US 9319075

I'll search for this specific patent and any related 2026 appellate activity.

Let me retrieve the full claim set and check for any other 2026 appellate activity.

I have the key case law. Let me try to pin down dependent claims 2–5 text.

US 9,319,075 B1 — Analyst Summary

Search scope note: I searched for the literal identifier 9319075 and did not substitute similar numbers. Results below are from Google Patents, Justia, uspto.report, Unified Patents, Docket Alarm, the PTAB Final Written Decision in IPR2022-01223, the CAFC opinion in No. 25-1090, and the CAFC's own docket page. I did not run an exhaustive line-by-line sweep of every CAFC 2026 docket entry; I confirmed the '075 patent's appellate activity via targeted searches plus the Federal Circuit's published opinion list.

Also flagged: you stated today is April 26, 2026. My retrieval environment returned material dated August 27–28, 2026 (including a CAFC opinion in this very case). I report what the sources show rather than reconciling the dates.

1. Bibliographic data

Uncertainty on priority date: the patent's own text traces the earliest priority to June 4, 1999, and the Federal Circuit opinion in No. 25-1090 says the '075 patent "claim[s] priority to a 1999 application." However, Google Patents displays a "prior art date" of 1996-12-16 and Unified Patents lists a 1996-12-15 priority date. These conflict with the intrinsic record in the patent document. Treat 1999-06-04 as the best-supported date, with the 1996 entries unexplained.

2. Abstract (as printed on the face of the patent)

"A method and apparatus in which multiple Internet Protocol (IP) based wireless data transmissions are simultaneously provided between a wireless device and a server, including providing multiple antennas, multiple T/R units, multiple processors and multiple I/O ports on the wireless device. The method includes receiving multiple IP data packets on the I/O ports at substantially the same time, and sending multiple data packets from the wireless device to the server, whereby the transmission rate between the wireless device and the server is increased."

3. Independent claim in plain language

Only one independent claim — claim 1, an apparatus ("device") claim. (The Federal Circuit opinion states: "Claim 1 of the '075 patent is the only independent claim at issue here." The patent has 5 claims total.)

Claim 1 is long and dense. Broken into its lettered elements as the Federal Circuit reproduced them:

Preamble — "An IP-enabled communication device for multiplexing signals comprising:"
This is a device claim, not a method claim, framed around multiplexing. The preamble term "multiplexing signals" proved decisive in litigation (see §5).

[1a] Multiple radios with multi-frequency, multi-protocol support. The device has a plurality of wireless communication units and, as a whole, supports multiple transmit and receive frequencies and multiple wireless protocols. Plainly: not a single-radio phone — several radios, more than one band, more than one air interface standard.

[1b] Separate antenna sets, separate networks. A first wireless communication unit is coupled to a first set of antennas operating on a first network; a second unit couples to a second set of antennas on a second network. Plainly: each radio has its own antenna group and its own network.

[1c] RF communication. At least one of the units is configured for radio-frequency communication.

[1d] Complementary frequency relationship. The first unit is configured to operate at a lower frequency than the second unit, such that the two "operate as complementary systems and reduce interference with each other." Plainly: the two radios are deliberately separated in frequency so they don't step on each other.

[1e] Failover / signal-strength-responsive path steering. The first unit talks to a remote server over the first network path; the second unit communicates with that same remote server over the second path in response to a change in signal strength and/or connectivity of either unit. Plainly: automatic switching (or adding) of a second path when the first degrades — the "in response to" trigger was a live dispute point in the IPRs.

[1f] Simultaneous video/audio with per-path optimization. Video or audio can be accessed simultaneously, "with performance optimized for each through dedicated or multiplexed paths."

[1g] and [1h] Simultaneous multi-path IP transport. In communication with the remote server, the mobile device (i) receives multiple IP data packets on a plurality of ports at substantially the same time and (ii) sends multiple data packets to the server, allowing multiple simultaneous communication paths between device and server. Elements [1g] and [1h] are substantively duplicative — the same limitation appears twice in the claim, which the PTAB and Federal Circuit both reproduced without comment. Plainly: the device aggregates parallel IP connections to the same server to raise throughput.

Note on claim construction: the PTAB and the Federal Circuit construed "multiplexing signals" (the preamble term, which the Board treated as limiting) to mean "combining multiple signal streams into one" — requiring combination into a composite before simultaneous transmission over a shared pathway. A "switching" reading (different signals merely sharing a path sequentially) was rejected. (Sources: PTAB Final Written Decision, IPR2022-01223; Apple Inc. v. Smart Mobile Techs. LLC, No. 25-1090, slip op. (Fed. Cir. Aug. 27, 2026).)

4. Dependent claims 2–5 — not verbatim-verified

I was unable to retrieve the authoritative verbatim text of claims 2, 3, 4, and 5 from the sources available to me. The Google Patents rendering of the claim set was truncated in my retrieval, and my targeted searches for the dependent-claim language returned unrelated patents.

What I can state with confidence from the PTAB and CAFC records:

• The patent has 5 claims total.
• Apple's IPR petition in IPR2022-01223 challenged claims 1–3 and 5. Claim 4 was not challenged in that proceeding. The Federal Circuit affirmed the Board's finding that claims 1–3 and 5 were not shown unpatentable.
• I do not have verified subject matter for claims 2–5 and will not guess at their content.

5. Litigation and 2026 appellate posture

District court. Smart Mobile Technologies LLC v. Apple Inc., No. 6:21-cv-00603 (W.D. Tex., filed June 11, 2021) — part of a 13-patent assertion. The case was transferred out of Texas (docketed as 3:23-cv-04168 or 4:23-cv-04168, N.D. Cal.) and stayed pending the IPRs. (Minor source conflict: the Federal Circuit opinion says the case was "transferred … to the Central District of California," while IPWatchdog and the Google Patents case link indicate the Northern District of California. Unresolved.) Google Patents also links a CAFC case 25-1090 and a Darts-IP "first worldwide family litigation" record (family 34221038).

IPR. Apple Inc. v. Smart Mobile Technologies LLC, IPR2022-01223 (PTAB). Institution was denied (Jan 30, 2023), then granted on rehearing. Real parties in interest included Apple, Samsung Electronics Co., Ltd., and [Samsung Electronics America, Inc.](/litigations/by-plaintiff/Samsung%20Electronics%20America%2C%20Inc.) Final Written Decision issued September 25, 2024, determining no challenged claim (1–3, 5) unpatentable. The Board directed briefing on a construction of "multiplexing."

Federal Circuit (the 2026 activity). Apple Inc. v. Smart Mobile Technologies LLC, No. 25-1090 (Fed. Cir.), opinion issued August 27, 2026, nonprecedential, authored by Judge Stark for a panel with Judges Reyna and Hughes. The court affirmed the Board — Apple failed to show claims 1–3 and 5 of the '075 patent obvious over the Yegoshin / Bernard combination. The court adopted the "combining multiple signal streams into one" construction of "multiplexing signals" and incorporated its analysis from the companion appeal. Oral argument audio was posted June 2, 2026.

Companion appeal. Apple Inc. v. Smart Mobile Technologies LLC, No. 24-1607 (Fed. Cir., also decided Aug. 27, 2026), addressing U.S. Patents 8,842,653 and 9,019,946 — same family, same specification. There, Apple's appeal was affirmed but Smart Mobile's cross-appeal on claim 17 was vacated and remanded for an APA/motivation-to-combine defect (the Board did not engage Smart Mobile's "bottlenecking" evidence).

Related IPRs on sibling patents (not the '075 patent): IPR2022-00766, -01004, -01005, -01222, -01248, -01249.

Key URLs:

• https://patents.google.com/patent/US9319075/en
• https://patents.justia.com/patent/9319075
• https://uspto.report/patent/grant/9,319,075
• https://portal.unifiedpatents.com/patents/patent/US-9319075-B1
• https://portal.unifiedpatents.com/ptab/case/IPR2022-01223
• https://www.cafc.uscourts.gov/08-27-2026-25-1090-apple-inc-v-smart-mobile-technologies-llc-opinion-25-1090-opinion-8-27-2026_2746224/
• https://law.justia.com/cases/federal/appellate-courts/cafc/25-1090/25-1090-2026-08-27.html
• https://ipwatchdog.com/2026/08/27/cafc-delivers-blow-to-apple-in-appeal-of-ptab-disputes-with-smart-mobile/
• https://gravesshaw.com/gs-obtains-ipr-win-for-smart-mobile/

6. Explicit uncertainty register

1. Issue date: 2016-04-19 (majority of sources, including the PTAB's own statement) vs. 2016-04-18 (Unified Patents). Majority date used.
2. Filing date: 2015-04-13 vs. 2015-04-12 (Unified Patents).
3. Priority date: 1999-06-04 per the patent's cross-reference and the Federal Circuit; 1996-12-15/16 per Unified Patents and Google Patents' metadata. Unreconciled.
4. Dependent claims 2–5 text: not retrieved; not fabricated.
5. Transfer district: C.D. Cal. vs. N.D. Cal. conflict in sources.
6. Date context: the retrieved CAFC disposition (Aug. 27, 2026) postdates your stated "current date" of April 26, 2026. If your date is authoritative, the 25-1090 opinion and the Aug. 2026 round of coverage would be future-dated and should be treated as such.
7. Claim 1 anomaly: elements [1g] and [1h] appear to recite the same limitation twice. This is reproduced faithfully from the Federal Circuit's quotation of the claim, but I have not independently confirmed it against the printed patent column.

I'll research litigation involving US patent 9319075. Let me search multiple sources.

Let me gather more details on the specific cases.

Let me get details on the Federal Circuit appeal and the current status of the district court cases.

Litigation Involving U.S. Patent No. 9,319,075 ("the '075 patent")

The patent. U.S. 9,319,075 B1, "Wireless devices with transmission control and multiple internet protocol (IP) based paths of communication," inventors Raman K. Rao, Sanjay K. Rao, and Sunil K. Rao; granted April 19, 2016. Current assignee of record: Smart Mobile Technologies LLC (originally IP Holdings, Inc.). All litigation below traces to the same enforcement campaign by Smart Mobile against mobile-device makers.

I limited my search to the exact identifier 9319075 / 9,319,075 and did not include results for similar numbers (e.g., 9,319,079, 9,319,075-family siblings such as 8,842,653, 9,019,946, etc.).

1. District court litigation

Smart Mobile Technologies LLC v. Apple Inc.

The '075 patent was asserted as one of 13 patents in the complaint (attached as Exhibit K). Apple moved to transfer venue under 28 U.S.C. § 1404(a); the transfer was granted, and the case was reassigned to Judge Alsup. On September 19, 2023, Judge Alsup granted Apple's unopposed motion to stay the case "pending issuance of the Final Written Decisions in the instituted IPR proceedings against the Asserted Patents." Reporting as of the Federal Circuit's August 2026 decisions indicates the N.D. Cal. case remains stayed.

Note on a source conflict: the docket numbers and orders consistently show the Northern District of California (3:23-cv-04168, Judge Alsup, San Francisco). One secondary article (VitalLaw, Aug. 2026) states the case was transferred to the Central District of California. The court records I found do not support the Central District reference; I treat N.D. Cal. as correct.

Smart Mobile Technologies LLC v. Samsung Electronics Co., Ltd. et al.

Important caveat: I could not confirm that the '075 patent itself was asserted against Samsung. The complaint/venue record tied to the '075 patent by Google Patents and Unified Patents lists only the Apple W.D. Tex. case. Moreover, Smart Mobile's Preliminary Infringement Contentions in the Samsung case list twelve patents (per Exhibits A–L: 8,442,501; 8,472,936; 8,472,937; 8,761,739; 8,824,434; 8,842,653; 9,019,946; 9,049,119; 9,084,291; 9,191,083; 9,614,943; 9,756,168) — the '075 patent is not among them. The two suits were declared "related matters" in the IPRs, but that relatedness appears to stem from the shared patent family, not identical asserted patents. Treat the Samsung case as family-adjacent, not a confirmed '075 assertion.

2. PTAB — Inter Partes Review (this is an IPR on the '075 patent itself)

Apple Inc. v. Smart Mobile Technologies LLC, IPR2022-01223

Apple challenged claims 1–3 and 5 over combinations of prior art including Yegoshin and Bernard. The Board found Apple did not prove the claims unpatentable. Apple appealed.

Source conflict flagged: one PTAB document I retrieved carries a "Decision Denying Institution of Inter Partes Review" caption bearing the same IPR number and patent. I cannot fully reconcile that caption with the Unified Patents record (institution 2023-09-28) and the September 25, 2024 Final Written Decision that Apple then appealed. The existence and content of the FWD and the ensuing appeal (below) are well corroborated; the "denial" caption may be an OCR/mislabel artifact or relate to an earlier petition.

3. Federal Circuit appeals

Apple Inc. v. Smart Mobile Technologies LLC, No. 25-1090 (Fed. Cir.)

The court affirmed the Board, holding that substantial evidence supported the finding that Apple failed to prove claims 1–3 and 5 of the '075 patent unpatentable. The decision turned on the construction of "multiplexing signals" — construed to mean "combining multiple signal streams into one," thus excluding signals merely sent sequentially over the same path. The companion appeal, No. 24-1607 (concerning U.S. 8,842,653 and 9,019,946), was decided the same day: Apple's appeal was affirmed, but Smart Mobile's cross-appeal was vacated and remanded on an Administrative Procedure Act "bottlenecking" issue.

Summary of the '075 patent's litigation footprint

The '075 patent's known enforcement history is a single dispute chain against Apple, plus the associated IPR and appeal:

1. W.D. Tex. 6:21-cv-00603 (filed 6/11/2021) → transferred → N.D. Cal. 3:23-cv-04168 (Judge Alsup), stayed 9/19/2023.
2. IPR2022-01223 (petition 7/6/2022; FWD 9/25/2024) — Apple/Samsung RPI; no challenged claim held unpatentable.
3. Fed. Cir. No. 25-1090 — affirmed the PTAB on 8/27/2026.

No other district-court cases or PTAB proceedings asserting the '075 patent itself were found. The Samsung case (6:21-cv-00701) is part of the same patent family but does not confirmably involve the '075 patent.

Sources:

• Google Patents, US9319075B1 (family litigation links): https://patents.google.com/patent/[US9319075](/patent/US9319075)/en
• Unified Patents, patent record: https://portal.unifiedpatents.com/patents/patent/US-9319075-B1
• Unified Patents, IPR2022-01223: https://portal.unifiedpatents.com/ptab/case/IPR2022-01223
• CourtListener, N.D. Cal. docket 3:23-cv-04168 (stay order, Doc. 146): https://www.courtlistener.com/docket/67698985/smart-mobile-technologies-llc-v-apple-inc/
• Docket Alarm, W.D. Tex. 6:21-cv-00603 records: https://www.docketalarm.com/search/?q=patent%3A%289319075%29
• CAFC, 25-1090 opinion (Aug. 27, 2026): https://www.cafc.uscourts.gov/08-27-2026-25-1090-apple-inc-v-smart-mobile-technologies-llc-opinion-25-1090-opinion-8-27-2026_2746224/
• Fed Circuit Blog, Opinions & Orders (Aug. 27, 2026): https://fedcircuitblog.com/2026/08/27/opinions-orders-august-27-2026/

Two caveats I want to be explicit about: (1) Case statuses change quickly; the N.D. Cal. stay may have been lifted or the case dismissed after the IPRs concluded — the latest reporting I found still described it as stayed. (2) I could not verify the '075 patent's presence in the Samsung suit; if you need that confirmed, the operative Samsung infringement contentions and complaint would be the authoritative check.

I'll investigate the discrepancy between the ODP "no proceedings" block and the patent page's explicit reference to IPR2022-01223, plus the Federal Circuit docket.

The ODP block says "no proceedings," but the patent page itself and multiple public sources show an IPR. Let me pin down the procedural history — particularly an apparent denial-then-institution oddity.

PTAB / AIA Trial Analysis — U.S. Patent No. 9,319,075 B1

⚠️ Source conflict flagged up front. The supplied structured block states that the USPTO ODP API returns no AIA trial proceedings for this patent. That is not correct as a matter of public record. Google Patents' own family page for US9319075 lists "PTAB case IPR2022-01223 filed (Final Written Decision)" with an FWD, and the Board's papers, Docket Alarm, Unified Patents' PTAB portal, and two Federal Circuit opinions all confirm a full, tried IPR. I proceeded on the documentary record rather than the block. Treat the ODP block as an ingest/indexing gap (the case terminated 2024-09-25 and the patent has since been listed as expired — plausible causes of a stale or scope-limited ODP response), not as evidence of no PTAB activity. Anyone relying on the ODP block alone in a freedom-to-operate or defense posture would be materially misinformed.

Proceedings overview

**One AIA trial proceeding is on file for the '075 patent — Apple Inc. v. Smart Mobile Technologies LLC, IPR2022-01223 — and it is a total loss for the petitioner: institution was initially denied, then granted on rehearing, and the Final Written Decision of 2024-09-25 held that no challenged claim (claims 1, 2, 3, 5) is unpatentable, which the Federal Circuit affirmed on 2026-08-27 in Appeal No. 25-1090. Status count: active 0; claims invalidated 0; claims sustained 1 (0 independent claims canceled); settled 0; institution denied 1 (denied at Paper 13, but reversed by the Board itself on rehearing at Paper 15). Bottom line for a defendant: this is a hardened patent, not a paper tiger. Independent claim 1 and dependent claims 2, 3, and 5 have survived a fully tried IPR and Federal Circuit review; the asserted art (Yegoshin, Johnston, Bernard, Preiss) has been litigated to final judgment and rejected. Your realistic attack surface is claim 4 (never challenged), non-IPR invalidity theories (§ 101 / § 112) that IPR cannot reach, or new art that is not one of those four references — and you may also be estopped if you are a privy of Apple or Samsung.

IPR2022-01223 — Apple Inc. (with Samsung Electronics Co., Ltd. and Samsung Electronics America, Inc. as named real parties-in-interest) v. Smart Mobile Technologies LLC

• Patent: U.S. Patent No. 9,319,075 B1 ("Wireless devices with transmission control and multiple internet protocol (IP) based paths of communication") — the only independent claim is claim 1; the patent contains 5 claims total.
• Type: Inter Partes Review (35 U.S.C. §§ 311–319). No PGR, no CBM, and no second IPR on this patent appears on file. Caveat: I could not positively exclude a pre-2018 CBM/PGR or a second, never-instituted petition using the sources available; what I can say is that no such proceeding appears in the Board docket, Docket Alarm's patent-scoped search for patent:(9319075), or Unified Patents' PTAB record for the '075.
• Filed: 2022-07-06 (Paper 2, Petition).
• Status: Terminated — Final Written Decision (Paper 35, 2024-09-25): "Determining No Challenged Claim Unpatentable 35 U.S.C. § 318(a)." Docket status: Final Written Decision – Appealed. Patent-level status: Expired – Fee Related (see "Defensive value" / next steps for why this matters).
• Judge panel:
  • Papers 13 and 15 (denial and rehearing/institution): Hyun J. Jung, Nathan A. Engels, and Paul J. Korniczky, APJs (Korniczky authored).
  • Paper 20 — Order Panel Change (2024-01-08) — reconfigured the panel; the oral hearing held 2024-07-10 was conducted before Jameson Lee, Hyun J. Jung, and Nathan A. Engels, APJs (transcript at Paper 34).
  • The docket metadata aggregates five judges across the case's life (Jung, Lee, Engels, Korniczky, Ullagaddi). I could not confirm from the sources I reached which three judges authored the FWD itself — do not quote a panel for the FWD without pulling Paper 35 directly.
• Petition grounds: A single ground — § 103 obviousness of claims 1–3 and 5 over the combination of Yegoshin (US 6,711,146 B2), Johnston (US 5,784,032), Bernard (US 5,497,339), and Preiss (US 6,031,503). No § 102 anticipation ground and no § 112 ground (as expected — § 112 is not available in IPR). Claim 4 was not challenged.
• Institution decision: Denied at Paper 13, then instituted 2023-09-28 on rehearing — this is the unusual and important part of the file history:
  • Paper 13 (Institution Decision, date not confirmed in my sources but in the record before 2023-05-26): the Jung/Engels/Korniczky panel found no reasonable likelihood of prevailing, because Apple had not adequately shown that Yegoshin discloses the claimed "multiplexing [of] signals." The panel relied on Yegoshin's own disclosure that "if engaged with an IP call, an incoming cell call would get a busy signal and so on, or it would be redirected to the IP call point" — i.e., Yegoshin teaches alternation/redirection, not simultaneous combination of signal streams. (Board decision, IPR2022-01223)
  • Paper 14: Apple filed a Request for Rehearing under 37 C.F.R. § 42.71(d).
  • Paper 15 — Decision Granting Petitioner's Request for Rehearing; Granting Institution of Inter Partes Review (2023-09-28): the same three-judge panel granted rehearing and instituted trial on all challenged claims on the only ground raised in the Petition. (Paper 15 PDF)
  • Procedural takeaway: this patent was on the verge of escaping IPR entirely, and only the Board's own reconsideration put it into trial. That history is a template a new petitioner can try to exploit (or must expect the patent owner to exploit against it).
• Trial milestones: Order Setting Oral Argument (Paper 29) 2024-05-29; oral hearing 2024-07-10; hearing transcript entered 2024-09-24 (Paper 34); FWD 2024-09-25 (Paper 35) — 363 days after institution, inside the statutory one-year deadline of § 316(a)(11). (Hearing transcript PDF)
• Final Written Decision (verdict at claim level):
  • All challenged claims — claims 1, 2, 3, and 5 — were held NOT unpatentable. No independent claim canceled, no dependent claim canceled. The FWD is styled "Determining No Challenged Claim Unpatentable 35 U.S.C. § 318(a)."
  • Claim 4 was not addressed — it was not challenged. Do not treat claim 4 as adjudicated in either direction.
  • The FWD turned on the construction of "multiplexing signals" in the preamble of claim 1 ("An IP-enabled communication device for multiplexing signals"). The Board effectively adopted the patent owner's narrow reading — combining multiple signal streams into one, requiring combination into a composite before simultaneous transmission over a shared pathway — and found the asserted art (Yegoshin, and Bernard's second embodiment in particular) did not disclose it.
  • Patent owner's counsel (Graves & Shaw LLP) publicly summarized the result: "the U.S. Patent Trial and Appeal Board issued a Final Written Decision confirming the patentability of all challenged claims … including … U.S. Patent No. 9,319,075." (Graves & Shaw)
• Settlement / termination: None. The case was terminated by FWD on the merits, not by settlement or adverse judgment. No termination-on-settlement papers appear on the docket.
• Appeal: Yes — appealed by Apple.
  • Docket: Apple Inc. v. Smart Mobile Technologies LLC, No. 2025-1090 (Fed. Cir.), appeal from PTAB IPR2022-01223. Oral argument audio posted 2026-06-02. (CAFC docket page)
  • Issues on appeal: (1) claim construction — whether "multiplexed/multiplexing signals" covers Apple's broad "switching" reading (multiple signals sharing a physical pathway, even sequentially) or the Board's narrower "combining multiple signal streams into one"; and (2) whether the Board's obviousness determination was supported by substantial evidence.
  • Disposition — AFFIRMED (2026-08-27), nonprecedential, authored by Circuit Judge Stark for a panel of Reyna, Hughes, and Stark: "Apple… appeals from a final written decision of the … Board, rejecting Apple's arguments that claims 1-3 and 5 of … U.S. Patent No. 9,319,075 … are unpatentable as obvious. We affirm." The court agreed with the Board's construction and incorporated "the discussion of the pertinent technology" from the companion opinion on U.S. Pat. Nos. 8,842,653 and 9,019,946 (No. 24-1607, decided the same day). (Opinion text)
  • No cross-appeal in 25-1090 (Smart Mobile had nothing to appeal — it won everything). The cross-appeal activity in the family was in the companion 24-1607 appeal, on other patents, where the court vacated the Board's judgment of unpatentability of claim 17 and dependents of the '653/'946 patents and remanded for consideration of Smart Mobile's "bottlenecking" evidence — that does not touch the '075.
• Defensive value: Very low for the defense. Claim 1 — the only independent claim — has been tested on the merits, sustained, and affirmed on appeal. Any obviousness theory built on the Yegoshin/Johnston/Bernard/Preiss combination is now a dead end; re-running it at the Board invites § 325(d) and General Plastic discretionary denial, and re-running it in district court runs into issue-preclusion-type arguments and the Board's and CAFC's construction of "multiplexing signals." If you are Apple or Samsung (or their privy), you are also barred by § 315(e)(2) today. Conversely, the file history gives the patent owner a strong story — "the patent survived a fully tried IPR and a Federal Circuit appeal" — and tells you the only realistic offsets are (i) the patent's apparent expired status, and (ii) invalidity theories IPR cannot reach.

Strategic summary

Claim status — what is canceled vs. sustained vs. untested. For the '075 patent there are no canceled claims. Claims 1, 2, 3, and 5 are SUSTAINED — challenged, tried, decided in the patent owner's favor on 2024-09-25, and affirmed by the Federal Circuit on 2026-08-27. Claim 4 is UNTESTED — Apple never challenged it, so the Board never addressed it; it is dependent on claim 1 and therefore inherits claim 1's survival, but nothing in the FWD or the CAFC opinion validates or invalidates it independently. If a demand letter or complaint asserts claims 1, 2, 3, or 5, you are facing a patent whose core validity has already been adjudicated against a sophisticated, well-funded petitioner. Note also that the '075 family is under concurrent attack on other members, and the results are genuinely mixed: the Board invalidated claims 1–3, 5, 6, 10, 13, and 16–18 of the '937 patent (IPR2022-00982), and the Federal Circuit applied issue preclusion to hold claim 1 of the '936 patent unpatentable (No. 24-1352, 2026-01-21) — but none of that reaches the '075, which has a different independent claim and, as construed, a meaningful "multiplexing" limitation that the Board found the art does not disclose.

Estoppel landscape. Because the patent owner won, § 315(e)(2) estoppel is triggered against Apple Inc., Samsung Electronics Co., Ltd., and Samsung Electronics America, Inc. — all three are named as real parties-in-interest on the IPR2022-01223 Petition — and their privies. As of the FWD date, those parties cannot assert in a civil action or ITC proceeding any ground they raised or reasonably could have raised, meaning the entire Yegoshin / Johnston / Bernard / Preiss combination and any other patents or printed publications a skilled searcher would have located on claims 1–3 and 5. Practically, for Apple and Samsung the IPR route on this patent is closed. (There is a district-court split on whether § 315(e)(2) attaches at the FWD or only after appeal finality; here that question is academic — the appeal is decided and the FWD is final, so estoppel is live regardless of which rule the court follows.) For a non-privy defendant, estoppel does not apply, but the practical doors are narrow: (i) Section 315(b)'s one-year bar runs from service of a complaint; (ii) § 325(d) and General Plastic discretionary-denial factors weigh heavily against a follow-on petition asserting the same art and the same "multiplexing" theory the Board already rejected; and (iii) the Board's construction of "multiplexing signals" is now affirmed law for this claim — any new art must disclose combining multiple signal streams into one composite for simultaneous transmission, not merely sequential sharing of a path. What remains genuinely open to a non-privy is what IPR structurally cannot reach: § 101 eligibility and § 112 written-description / enablement / indefiniteness attacks in district court, plus prior-art system/product art under § 102(a)/(b), which is not a "patent or printed publication" for estoppel purposes to the extent it was never available at the Board.

Pattern signals. Apple is the petitioner of record, but this was a coordinated Apple+Samsung campaign against the Smart Mobile portfolio — the same Apple petitions across the family named Samsung as an RPI, and Samsung separately filed its own IPRs (IPR2022-00766, -01003, -01004, -01005, -01248, -01249) on sibling patents. So the "same petitioner filed multiple IPRs" signal is yes, across the family, and the corporate bloc matters for estoppel and privity analysis. There is no defensive aggregator in the chain. The references to "Unified Patents" in the patent's public metadata are data-license attributions ("Unified Patents PTAB Data," "Unified Patents Litigation Data") for the portal that indexed the case — Unified Patents did not petition in IPR2022-01223 and is not a party. The patent owner, by contrast, has litigated aggressively and effectively: it defended the '075 to a complete win at the Board, through a rehearing rescue of an institution denial, and through Federal Circuit affirmance, while also successfully resisting institution outright in this case's first pass. It has also been on the receiving end of adverse family rulings and litigated those appeals rather than settling.

Recommended next steps

1. If you are a defendant and the patent has claims invalidated — it doesn't here. There is nothing to quote for your side. The operative dispositions you must work around are:
   • FWD (Paper 35, 2024-09-25): "Final Written Decision Determining No Challenged Claim Unpatentable 35 U.S.C. § 318(a)" — claims 1–3 and 5 sustained. Docket entry via Docket Alarm, IPR2022-01223; pull the paper itself through PTAB E2E or the Unified Patents PTAB portal.
   • CAFC No. 25-1090 (2026-08-27), nonprecedential: "We affirm." — opinion.
   • Do not represent to a court or an adversary that any claim of the '075 was canceled. That is wrong, and the distinction between this patent and the invalidated siblings ('501, '739, '937, '936) is exactly where the analysis turns.
2. Active proceedings: none. IPR2022-01223 is terminated; the appeal is decided. There are no PTAB trial-stage milestones to calendar. The only live timers are appellate: the 14-day window for panel/en banc rehearing from 2026-08-27 has effectively run, leaving only Apple's Supreme Court certiorari window (roughly 90 days from judgment, i.e., on or about 2026-11-25). A cert petition would not automatically stay the FWD or suspend its estoppel effect, so treat the Board's judgment as final for defense-planning purposes.
3. Attack the patent's expired status first — this is likely your most valuable, least-litigated point. Google Patents and Unified Patents both record the '075 as expired on/around 2019-06-04 (20 years from the 1999-06-04 application), and Google shows a legal status of "Expired – Fee Related." If that is accurate, a damages case is limited to pre-expiration infringement within the 35 U.S.C. § 286 six-year lookback, and no injunction is available. Verify directly against the USPTO Patent Center maintenance-fee and expiration records before relying on it — I am reporting what the public records state, not offering a legal conclusion, and the family's 1996/1999 priority chain makes the expiration calculus non-obvious.
4. If you are a non-privy and still want a validity challenge, target the gaps, not the walls: (a) claim 4, which has never been construed or litigated; (b) § 101 Alice/Mayo and § 112 — theories no IPR can hear and therefore no estoppel can reach; and (c) prior-art patents/printed publications that are not Yegoshin, Johnston, Bernard, or Preiss, supported by art that could not reasonably have been found in 2022 — all while expecting the Board to weigh § 325(d) and General Plastic against you given the identical "multiplexing" issue was already decided.
5. If you are Apple, Samsung, or a supplier/contract manufacturer/co-defendant in privity with them, assume § 315(e)(2) estoppel applies now and do not plan on a Board challenge to claims 1–3 or 5 — plan the defense around non-estopped theories, license/settlement valuation against an expired patent, and the preclusive effect of the Board's "multiplexing signals" construction in the stayed litigation.
6. Litigation context to monitor (not PTAB, but it drives the demand side): Smart Mobile Technologies LLC v. Apple Inc., originally W.D. Tex. 6:21-cv-00603 (filed 2021-06-11), transferred and docketed as 4:23-cv-04168 (CourtListener). The case was stayed pending the IPRs (UniCourt docket entry: Order Granting motion to stay). Sources conflict on the transferee court — Google Patents and CourtListener reflect the Northern District of California (4:23-cv-04168), while IPWatchdog and Bloomberg's IP Law Daily describe a transfer to the Central District of California. Confirm the current venue from the docket before citing it. With the '075 IPR now final in the patent owner's favor, expect the patent owner to press to lift the stay as to the surviving '075 claims while the invalidated siblings drop out.

I'll research the assignment record for US 9,319,075, focusing on reel/frame data and correspondents of record.

Let me search for the specific assignment records and correspondents.

Let me try to pull the USPTO Assignment Center records directly.

US 9,319,075 B1 — Assignment / Chain-of-Title Analysis

Retrieval caveat (read first). My tools returned search-index snapshots, not the live USPTO Assignment Center record. I was able to confirm the sequence and parties of the recorded assignments (Google Patents "reassignment" legal events, which mirror the USPTO assignment record, plus Unified Patents), but I was not able to retrieve reel/frame numbers, execution dates, or correspondent-of-record for any link in this chain. I flag those as explicit data gaps below rather than estimating them. The one item of corroborating source text I did capture is the defendant-side litigation record (Smart Mobile v. Apple transfer order), quoted where relevant.

Inventors

Employer at filing: not determinable from the record — the application was filed by "Individual" (Google Patents assignee field). There is no corporate employer of record. The Raos appear to be a family group (likely father/sons); Rekha K. Rao, who appears as an assignor on 2020-09-24, does not appear as a named inventor and is consistent with an heir/estate party for the deceased Raman K. Rao (probate conveyance of an inventor's residual interest).

Unusual pattern — flag: all three inventors assigned within ~3 months of the 2015-04-13 filing (first assignment 2015-07-22), and the two surviving inventors ended up as the sole members of the ultimate assignee. This is not the classic "inventors depart the operating company → fire-sale" pattern; it is closer to inventor self-monetization — the inventors never had an operating-company employer and instead built the holding chain themselves. Note also the 2020-09-24 re-execution of assignments from all three Rao parties individually, ~5 years after the original 2015-07-22 assignment — a chain-of-title cleanup pattern that typically precedes an assertion campaign.

Original assignee

• On the face of the patent: Google Patents lists "Individual" as original assignee (the Rao inventors as applicants). Unified Patents, however, lists "Original Assignee: IP Holdings Inc." — these two sources conflict; see uncertainty register.
• At issuance (2016-04-19): the last recorded pre-grant assignment (2015-07-28) ran to Smart Mobile, Inc., so Smart Mobile, Inc. was assignee of record at grant.
• IP Holdings, Inc. — a holding entity; no products in commerce evidenced anywhere in the record. Its only documented function in this chain is to sit between the individual inventors and Smart Mobile, Inc. for ~6 days (2015-07-22 → 2015-07-28).
• Smart Mobile, Inc. / Smart Mobile Technologies LLC — the operative monetization entity. It does not ship a product embodying the claims. The most probative evidence is the Smart Mobile v. Apple transfer order itself: "All of Smart Mobile's operations, including all of its efforts to monetize the asserted patents, are in Texas." The court also credited that Smart Mobile had an electronics project lab and prototypes and had "tested software applications" — i.e., R&D-stage activity, not commercial shipment.
• Current status: Smart Mobile Technologies LLC — operating as an active patent assertion entity (13 patents asserted across the campaign; classified "NPE (Individual)" by Unified Patents). Patent itself is Expired – Fee Related (anticipated expiration 2019-06-04).

Assignment timeline

Chain-of-title integrity note: I have the parties and sequence only. Reel/frame, execution dates, and correspondents were not retrievable and are marked [NOT RETRIEVED]. Do not treat the recording dates below as execution dates.

• 2015-07-22 (executed [NOT RETRIEVED]) / recorded 2015-07-22 — Reel [NOT RETRIEVED]

  • Conveyance: Assignment of assignors' interest
  • Assignor: RAO, RAMAN K.; RAO, SANJAY K.; RAO, SUNIL K. (all three inventors)
  • Assignee: IP HOLDINGS, INC.
  • Correspondent: [NOT RETRIEVED] — data gap; cannot run the repeat-correspondent test on this chain
  • Context: Founder-side aggregation — the three inventors sweep their rights into a common holding vehicle 3 months after filing.

• 2015-07-28 (executed [NOT RETRIEVED]) / recorded 2015-07-28 — Reel [NOT RETRIEVED]

  • Conveyance: Assignment of assignors' interest
  • Assignor: IP HOLDINGS, INC.
  • Assignee: SMART MOBILE, INC.
  • Correspondent: [NOT RETRIEVED]
  • Context: Internal reorg / holding-vehicle rotation — the patent passes through IP Holdings in only 6 days, leaving Smart Mobile, Inc. as assignee of record at the 2016-04-19 grant.

• 2020-09-24 (executed [NOT RETRIEVED]) / recorded 2020-09-24 — Reel [NOT RETRIEVED] (three separate recordings, one per assignor)

  • Conveyance: Assignment
  • Assignors: RAO, REKHA K., MRS.; RAO, SANJAY K., MR.; RAO, SUNIL K., MR. (separate instruments)
  • Assignee: SMART MOBILE INC
  • Correspondent: [NOT RETRIEVED]
  • Context: Chain-of-title cleanup / probate conveyance — five years after the original inventor assignments, the three Rao parties re-convey individually; consistent with curing an evidentiary gap (and with estate transfer of the deceased inventor's interest) ahead of assertion.

• 2020-09-26 (executed [NOT RETRIEVED]) / recorded 2020-09-26 — Reel [NOT RETRIEVED]

  • Conveyance: Assignment
  • Assignor: IP HOLDINGS, A CALIFORNIA CORPORATION
  • Assignee: SMART MOBILE INC
  • Correspondent: [NOT RETRIEVED]
  • Context: Chain-of-title cleanup — a second, overlapping conveyance from the IP Holdings vehicle two days after the individual confirmations. The burst of four recordings in three days is the classic "fix the record before filing suit" sequence.

• 2020-11-12 (executed [NOT RETRIEVED]) / recorded 2020-11-12 — Reel [NOT RETRIEVED]

  • Conveyance: Assignment
  • Assignor: SMART MOBILE, INC.
  • Assignee: SMART MOBILE TECHNOLOGIES LLC (current assignee)
  • Correspondent: [NOT RETRIEVED]
  • Context: Transfer to the asserting entity — final step that moves the portfolio into the LLC that filed the 2021 campaign.

(No security agreements, mergers, licenses, releases, or corrections were surfaced in the record. No bankruptcy conveyance appears.)

Timeline diagram

timeline
    title Ownership of US 9319075
    1999 : Earliest priority date claimed
    2015 : Filed by Rao inventors
         : Assigned to IP Holdings Inc
         : Reassigned to Smart Mobile Inc
    2016 : Patent issued
    2019 : Patent term expires
    2020 : Ownership reconfirmed to Smart Mobile Inc
         : Transferred to Smart Mobile Technologies LLC
    2021 : Suit filed against Apple
    2022 : Apple IPR petition filed
    2024 : PTAB finds no claim unpatentable
    2026 : CAFC affirms PTAB

NPE / troll-pattern signals

Verdict

NPE — high confidence.

Four signals are supported by concrete record evidence rather than naming: (i) the two-hop holding-vehicle sequence inventors → IP Holdings, Inc. (2015-07-22) → Smart Mobile, Inc. (2015-07-28, 6-day transit) → Smart Mobile Technologies LLC (2020-11-12), with a five-recording cascade between 2020-09-24 and 2020-11-12 immediately preceding the enforcement campaign; (ii) Unified Patents' explicit "NPE (Individual)" classification of the current assignee; (iii) the court's own finding in Smart Mobile v. Apple that "all of its efforts to monetize the asserted patents" constitute Smart Mobile's operations, with no commercial product; and (iv) a 13-patent, multi-defendant assertion campaign against Apple and Samsung. Per your rules I do not raise this beyond "high" confidence via naming alone, and I explicitly withhold the two signals I could not test — repeat correspondent (#3) and the exact 6-month pre-suit window for this patent (#5) — because the reel/frame and correspondent fields were not retrievable.

Verify at USPTO Assignment Center (search by patent number; both front-ends surface the same records):

• https://assignment.uspto.gov/patent/index.html
• https://assignmentcenter.uspto.gov/
• Underlying patent record: https://patents.google.com/patent/US9319075/en (Legal Events → Reassignments)

What a follow-up on the live Assignment Center should capture, given the gaps: (1) reel/frame for each of the 7 recorded links; (2) correspondent of record for the 2015-07-22, 2015-07-28, 2020-09-24 ×3, 2020-09-26, and 2020-11-12 recordings — if a single attorney/firm filed all seven, that converts signal #3 from "unclear" to "present" and would independently reinforce the high-confidence verdict; (3) execution dates, to resolve whether the 2020-11-12 transfer falls inside six months of the first '075 infringement suit.

Cross-reference to prior sections

• The prior summary listed the same seven post-filing assignment events sourced from Google Patents; this section adds the party-relationship findings (Sunil and Sanjay Rao as sole members of the current assignee; Raman K. Rao deceased; Rekha K. Rao as the 2020 non-inventor assignor) and converts the events into the chain-of-title cascade analysis.
• One contradiction to flag: the prior section states the "Original assignee: Individual." Unified Patents' patent record states "Original Assignee: Ip Holdings Inc." while Google Patents' current-assignee field names Smart Mobile Technologies LLC. These are three different answers to "who owned it at grant." The assignment sequence resolves it — Smart Mobile, Inc. was assignee of record at the 2016-04-19 grant — but the "Individual" vs. "IP Holdings Inc." labels in the two databases are unreconciled.
• No new reel/frame, correspondent, or execution-date data was obtained in this pass beyond what the earlier section already reported. If the earlier section implied those fields were available, that impression should be corrected: they are not.

I'll search for the USPTO record and the citation lists for US 9,319,075.

Let me pull the actual front-page citation list from multiple sources.

Scope and retrieval limitations (read first)

You asked me to search the USPTO database for the specific number 9319075 and then walk each patent citation on that patent. Here is exactly what I was and was not able to establish, before I give you the substantive analysis:

What I could not retrieve. The front-page "(56) References Cited" list of US 9,319,075 B1 itself was not retrievable in this session. The authoritative patent text supplied to me is truncated — it stops mid-claim ("What is c…") and contains no front-matter citation table. My searches routed through Google Patents, Justia, uspto.report, and Unified Patents mirrors rather than a direct USPTO Patent Public Search / PatentCenter query (my only tool is web search). I therefore cannot give you a verbatim reproduction of the Examiner-cited references on the '075 face, and I will not reconstruct one by inference.

A trap I want to flag. Search results repeatedly surfaced lines like "Patent Citations (2)", "Patent Citations (9)", and "Patent Citations (72)" alongside US9319075B1. Those counts belong to other patents — US 8,977,323, US 9,071,454, and US 9,019,946 respectively — which merely list US 9,319,075 as a family member in their own "related applications" tables. Those are siblings under the same priority chain, not prior art against the '075 patent. Anyone (or any automated crawler) that reads those counts as the '075's citation list is reading the wrong table. This is worth stating because it is exactly the error the task's "do not return similar numbers" instruction is guarding against.

What I did retrieve. Two independent categories of reference material, which I treat separately below because their evidentiary weight is very different:

1. The actual adversarial art — the references Apple actually petitioned on in IPR2022-01223 and lost on at the Federal Circuit.
2. Machine-generated candidate art from a third-party automated invalidity/claim-chart page that literally names US9319075B1.

A hypothesis that builds on the earlier section's uncertainty register. The prior section flagged an unreconciled conflict: the patent's own cross-reference and the Federal Circuit say priority runs to 1999-06-04, while Google Patents shows a "prior art date" of 1996-12-16 and Unified Patents shows 1996-12-15. The candidate references I retrieved cluster tightly in 1993–1996. That clustering suggests the 1996 date in the metadata is being derived from a cited reference's date (a "prior art date" field populated from the earliest cited reference), not from the '075's own priority claim. That would explain the conflict rather than leave it mysterious. I cannot confirm this without the front-page table — which is precisely the artifact I could not obtain. Flagging it as a hypothesis, not a finding.

Part 1 — The prior art actually litigated against US 9,319,075

This is the highest-confidence answer to "most relevant prior art," because it is the art a sophisticated petitioner (Apple, with Samsung as RPI) chose and a tribunal adjudicated.

Honest gap: I was unable to pull the full citations (patent/publication numbers, dates, assignees) for Yegoshin and Bernard within my search budget. Their names are confirmed from the IPR and appellate record; their bibliographic details are not, and I will not fill them in from memory. If you need those numbers, they will be in the IPR2022-01223 petition and FWD.

Two points of substance for your §102 question. First, note the ground was §103, not §102. Neither Yegoshin nor Bernard was asserted as a single-reference anticipation of claim 1 — Apple needed a combination, and the combination still failed. Second, the "multiplexing signals" construction ("combining multiple signal streams into one") is what did the work: it forecloses the reading in which two radios merely share a pathway sequentially. That construction is the single most important fact for any prior-art analysis of this claim, and it applies with equal force to every candidate reference below.

Part 2 — Candidate references from the automated invalidity chart naming US9319075B1

Provenance and caution. These come from a page headed "Invalidity Claim Chart for Patent US9319075B1" (Leegal AI, s3.amazonaws.com/leegal-ai-website/invalidity_search/benchmark/claim_charts/US9319075B1.html). This is a third-party automated search artifact, not an official USPTO record and not part of the '075's prosecution file. Its own language gives it away: every entry says the reference "is relevant to the claim features" — that is §103 framing, not §102. Treat the list as a lead-generation set, not a citation list. Dates shown may be filing or priority dates rather than publication dates; I have flagged where that is likely.

Also note: no inventor name for the first reference is a chart author — "Jonathan F. Nethercott, Guy A. Cooper, Jonathan A." are inventors of a reference, and the snippet did not display that reference's patent number. I will not supply a number.

Part 3 — Direct answer on §102 anticipation

No reference I retrieved anticipates any claim of US 9,319,075 under 35 U.S.C. § 102. That is a considered conclusion, not a placeholder:

• §102 requires a single reference disclosing every limitation, arranged as claimed. The candidate references above each map to one or two elements of claim 1 at most, and the chart's own phrasing ("relevant to the claim features") concedes it is arguing obviousness.
• Claim 1 is not a thin claim. It demands, in combination: a plurality of wireless units with multiple protocols, separate antenna sets on separate networks, a directed lower-than relationship between the two frequencies operating "as complementary systems," both units talking to the same remote server with the second path engaged in response to a change in signal strength or connectivity, simultaneous video/audio with per-path optimization, and multiple IP data packets on a plurality of ports at substantially the same time to that server. No single one of these ten references spans that.
• The construction is the choke point. "Multiplexing signals" = "combining multiple signal streams into one." The automatic-and-parallel reading the claim needs is exactly what a "switching/sequential sharing" reference cannot supply. Several of the candidates above (items 3, 6, 9) are strongest precisely on the limitations the claim does not depend on, and weakest on [1e], [1g], and [1h].
• [1e]'s "in response to" trigger was, per the earlier section, itself a live dispute point. Item 9 (WO 97/12456) is the best §102-flavored candidate on that lone element — and it is a wired router invention.
• Empirical confirmation. The one tribunal that tested this claim against real art (IPR2022-01223) found claims 1, 2, 3, and 5 not unpatentable, and the Federal Circuit affirmed on August 27, 2026. That outcome is consistent with — though it does not by itself prove — the absence of anticipating art.

Which claims are in play at all: claims 1–3 and 5 (the only ones ever challenged); claim 4 has never been challenged in the proceeding I found. I remain unable to state the subject matter of dependent claims 2–5; consistent with the earlier section, I will not characterize limitations I have not read. That means the per-claim §102 mapping above is necessarily claim-1-only, and any mapping onto 2–5 would be guesswork.

Part 4 — Uncertainty register (additions to the earlier one)

1. The '075's own front-page citation list was never retrieved. Everything in Part 2 is third-party automated candidate art, not the Examiner-cited record. This is the single largest gap in this response.
2. Full citations for Yegoshin and Bernard were not verified — names only.
3. Dates in the candidate table may be filing/priority rather than publication dates, which matters for §102(b) versus §102(e) categorization. I flagged the two most suspicious (items 4 and 6).
4. Item 7 (US 6,006,105) is likely a chart artifact (verbatim duplicate of item 2's description). Items 5 and 10 cannot be assessed on the retrieved text.
5. Pre-AIA §102 governs. The application (14/685,592, filed 2015-04-13) is a continuation claiming priority into the 1999 chain — before March 16, 2013 — so AIA §102 does not apply. All the 1993–1996 candidates would clear the §102(b) one-year bar against a 1999-06-04 priority, but not against a 1996-12-16 date for the later items. The priority-date question is therefore outcome-determinative, and it is unresolved (see the hypothesis in my opening).
6. Date context, unresolvable as before. You state today is April 26, 2026. My retrieval environment returned the CAFC disposition in No. 25-1090 dated August 27, 2026. I report the sources as they stand rather than reconciling them; if your date is authoritative, that opinion is future-dated material and should be treated as such.

Bottom line: the most relevant prior art for US 9,319,075 is, on the evidence actually tested, the Yegoshin/Bernard combination from IPR2022-01223 — and it failed to invalidate claims 1–3 and 5. The most relevant single-reference art by subject matter among the candidates is US 4,748,655 (multiplexing plus complementary systems to reduce interference), with US 6,035,212 strongest on multi-frequency/multi-protocol and WO 97/12456 strongest on condition-responsive path switching — but each is an obviousness reference, and none anticipates claim 1 under §102.

I'll pull the prior-art record for the '075 patent — both its own cited references and the references the PTAB/CAFC actually litigated.

Now let me pull the specific prior-art citations listed on the '075 patent's Google Patents page, and the invalidity claim chart for it.

Let me get substance on the key references — both those the examiner cited and those the PTAB litigated.

US 9,319,075 B1 — Obviousness Analysis Under 35 U.S.C. § 103

Scope note. You asked me to work from the Prior Art section of the page. The Google Patents rendering truncated that section in my retrieval: I recovered the "Patent Citations (72)" table only in part (roughly the first twenty entries, via the family page rendering), plus a third‑party invalidity chart that reproduces a set of references with element-by-element rationales. I did not retrieve a discrete Google Patents "Prior Art" search-result block for this patent, and I did not retrieve the full 72-entry citation table. Everything below is built from (a) the partial citation table I did see, (b) the claim-chart reference set, and (c) the references Apple actually litigated in IPR2022‑01223 (Yegoshin, Bernard, Johnston, Billström). I flag identification gaps explicitly rather than filling them in.

Note also: the earlier section's date/priority uncertainty carries forward unchanged. I treat 1999‑06‑04 (the CIP root, application 09/281,739) as the presumptive critical date, consistent with the patent's own cross-reference block and the Federal Circuit's statement that the '075 "claim[s] priority to a 1999 application." The 1996‑12‑16 "prior art date" that Google Patents displays remains unexplained. Every reference I use below predates 1999‑06‑04 on its face, so the analysis holds under either reading of the 1996 metadata — but not if the priority claim to 1999 were to fail for § 112 reasons, in which case the critical date shifts to 2004‑09‑13 and the art base expands enormously (see §7).

1. The claim to be assessed

Claim 1 is the only independent claim; claims 2–5 depend from it. Per the prior section, the elements are lettered 1[pre] through 1[h], with the preamble ("An IP-enabled communication device for multiplexing signals") treated as limiting and the term "multiplexing signals" construed as "combining multiple signal streams into one." That construction is the hinge on which this entire § 103 analysis turns.

Two anomalies flagged in the prior section remain relevant to obviousness and I do not re-derive them: (i) elements [1g] and [1h] appear to recite the same limitation twice; and (ii) there is a genuine tension between the preamble's requirement that signals be combined into one and element [1h]'s requirement of "multiple simultaneous communication paths."

2. Level of ordinary skill and the legal frame

A POSITA here is a wireless-systems engineer (B.S.E.E. plus ~2–4 years, or equivalent) familiar with cellular air interfaces, WLAN, RF front ends, and packet-data protocols. The relevant art is portable wireless voice/data devices — the same field as every reference below, so no analogous-art or field-of-invention dispute arises.

KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), governs. The claim is a combination claim whose elements were each individually known; the question is whether the arrangement would have been obvious. Two KSR considerations bite hard here: (1) design incentives and market forces; and (2) whether the combination is "the mere aggregation of known elements according to known methods to yield predictable results." The '075's own specification supplies both.

3. Threshold point: substantial admitted prior art in the '075's own Background

Before reaching the references, note that the patent's Background section admits two of the most distinctive limitations:

This is applicant-admitted prior art — usable under § 103 without a documentary reference. It means the inventive contribution, if any, collapses to: integrating an admitted dual-band complementary failover arrangement into a single handheld device, plus aggregating IP packet flows across it. That is a narrow space, and it is exactly the space KSR addresses.

4. Element-by-element coverage table

References marked ★ are on the face of the patent-family citation table I retrieved; references marked † come from the invalidity claim chart; references marked ‡ are from the IPR record.

5. Ground-by-ground § 103 analysis

Ground 1 — Yegoshin in view of Bernard (the combination Apple actually ran)

Yegoshin (US 6,711,146 B2, ‡ EX‑1004) is a dual-radio phone with cellular and WLAN communication ports (11, 13), client software 19 allowing network selection, and express description of the phone taking calls "via cellular path while receiving other calls via IP path." Bernard (US 5,497,339, ‡ EX‑1007) is a PDA cradle exposing cellular telephone interface 720, land phone interface 724, phone modem interface 722, and packet radio 124, feeding a communication server 750, packet interface 752, packet distributor 754, and a single serial interface 701 to application server 710. Bernard's first embodiment (Fig. 13) expressly uses a decoder/multiplexer.

Motivation to combine (as Apple framed it): two alternatives — (a) phone-plus-cradle (physically connect Yegoshin's phone to Bernard's cradle), or (b) phone-only (integrate Bernard's multi-network circuitry into Yegoshin's phone). The stated rationales: Bernard's own teaching that its cradle provides wireless connections not available at the phone itself; and the benefit of multi-network connectivity without a separate cradle. Yegoshin supplies an independent pull — reducing "costly cell charges" by using the IP network when available — and Bernard supplies the express statement that "an alternative interconnection could be used that would allow multiple connections to be established simultaneously."

Assessment. For limitations [1a]–[1e], [1g] and [1h], this combination is strong on paper: two radios, two antenna paths, two networks, a shared interface, a packet interface and distributor, and an express suggestion of simultaneous multi-connection operation. It is a textbook KSR aggregation of known elements in the same field with a predictable benefit (higher aggregate throughput, per-application routing).

But it failed, and the reason matters. The PTAB construed "multiplexing signals" as "combining multiple signal streams into one" and found:

• Yegoshin teaches the opposite of simultaneous use — "[i]f engaged with an IP call, an incoming cell call would get a busy signal … or it would be redirected." The Board read the calls as consecutive, not simultaneous (Institution Decision at 24–26; Final Written Decision, IPR2022‑01223).
• Bernard's second embodiment (the one relied on) expressly states "[i]n this second embodiment, only one of the four above-described connections can be established at a time" (Bernard, 26:56–58). Only Bernard's first embodiment (Fig. 13) describes a multiplexer.
• Apple never identified Bernard's "alternative interconnection," so the Board declined to credit it.
• Apple's "switching" reading of multiplexing was rejected; a two-completed-calls-one-year-apart theory (Dr. Jensen's deposition) was self-defeating.

The Federal Circuit affirmed on substantial-evidence review and adopted the Board's construction verbatim: "we agree with the Board's implicit construction of 'multiplexed signals' as meaning 'combining multiple signal streams into one' and not including multiple signals sequentially using the same path." (CAFC, No. 25‑1090, and companion No. 24‑1607.)

Analyst conclusion on Ground 1: the combination would render [1a]–[1e] and [1g]–[1h] obvious if "multiplexing signals" is construed per Apple's broader district-court position ("to interleave or simultaneously transmit two or more messages on a single communications channel"). Under the narrower construction the Board adopted and the CAFC affirmed, the preamble limitation is not met, and the ground fails regardless of how strong the remaining elements are. This is a construction-driven loss, not an evidentiary one.

Ground 2 — US 5,121,391 as primary (the ground that was never run)

This is the most striking gap in the record, and it comes straight out of the patent's own citation table: US 5,121,391 ("Subscriber RF telephone system for providing multiple speech and/or data signals simultaneously over either a single or a plurality of RF channels," Int'l Mobile Machines, 1985‑03‑20 priority / 1992‑06‑09) teaches simultaneous multiple speech and data signals over a single channel or a plurality of channels.

That is not merely close to "combining multiple signal streams into one" — it is a direct textual hit on the narrow construction that defeated Ground 1, together with a direct hit on [1g]/[1h] (multiple streams at the same time across multiple channels). Because it issued in 1992, it is § 102(b) art against a 1999 critical date.

Combine it with:

• † US 6,035,212 (Rostoker/Daane/Jaggi, priority 1996‑08‑02) — a wireless device that self-adapts across multiple frequency bands and protocols → [1a];
• † US 4,748,655 (Thrower/Munday/Gill, 1984‑07‑25) — portable telephones via gateways supporting multiple channels through frequency, time, or code division multiplexing → [1b], [1c], [1d];
• ★ US 5,828,658 (Arraycomm) and ★ US 5,507,035 (IBM) — multiple antennas / antenna diversity in the same device → [1b].

Motivation: the '075's own stated problem — a 14.4 kbps wireless ceiling — and its own admission that throughput "is gated by the speed of the transmit and receive units [and] is improved by the parallel paths." A POSITA seeking higher throughput had every reason to parallelize converters, channels and T/R chains, which the '391 reference expressly enables. Frequency separation to avoid self-interference is a standard design technique, and the Background admits the low/high-band complementary arrangement outright.

Assessment: materially stronger than Ground 1 against the operative construction, and it avoids the Yegoshin "busy signal" teaching-away problem entirely because '391 is not about selecting between networks — it is about using several at once.

Ground 3 — IP-layer aggregation: Bernard + US H1641 + WO 97/12456 (and the "downward multiplexing" treatise evidence)

• † US H1641 H (Sharman, 1993‑11‑30) — connecting a mobile end system to packet data and telephone networks, enabling simultaneous communication over multiple networks → [1g]/[1h] with a packet-data orientation the other references lack.
• † WO 97/12456 A1 (Bowman-Amuah, 1995‑09‑26) — multiple protocols and primary and secondary routes (frame relay, internet) with routers performing protocol conversion and path switching based on network conditions → [1e] almost verbatim, plus [1a].
• † WO 96/28904 A1 (Kostreski/Schneider, 1995‑03‑16) — combining a broadband digital wireless network with call connections through an existing cellular network → the two-network architecture of [1b].
• The † Nethercott/Cooper reference (1995‑06‑02) — 1.5–2.5 GHz system with multiple frequency channels, modem cards, and RF units → [1d] and [1a]. Caveat: the snippet I retrieved did not return its publication number; I am not guessing at it.

A particularly useful data point sits inside the Board's own final written decision (IPR2022‑01249, quoted at length in the panel record): Apple's technical treatises (Ex. 1012, pp. 506–508) describe "downward multiplexing" — "hav[ing] the transport layer open multiple network connections and distribute the traffic among them on a round-robin basis." That is precisely the claimed technique of splitting a data flow across simultaneous IP paths to raise aggregate rate. The Board discounted it only because Apple failed to tie it to Yegoshin's call-handling passage. Used as the motivation-and-enablement reference behind a Ground‑2 primary, that evidence is far more effective.

Ground 4 — Split-path secondary references (weak, but useful for [1f])

† EP 1,814,352 B1 (Klok/Koster/Dirksen, 1995‑09‑08) teaches user information over a first path and selection information over a second path (satellite/cable vs. PSTN/ISDN). That supports [1b] and the "per-path optimization" idea in [1f], but it is a split of function rather than an aggregation of streams, so it does not help on the preamble. Use as a secondary, not primary.

6. Dependent claims 2–5

The prior section was unable to retrieve verbatim claim text for 2–5, and I likewise decline to fabricate it. What the record does establish:

• Claims 1–3 and 5 were challenged in IPR2022‑01223; claim 4 was not challenged. Apple failed on 1–3 and 5, and that judgment was affirmed.
• For the sibling '946 patent, the CAFC opinion of Aug. 27, 2026 refers to "dependent claim 2 of the '946 patent … as requiring simultaneous transmission of multiple signals." That is a different patent with a different claim set and must not be imported into the '075's claim 2 — but it is a useful signal that the family's dependent claims tend to add simultaneity limitations.

Structural prediction, clearly labeled as prediction: if any of claims 2–5 adds a simultaneity requirement, an antenna-count requirement, a multiple-IP-address requirement, or an optical/universal-I/O requirement, then Ground 2 (US 5,121,391 + US 6,035,212 + US 4,748,655) maps more cleanly onto them than Ground 1 did, for the reasons above. I do not represent that mapping as verified.

7. Motivation-to-combine — consolidated articulation

The strongest KSR-compliant rationales, in order of persuasiveness:

1. Admitted prior art does the heavy lifting. [1d] and [1e] are effectively recited in the patent's own Background. Only integration remains, and integration into a handheld was itself the acknowledged trend.
2. Common problem / same field. Every reference addresses the throughput, coverage, or cost limitations of portable wireless voice/data. No field-change, no analogous-art gap.
3. Express suggestions in the references. Bernard: an alternative interconnection "could … allow multiple connections to be established simultaneously." Yegoshin: the program "may negotiate the best possible connection … Selection of the network could be according to an order of preference, by availability." Both are affirmative modification prompts, not hindsight.
4. Design incentive to frequency-separate. Low/high-band operation to avoid mutual interference is a standard coexistence technique — and one the applicant admitted.
5. Predictable result. Aggregating parallel links to increase throughput was a well-understood technique (inverse multiplexing; "downward multiplexing"). KSR treats improvement-by-known-technique as obvious.
6. Reasonable expectation of success. All components — multiple RF chains, multiple antennas, multiple processors, multiple I/O ports — are conventional; the '075 itself concedes the invention "is possible due to advances in the art which allow the necessary components to be integrated."

Critical date sensitivity: if the 1999 priority claim were to fail for lack of § 112 support, the critical date moves to 2004‑09‑13, and the art base widens dramatically — including, e.g., the 1996 Multilink PPP inverse-multiplexing publication (not from this page's prior-art section; noted only as an illustration of what becomes available).

8. Countervailing evidence — and why it won

I would be doing you a disservice to present only the affirmative case. The following defeated or constrained these grounds:

Net. For the '075 specifically, Apple lost on a claim-construction + substantial-evidence basis, not on a failure to identify combinable art. The same panel vacated the Board's obviousness finding on sibling claim 17 because the Board failed to engage the bottlenecking evidence — i.e., the court accepted that the motivation-to-combine weighing was inadequately reasoned there.

9. Bottom line

• Under the operative construction ("combining multiple signal streams into one"), Ground 1 (Yegoshin + Bernard) does not render claim 1 obvious — and both the PTAB and the Federal Circuit so held, affirmed.
• Ground 2 — US 5,121,391 as primary, with † US 6,035,212 and † US 4,748,655 — is the strongest unrun ground. It supplies simultaneous multi-stream, multi-channel operation (i.e., "combining multiple signal streams into one" on its own terms), multi-band/multi-protocol adaptation, and multiple antennas, and it avoids Yegoshin's consecutive-use teaching away. Under KSR, the motivation is supplied by the patent's own admitted prior art and its own stated problem.
• Ground 3 — adding † US H1641 and † WO 97/12456 — closes the IP-packet and signal-strength-triggered-path-switch elements without needing Yegoshin's call-handling passages.
• Claim 1 carries internal tension between a preamble requiring combination into one and element [1h] requiring "multiple simultaneous communication paths," plus the apparent duplication of [1g]/[1h]. That tension is a § 112(b) exposure that also complicates any § 103 mapping — a challenger should address it head-on rather than paper over it.
• Claims 2–5: text not verified; no analysis offered beyond structural prediction. Claim 4 was never challenged in the IPR.

10. Uncertainty register (new items only)

1. Publication number of the "Nethercott/Cooper" reference (1995‑06‑02) was not returned; not guessed.
2. Yegoshin's exact priority date was not retrieved; I relied on its position in the IPR record as pre‑1999 art, which is consistent with its March 23, 2004 issue date and PCT-based lineage, but is not independently verified here.
3. US 5,121,391's assignee appears in the citation table as "International Mobile Machines"; I report it as listed.
4. Full 72-entry citation table for the '075 was not retrieved — only a partial rendering. My reference set is therefore not exhaustive of the patent's own cited art, and there may be additional on-point references in the unretrieved portion.
5. The Aug. 27, 2026 CAFC opinions (Nos. 25‑1090 and 24‑1607) appear in multiple retrieved sources, including the court's own published opinion PDF, despite the stated current date of April 26, 2026. This is the same conflict the prior section flagged; I am not reconciling it, only reporting it.
6. Sibling-patent claim 2 ('946) is referenced above solely as a family-wide signal; it is not the '075's claim 2.

Key URLs: https://patents.google.com/patent/US9319075/en · http://s3.amazonaws.com/leegal-ai-website/invalidity_search/benchmark/claim_charts/[US9319075B1](/patent/US9319075B1).html · https://www.docketalarm.com/cases/PTAB/IPR2022-01223/Apple_Inc._v._Smart_Mobile_Technologies_LLC/ · http://lowensteinweatherwax.com/wp-content/uploads/2024/01/2024-01-23-IPR2022-01248-FWD.pdf · http://business.cch.com/ipld/AppleIncSmartMobileTechnologiesLLCFedCir20260827082826.pdf · https://ipwatchdog.com/2026/08/27/cafc-delivers-blow-to-apple-in-appeal-of-ptab-disputes-with-smart-mobile/