA output after lock, which is conceptually the '875's first/second-time-period behavior. These are not in the '875 citation list and I have not independently verified them; flag as unverified leads worth a formal search.

2.3 US 6,526,265 B1 — Skyworks (translation-loop architecture; same inventor family)

• Full citation: U.S. Patent No. 6,526,265 B1, "Wireless transmitter having a modified translation loop architecture," appl. 1999-09-14 (U.S. Ser. 09/398,911), issued 2003-02-25, Skyworks Solutions, Inc. (inventors Morten Damgaard, Rahul Magoon — note Damgaard is also an inventor on the '875 under analysis). Later continuation: US 6,952,569 B2.
• Brief description (verified): A translation-loop transmitter: a VCO generates the transmit signal; a modulated LO source generates a modulated LO signal; a mixer mixes the transmit signal with the LO to produce an IF signal; first and second dividers produce comparison signals; a phase detector compares their phases and outputs a control voltage to the VCO. The invention's improvement is the frequency plan (f_Tx = f_LO·(KM±N)/M) that eliminates low-frequency "zero-crossing" spurs and removes the need for a quadrature mixer.
• Which claims it potentially anticipates under § 102: It supplies the "translation loop having a phase locked loop" and "first mixer" elements of claims 1(a)/12 and 23/28, and it is the same corporate family as the '875. But it reaches no independent claim in full.
• Why it does not anticipate: It contains no power amplifier feedback path at all — no sampling of the PA output, no second mixer receiving the PA output, no phase detector comparing PA feedback to the modulated signal, and no phase shifter in the input path. It is a frequency-generation architecture.
• Verdict: No § 102 anticipation. Useful as evidence of the translation-loop environment and as a § 103 reference for the "translation loop" element; also probative background that the '875 and '265 share a common assignee/inventor and a common 1999–2000 development program.

2.4 US 6,420,940 B1 — Philips (phase modulator + PLL structure)

• Full citation: U.S. Patent No. 6,420,940 B1, "Transmitter with a phase modulator and a phase locked loop," GB priority 1998-12-14; PCT/EP99/10051 filed 1999-12-13; U.S. issue 2002-07-16; Koninklijke Philips Electronics N.V. (inventors Brian J. Minnis, Pascal Philippe).
• Brief description (verified): A transmitter with a phase modulator and a PLL "having a relatively high powered VCO." The PLL's phase-sensitive detector compares a phase-comparison frequency derived from the VCO output with a phase-modulated IF carrier derived from the phase modulator. The phase modulator uses four quadrature phase components of a reference and random-interpolation phase selection (sigma-delta).
• Which claims it potentially anticipates under § 102: Structurally analogous to the '875's first (translation) loop: a modulated IF reference compared against a divided/frequency-derived VCO signal, with the upconverted result fed to a power stage. It thus touches claims 1(a)/12 and 23/28's "translation loop … providing a frequency specific modulated signal" and the phase-detector portion.
• Why it does not anticipate: No power amplifier (the high-power VCO drives the output directly), therefore no PA-derived feedback, no second mixer, no first/second time periods, no PA phase-error correction, and no input phase shifter driven by a PA-feedback phase detector.
• Verdict: No § 102 anticipation; § 103 reference for the translation-loop / phase-comparison structure.

2.5 US 5,825,257 A — TRL (VCO-output sampling + phase-difference detection)

• Full citation: U.S. Patent No. 5,825,257 A, "GMSK modulator formed of PLL to which continuous phase modulated signal is applied," appl. 1997-06-17 (Ser. 08/777,772), issued 1998-10-20, Telecommunications Research Laboratories (Klymyshyn, Kumar, Mohammadi).
• Brief description (verified): A direct-GMSK modulator: a continuous-phase-modulated reference signal drives a PLL; a coupler (19) samples the VCO output and feeds a phase-difference detector (15) whose other input is the continuous-phase-modulated signal; the error signal drives the high-power VCO control port. The loop also provides Gaussian spectral shaping.
• Which claims it potentially anticipates under § 102: It squarely meets the "detect a portion of an output of the translation loop" element (claims 1(d), 23) and the phase-detector-comparing-modulated-signal-to-loop-output step, since it compares the CPFSK reference to the sampled VCO output.
• Why it does not anticipate: No power amplifier and no PA-output feedback. The loop closes only on the VCO. There is no second mixer, no second time period, no PA-phase correction, and no PA-derived phase shifter control. (Note also that its "phase shifter" is the modulator's own fractional CPM phase shifter, not a correction element driven by PA feedback.)
• Verdict: No § 102 anticipation; a good § 103 reference for the loop-feedback-detection and phase-comparison elements of claims 23/28.

2.6 US 6,377,784 B2 — Tropian (polar / high-efficiency modulated amplifier)

• Full citation: U.S. Patent No. 6,377,784 B2, "High-efficiency modulation RF amplifier," appl./priority 1999-02-09, issued 2002-04-23, Tropian, Inc.
• Brief description (moderate confidence; title is self-describing and consistent with the '875's art discussion): A high-efficiency modulated RF amplifier in the envelope-restoration / polar family, in which phase modulation and amplitude modulation are separated and the AM is applied to the amplifier (supply/control) to keep the amplifier efficient.
• Which claims it potentially anticipates under § 102: Relevant to the "second modulated signal to the power amplifier" element (claims 1(c)/12; dependent claims 5/16 — AM signal) and to the AM-to-PM problem. It does not disclose the translation-loop-plus-PA-feedback architecture.
• Why it does not anticipate: No translation loop with the claimed dual feedback, no first/second time periods, no PA-output phase detector driving an input phase shifter.
• Verdict: No § 102 anticipation; § 103 reference for the polar AM-injection element.

2.7 US 5,880,633 A — Motorola (high-efficiency PA / EER)

• Full citation: U.S. Patent No. 5,880,633 A, "High efficiency power amplifier," priority 1997-05-08, issued 1999-03-09, Motorola, Inc.
• Brief description (moderate confidence): A high-efficiency amplifier arrangement in the envelope-elimination-and-restoration (Kahn/EER) tradition — phase-modulated carrier through a saturated/non-linear amplifier with the envelope applied separately — addressing AM/PM interaction.
• Which claims: Potentially touches claims 1(c)/5 (second/AM signal to the PA) and the general "non-linear but efficient amplifier + separate AM path" premise. No translation loop, no PA-feedback phase correction, no time-period switching.
• Verdict: No § 102 anticipation; § 103 background. Note the overlap concern: US 5,880,633 and US 6,377,784 teach essentially the same EER premise, and either could be argued into a § 103 combination with the Nokia '704 or Ericsson '442 references.

2.8 US 6,236,267 B1 — IBM · US 6,259,318 B1 — Motorola · US 6,275,685 B1 — Nortel · US 5,420,536 A — Victoria University · US 5,570,063 A — Spectrian (linearization / predistortion cluster)

• US 6,236,267 B1 (IBM; priority 1998-12-29; issued 2001-05-22) — Linearization for power amplifiers using feed-forward and feedback control. Feed-forward + feedback linearization.

• US 6,259,318 B1 (Motorola; priority 1999-05-28; issued 2001-07-10) — Method for extending the linear range of an amplifier (predistortion/gain-expansion technique).

• US 6,275,685 B1 (Nortel Networks; priority 1998-12-10; issued 2001-08-14) — Linear amplifier arrangement (feedback/Cartesian-style linearization).

• US 5,420,536 A (Victoria University of Technology; priority 1993-03-16; issued 1995-05-30) — Linearized power amplifier.

• US 5,570,063 A (Spectrian, Inc.; priority 1995-05-18; issued 1996-10-29) — RF power amplifier with signal predistortion for improved linearity.

• Descriptions (moderate confidence, derived from titles and the '875's characterization): These form the generic linearization-by-predistortion-and/or-feedback cluster. They teach correcting amplifier nonlinearity by monitoring amplifier behavior and adjusting an input, but at baseband/predistortion level, without a translation-loop upconverter and without the polar dual-feedback structure.

• Which claims they potentially anticipate under § 102: None of the independent claims. At most they touch the generic "close a correction loop around the power amplifier" concept underlying claim 11/22/27/32 ("second feedback loop").

• Why they do not anticipate: No translation loop; no phase detector comparing a modulated IF signal to the PA output; no phase shifter in the input path controlled by that comparison; no first/second time-period dual feedback.

• Verdict: No § 102 anticipation. These are the references that map most cleanly to the patent's H03F 1/3241 / 1/3247 classification (predistortion with feedback acting on the predistortion circuits) and are § 103 material for the general predistortion/feedback element.

2.9 US 4,600,892 A — Hughes · US 5,737,703 A — Nokia · US 5,740,521 A — Nokia (general/background)

• US 4,600,892 A (Hughes Aircraft; priority 1984-06-28; issued 1986-07-15) — RF amplifier with frequency spectrum control. Very early amplifier-spectrum-shaping art; the oldest cited reference.
• US 5,737,703 A (Nokia Mobile Phones; priority 1994-12-23; issued 1998-04-07) — Multi-mode radio telephone which executes handover between different systems. A multimode-handset/handover patent, not an amplifier-architecture patent; cited for general "one handset for multiple systems" context (relevant to the '875's EDGE/multimode motivation).
• US 5,740,521 A (Nokia Mobile Phones; priority 1994-11-14; issued 1998-04-14) — Method and circuit for creating frequencies for a radio telephone. Per the description of US 6,480,704, this reference concerns using a phase-locked loop to correct an intermediate-frequency oscillator error; i.e., translation/IF correction frequency-generation art.
• Which claims they potentially anticipate under § 102: None. US 4,600,892 and US 5,737,703 reach no claim element of substance. US 5,740,521 is a weak § 103 candidate for the "translation loop / IF correction" element only.
• Verdict: No § 102 anticipation; background/§ 103-only relevance.

3. Ranked relevance summary

4. Key takeaway and confidence

• No single cited reference anticipates any independent claim (1, 12, 23, 28, 33) of US 6,650,875, because each independent claim requires the combination of a translation loop, a PA-derived second feedback path applied during a second time period, a phase detector comparing the modulated signal to the PA feedback, and an input-path phase shifter — a combination no cited reference shows in one document.
• The most likely basis of patentability (and of any validity challenge) is § 103: the closest combination would be US 6,480,704 (Nokia) as primary with US 6,295,442 (Ericsson) and/or US 6,526,265 (Skyworks)/US 6,420,940 (Philips) as secondary references, against the dual-feedback/time-period and translation-loop limitations.
• Verification caveat: I verified four references directly (11, 13, 14, 15) plus US 5,825,257; the remainder are characterized from the authoritative citation list, the '875 specification, and general knowledge. I was unable to complete a full live USPTO pull for all fifteen within this session, so dates and descriptions for items 1–10 and 12 should be confirmed against the official PatentCenter records before being relied on in a filing or opinion.
• Unverified lead (worth a formal search): US 3,900,823 (Sokal) and US 5,430,416 (Black), surfaced in the Ericsson '442 background via DE 699 16 886 T2, appear to teach switching PLL feedback from the VCO to the PA output after lock — directly pertinent to the '875's first/second-time-period limitation (claims 8/19/24/29). These are not among the 15 cited references and I did not verify them directly.

I'll ground this analysis in the actual prior-art references from the patent record before constructing the § 103 combinations. Let me verify the key references.

Let me verify a few more of the cited references that are likely central to specific claim limitations.

Obviousness Analysis — U.S. Patent No. 6,650,875 under 35 U.S.C. § 103

Builds on the prior Patent summary and Litigation summary sections. No contradictions with those sections were found; the claim overview and the cited-reference list there are used as the starting point and not repeated.

One cross-reference correction to flag: the earlier Litigation summary described US 6,754,471 as a "family relative." That number does not appear in the record for this patent; the family relatives actually on the record are US 6,952,569 B2 and US 2003/0139148 A1 (Damgaard, "Wireless transmitter having a modified translation loop architecture") plus US 6,792,282. I flag the earlier reference as unsupported rather than harmonizing it.

1. Governing standard and framing

• Statute: Because the '875 patent's filing/priority date is 2000-08-30 — before the 2013-03-13 effective date of the AIA first-inventor-to-file provisions — pre-AIA § 103(a) governs.
• Framework: Graham v. John Deere Co., 383 U.S. 1 (1966) (scope/content of prior art; differences; PHOSITA level; secondary considerations) and KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) (a combination is obvious where the improvement is a predictable use of prior-art elements according to known methods, or is an "obvious to try" variation on a finite set of identified, predictable solutions, or where a design need / market pressure supplies the impetus).
• PHOSITA: a B.S. in electrical engineering (or equivalent) with ≈3–5 years in RF/wireless transmitter design, familiar with polar/PM+AM transmitters, translation-loop (offset-PLL) upconverters, PLL loop dynamics, burst-mode (GSM/EDGE) operation, and the AM-to-PM phenomenon in non-linear PAs.
• Procedural posture: all 15 references below were before the examiner and are cited on the face of the '875 patent. Obviousness therefore must rest on showing a combination the examiner did not appreciate (or a motivation the examiner discounted), not on newly discovered art. That is a meaningful hurdle and is discussed in § 7.

Claim-construction note that materially affects scope: independent claim 33 is a classic means-plus-function claim (pre-AIA § 112 ¶ 6). Its "means" are limited to the corresponding structures disclosed — phase detectors 201/209, mixers 212/214, phase shifter 220, TX VCO 207, variable gain element 312/306, etc. — and equivalents. That narrows claim 33 relative to claims 1/12/23/28, which are structural/step claims not so limited.

2. § 102 prior-art qualification of the cited references

All of the following have a filing date or § 102(e) date before 2000-08-30, so all qualify as prior art (basis in parentheses). Where a reference issued after the '875 filing, it is § 102(e)-only art.

Caveat on US 6,526,265 (important). This reference is commonly owned with the '875 (both Skyworks family; Damgaard is a named inventor on both, and the '875 filed from Conexant's wireless/RF group that merged into Skyworks). Pre-AIA § 103(c) disqualifies subject matter that qualifies as prior art only under § 102(e)/(f)/(g) where, at the time the invention was made, the subject matter and the claimed invention were commonly owned or subject to an obligation of assignment to the same person. If both applications were commonly owned in August 2000, US 6,526,265 cannot be used in a § 103 combination. Because I cannot establish the 2000 ownership record with confidence from the sources available, I treat '265 as a corroborating reference only and build the primary combination on third-party art. This is a genuine weakness in any IPR-style challenge that leans on '265.

3. What each reference discloses (verified where the text was retrieved)

3a. US 6,295,442 (Ericsson) — the pivotal reference

Verified from the Google Patents/Justia/SumoBrain texts. It discloses a PM+AM ("polar") transmitter in which:

• a phase-modulation command and a separate amplitude-modulation command are developed (blocks 22, 24);
• the phase command drives a phase-modulated oscillator 28, expressly described as "a voltage-controlled oscillator (VCO) in a phase-locked loop (PLL), whereby either the reference signal is phase modulated with a complex vector (I/Q) modulator, or the phase modulation is impressed within the loop";
• the resulting RF input signal on line 30 is amplified by power amplifier 32, and the AM command is applied to the PA control input (block 24 → PA 32) — i.e., the exact polar split of '875 claims 4/5/16;
• a phase-correction block 34 corrects the phase command based on the amplitude command to fix AM-induced phase error (AM-to-PM);
• critically for the "first/second time period" limitation, the Background states: "This problem has been addressed by including the amplitude and modulated power amplifier within the total phase locked loop. … The feedback point from the voltage-controlled oscillator can be shifted to the output of the power amplifier only after the loop has locked and is on frequency."

That passage is a direct description of the '875's claim-1 steps (e)–(f): translation-loop-output feedback during a first period; PA-output feedback during a second period. Ericsson also identifies the drawback ('875's motivation for the overlap of claims 8/19/24/29): "This shift can cause disruptions of the loop and cause the loop to lose lock."

• Ericsson further discloses closed-loop PA phase correction: claims 19–21 recite "determining actual phase modulation of the RF output signal and … comparing the actual phase modulation to the phase modulation error to vary the phase modulation command" — i.e., sense the PA output phase and feed back to correct the phase command ('875 claim-1 steps (g)–(i)).
• And PA-output power sensing: detector 38 measures "the amplitude or power level at the output of the power amplifier" (FIG. 3), and claim 24 recites "monitoring amplitude of the RF output signal."

3b. US 5,825,257 (TRL) — the translation-loop / modulated-reference teaching

Verified. Discloses a PLL (phase detector + loop filter + high-power VCO) to which a continuous-phase-modulated signal is applied as reference, with an optional frequency divider in the feedback path, the PLL providing frequency tracking plus spectral shaping of the modulated output. This is the classic translation-loop = PLL receiving a modulated signal and producing a frequency-specific modulated output structure of claim 1 step (a). It also discloses a "360° linear phase shifter" / fractional CPM modulator 13 controlled by a control voltage 6 — a voltage-controlled phase shifter on the modulation input, which reads on '875's phase shifter 220.

3c. US 6,420,940 (Philips) — translation loop + high-power VCO + PA

Verified. Discloses a phase modulator producing a modulated IF carrier applied to a transmit loop (PLL) with a relatively high-powered VCO 16, a phase-sensitive detector comparing a VCO-derived signal to the phase-modulated IF carrier, and the VCO output coupled to a power amplifier 18 and then the antenna. This is, in substance, the '875 FIG. 2 front end (phase shifter → phase detector 201 → VCO 207 → PA 134) without the PA-output secondary correction loop.

3d. US 6,377,784 (Tropian / McCune) — AM injection into the PA for efficiency

Verified. Discloses high-efficiency amplitude modulation of a hard-limiting/switch-mode PA by varying the PA's operating voltage/control input, i.e., AM applied to the PA control port while phase is carried on the RF input (polar). Abstract: "control the operating voltage of the power amplifier with sufficient bandwidth to faithfully reproduce the desired amplitude modulation waveform." It expressly recites "an amplitude varying circuit 203 such as a variable gain amplifier (VGA) or an attenuator, controlled in accordance with a power level signal" — read directly on '875 claims 2/3/13/14 and step (c) of claim 1.

3e. US 6,526,265 (Skyworks) — translation loop with modulated LO source

Verified. Discloses a VCO producing transmit signal f_Tx, a modulated signal source producing modulated LO f_LO, a mixer mixing transmit and LO to an IF, dividers M and N, and a phase detector comparing the divided signals and driving the VCO — a translation loop with a PLL and a modulated input (claim-1 preamble). Corroborative only, per the § 103(c) caveat.

3f. Lower-confidence references (titles/abstracts only)

• US 5,880,633 (Motorola, "High efficiency power amplifier") — efficiency/linearity techniques; peripheral.
• US 6,236,267 (IBM, "Linearization for power amplifiers using feed-forward and feedback control"); US 6,275,685 (Nortel, "Linear amplifier arrangement"); US 6,259,318 (Motorola); US 5,570,063 (Spectrian, predistortion); US 5,420,536 (Victoria Univ., linearized PA); US 4,600,892 (Hughes) — all teach feedback/feed-forward linearization of a PA, i.e., the general concept of sensing PA output and correcting distortion.
• US 6,480,704 (Nokia, "forming a transmission signal") and US 5,740,521 (Nokia) — upconversion/frequency-plan architecture; US 5,737,703 (Nokia) is multi-mode handset control and is essentially non-informative for these claims.

Honest uncertainty flag: I retrieved full text for US 6,295,442, US 5,825,257, US 6,420,940, US 6,377,784, and US 6,526,265. For US 5,880,633, US 6,236,267, US 6,275,685, US 6,259,318, US 5,570,063, US 5,420,536, US 4,600,892, US 6,480,704, US 5,740,521, and US 5,737,703 I am working from title/abstract/classification only. Column-level pin cites for those references should be verified before filing.

4. Claim-by-claim obviousness mapping

Combination 1 (primary): Ericsson '442 + TRL '257 + Tropian '784

Motivation is express, not speculative: '442 itself frames the problem (AM-PM distortion in a PM+AM transmitter; PA inside the PLL; feedback handoff) that '257 and '784 solve mechanically.

Claim 1 (method):

Claim 12 / Claim 28 (systems): same mapping in structural terms — first mixer = '257/'940 PLL feedback mixer; second mixer = '442's PA-output sense path downconverted for phase comparison; phase detector = '257/'940 phase-sensitive detector; phase shifter = '257 shifter 13; variable gain element = '784 VGA/attenuator. Note the literal antecedent quirk already flagged in the earlier summary (claim 28 recites "the first modulated signal" against an antecedent of "a modulated signal") — a § 112 defect, not an obviousness point.

Claim 23 / Claim 33 (generalized method / means-plus-function): same as claim 1, minus step (c). These are easier to invalidate than claim 1, because the only element arguably added by the AM injection is removed. The '442 + '257 combination alone maps claims 23/28/33 nearly element-for-element — in fact, US 6,295,442 standing alone is a colorable § 102 anticipation of claims 23/28/33, since it discloses the VCO↔PA feedback handoff, PA-output phase sensing, and phase-command correction, all within a PLL-based polar transmitter. I rate that a colorable rather than clean anticipation because '442's phase comparison is expressed partly as a table-based feed-forward correction with optional monitoring, and it does not use the words "translation loop," "first/second time period," or "phase shifter"; a § 103 framing over '442 + '257 is the safer and stronger position.

Dependent-claim anchors

• Claims 2/3, 13/14 (VGA / attenuator): '784 ("amplitude varying circuit … a variable gain amplifier (VGA) or an attenuator"); the '875's own variable gain element 312 is described as "a variable gain amplifier or an attenuator."
• Claims 4, 15 (first signal is PM) / Claims 5, 16 (second is AM): '442 (separate PM command → oscillator, AM command → PA), '784, '940.
• Claims 6/7, 17/18 (variable gain element on RF vs. baseband representing RF power): both modes are known — '784 injects AM at the PA control (RF/control domain) whereas '442's amplitude-correction table operates on the baseband amplitude command; the '875 FIG. 2 element 306 operates on the baseband DC output of log detector 301 (claim 7) and FIG. 7 element 402 on an IF signal. Selecting the domain is a routine design choice with a known trade-off.
• Claims 8, 19, 24, 29 (first and second time periods overlap): the best nonobviousness candidate. But '442 expressly teaches the disruption/lock-loss hazard of the abrupt shift; a POSITA reading '442 would recognize that ramping in the PA feedback before/while removing the VCO feedback (overlap) is the natural way to avoid that hazard. Under KSR, that is an "obvious to try" narrowing of the transition with a predictable benefit.
• Claims 9, 25, 30 (second variable gain element applies PA output to the loop): '784's VGA/attenuator, plus the '875's own AM-reconstruction element (312 driven by inverse-AM on line 311; IF variable gain element 402 in FIG. 7). The generic concept of a variable gain element in the feedback path is well known (see also '442's amplitude-correction path and '685/'267 feedback linearizers).
• Claims 10/11, 21/22, 26/27, 31/32 (first and second feedback loops): inherent in any '257/'940 PLL plus '442 PA-in-loop arrangement.

5. Motivation to combine (the KSR analysis)

1. Same field, same problem. All references are in RF polar/burst-mode transmitter design. '442 and '940 are GSM-specific; '784 addresses EDGE/TDMA efficiency; '257 addresses GMSK/GSM modulation. Combining PLL-based upconversion with PA feedback linearization is the shared subject matter.
2. Express teaching in the primary reference. US 6,295,442 does not merely make the combination possible; it identifies the combination as the known solution ("This problem has been addressed by including the amplitude and modulated power amplifier within the total phase locked loop") and identifies the exact feedback-point handoff ('875 steps (e)–(f)) as an existing technique. That is a textbook TSM under pre-AIA § 103(a) and KSR.
3. Reasonable expectation of success. Each element was independently known and proven; the combination is a "predictable use of prior-art elements according to known methods" — the KSR paradigm. The '875 adds no unexpected result; the spec itself merely asserts the intended benefit (phase-error cancellation, efficiency).
4. Design need / market pressure. GSM/EDGE requirements (low modulation phase error — '940 quantifies "<5° rms and 20° peak" — plus PA efficiency and burst ramp-up) supply the impetus, per KSR.
5. Not teaching away. '442's warning that the abrupt handoff "can cause disruptions of the loop and cause the loop to lose lock" is a caution about a known implementation detail, and it simultaneously teaches the technique exists; it does not disparage the '875 solution (overlap). A caution that a known technique has a known drawback is generally not a teaching away — it is a design problem the '875 solves in the predictable way (gradual/overlapping transition).

6. Secondary considerations (Graham factor 4)

None are established on the record available to me. Forward-citation volume (the '875 is cited by ~19 publications, including Skyworks' own US 6,792,282, US 6,952,569 and Trinh's US 7,218,951) shows the disclosure sits in an active technical family, but citation is not evidence of commercial success or nexus. The patent expired 2021-12-27 (per the earlier status finding), so any present-day commercial-nexus argument is moot. Absent evidence of unexpected results tied to the overlap limitation (claims 8/19/24/29), secondary considerations do not, on this record, overcome the § 103 case for claims 1/12/23/28/33.

7. Realistic weaknesses in the § 103 position

• All references were before the examiner. Getting an IPR institution (or vacatur) requires a combination the examiner did not make. The strongest such combination is '442 (as the PA-phase-correction/handoff teaching) + '940 or '257 (translation loop) + '784 (AM injection) — none of which the record shows were combined.
• § 103(c) risk on US 6,526,265. Its value as primary art is compromised by the common-ownership disqualification (see § 2 caveat). Do not build a challenge on '265 without the ownership record.
• The "phase shifter" mapping is functional. '442 corrects phase by summing a correction into the command; '257's controlled phase shifter is the cleaner structural read on element 220. A petitioner should lead with '257 for that limitation.
• The overlap limitation (claims 8/19/24/29) is the hard target. Claims 1/12/23/28/33 are the vulnerable ones; if any claim survives, it is likely to be the overlap-dependent claims, and only if the patentee can show the overlap produces an unexpected result (none is evident in the record).
• Means-plus-function claim 33 is narrow (limited to the disclosed structures and equivalents) and thus harder to invalidate broadly than claims 1/12/23/28.
• Anticipation vs. obviousness: I would plead § 102 over US 6,295,442 for claims 23/28/33 as a secondary count, with § 103 over '442 + '257 (+ '784) as the primary count for all independent claims. The § 102 count is colorable but not clean (see § 4).

Bottom line

On the record, claims 1, 12, 23, 28 and 33 are rendered obvious by the combination of US 6,295,442 + US 5,825,257 + US 6,377,784 (with US 6,420,940 as an interchangeable translation-loop substitute for '257 and US 6,526,265 as corroboration subject to the § 103(c) caveat). The dependent claims are obvious over the same references plus the routine-design-choice references (US 6,236,267 / US 6,275,685 / US 5,570,063 for feedback linearization; US 6,377,784 for the VGA/attenuator). The only materially defensible claim family is the "overlapping first and second time periods" group (8/19/24/29), and even that is exposed under KSR because US 6,295,442 itself identifies the lock-disruption problem that the overlap predictably solves.

Caveat, restated: column-level pin cites are verified for '442, '257, '940, '784 and '265, but not for the remaining nine references, which I mapped from titles/abstracts/classifications. Confirm those before relying on this analysis in a filing.