US 9608675

I'll research this patent across the provided sources and live search.

Let me verify the litigation status and related family details with additional searches.

US Patent 9,608,675 — Analyst Summary

1. Identification

Source of record: the full text at https://patents.google.com/patent/US9608675/en (fetched 2026‑09‑26), cross-checked against FreePatentsOnline (https://www.freepatentsonline.com/9608675.html) and Espacenet's claims page for US9608675B2.

Family note (for completeness only): the same specification continues in US 11,133,833 B2 (app. 15/916,101), US 11,641,215, US 11,996,872, and US 12,562,763; foreign counterparts include EP 2 954 737 B1, EP 3 203 787 A1, CN 104982077 B, JP 6490596 B2, KR 102182001 B1.

2. Abstract (verbatim)

"Techniques for generating a power tracking supply voltage for a circuit (e.g., a power amplifier) are disclosed. The circuit may process multiple transmit signals being sent simultaneously on multiple carriers at different frequencies. In one exemplary design, an apparatus includes a power tracker and a power supply generator. The power tracker determines a power tracking signal based on inphase (I) and quadrature (Q) components of a plurality of transmit signals being sent simultaneously. The power supply generator generates a power supply voltage based on the power tracking signal. The apparatus may further include a power amplifier (PA) that amplifies a modulated radio frequency (RF) signal based on the power supply voltage and provides an output RF signal."

3. Independent claims — plain language

The specification supports envelope/power tracking of a single PA driven by multiple simultaneous (carrier-aggregated) transmit signals, with one aggregate tracking signal rather than one tracker per carrier. The independent claims read on that idea; the "inventive" arithmetic is largely in the dependents (Eq. 1 and Eq. 2 of the spec).

Claim 1 (apparatus). Three elements:

• a power tracker that receives the I/Q components of a set of carrier-aggregated transmit signals being sent simultaneously and computes one (single) power tracking signal from a combination of those I/Q components — with the express proviso that the aggregated signals are OFDM or SC-FDMA signals;
• a power supply generator producing one supply voltage from that single tracking signal; and
• a power amplifier that receives that one supply voltage and the multiple aggregated signals together and emits one output RF signal.
  In substance: compute one envelope/voltage estimate from all streams, generate one PA rail, and amplify the combined multi-carrier signal with a single PA.

Claim 18 (method). The method counterpart — "determining a single power tracking signal based on a plurality of inphase (I) and quadrature (Q) components of a plurality of carrier aggregated transmit …" — i.e., the same determination/generation steps recited as a process. Caveat: the publicly retrieved excerpts truncate this claim mid-sentence; I do not have the verified full text of claims 18–27 and will not paraphrase beyond what is quoted.

Claim 28 (means-plus-function apparatus). Same three functional blocks, drafted in §112(f) form: "means for determining a single power tracking signal…" (again with the OFDM/SC-FDMA proviso), "means for generating a single power supply voltage…", and "means for receiving the single power supply voltage and the plurality of carrier aggregated transmit signals… and producing a single output RF signal." (Text as reproduced in the PTAB institution decision for IPR2018‑01329, https://www.finnegan.com/a/web/cJC89ooTEfGAJyvNx6JnJ3/2SNQu2/ipr201801329_intel_v_qualcomm.pdf.)

Notable dependents: cl. 2 (tracking from overall power); cl. 3 (sum of per-signal powers); cl. 4 (sum of per-signal voltages); cl. 5 (per-carrier transmit circuits + summer into one PA); cl. 6 (digital modulated-IF path, single transmit circuit); cl. 7‑10 (power-tracking amplifier; switcher sensing PT-amp current; boost converter; boosted vs. battery rail); cl. 11‑12 (different carrier frequencies; tracking bandwidth smaller than aggregate carrier bandwidth); cl. 13‑15 (intra-band CA, contiguous / non-contiguous); cl. 16 (square each I/Q, sum, take square root — spec Eq. 1); cl. 17 (sum of per-signal √(I_k²+Q_k²) — spec Eq. 2).

4. Litigation and validity posture (this is the material part)

PTAB — Intel's six IPRs. Intel Corporation filed six petitions on 2018‑07‑03: IPR2018‑01326, ‑01327, ‑01328, ‑01329, ‑01330, ‑01340, all against app. 13/764,328. Institution 2019‑01‑15; final written decisions 2020‑01‑14. Four petitions relied on EP 2,442,440 A1 ("Yu," Alcatel-Lucent) and two on the Chen et al. dual-band envelope-tracking paper.

• First CAFC appeal (2021): In Qualcomm Inc. v. Intel Corp., No. 20‑1589 (Fed. Cir. July 27, 2021), the court vacated and remanded all six decisions because the Board adopted a sua sponte construction of "a plurality of carrier aggregated transmit signals" (dropping the parties' agreed "increased bandwidth for a user" requirement) without APA notice. It rejected Qualcomm's §112(f) construction argument.
• Remand decisions (2022): On 2022‑03‑23 the Board again held all challenged claims unpatentable.
• Second CAFC appeal (2024): Appeals were docketed as 22‑1824, 22‑1825, 22‑1826, 22‑1828, 22‑1829, 22‑1830. In a nonprecedential opinion issued September 23, 2024 (No. 22‑1824, 10 pages, https://cafc.uscourts.gov/opinions-orders/22-1824.OPINION.9-23-2024_2389024.pdf), the court affirmed that claims 1‑15, 17‑25 and 27‑33 are unpatentable as obvious over Yu, holding substantial evidence supported the Board's motivation/expectation‑of‑success findings that Yu's PA (Fig. 2; ¶34) is mobile-device appropriate; and it dismissed Intel's cross-appeal on the Chen‑based petitions. Per WilmerHale, that decision invalidated 31 claims and resolved the last dispute from Qualcomm's 2017-era power-amplifier assertions (https://www.wilmerhale.com/.../20241023-wilmerhale-and-intel-score-significant-victory-at-the-federal-circuit). The Board recorded the Federal Circuit mandate on 2024‑11‑14 (Unified Patents docket for IPR2018‑01327).

District court / ITC context: Qualcomm's earlier enforcement included ITC Inv. No. 337‑TA‑1065 and S.D. Cal. No. 3:17‑cv‑01375 (Qualcomm v. Apple); Apple separately sought a declaratory judgment of invalidity/non-infringement of the '675 patent in the S.D. Cal. Qualcomm–Apple litigation (counterclaim text retrieved from the 8:20‑cv‑00048 docket).

Family litigation: Google Patents' litigation module lists an EPO/opposition track as well — a Boards of Appeal decision discussing the European counterpart and its "overall power" vs. "overall voltage" distinction (https://www.epo.org/boards-of-appeal/decisions/pdf/t221996eu1.pdf).

Practical takeaway: although Google Patents' status field still reads "Active," the claims of the '675 patent have been finally adjudicated unpatentable over Yu for claims 1‑15, 17‑25, 27‑33, with the Federal Circuit mandate having issued. Freedom-to-operate analysis should therefore focus on the live continuations in the family (US 11,133,833; 11,641,215; 11,996,872; 12,562,763), not on the '675 claims.

5. CAFC 2026 dockets — specific answer

I found no 2026 Court of Appeals for the Federal Circuit docket activity for US 9,608,675. The '675 appeals (22‑1824 et seq.) were decided September 23, 2024, and the mandate returned to the PTAB on November 14, 2024. Filings from 2025‑2026 that surfaced in searching concern unrelated matters (e.g., Qualcomm v. Apple, No. 23‑1208, April 23, 2025; Golden v. Qualcomm, 23‑1818). No rehearing, en banc, or certiorari activity on the '675 patent appeared in the sources retrieved.

6. Explicit uncertainties

1. Petitioner identity. The IPR/PACER-derived sources (RPX Insight, Unified Patents PTAB portal, Finnegan's posting of the IPR2018‑01329 decision, and the AU IP Brief) identify Intel as petitioner in all six IPRs. One litigation table retrieved from a district-court exhibit lists Apple Inc. as petitioner for IPR2018‑01326, which conflicts with those records; I could not reconcile it, so treat "Intel" as well-supported but not perfectly unanimous.
2. Claim 18–27 text. The publicly retrieved excerpts truncate the method claim; I have not verified claims 18‑27 verbatim.
3. Claim 16/26 treatment. The reported affirmance set (1‑15, 17‑25, 27‑33) excludes claims 16 and 26; I could not confirm whether those were simply not challenged in the Yu IPRs or were addressed elsewhere.
4. Third-party "2026" AI-generated litigation write-ups. Several PatSnap blog pages surfaced that describe the appeal as "Qualcomm and Intel" co-appellants and give mutually contradictory characterizations ("patent survivability strengthened" vs. "Qualcomm loses ground; claims remain cancelled"). These pages are internally inconsistent and conflict with the actual Federal Circuit opinion and PTAB docket; I have not relied on them.
5. Dates/spelling. Minor source inconsistencies exist (e.g., inventor "Dorosenco" vs. "Dorsenco"; a family listing showing a priority date of 2013‑02‑10 for a continuation). The controlling date is the 2013‑02‑11 filing date on the face of the '675 patent.
6. I did not access USPTO PatentCenter/Global Dossier directly; USPTO-side bibliographic data above is taken from the patent text and mirror databases (Google Patents, FreePatentsOnline, Espacenet).

I'll research litigation involving US patent 9608675. Let me search multiple sources.

Let me dig deeper into the specific cases and their outcomes.

Litigation Involving U.S. Patent No. 9,608,675 ("the '675 patent")

Patent identified literally as: US 9,608,675 B2, Application No. 13/764,328, "Power tracker for multiple transmit signals sent simultaneously," inventor Alexander Dorosenco, assignee Qualcomm Incorporated, priority/filing 2013‑02‑11, granted 2017‑03‑28.

Below is what I can confirm from the sources retrieved, with confirmed items separated from items where the record is ambiguous. I have not auto-corrected any patent numbers, case numbers, or dates.

1. International Trade Commission — Investigation No. 337‑TA‑1065

Key point: The '675 patent appears on the ITC complaint caption, but the '675 claims were dropped from the investigation before hearing and were never adjudicated by the Commission.

Source: USITC pub. 4981 (337‑TA‑1065); Federal Register 84 FR 12292 (Apr. 1, 2019); https://portal.unifiedpatents.com/litigation/International%20Trade%20Commission/case/337-TA-1065

2. Qualcomm Incorporated v. Apple Inc., S.D. Cal. No. 3:17‑cv‑01375

Sources: CourtListener docket 6088510 (https://www.courtlistener.com/docket/6088510/qualcomm-incorporated-v-apple-incorporated/); ECF No. 750, S.D. Cal. 3:17‑cv‑01375; Jones Day case description.

3. German enforcement actions (Mannheim and Munich)

Sources: aggregator/press reporting on the Qualcomm–Apple German complaints (July 2017); Google Patents family litigation entries (Darts‑IP "first worldwide family litigation filed").

4. PTAB — Inter Partes Review (Intel Corp. v. Qualcomm Inc.)

Related PTAB cases. Google Patents' litigation data for this patent lists six PTAB cases: IPR2018‑01326, IPR2018‑01327, IPR2018‑01328, IPR2018‑01329, IPR2018‑01330, and IPR2018‑01340. The Federal Circuit described the appeal as arising from "six petitions for inter partes review challenging the validity of a single Qualcomm patent."

Source conflict to flag: A table filed in Apple Inc. v. Qualcomm (C.D. Cal. 8:20‑cv‑00048, Doc. 21‑2) lists IPR2018‑01326, patent 9608675, petitioner Apple Inc., patent owner Qualcomm Incorporated, filed 2018‑07‑03, FWD 2020‑01‑14. The Unified Patents PTAB record for the parallel case IPR2018‑01327 identifies the petitioner as Intel Corp. I report both as found rather than reconciling them; the petitioner attribution for IPR2018‑01326 should be verified against the PTAB docket directly.

Sources: https://portal.unifiedpatents.com/ptab/case/IPR2018-01327 ; https://services.patexia.com/lawsuits/shorturl-id-[116001](/patent/116001)

5. Appeals — U.S. Court of Appeals for the Federal Circuit

(a) Qualcomm Inc. v. Intel Corp., No. 2020‑1589 (Fed. Cir. July 27, 2021). Appeal from IPR2018‑01327. The Federal Circuit held the PTAB violated the Administrative Procedure Act by adopting, in its final written decision, a claim construction of "a plurality of carrier aggregated transmit signals" that neither party had proposed, and remanded. This is the appeal that produced the March 23, 2022 remand FWD.

(b) Qualcomm Inc. v. Intel Corp., Nos. 2022‑1824, ‑1825, ‑1826, ‑1828, ‑1829, ‑1830 (Fed. Cir.). Six appeals arising from the '675 IPRs. Qualcomm was the appellant challenging the adverse unpatentability findings, and Intel cross‑appealed on the "Chen" grounds where it had lost below. Disposition: "AFFIRMED IN PART AND DISMISSED IN PART" (reported date September 23, 2024 — treat the exact date as secondary‑sourced). The opinion states: "we affirm the Board's final written decisions holding claims 1‑15, 17‑25, and 27‑33 of the '675 patent unpatentable as obvious." Costs were awarded to Intel. Intel's Chen‑based cross‑appeal was dismissed as moot in light of the affirmance.

Practical effect: claims 1–15, 17–25, and 27–33 of the '675 patent were held unpatentable and that holding was affirmed. On the face of the opinion, claims 16 and 26 were not within the affirmed invalidation (i.e., they appear not to have been part of the Yu‑based grounds decided). I flag this as a reading of the reported claim ranges, not as a legal conclusion about surviving claim scope.

Sources: CourtListener opinion (https://www.courtlistener.com/opinion/[10124057](/patent/10124057)/qualcomm-incorporated-v-intel-corporation/); KTS Law post‑grant blog (https://ktslaw.com/en/Blog/Post-Grant-Proceedings/2021/8/The-Federal-Circuit-Further-Defines-the-Contours); Google Patents litigation links for CAFC 22‑1824 through ‑1830.

6. Things I could NOT confirm / cautions

1. No district‑court infringement action by Qualcomm against Intel on the '675 patent was found. The "Qualcomm v. Intel" matter is an IPR/appellate proceeding, not a district‑court infringement suit. Some third‑party litigation‑analytics summaries (e.g., PatSnap) describe it as a "Qualcomm v. Intel" patent appeal; the underlying record is PTAB + Federal Circuit.
2. A different U.S. Patent No. 9,608,675‑type '765 patent exists in PTAB filings concerning a "cooperative conversational voice user interface" (VB Assets / VoiceBox / IPR2025‑00866 et seq., asserted against Samsung, Amazon, SoundHound, Apple, and others). That is not the Qualcomm power‑tracker patent and involves a different patent number. I flag this because search results on "'765 patent" mix the two; do not merge them.
3. Google Patents legal status for US 9,608,675 B2 still displays "Active, adjusted expiration 2033‑06‑25." That status field is a data‑provider assumption and does not reflect the affirmed PTAB invalidation of claims 1‑15, 17‑25, 27‑33. Do not rely on the "Active" label alone.
4. German outcomes specific to the '675‑based action are unverified in the sources retrieved.
5. Patent owner's identity throughout all U.S. proceedings above is Qualcomm Incorporated (original assignee); no assignment to a monetization entity appears in the record for this patent.

Summary Table

Bottom line: The only adjudicated U.S. litigation on the merits of US 9,608,675 is the Intel‑initiated IPR cluster, which ended with the Federal Circuit affirming that claims 1‑15, 17‑25, and 27‑33 are unpatentable. The Qualcomm‑as‑plaintiff enforcement actions (ITC 337‑TA‑1065, S.D. Cal. 3:17‑cv‑01375, and the German actions) either did not reach the '675 merits or ended in the April 2019 global Qualcomm–Apple settlement.

I'll research the PTAB proceedings and their outcomes before drafting.

Significant finding: there appear to be six Intel IPRs. Let me dig into outcomes.

Now let me pin down the petitioner identities, per-IPR claim sets, and the second-round appeals.

AIA Trial Proceedings — US 9,608,675 ("Power tracker for multiple transmit signals sent simultaneously")

Data-gap flag first. The "PTAB proceedings on file" block states the ODP API returns no AIA trial proceedings for this patent. That is wrong. Six Intel-filed IPRs are on file, each with two rounds of Final Written Decisions, and the Federal Circuit reached a merits disposition in 2024. Google Patents' own family/litigation record for this patent lists all six PTAB case numbers (https://patents.google.com/patent/US9608675/en), and each is corroborated by RPX Insight, Patexia, Unified Patents' docket portal, and the Federal Circuit opinions. I have used those six numbers only — none are invented.

Proceedings overview

There are six AIA proceedings on file — all Inter Partes Reviews, all filed by Intel Corporation on 2018-07-03, all instituted 2019-01-15, all going to Final Written Decision (original FWDs January 2020), all vacated and remanded by the Federal Circuit on 2021-07-27, all re-decided on remand in March 2022 with the challenged claims again held unpatentable (except the Chen-based grounds), and the resulting second round of appeals affirmed on 2024-09-23 as to the Yu-based invalidity findings. Bottom line for a defendant today: claims 1–15, 17–25 and 27–33 of the '675 patent were held unpatentable and that holding now has appellate finality — if a demand letter or complaint cites any of those claims, the assertion is dead on arrival. Only claims 16 and 26 were never challenged; both are dependent claims that cannot survive their canceled parents.

Caveat I will not paper over: I could not retrieve the full text of the 2024-09-23 opinion or the individual remand FWDs for IPR2018-01328 and IPR2018-01330. Two points therefore need verification in PTAB E2E / Patent Center before you rely on them: (i) the exact claim-level disposition for claims 6 and 7–15 in the two non-consolidated cases, and (ii) whether the § 318(b) cancellation certificate has issued — Google Patents still labels the patent "Active, expires 2033-06-25," which is irreconcilable with a full claim cancellation and is likely a stale auto-generated status field.

IPR2018-01326 — Intel Corporation v. Qualcomm Incorporated

• Type: Inter Partes Review
• Filed: 2018-07-03
• Status: "Final Written Decision" (Google Patents family record); RPX Insight: Final 2022-03-23, "All Claims Unpatentable"; consolidated with 01327/01329/01340 on remand
• Judge panel: Amanda F. Wieker, Michelle N. Wormmeester, Scott B. Howard (RPX lists Judge Wieker "+4" — the panel sat across the related cases)
• Petition grounds: § 103 obviousness. Ground 1 combined Yu (EP 2,442,440) with Wang, "Design of Wide-Bandwidth Envelope-Tracking Power Amplifiers"; a separate ground combined Chen, "Hybrid Envelope Tracking" with other references. Petition exhibits also included Choi (MTT 2010/2011), Dahlman 4G: LTE/LTE-Advanced for Mobile Broadband (2011), Eliezer US 2009/0004981, Oga US 2013/0034186, Sauter From GSM to LTE (2011), and Mathe US 6,300,826 (the Kenington/Wang-class envelope amplifier patent Qualcomm itself cites in the spec). No § 112 ground — the Board has no authority to cancel for indefiniteness.
• Institution decision: Instituted 2019-01-15 on all challenged claims.
• Final Written Decision: Original FWD 2020-01-14: challenged claims held unpatentable. Vacated by the Federal Circuit (see Appeal). FWD on Remand, 2022-03-23: the Board again concluded — in the consolidated 01326/01327/01329/01340 remand decision — "Petitioner has shown by a preponderance of the evidence that claims 1–15, 17–25, and 27–33 of the '675 patent are unpatentable" (verbatim from the remand FWD, https://insight.rpxcorp.com/federal_circuit/[1146501](/patent/1146501)/rpx-ptab/IPR2018-01340/IPR2018_01340___Patent_Owner_s_Notice_of_Appeal_54771e86ca469120b614573c72c3afdb.pdf). Judgment title verbatim: "Final Written Decision on Remand Determining All Challenged Claims Unpatentable."
• Settlement / termination: None. No adverse-judgment or joint-stipulation termination on the docket.
• Appeal: First appeal 2020-1589 (one of 2020-1589 through 2020-1594 for the six cases) → Qualcomm Inc. v. Intel Corp., 6 F.4th 1256 (Fed. Cir. 2021) (Moore, Reyna, Stoll; 2021-07-27) — vacated and remanded because the Board sua sponte omitted the parties' agreed "increase bandwidth" requirement from its construction of "a plurality of carrier aggregated transmit signals" without APA notice; the court affirmed the Board's means-plus-function construction of the claim 28 "means for determining a single power tracking signal" limitation (power tracker 582 as corresponding structure; no algorithm required, because power tracker 582 is circuitry, not a general-purpose computer). Second appeal 22-1824 (one of 22-1824/1825/1826/1828/1829/1830) → decided 2024-09-23, https://www.courtlistener.com/opinion/[10124057](/patent/10124057)/qualcomm-incorporated-v-intel-corporation/ — the court affirmed the Board's obviousness conclusion over Yu as supported by substantial evidence (including the motivation-to-modify finding turning on Yu's Fig. 2 and ¶ 34 teaching use in "wireless transceivers of mobile terminals"), and rejected the Chen-based attacks. PACER disposition recorded for the companion appeals (22-1824, 22-1828) as "AFFIRMED IN PART AND DISMISSED IN PART."
• Defensive value: The full claim set that Intel challenged in this case is dead and the invalidity holding is appellate-final. Any infringement theory built on claims 1–15, 17–25 or 27–33 is untenable; lead with the remand FWD and the 2024 opinion in any pre-suit response.

IPR2018-01327 — Intel Corporation v. Qualcomm Incorporated

• Type: Inter Partes Review
• Filed: 2018-07-03
• Status: "Final Written Decision Appealed"; institution 2019-01-15; termination date 2022-03-23 (Unified Patents portal, https://portal.unifiedpatents.com/ptab/case/IPR2018-01327)
• Judge panel: Wieker, Wormmeester, Howard (same panel)
• Petition grounds: § 103 — Yu + Wang as to claims 11, 17, 27 and 33, and Chen-based grounds; challenges directed to claims 7–15, 17, 23–25, 27 and 33 ("Petitioner challenges claims 7-15, 17, 23-25, 27, and 33 of the '675 patent," FWD at *C, https://www.finnegan.com/a/web/jKBJw1oyMzUaaceCEPfQqR/2SNP3A/ipr201801327_intel_v_qualcomm.pdf)
• Institution decision: Instituted 2019-01-15.
• Final Written Decision: Original FWD 2020-01-14 — claims unpatentable. On remand the Board specifically rejected Qualcomm's disavowal argument and adopted Intel's construction of "generates the single power tracking signal based on a combination of the plurality of I and Q components" ("using a combination derived from the plurality of I and Q components"), holding that this encompasses both Equation 1 and Equation 2 embodiments, and that "claims 11, 17, 27, and 33 would have been obvious over Yu and Wang." Remand judgment: all challenged claims unpatentable (2022-03-23).
• Settlement / termination: None.
• Appeal: Vacated/remanded (6 F.4th 1256); second appeal within 22-1824…22-1830; 2024-09-23 affirmance on the Yu ground.
• Defensive value: Strong. Claims 7–15 and 17/23–25/27/33 are canceled, and the 2024 affirmance on Yu is now the controlling merits ruling. Note the construction holding cuts against defendants who might have hoped to read "combination of" narrowly — but it is moot given cancellation.

IPR2018-01329 — Intel Corporation v. Qualcomm Incorporated

• Type: Inter Partes Review
• Filed: 2018-07-03
• Status: Final Written Decision; consolidated with 01326/01327/01340 on remand (remand FWD header lists all four)
• Judge panel: Wieker, Wormmeester, Howard
• Petition grounds: § 103 — Yu/Wang and Chen-family grounds; claim 28's means-plus-function limitation was squarely at issue. FWD: https://www.finnegan.com/a/web/6JLDxmHZhPQ2Ye1881pwn5/2SNPdg/ipr201801329_intel_v_qualcomm.pdf
• Institution decision: Instituted 2019-01-15.
• Final Written Decision: FWD 2020-01-14 holding the challenged claims unpatentable; remand FWD 2022-03-23 again determining all challenged claims unpatentable (claims 1–15, 17–25, 27–33 across the four consolidated cases). The Board treated "means for determining a single power tracking signal" (claim 28) as § 112(f) and identified power tracker 582 as the corresponding structure — a finding the Federal Circuit expressly upheld in 2021.
• Settlement / termination: None.
• Appeal: First appeal vacated/remanded; second appeal in the 22-18xx cluster; affirmed on the Yu ground 2024-09-23.
• Defensive value: Claim 28 and its dependents are canceled. The surviving appellate holding on the means-plus-function issue is the one piece of this proceeding that still has any analytical value to a defendant (it forecloses a "no corresponding structure ⇒ indefinite ⇒ claim can't be invalidated" escape hatch).

IPR2018-01340 — Intel Corporation v. Qualcomm Incorporated

• Type: Inter Partes Review
• Filed: 2018-07-03
• Status: Final Written Decision; consolidated for remand with 01326/01327/01329
• Judge panel: Wieker, Wormmeester, Howard
• Petition grounds: § 103, Yu/Wang and Chen-family references (same exhibit set as the other Intel petitions).
• Institution decision: Instituted 2019-01-15.
• Final Written Decision: Original FWD 2020-01-14 (claims unpatentable); remand FWD 2022-03-23 — all challenged claims unpatentable. Qualcomm filed a Notice of Appeal from this remand FWD (document at the RPX link above), i.e., this is the source of one of the 22-18xx appeals.
• Settlement / termination: None.
• Appeal: 22-18xx cluster; 2024-09-23 affirmance on the Yu ground.
• Defensive value: Same as the other three Yu-ground cases — the claims are gone.

IPR2018-01328 — Intel Corporation v. Qualcomm Incorporated

• Type: Inter Partes Review
• Filed: 2018-07-03
• Status: "Final Written Decision" (Patexia, https://services.patexia.com/lawsuits/Intel-Corporation-et-al-v-Qualcomm-Incorporated-id-[116004](/patent/116004)); institution 2019-01-15; decision 2020-01-14; Federal Circuit appeal recorded as 2020-1589
• Judge panel: Michelle N. Wormmeester (writing), Amanda F. Wieker, Scott B. Howard
• Petition grounds: § 103 — Yu/Wang and Chen/Hybrid-Envelope-Tracking grounds. Claims challenged in the petition (per Patexia): 1, 2, 3, 5, 7–15, 17–21, 23–25, 27.
• Institution decision: Instituted 2019-01-15.
• Final Written Decision: 2020-01-14 — all challenged claims unpatentable; vacated and remanded 2021-07-27. I could not verify the claim-level result of this case's FWD on remand vs. the two cases not included in the four-case consolidated remand decision. The 2024 opinion distinguishes "the Yu IPRs" (where Intel prevailed) from "the Chen IPRs" (where the Board found "Intel had failed to show by a preponderance of the evidence that any claims of the '675 patent were unpatentable over Chen"), but the opinion excerpt I obtained does not map proceeding numbers to those groupings. Do not assume; pull the FWDs.
• Settlement / termination: None.
• Appeal: 2020-1589 (first round, vacated/remanded); then one of 22-1824/1825/1826/1828/1829/1830 (second round, decided 2024-09-23).
• Defensive value: Partial but useful — the claims challenged here are the same claims swept up in the Yu-ground cases, so the practical effect is the same. If your theory depends on claims 6 or 7–15 specifically, confirm the Yu ground reached them before relying on this proceeding alone.

IPR2018-01330 — Intel Corporation v. Qualcomm Incorporated

• Type: Inter Partes Review
• Filed: 2018-07-03
• Status: Final Written Decision (Google Patents family record); instituted 2019-01-15
• Judge panel: Same core panel (Wieker, Wormmeester, Howard) — not independently verified for this case
• Petition grounds: § 103, Yu/Wang and Chen-family references.
• Institution decision: Instituted 2019-01-15.
• Final Written Decision: 2020-01-14 (claims unpatentable); vacated and remanded 2021-07-27. Remand FWD not verified — this case is not part of the four-case consolidated remand judgment (01326/01327/01329/01340).
• Settlement / termination: None.
• Appeal: One of the 22-18xx cluster; 2024-09-23 decision.
• Defensive value: Same caveat as 01328 — verify the remand disposition directly.

Strategic summary

Claim status. Treated as a whole, the six IPRs challenged and invalidated claims 1–15, 17–25 and 27–33 of the '675 patent — every independent claim (1, 18, 28, 33) and almost every dependent claim. Those holdings were entered in January 2020, re-entered on remand in March 2022 after the Federal Circuit's APA-based vacatur, and affirmed on the merits on 2024-09-23 on the Yu-based § 103 ground. Claims 16 and 26 were never challenged, but claim 16 depends from claim 1 and cannot survive independently; assume the patent has no enforceable claim scope. Verify the status of claims 6 and 7–15 against the individual remand FWDs and confirm whether the § 318(b) certificate has issued — the stale "Active" label on Google Patents is not evidence to the contrary.

Estoppel. Intel and its privies are barred under 35 U.S.C. § 315(e)(2) from raising in civil actions any ground they raised or reasonably could have raised in these IPRs — but that is academic now that the claims are canceled. For a new defendant being asserted against, no IPR estoppel applies in either direction: you are not a privy of Intel, and the patent owner cannot assert claims that a final, affirmed FWD held unpatentable. You do not need prior art at all for those claims — the FWDs and the 2024 opinion do the work. If Qualcomm asserts claims 16/26, your first move is a motion directed to the dependency problem plus the practical impossibility of asserting a dependent claim whose sole independent claim is statutorily canceled.

Pattern signals. One petitioner (Intel) filed all six IPRs on a single day against one patent — a coordinated, portfolio-level attack, consistent with Intel's 2018–2019 modem/turbo-code campaign against Qualcomm. Qualcomm litigated to the hilt: it appealed all six FWDs in 2021, won a vacatur on procedural grounds, lost on the merits at the Board on remand, then appealed all six remand FWDs again and lost again in 2024. There is no defensive aggregator (Unified Patents) and no Apple-filed IPR in this chain: the Unified Patents data in the prompt template is a license attribution, not petitioner identity. One third-party litigation table (a Lex Machina-derived chart filed in Apple v. Masimo-style litigation, Docket 21-2, 8:20-cv-00048 (C.D. Cal.)) lists "Apple Inc." as petitioner in IPR2018-01326; every primary patent source (RPX, Patexia, Unified Patents, the Federal Circuit's own opinion) attributes the six IPRs to Intel. Treat the Apple entry as a likely data-entry error, but confirm in E2E if Apple's potential privity matters to you. Context: the '675 patent was asserted by Qualcomm against Apple at the ITC, Inv. No. 337-TA-1065, in S.D. Cal. 3:17-cv-01375, and in German courts (Munich/Mannheim) — the PTAB campaign ran parallel to that.

Recommended next steps

1. Get the operative documents. The controlling dispositive text is the 2022-03-23 FWD on Remand in IPR2018-01326/01327/01329/01340 ("Final Written Decision on Remand Determining All Challenged Claims Unpatentable") and the 2024-09-23 Federal Circuit opinion (https://www.courtlistener.com/opinion/10124057/qualcomm-incorporated-v-intel-corporation/), which affirms on Yu. Pull both from PTAB E2E / CourtListener, plus the two remand FWDs for IPR2018-01328 and IPR2018-01330, which I could not verify.
2. Confirm cancellation status. Check Patent Center for the § 318(b) certificate. If no certificate has issued despite the 2024 affirmance, that is a procedural loose end worth raising early; if it has, note the cancellation date.
3. No live PTAB milestones to track. Every one of the six proceedings is terminated (last docket activity mid-2022; appeals resolved 2024-09-23). There is no institution-deadline clock, no oral hearing, no 1-year FWD clock running on this patent.
4. Pre-suit response posture. If you receive an assertion letter citing claims 1–15, 17–25 or 27–33, respond with the remand FWD and the 2024 affirmance; a complaint on those claims is subject to Rule 11 exposure. If the assertion cites claims 16 or 26, attack on dependency/survival grounds and note that a patent owner cannot enforce a dependent claim whose parent has been canceled.
5. Do not rely on secondary sources' narrative. Several AI-generated litigation summaries (e.g., the PatSnap "Qualcomm and Intel co-appellants" pieces) contain plainly wrong statements about who filed what and who appealed. Use PTAB E2E, the CAFC docket, and CourtListener only.

I'll research this patent's assignment record and litigation history before writing. Let me run several searches in parallel.

Let me pin down the assignment record specifics and confirm whether any post-issuance transfers exist.

Inventors

Alexander Dorosenco — sole named inventor (spelled "DOROSENCO, Alexander" in the USPTO assignment event; rendered "Alexander Dorsenco" in the later continuation US 11,133,833). Residence of record: San Diego, CA (later El Cajon, CA).

• Employer at filing: Qualcomm Incorporated. The application was filed 2013-02-11 naming Qualcomm as applicant, and the inventor's rights were assigned to QUALCOMM INCORPORATED (Google Patents legal event 2013-05-02, "ASSIGNMENT OF ASSIGNORS INTEREST"). Dorosenco is credited with 20 Qualcomm patents across 2015–2025 (Patent Leaderboard), i.e., he remained a Qualcomm engineer long after this filing.
• Unusual-departure pattern: Not present — the inverse. The inventor did not leave the assignee within 12 months; he continued filing Qualcomm applications into the mid-2020s. There is no "orphaned inventor precedes a fire-sale" signature here.

Original assignee

Qualcomm Incorporated (recorded as "QUALCOMM INCORPORATED"), 5775 Morehouse Drive, San Diego, CA 92121-1714.

• Primary line of business: Semiconductor / wireless R&D and licensing. Qualcomm operates through QCT (chipsets; Snapdragon baseband and RF front-end modules) and QTL (technology licensing). In its own S.D. Cal. pleading Qualcomm described itself as a ~20,000-employee U.S. multinational and "holder of US9608675B2."
• Product embodying the claims: Yes. Claim 1 covers a power tracker + single power-supply generator + PA that receives a single supply voltage and multiple carrier-aggregated OFDM/SC-FDMA signals to emit one output RF signal — i.e., envelope/power tracking for carrier aggregation, which is core RF front-end technology in Qualcomm's modem chipset line.
• Current status: Operating. Publicly traded (NASDAQ: QCOM), solvent, active as record owner. Note that Qualcomm Technologies, Inc. ("QTI") is a wholly owned subsidiary that "operates substantially all of Qualcomm's product and services business," but no assignment of record to QTI appears — the record owner remains Qualcomm Incorporated.
• Family: All continuations remain Qualcomm-assigned — US 11,133,833 (from 15/916,101), US 11,641,215 (from 17/486,669), US 11,996,872 (from 18/141,891), US 12,562,763 (from 18/672,458), plus EP 2,954,737 B1 / EP 3,203,787 A1, CN 104982077 B, JP 6490596 B2, KR 102182001 B1.

Assignment timeline

Data-access caveat (read first): The sources reachable in this session (Google Patents legal events, Espacenet/INPADOC family data, litigation dockets) expose only one recorded reassignment event for US 9,608,675 and do not expose reel/frame numbers or the recording correspondent. I was not able to query the USPTO Assignment Center (https://assignmentcenter.uspto.gov/) directly. Everything below is grounded in the legal-events data provided; I have deliberately not invented reel/frame numbers or correspondent names.

Chronological record:

• Executed on/before 2013-05-02 / recorded 2013-05-02 — Reel/frame not retrievable from available sources

  • Conveyance: Assignment (reassignment record)
  • Assignor: DOROSENCO, Alexander (sole inventor)
  • Assignee: QUALCOMM INCORPORATED
  • Correspondent: Not retrievable. Google Patents legal events does not carry the recording correspondent. (The patent's prosecution firm of record is Haynes and Boone, LLP, Dallas, TX, per FreePatentsOnline; that is not the assignment recording agent and should not be conflated with one.)
  • Context: Original employee invention assignment. Filed ~2.5 months after the 2013-02-11 priority filing; Qualcomm was already the applicant of record at filing. This is a routine employer→inventor capture, not an acquisition.

• No post-issuance assignment of record. After issuance (2017-03-28) the legal-events feed shows only: continuation filings (2017-02-27, 2018-03-08, 2021-09-27, 2023-05-01, 2024-05-23), the grant, and the litigation events. There is no transfer to any LLC, no change-of-name record, and no security interest. Current assignee (per Google Patents): Qualcomm Inc. Status: Active, adjusted expiration 2033-06-25.

Conclusion for this section: the Assignment Center chain is a single link — inventor → Qualcomm — and nothing else. That absence of later records is itself the finding: the original assignee still owns the patent.

Timeline diagram

timeline
    title Ownership of US 9608675
    2013 : Filed by Qualcomm 11 Feb
         : Inventor assignment recorded 02 May
    2017 : Patent issued 28 Mar
         : Qualcomm sues Apple in S D Cal
         : Qualcomm files ITC complaint
    2018 : Intel files IPR petitions
    2019 : ITC finds claim invalid
         : S D Cal case dismissed
    2022 : Federal Circuit appeals filed

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. No assignment to any "IP / Patents / Licensing / Holdings / Ventures" entity exists. The only recorded link is DOROSENCO → QUALCOMM INCORPORATED (2013-05-02). Record owner remains a named operating parent, not a single-purpose LLC.

2. Known asserter in the chain — Not present. Neither Qualcomm Incorporated nor any listed NPE (Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Conversant/Mosaid, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Spangenberg entities) appears as assignee at any point. The patent-owner of record is an operating company that appears on the defendant side of NPE trackers, not the plaintiff-NPE side.

3. Repeat correspondent across the chain — Not present / unclear. With only one recorded link there is no chain across which a correspondent could recur, and the recording correspondent is not exposed in the sources available. Verified counsel in the litigation are Qualcomm's (Jones Day — Karen P. Hewitt, Randall E. Kay; Quinn Emanuel — David A. Nelson) versus Intel's (Wilmer Cutler Pickering Hale & Dorr — Sarah B. Petty, David L. Cavanaugh). These are litigation counsel, not assignment correspondents — no recurrence finding can be drawn from them.

4. Cascading transfers — Not present. Zero chained assignments; a single 2013 conveyance over a 13-year (2013–2026) record.

5. Pre-litigation transfer — Not present. The sole assignment (recorded 2013-05-02) precedes the first suits (ITC complaint 2017-07-07; S.D. Cal. filed 2017-07-06) by roughly four years. There is no last-minute assignment arranged to set venue or standing; Qualcomm sued as the original assignee, exactly as it pleaded ("Qualcomm is the original assignee").

6. Bankruptcy fire-sale — Not present. No Chapter 7/11, no estate sale. Qualcomm remained financially strong throughout.

7. Privateering — Not present. Qualcomm asserted in its own name, not through an NPE proxy: In re Certain Mobile Electronic Devices…, Inv. No. 337-TA-1065 (complaint 2017-07-07; instituted 2017-08-08, Apple sole respondent; '675 among six asserted patents), and Qualcomm Inc. v. Apple Inc., S.D. Cal. 3:17-cv-01375 (filed 2017-07-06; the '675 is listed in the court's patent-report along with '936, '558, '658, '949, '490). Qualcomm also asserted the '675 family in Germany (Mannheim complaint filed 2017-07-17, cited as IPR2018-01327 Ex. 1119).

8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. Caution: the Google Patents page lists "PTAB case IPR2018-01326…01330 filed," which is Unified Patents' litigation-data feed, not a Unified ownership interest. The Unified Patents case page for IPR2018-01327 shows Petitioner: Intel Corp. (Large Operating Company) and Patent Owner: Qualcomm (Petitioner counsel Wilmer Cutler Pickering Hale & Dorr; Patent Owner counsel Jones Day; Final Written Decision appealed; termination 2022-03-23). Intel's IPRs are a competitor's validity challenge, not defensive-aggregation ownership.

Verdict

Operating-company assertion.

Justification: Qualcomm Incorporated is the sole recorded assignee — the chain consists of one 2013-05-02 inventor-to-Qualcomm assignment and nothing more, with the record owner unchanged through grant (2017-03-28) and to the present (Active; adjusted expiration 2033-06-25). Qualcomm ships products practicing claim 1 and asserted the '675 in its own name against a direct competitor, Apple, at the ITC (337-TA-1065, filed 2017-07-07) and in S.D. Cal. 3:17-cv-01375 (filed 2017-07-06), while Intel (a competitor, not a defensive aggregator) challenged validity in IPR2018-01326–01330. No shell entity, no known NPE, no cascading or pre-suit transfer appears anywhere in the record.

Primary-source limitation and verification: I could not reach the USPTO Assignment Center in this session, so reel/frame and recording-correspondent fields are unverified and were deliberately left blank rather than guessed. Verify the single-link chain at the USPTO Assignment Center search (https://assignmentcenter.uspto.gov/) and cross-check the legal-events record at https://patents.google.com/patent/US9608675/en.

I'll research US patent 9608675 and its prior art citations. Let me start with targeted searches.

Let me get the definitive citation list from the patent's front page and Google Patents.

Let me pull the clean citation lists from Justia and FreePatentsOnline.

Prior-Art Analysis — U.S. Patent No. 9,608,675

1. Patent identified (literal match only)

Important disambiguation note (operating rule): the alphanumeric string 9608675 also appears as UK application GB 9608675.6 (granted as GB 2312518 B, "Warning device for live electricity," Taylor, filed 1996‑04‑26) and as a CNV record at TCAG. These are unrelated to the US patent and are excluded. I searched specifically for US 9,608,675.

2. What "patent citations" means here — two different lists

The record contains two distinct citation sets, and they do not agree with each other. I report both rather than silently merging them.

• List A — Front-page (56) "References Cited" of US 9,608,675, as reproduced on FreePatentsOnline (https://www.freepatentsonline.com/9608675.html). This is the U.S. prosecution/IDS list.
• List B — "Patent Citations" shown by Google Patents for this family (visible on the CN and JP family members, e.g. https://patents.google.com/patent/CN104982077B/en and https://patents.google.com/patent/JP6490596B2/en), which lists only four references: GB 2476393 A, WO 2012/110409 A1, US 2012/0326783 A1, GB 2488380 A.

Flag: I could not retrieve the complete front-page (56) list in one continuous block — the FPO listing I captured jumps from US 2006/0264186 to US 6,009,090 (some intermediate entries are truncated). Treat List A as incomplete. Also, one entry on List A (US 2014/0111275, published 2014‑04‑24) postdates the 2013‑02‑11 priority date, so it cannot be §102 prior art — it was likely cited during a later prosecution/IDS event and I flag it as an anomaly.

3. List A — References Cited (US front page)

U.S. patent documents

(The FPO block between US 2006/0264186 and US 6,009,090 is truncated in the retrieved text; additional U.S. entries likely exist in the complete (56) list, e.g. US 6,300,826 (Mahe et al.), which is discussed in the specification itself as background for the power-tracking amplifier. I could not confirm the full intermediate list — flagged as unknown rather than invented.)

Foreign patent documents

Other references (non-patent)

4. List B — Google Patents "Patent Citations" (family view)

The four references shown by Google Patents for the family:

1. GB 2476393 A (Ubidyne) — 2011‑06‑22 — Multiple envelope tracking for an antenna array.
2. WO 2012/110409 A1 — Nujira Ltd. — 2012‑08‑23 — Power control. (US equivalent: US 2012/0229208 A1 = Wimpenny, above.)
3. US 2012/0326783 A1 (Qualcomm) — 2012‑12‑27 — Low-voltage power-efficient envelope tracker.
4. GB 2488380 A (Nujira) — 2012‑08‑29 — MIMO amplifier array with common envelope-tracking supply.

This list is a subset consistent with List A.

5. Most relevant prior art for § 102 anticipation

A reference anticipates under § 102 only if it discloses every element of the claim in a single disclosure. Applying that test to independent claim 1 (power tracker computing a single power-tracking signal from I/Q components of carrier-aggregated simultaneous transmit signals → single supply voltage → single PA producing one output RF signal):

• GB 2488380 A (Nujira) — strongest candidate against claim 1. Discloses a common envelope-tracking supply for multiple transmit paths. But it appears to track per-path envelopes/envelope replica rather than computing one tracking signal from the summed I/Q of carrier-aggregated signals, and it uses an amplifier array rather than a single PA — so it likely fails the "single PA / single output RF signal" and "I/Q combination" limitations. Best characterized as § 103 art, not § 102.
• US 2011/0151806 A1 / GB 2476393 A (Kenington, Ubidyne) — discloses multiple envelope trackers (one per element), the opposite of the claimed single tracking signal. § 103 art (especially against claims 7–10), § 102 unlikely for claim 1.
• US 2012/0326783 A1 (Mathe/Mahe, Qualcomm) and US 2012/0326686 A1 (Dai) — directed to the power-supply/amplifier side (claims 7–10), not to I/Q combination for carrier-aggregated signals. In combination they anticipate-related claim 7/8/9 structure.
• US 2012/0229208 A1 / WO 2012/110409 (Wimpenny, Nujira) — envelope tracking per band/carrier; relevant to claims 1, 11, 13.
• US 2010/0291963 A1 (Patel) and US 2012/0033656 A1 (De Maaijer) — simultaneous multi-signal transmission, but no I/Q-derived single power-tracking supply; § 103 art at best for claim 1's "sent simultaneously" limitation.

Bottom line: On the record retrieved, none of the cited references alone appears to anticipate claim 1 under § 102. Their principal force is under § 103, particularly the Kenington/Ubidyne (GB 2476393) + Nujira (GB 2488380) + Qualcomm envelope-tracker (US 2012/0326783) combination. Dependent claims 7–10 (power tracking amplifier, switcher, boost converter, VBAT/VBOOST selection) are the ones most exposed to § 102-style single-reference attacks because they map directly onto the Qualcomm envelope-tracker and tracking-supply references (US 2012/0326783; US 2011/0193629; US 2012/0326686).

6. Additional relevant art from the invalidity proceedings (not "citations," but prior art of record)

Because this patent was heavily litigated, the following were asserted as prior art (relevant to any § 102/§ 103 analysis) in the IPRs filed against it — IPR2018‑01326, ‑01327, ‑01328, ‑01329, ‑01330, ‑01340 (Petitioner: Intel/Unified Patents), per https://portal.unifiedpatents.com/[ptab](/ptab)/case/IPR2018-01327:

• Choi, IEEE MTT-S 2010 (envelope tracking)
• Wang, Design of Wide-Bandwidth Envelope-Tracking Power Amplifiers
• Dahlman, 4G: LTE/LTE-Advanced for Mobile Broadband (2011)
• Kim, IEEE MTT 2011
• Yu, EP 2 442 440
• Eliezer, US 2009/0004981
• Oga, US 2013/0034186
• Mahe, US 6,300,826 (the reference cited in the patent's own specification)
• Myers, US 5,929,702
• Wimpenny, US 2012/0229208
• Sebire, US 8,498,666
• Pelletier, US 2012/0140743

(These are the grounds papers listed at the URL above; the earlier single-reference ± combination structure is consistent with § 103 challenges rather than clean § 102 anticipation of claim 1.)

Related litigation of record on the Google Patents page: ITC 337‑TA‑1065; CAFC Nos. 22‑1824/1825/1826/1828/1829/1830; S.D. Cal. 3:17‑cv‑01375.

7. Confidence and caveats

• High confidence: patent number, title, inventor, assignee, dates, claim set, and the four-reference Google Patents family citation list (List B).
• Medium confidence: the itemized List A — it is a faithful transcription of what FreePatentsOnline renders, but the block is truncated (entries between US 2006/0264186 and US 6,009,090 are missing), and it contains one entry (US 2014/0111275) that postdates the priority date.
• I did not verify the actual disclosure content of each reference against claim 1 element-by-element; the § 102/§ 103 assessments above are based on the references' titles/abstracts and their characterized subject matter, not on full-text readings or file-wrapper examiner rejections. I explicitly cannot confirm from the retrieved material which references were applied in a § 102 rejection during prosecution — that would require the USPTO file wrapper (Patent Center) for application 13/764,328, which I did not retrieve.

Suggested authoritative sources for verification:

• USPTO PatentCenter / Global Dossier for application 13/764,328 (full (56) list and examiner citations) — https://patents.google.com/patent/US9608675/en (USPTO links)
• https://www.freepatentsonline.com/9608675.html
• https://patents.google.com/patent/CN104982077B/en and https://patents.google.com/patent/JP6490596B2/en (family citation lists)
• https://portal.unifiedpatents.com/ptab/case/IPR2018-01327 (IPR grounds/references)

I'll research the prior art cited on this patent to ground the obviousness analysis.

Let me look at the actual IPR grounds and key prior-art references to ground the § 103 analysis.

Let me identify the specific references the PTAB relied on in the IPRs.

Obviousness Analysis of U.S. Patent No. 9,608,675 Under 35 U.S.C. § 103

Patent: US 9,608,675 B2 ("Power tracker for multiple transmit signals sent simultaneously"), App. No. 13/764,328, filed 2013-02-11, priority 2013-02-11, inventor Alexander Dorosenco, assignee Qualcomm Inc.

Note on sourcing: The authoritative full text I was given does not itself contain a "Prior Art" listing; the prior-art references are those cited on the face of the patent (as reproduced on FreePatentsOnline's patent page) plus the references litigated in the IPRs. All references below are dated before the 2013-02-11 priority date unless noted. Where the live record (PTAB/Federal Circuit) contradicts a training assumption, I rely on the live record.

I. Legal framework

Obviousness is a question of law based on the four Graham v. John Deere factual inquiries: (1) scope and content of the prior art; (2) differences between the prior art and the claims; (3) level of ordinary skill; and (4) secondary considerations. KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007). A combination of references renders a claim obvious when the references are analogous, the combination is of known elements to achieve predictable results, and a POSITA would have had a motivation to combine with a reasonable expectation of success. Critically, the record here is not hypothetical — the PTAB reached a § 103 conclusion on this exact patent, and the Federal Circuit affirmed at least the principal ground.

II. Level of ordinary skill in the art (POSITA)

A POSITA here would be an RF/analog circuit designer with a bachelor's degree in electrical engineering (or equivalent) and ~3–5 years of experience in power-amplifier (PA) and envelope/power-tracking design for wireless transceivers, including carrier aggregation and quadrature (I/Q) modulation. The Board's IPR findings employed substantially this level of skill.

III. Claim 1 and the point of novelty

Claim 1 requires:

1. a power tracker determining a single power-tracking signal based on I and Q components of a plurality of carrier-aggregated transmit signals sent simultaneously, the signal being generated from a combination of the I/Q components, where the signals comprise OFDM or SC-FDMA signals;
2. a power supply generator generating a single supply voltage from that signal; and
3. a power amplifier receiving the single supply voltage and the simultaneous carrier-aggregated signals to produce a single output RF signal.

The alleged invention (per the specification) is the shift from the FIG. 4 architecture (one PA + one tracker per transmit signal) to a single PA with a single common power tracker across all simultaneous signals (FIGS. 5–6). The two disclosed computations are Eq. (1) (sum of per-signal powers, then √) and Eq. (2) (sum of per-signal magnitudes √(I²+Q²)). The specification concedes (as the Board noted, Ex. 1001, 8:33–36) that both are merely "exemplary."

This is a classic combination of known elements case. The two building blocks — (a) common/shared envelope tracking for multiple concurrent signals driving one PA, and (b) computing a PA supply control signal from baseband I/Q — were each independently known.

IV. Ground A — Yu in view of Wang (the ground the Federal Circuit affirmed)

Yu discloses a power-amplifier system in which multiple input signals S₁, S₂, … are transformed by a signal-processing unit into an RF signal S_RF fed to a single power amplifier PA, and a control unit (DSP) derives a control signal CTRL based on the multiple input signals to control the PA's supply voltage module. Yu expressly states its principle "may be considered as an extension to the known principle of envelope-tracking amplifiers," and is especially advantageous "where more than one input signal S₁, S₂, … is to be processed to obtain said RF signal S_RF."

Wang discloses an envelope-tracking PA system in which the input is a complex baseband signal whose amplitude is A = (I² + Q²)^{1/2} — i.e., deriving the supply-control quantity from the I and Q components. This is precisely the "combination of the plurality of I and Q components" limitation, and the mathematical foundation of the patent's Eq. (1)/(2).

Why the combination is obvious. Yu supplies every structural element of claim 1 except the specific "I/Q-derived" nature of the tracking signal; Wang supplies that with a one-for-one substitution of a known envelope-detection technique (I/Q magnitude) for Yu's generic envelope derivation. A POSITA seeking to implement Yu's multi-signal supply control would naturally compute the controlling envelope from the I/Q baseband of each signal — the standard, well-understood modem interface — yielding the claimed "single power tracking signal based on a combination of the plurality of I and Q components." The Board so held, and the Federal Circuit affirmed the Board's finding that a POSITA "would have been motivated to modify Yu" and that the Yu-based claims were unpatentable (No. 22-1824, Op. 9/23/2024). This ground alone disposes of claims 1–5, 17–25, and 27–33 as previously held, with the dependent limitations supplied as set out in §VII.

Caveat: The precise identity/citation of "Yu," "Wang," and "Chen" as litigated is from the IPR record (Ex. 1001/1004/1005, 1104/1205/1212) and secondary reporting; I could not independently verify every exhibit number from the two search snippets. The substance of each reference and the Board/Court's reliance is corroborated by the Federal Circuit opinion text.

V. Ground B — Chen in view of Wang (concurrent dual-band envelope tracking)

Chen (a paper) proposes a hybrid envelope-tracking scheme for concurrent dual-band PAs: input signals 1 and 2 at different frequencies are envelope-detected (detectors 1, 2), weighted, and added by an envelope combiner whose output feeds an envelope amplifier that modulates the supply of a single dual-band PA; on the parallel RF path the signals are upconverted and added by a power combiner feeding the same dual-band PA. That maps almost verbatim onto claims 1, 5, and 6:

• envelope detectors/combiner → "power tracker … based on a combination of the plurality of I and Q components";
• single envelope amplifier → "power supply generator … single power supply voltage";
• dual-band PA fed by the power combiner → "power amplifier … produce a single output RF signal";
• upconverters + power combiner → claim 5; the combined multi-band input → claim 6.

Wang again supplies the I/Q magnitude computation. Motivation to combine: Chen and Wang are in the same field (ET PAs) and would have been recognized as solving the same problem — computing an ET control signal for a multi-band/complex-modulated PA. Substituting Wang's I/Q-based envelope derivation into Chen's envelope combiner is a predictable use of a known technique to achieve a known result (KSR).

The Board initially found the Chen-based claims unpatentable, but (per the 9/23/2024 opinion) later concluded Intel had not shown Chen's teachings were applicable to a mobile device — so the Chen ground is weaker / failed on that specific record. Because the "mobile/handset" limitation is not recited in most claims and is at most an intended use, I would still assert the Chen ground for the apparatus claims, while noting the litigation posture.

VI. Ground C — Kenington in view of Wang (and Mahe for claims 7–10)

Kenington (US 2011/0151806 A1; GB 2476393 A; and the MIMO-related GB 2488380 A) discloses a multiple envelope-tracking system for an antenna array with a plurality of transmission paths and a single variable common power supply driven by a common envelope signal "representative of the envelope of one of the transmission path signals, or of even all of the transmission path signals." This is the core inventive concept the '675 patent claims as its own — a single supply generator serving multiple simultaneous transmit paths. Indeed, Kenington was cited during prosecution, and the Examiner found "Kenington teaches the power tracking signal based on I and Q" (Ex. 1002 at 203, quoted in the IPR decisions).

Adding Wang (I/Q envelope computation) to Kenington yields claim 1's "single power tracking signal based on a combination of the plurality of I and Q components"; the motivation is atextual and strong: Kenington itself touts the efficiency benefit (and DAC interface) of a shared envelope-tracking supply. Kenington/GB 2488380 additionally renders claim 11 (multiple carriers at different frequencies) and the common-supply concept obvious.

For claims 7–10 (power tracking amplifier, switcher sensing current, boost converter, battery/boost selection), the Mahe reference (US 6,300,826, "Apparatus and Method for Efficiently Amplifying Wideband Envelope Signals"), which the '675 specification itself cites as the implementation of power-tracking amplifier 810, discloses the linear power-tracking amplifier plus switcher/current-sensing hybrid, and Mathe (US 2012/0326783, "Low-voltage power-efficient envelope tracker") and Dai (US 2012/0326686, "Power supply generator with noise cancellation") disclose boost/battery-selection supply generators. Combining these known supply topologies with Yu/Chen/Kenington is a routine design choice with predictable results (KSR).

VII. Dependent-claim mapping

VIII. Motivation to combine / KSR rationales

1. Same field, same problem, same solution. Yu, Chen, Wang, Kenington, Mahe, Mathe all address ET/power-tracking for complex-modulated (OFDM/SC-FDMA) PAs; all are reasonably pertinent to the problem the '675 patent addresses (efficiently supplying a PA transmitting simultaneous multi-carrier signals).
2. Predictable results. Summing envelopes/powers and driving a single common supply is a straightforward aggregation; the patent itself concedes both equations are "exemplary" and that bandwidth narrowing is a recognized byproduct.
3. Express teaching in the art. Yu expressly teaches its benefit for "more than one input signal"; Kenington expressly teaches sharing one variable supply across multiple paths; Chen expressly teaches an envelope combiner for concurrent dual-band operation. These are explicit "combination hints," the strongest form of motivation.
4. Known substitute technique. Deriving the controlling quantity from baseband I/Q (Wang) is the standard, well-understood modem interface; its substitution for a generic envelope detector is the paradigm of obviousness.

IX. Secondary considerations

The record I have does not contain evidence of unexpected results, industry praise, licensing due to the merits, or long-felt need tied to the claimed combination. Commercial success in carrier-aggregation handsets is attributable to many factors (and is coextensive with the prior art's teachings), so it would not rebut the strong prima facie case. Notably, the asserted commercial/industry significance did not prevent the PTAB from finding the claims unpatentable.

X. Litigation posture (live record)

• Intel filed six IPRs (IPR2018-01326, -01327, -01328, -01329, -01330, -01340). The Board's initial final written decisions held all challenged claims unpatentable.
• On appeal, the Federal Circuit vacated and remanded the "plurality of carrier aggregated transmit signals" construction on APA/notice grounds (Qualcomm v. Intel, No. 2020-1589, July 27, 2021) — a procedural, not merits, ruling on that issue.
• On remand (Nos. 22-1824 through 22-1830, decided Sept. 23, 2024), the court affirmed the Board's factual findings and the Yu-based unpatentability conclusion (claims 1–5, 17–25, 27–33), while the Chen-based ground failed on the "mobile device" motivation issue.
• Parallel ITC proceeding 337-TA-1065 and the Reexam/appeals family exist; a broader family continuation (US 11,133,833; 11,641,215; 11,996,872; 12,562,763) shares the same specification.

Bottom line: The claims of US 9,608,675 are prima facie obvious under § 103. The strongest, litigation-tested ground is Yu + Wang for claim 1 and its dependents (affirmed by the Federal Circuit). Chen + Wang and Kenington + Wang (+ Mahe for claims 7–10) provide independent, well-motivated grounds covering the same subject matter, with the enumerated secondary references supplying every dependent limitation.

Confidence caveats: I have high confidence in the claim scope, the reference content as described, and the KSR/§ 103 reasoning. I have somewhat lower confidence in the exact exhibit numbers/identities of the litigated "Yu," "Wang," and "Chen" references (they are drawn from IPR documents surfaced by search, not the authoritative patent text), and in the current claim-by-claim survival status after the 2024 appellate disposition — a post-remand claim-by-claim review of the Board's decisions would be required for a definitive FTO conclusion.