US 10567437

I'll search for information on US Patent 10,567,437, including its claims and any CAFC 2026 docket activity.

Let me search specifically for the granted claims of US10567437B2 and the CAFC dockets 25-1167 and 25-1168 linked to this patent.

US Patent 10,567,437 B2 — Summary

Important scope note: The Google Patents page for this patent links it to CAFC dockets 25-1167 and 25-1168. Those are the relevant appellate dockets. I did not find any final CAFC merits decision in those dockets as of the current date — the appeal was consolidated in late 2024, deactivated pending post-judgment motions, and reactivated in late December 2025, so it is pending/active in 2026. The April 2026 Federal Circuit activity reported in the news (the Cisco non-infringement appeal and the ITC "domestic industry" appeal) involves different Centripetal patents/cases, not 10,567,437.

Bibliographic data (verified via search results)

• Patent number: US 10,567,437 B2 (application no. 16/111,524; pre-grant publication US20190230128A1)
• Title: "Methods and systems for protecting a secured network"
• Inventors: Steven Rogers and Sean Moore
• Original assignee / current assignee: Centripetal Networks LLC (recorded assignment on 2019-06-05 to Centripetal Networks, Inc.; name change to Centripetal Networks, LLC in 2023)
• Filing date: August 24, 2018 (this application is a continuation)
• Earliest priority date: October 22, 2012 (parent U.S. App. 13/657,010 → US 9,137,205)
• Issue (grant) date: February 18, 2020
• Status: Active; Google Patents lists anticipated expiration as 2032-10-22 (post-Term-Adjustment-type date, assuming maintenance fees are paid)
• Continuation chain: 13/657,010 (US 9,137,205) → 14/698,560 (US 9,560,077) → 15/413,834 (US 10,091,246) → 16/111,524 (US 10,567,437)

Abstract (verbatim)

"Methods and systems for protecting a secured network are presented. For example, one or more packet security gateways may be associated with a security policy management server. At each packet security gateway, a dynamic security policy may be received from the security policy management server, packets associated with a network protected by the packet security gateway may be received, and at least one of multiple packet transformation functions specified by the dynamic security policy may be performed on the packets. Performing the at least one of multiple packet transformation functions specified by the dynamic security policy on the packets may include performing at least one packet transformation function other than forwarding or dropping the packets."

Plain-language overview of the independent claims

The granted claim text was not included in the materials provided, and my searches returned the claims as published in the pre-grant publication US20190230128A1 (the application that became this patent). The claims below are therefore from that published application; the claims as granted may have been amended during prosecution — treat them as the best available but not fully authoritative. The published claim set has two independent claims (claims 2 and 9), with claim 1 canceled:

• Claim 2 (method): A method of (a) provisioning a packet security gateway located at the boundary between a protected network and other networks with packet-filtering rules, where each rule pairs at least one packet-matching criterion tied to malicious network traffic with a corresponding packet-transformation function; and (b) configuring the gateway to receive packets through a communication interface that has no network-layer address, to drop any packets that match the criteria, and to modify a LAN switch's switching matrix so the switch also drops those matching packets. In plain terms: a stealthy (network-layer-invisible) inline security device that blocks malicious traffic and programs the adjacent LAN switch to do the same.

• Claim 9 (system): A system (at least one processor plus memory storing instructions) that performs the same provisioning and configuring steps as claim 2 — i.e., the claim covers a security-policy-management server/controller that sets up such a gateway and switch.

• Key dependent claims (for context): rules identifying spoofed source addresses (claim 3); rules based on subscription-service threat intelligence (claim 4); receiving rules over a network-layer-addressed management interface (claim 5); encapsulating matched packets with a header directing them to a different network address, e.g., a monitoring device (claim 6); Layer-2 VLAN-based rerouting (claim 7); and dynamically creating/altering criteria upon receipt of a policy information update (claim 8). Claims 10–15 mirror claims 3–8 for the system claim.

CAFC 2026 dockets (per Google Patents litigation metadata and Justia/RPX/DocketAlarm records)

• 25-1167 and 25-1168 — Centripetal Networks, LLC (fka Centripetal Networks, Inc.) v. Palo Alto Networks, Inc., U.S. Court of Appeals for the Federal Circuit.
  • Filed: November 13, 2024; originating from E.D. Va. case 2:21-cv-00137 (Centripetal v. Palo Alto Networks).
  • On December 11, 2024, the CAFC consolidated 25-1167 (lead) and 25-1168 (member) as appeal and cross-appeal and deactivated them pending post-judgment motions; the docket was reactivated by CAFC order on December 22, 2025, so the appeal is live in 2026. No 2026 merits decision was identified in my searches.
  • Patent 10,567,437 is listed among the Centripetal patents asserted in the underlying E.D. Va. litigation (along with, e.g., 10,091,246, 10,542,028, 10,567,343, 10,567,413, 10,659,573, 10,749,906, 10,757,126, 10,785,266).

Additional litigation/PTAB context (for completeness)

• IPR2021-01153 was filed against this patent; per the Google Patents metadata it was Not Instituted (Merits) — i.e., the PTAB declined to institute inter partes review on the merits.
• Related E.D. Va. cases referencing the patent family include 3:21-cv-00597, 1:21-cv-01051, and 1:21-cv-00313.
• Other reported 2026 Federal Circuit activity involving Centripetal (e.g., Centripetal Networks, LLC v. ITC, No. 24-1416, decided April 23, 2026, and the Cisco appeal argued April 2026) concerns different patents (e.g., the '370 patent and Cisco-asserted patents), not 10,567,437.

Uncertainties

1. Granted claim text — I could not confirm the exact granted independent claims of the B2 patent from the available sources; the overview above reflects the published application US20190230128A1.
2. Disposition of CAFC 25-1167/25-1168 — no 2026 outcome was located; the case was reactivated in December 2025 and remains pending per available docket data.
3. Maintenance-fee status — Google Patents marks the patent "Active" with anticipated expiration in 2032, but I did not independently verify current USPTO maintenance-fee status.

Litigation involving US Patent 10,567,437 ("the '437 patent")

Based on searches of PACER-derived dockets (Justia, Docket Alarm, RPX, UniCourt, Ex Parte), PTAB records, and contemporaneous press coverage (Reuters, IPWatchdog, Law360), the known litigation is set out below. All identifiers are reproduced literally.

A. District court litigation

1. Centripetal Networks, LLC (f/k/a Centripetal Networks, Inc.) v. Palo Alto Networks, Inc.

• Jurisdiction: U.S. District Court for the Eastern District of Virginia (Norfolk Division)
• Case No.: 2:21-cv-00137 (EWH-LRL) — Judge Elizabeth W. Hanes, Magistrate Judge Lawrence R. Leonard
• Filing date: March 12, 2021
• Plaintiff: Centripetal Networks, LLC
• Defendant: Palo Alto Networks, Inc.
• The '437 patent's role: Originally one of 13 patents asserted. By trial, only claim 8 of the '437 patent remained asserted, against PAN's Panorama (alone) and Panorama combined with Next-Generation Firewalls ("'437 Accused Products").
• Outcome/status:
  • January 31, 2024 jury verdict: $151.5 million total for Centripetal across four patents. For the '437 patent specifically, the jury awarded $37,875,000, finding direct infringement of claim 8 (NGFWs combined with Panorama), and rejecting PAN's invalidity defenses (the verdict form also shows the jury did not find induced infringement or willfulness).
  • Post-trial (November 2024, E.D. Va. Doc. 989): Judge Hanes granted PAN's JMOL motion in part, holding there was legally insufficient evidence to support the jury's infringement finding on the '437 patent, and directed entry of judgment as a matter of law of non-infringement of the '437 patent, reducing the jury award accordingly. The Court denied PAN's motion for a new trial and upheld the jury verdicts on the "Correlation Patents" ('903, '573, '797).
  • Current status: Final judgment entered; both sides appealed (see CAFC dockets below). The CAFC appeal was reactivated effective December 22, 2025, with appellant's brief due February 20, 2026. As of April 26, 2026, the appeal is pending; no CAFC merits decision was located.

2. Centripetal Networks, Inc. v. LookingGlass Cyber Solutions, Inc. et al.

• Jurisdiction: U.S. District Court for the Eastern District of Virginia (filed in Alexandria Division as 1:21-cv-01051; transferred to Richmond Division)
• Case No.: 3:21-cv-00597 — Judge David J. Novak
• Filing date: September 14, 2021 (transferred in September 16, 2021)
• Plaintiff: Centripetal Networks, Inc.
• Defendants: LookingGlass Cyber Solutions, Inc.; Gilman Louie; Alsop Louie Management LLC; Alsop Louie Capital 2, L.P.; Alsop Louie Partners 2, LLC
• The '437 patent's role: Asserted among five patents (also 10542028, 10757126, 10785266, 10735380). Causes of action included patent infringement, breach of contract, breach of fiduciary duty, and abetting breach of fiduciary duty.
• Status: Docket was "open" as of late 2021, but I could not verify the current status or any disposition from the sources retrieved. Treat the outcome as unconfirmed.

B. Federal Circuit appeals

3. Centripetal Networks, LLC v. Palo Alto Networks, Inc.

• Jurisdiction: U.S. Court of Appeals for the Federal Circuit
• Case Nos.: 25-1167 (lead, appeal) and 25-1168 (member, cross-appeal)
• Filing date: November 13, 2024 (appeal from E.D. Va. 2:21-cv-00137)
• Parties: Centripetal Networks, LLC (f/k/a Centripetal Networks, Inc.), Plaintiff-Cross-Appellant; Palo Alto Networks, Inc., Defendant-Appellant
• Status: Consolidated by CAFC order December 11, 2024 as appeal and cross-appeal; deactivated pending resolution of FRAP 4(a)(4) post-judgment motions; reactivated effective December 22, 2025 (appellant's brief due February 20, 2026). Pending as of April 26, 2026 — no merits decision located. This appeal encompasses the district court's post-trial rulings, including the JMOL of non-infringement of the '437 patent.

C. PTAB / IPR proceedings

4. Palo Alto Networks, Inc. v. Centripetal Networks, Inc., IPR2021-01153

• Jurisdiction: Patent Trial and Appeal Board
• Filing date: July 6, 2021
• Petitioner: Palo Alto Networks, Inc.; Patent Owner: Centripetal Networks, Inc.
• Claims challenged: 1–20 of the '437 patent
• Outcome: Institution DENIED (decision January 24, 2022; panel: J. John Lee, Brian J. McNamara, John P. Pinkerton). Google Patents metadata likewise codes this IPR as "Not Instituted – Merits."

5. Keysight Technologies, Inc. IPR petition against the '437 patent (institution denied)

• Per the parties' joint status report filed in 2:21-cv-00137 (E.D. Va. Doc. 329-8, filed March 29, 2023), Keysight Technologies, Inc. filed IPR petitions on patents asserted in that case and related patents; the row for 10,567,437 (challenging claims 1, 3, 5–8, 10, 12–15, and 17, 19–20) is listed as "Institution Denied." I did not locate the IPR case number for that particular petition in the sources retrieved, so the number is unverified. (The same filing identifies Keysight's IPR2023-00448 as a petition on related U.S. Patent No. 11,012,474, a grandchild of the '437 patent.)

D. Additional case number flagged in patent metadata (unverified)

• 1:21-cv-00313 (E.D. Va.) — Appears in Google Patents' litigation metadata for this patent as a Virginia Eastern District Court case. I could not verify the parties, filing date, or status, and it may simply be the original docket number of a case later renumbered. Do not rely on this entry without PACER confirmation.

Key caveats

• The biggest development affecting the '437 patent is the November 2024 JMOL: although the jury awarded $37.875M for claim 8, the district court entered JMOL of non-infringement of the '437 patent for insufficient evidence, and that ruling is now part of the pending CAFC appeal (25-1167/25-1168).
• No CAFC merits decision on 25-1167/25-1168 was located as of April 26, 2026; the appeal was reactivated in December 2025 and briefing was underway in early 2026.
• I did not verify current maintenance-fee status or whether the LookingGlass case (3:21-cv-00597) and the metadata-listed 1:21-cv-00313 have been resolved. PACER would be the authoritative source for those two items.

I'll research the PTAB proceedings on this patent. The structured data says the ODP returns no AIA trials, but the patent's own litigation metadata references IPR2021-01153 — so I need to verify that and check for any other proceedings via web search.

I found IPR2021-01153 (Palo Alto Networks v. Centripetal) — institution denied. Let me verify the panel, grounds, and check for any other proceedings against this patent or its direct parents.

I have solid confirmation of IPR2021-01153. Now let me get the actual institution decision reasoning and check for any other IPRs against this patent (or related family members that might cover the same claims).

Proceedings overview

There is one AIA trial proceeding on US 10,567,437 in the public record — IPR2021-01153 — and it was denied at institution (the only status that exists for it; no trial, no Final Written Decision). The bottom line for a defendant: zero claims have been canceled — all 20 claims survive, and the patent is arguably "hardened" in the sense that the only PTAB challenge (Palo Alto Networks' Jungck-based obviousness attack) failed at the threshold on 2022-01-24. But the claims have never been tested on the merits, no § 315(e) estoppel attaches (no FWD), and the co-pending E.D. Va. litigation (2:21-cv-00137) plus CAFC appeals 25-1167/25-1168 remain the live battlegrounds.

IPR2021-01153 — Palo Alto Networks, Inc. v. Centripetal Networks, Inc.

• Type: Inter Partes Review
• Filed: 2021-07-06 (notice of filing date accorded 2021-07-23)
• Status: Institution Denied — proceeding terminated 2022-01-24; petitioner's fee refund approved 2022-04-07 (per PTAB docket). Plain-English gloss: the Board never instituted trial, so there was no merits adjudication of any claim.
• Judge panel: Kevin F. Turner (APJ), Bryan F. Moore (APJ), Lynne E. Pettigrew (APJ) — as shown on the Decision Denying Institution (Paper 10) itself ("Before KEVIN F. TURNER, BRYAN F. MOORE, and LYNNE E. …"). ⚠️ Discrepancy flagged: Patexia lists a different panel (J. John Lee, Brian J. McNamara, John P. Pinkerton); that appears to be a data error — APJ McNamara sat on a different Centripetal/Palo Alto IPR (the encrypted-threat-detection patent at issue in CAFC No. 2023-2027), not this one. I would trust the Paper 10 caption over the secondary aggregator.
• Petition grounds (per Petition, Paper 2, and Paper 10):
  • Ground 1: claims 1–4, 6–11, 13–18, 20 — obvious under 35 U.S.C. § 103 over Jungck (US Patent Appl. Publ. No. 2007/0262741 A1);
  • Ground 2: claims 5, 12, 19 — obvious under § 103 over Jungck + RFC 2003 ("IP Encapsulation within IP," C. Perkins, IBM).
  • All 20 claims were challenged. Petitioner relied on a declaration of Dr. Vijay Madisetti (Ex. 1004) and, notably, a claim-comparison chart and the FWDs/institution decisions from Cisco's 2018 IPRs on Centripetal family members (Exs. 1012–1021).
• Institution decision: Denied — 2022-01-24, Paper 10 (Decision Denying Institution of Inter Partes Review, 37 C.F.R. § 42.108). The decision concludes: "For the reasons explained below, we do not institute an inter partes review in this proceeding." I could not retrieve the full text of Paper 10 to quote the panel's precise rationale; Google Patents' litigation metadata characterizes it as "Not Instituted – Merits" (i.e., a failure to show reasonable likelihood under § 314(a) rather than a discretionary Fintiv/§ 325(d) denial). The petition itself devoted a section to arguing the Board should not exercise discretion to deny under § 314(a) and § 325(d), so those were contested issues — but the available record points to a merits-based denial.
• Final Written Decision: None. Institution was denied, so no trial was conducted and no claim was canceled or confirmed.
• Settlement / termination: No settlement. The proceeding was terminated by the denial itself on 2022-01-24 (docket shows "Terminated Jan. 24, 2022").
• Appeal: Institution denials are not appealable under 35 U.S.C. § 314(d). Palo Alto nonetheless sought Director review of the denial; the USPTO refused (policy: no Director review of institution decisions), and PAN filed a mandamus petition in the Federal Circuit — In re Palo Alto Networks, Inc., No. 22-145 (Fed. Cir. 2022) (precedential). The CAFC denied mandamus, holding that the Director's delegation of institution authority to APJs and the policy of not accepting party requests for Director rehearing of non-institution decisions do not violate the Appointments Clause under United States v. Arthrex. The timing and docket entries are consistent with this mandamus arising from PAN's 2021–2022 Centripetal petitions (of which IPR2021-01153 is one), though the opinion addresses PAN's petitions collectively. No merits appeal exists because there is no FWD.
• Defensive value: For someone facing assertion of the '437 patent today, this proceeding cuts against an IPR-based defense: Palo Alto Networks — the same defendant in the E.D. Va. case and CAFC appeals 25-1167/25-1168 — threw its best Jungck/RFC 2003 combination at all 20 claims and got denied at the threshold. No claim is dead; any infringement theory built on claims 1–20 stands. A new petitioner would need materially different art, and PAN itself is now time-barred under § 315(b) (complaint served in 2021, more than one year before any new petition).

Strategic summary

Claims CANCELED vs. SUSTAINED vs. UNTESTED. No claims of 10,567,437 have been canceled — IPR2021-01153 challenged all 20 claims and was denied institution, so no claim has been "sustained" on the merits either; all 20 claims (1–20) are untested by a Final Written Decision. The only PTAB signal is the institution-stage denial, which found (per the available metadata and decision conclusion) that PAN did not demonstrate a reasonable likelihood of prevailing on any ground. The patent is therefore structurally intact — every claim is enforceable and presumptively valid.

Estoppel landscape. Because no FWD issued, no § 315(e)(2) estoppel attaches to anyone. PAN is free to press Jungck and Jungck+RFC 2003 in the E.D. Va. litigation, subject only to the district court's own assessment (it will know the PTAB already rejected that combination at the institution threshold). For any defendant currently being asserted against: the § 315(b) one-year bar means PAN (and its privies) cannot file another IPR on this patent — the E.D. Va. complaint (2:21-cv-00137) was filed in 2021. A different, non-time-barred petitioner could still petition with new art (Jungck-based grounds would also face § 325(d) if the art was before the Office), but the practical read is that the obviousness art PAN assembled has been tested and found wanting at the PTAB.

Pattern signals. Only one IPR has ever been filed on this patent, and it failed at institution. There is no Unified Patents IPR in the chain — Unified Patents appears only as the litigation-data aggregator in the Google Patents metadata, not as a petitioner. The petitioner was Palo Alto Networks, which is the adverse party in the parallel E.D. Va. case and the CAFC appeals 25-1167/25-1168 (consolidated; reactivated 2025-12-22). Centripetal's broader PTAB history is aggressive: Cisco filed multiple 2018 IPRs on Centripetal family members (IPR2018-01513, -01443, -01444, -01505, -01506 — all cited as exhibits in PAN's petition), and Centripetal secured a CAFC affirmance in the Cisco litigation (847 F. App'x 929, May 11, 2021). Separately — and not on this patent — the CAFC recently vacated and remanded a PTAB FWD in Centripetal Networks, LLC v. Palo Alto Networks, Inc., No. 2023-2027 (Fed. Cir. 2025-10-22), for the Board's failure to consider copying evidence, and rejected Centripetal's APJ-recusal challenge. That case involves a different Centripetal patent (encrypted network-threat detection) but confirms Centripetal's pattern of litigating PTAB outcomes aggressively.

Recommended next steps

• Do not cite any FWD for cancellation — there is none. No claim of 10,567,437 has been invalidated. Any demand or complaint citing claims 1–20 of this patent is not vulnerable to a "canceled claim" attack.
• For the record, the institution denial is Paper 10 in IPR2021-01153 (PTAB, 2022-01-24): available via the USPTO PTAB portal (search "IPR2021-01153") and mirrored at Docket Alarm (Paper 10 PDF; Petition PDF); the related mandamus ruling is In re Palo Alto Networks, Inc., No. 22-145, at CourtListener.
• If you are the defendant (PAN or a privy): you are § 315(b) time-barred from a new IPR on this patent; the realistic validity fight is in E.D. Va. (2:21-cv-00137) and on appeal (CAFC 25-1167/25-1168, reactivated 2025-12-22 — no merits decision yet as of this writing). Weigh Jungck-based arguments there knowing the PTAB already rejected them at the threshold.
• If you are a non-time-barred third party: a fresh petition would need new prior art; Jungck and RFC 2003 have been tried. Expect the Board to view repeat attacks on this patent skeptically, and expect Centripetal to litigate hard (see the 2023-2027 pattern).
• Milestones: there are no pending PTAB trial-stage milestones (no institution, no oral hearing, no FWD due date) — the proceeding is closed. The only live deadlines are in the district court and CAFC appeals, not at the PTAB.

Caveats: I could not retrieve the full text of Paper 10 to quote the panel's institution rationale verbatim, and the ODP structured data shows no AIA trials (the proceeding above was surfaced via web sources, which are consistent with the Google Patents "IPR2021-01153 — Not Instituted (Merits)" litigation metadata). The panel composition reflects the Paper 10 caption; the Patexia panel listing is flagged as likely erroneous.

I'll research the assignment record for US 10,567,437, starting with the USPTO Assignment Center and cross-referencing.

Let me dig into the specific assignment records and correspondent data.

Good leads. Now let me pin down the specific assignment reel/frame records and the security-interest chain.

Ownership-chain reconstruction — US 10,567,437 B2

Method note (read first). I could not reach assignmentcenter.uspto.gov directly in this session — the available tool is a general web search, and the Assignment Center's search results are rendered dynamically and are not indexed. I therefore reconstructed the chain from (a) the Google Patents "Legal Events" block in the authoritative patent text supplied for this analysis (which mirrors USPTO assignment-recordation events), and (b) litigation-record documents filed in the Centripetal cases, including a certified security agreement produced as a docket exhibit. I was not able to retrieve the reel/frame numbers or the recorded correspondent for either transaction. That is a gap, not a finding, and I flag it rather than manufacture numbers.

Inventors

Notes on pattern:

• Only two inventors on the '437 patent. Sibling members of the same family (e.g., US 9,565,213 and US 10,944,792) name four — Rogers, Moore, David K. Ahn, and Peter P. Geremia. The '437 is a pure continuation (App. 16/111,524 → 15/413,834 → 14/698,560 → 13/657,010), so the two-inventor set is exactly the inventorship of the 2012 priority application, carried forward.
• No inventor-departure pattern. Rogers and Moore were still with the company at the 2020 bench trial — eight years after the 2012 priority filing. There is no signal here of the "all inventors gone within 12 months" precursor to a portfolio fire-sale. To the contrary, the inventing team stayed and litigated.
• Assessment date for both is the 2012 priority filing; both are identified as Centripetal personnel in Centripetal's own litigation exhibits and testimony, so the "at filing" attribution is well grounded.

Original assignee

Centripetal Networks, Inc. (Herndon, VA at issue; later Portsmouth, NH per the face of later family members), a Delaware corporation.

• Product/embodiment: Yes — Centripetal shipped and sold products embodying the asserted claims. Its RuleGATE Threat Intelligence Gateway (including a "RuleGATE 2000 Gateway" data sheet, trial exhibit in its ITC complaint) and its CleanINTERNET managed service were both put in evidence as embodiment proof, and the company's site (centripetal.ai) markets them.
• Primary line of business: Network security — high-resolution, threat-intelligence-driven inline packet filtering at network boundaries. Described in the academic case literature (Berkeley Law / Haas "Heiden" case study) as "a VC-backed startup."
• Current status (as of the sources reviewed): Operating. The original assignee did not dissolve, go bankrupt, or get acquired. The only corporate change on the record is a change of name / form — Centripetal Networks, Inc. → Centripetal Networks, LLC, recorded 2023-01-20 per Google Patents legal events ("CHANGE OF NAME"). It remains the plaintiff of record in the suits asserting this patent, including the live Federal Circuit appeal in 25-1167/25-1168 (reactivated 2025-12-22).

Assignment timeline

Caveat restated: reel/frame numbers were not retrievable this session. Both entries below are corroborated by independent sources (Google Patents legal events; the 2023 "Change of Name" event is consistent with the assignee-name shift visible on the face of later family members such as US 11,477,237, which still says "Centripetal Networks, Inc., Portsmouth, NH").

• Executed date not retrieved / recorded 2019-06-05 — Reel NNNNNN/NNNN (not retrieved)

  • Conveyance: Assignment (USPTO-assigned code: "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
  • Assignor: Rogers, Steven; Moore, Sean (individually)
  • Assignee: Centripetal Networks, Inc.
  • Correspondent: not retrieved. Best available proxy: the attorney/agent of record on the face of this patent family is Banner & Witcoff, Ltd. (e.g., "Attorney, Agent, or Firm — Banner & Witcoff, Ltd." on US 9,137,205, US 10,944,792, US 11,477,237). That is the prosecution firm of record, not necessarily the recording correspondent — I could not confirm which entity filed the assignment, so I do not treat this as a recurrence finding.
  • Context: Internal — confirmation of the employee/founder-to-company assignment for the application that issued as the '437. Filed ~10 months after the 2018-08-24 application filing, ~3 years before the first suit asserting this patent.

• Executed date not retrieved / recorded 2023-01-20 — Reel NNNNNN/NNNN (not retrieved)

  • Conveyance: Change of Name (not an assignment of title to a third party)
  • Assignor: Centripetal Networks, Inc.
  • Assignee: Centripetal Networks, LLC
  • Correspondent: not retrieved.
  • Context: Internal corporate reorganization / conversion of form. Same beneficial owner; not a transfer to a licensing entity.

Additional title-clouding item (recorded against the family, not confirmed as recorded against this patent). In Centripetal Networks, Inc. v. Keysight Technologies, Inc., 2:17-cv-00383 (E.D. Va.), a docket exhibit reproduces a USPTO "NEW ASSIGNMENT / NATURE OF CONVEYANCE: SECURITY INTEREST" filing (EPAS ID PAT4374169) under a Patent and Trademark Security Agreement executed 2017-04-17 between Centripetal Networks, Inc. (Delaware corporation), as Grantor, and Douglas A. Smith, an individual, as Grantee, in connection with a Note and Warrant Purchase Agreement. The reciting party address of record was 12770 Merit Drive, Suite 800, Dallas, TX 75251 (with a 617-area code contact). The collateral list for that recording comprises 14 properties, including US 9,137,205 and US 9,560,077 — the '437's direct ancestors — plus an application number. Because the '437 did not exist in April 2017, it could only have been captured by an "after-acquired patents" collateral clause, if at all. I found no record of a release, and no record specifically encumbering 10,567,437. Treat this as a venture-debt securitization affecting the family; status as to the '437 unclear — verify on Assignment Center before relying on clean title.

Timeline diagram

timeline
    title Ownership of US 10567437
    2012 : Priority application filed by Centripetal Networks Inc
    2017 : Family patents pledged as collateral to a lender
    2018 : Continuation application filed
    2019 : Founders assign interest to Centripetal Networks Inc
    2020 : Patent issued
    2023 : Name change to Centripetal Networks LLC

NPE / troll-pattern signals

1. Shell-entity transfer — not present.
The only post-issuance ownership event is the 2023-01-20 change of name (Centripetal Networks, Inc. → Centripetal Networks, LLC), recorded as a Change of Name, not an assignment to a new party. There is no "IP / Patents / Licensing / Holdings / Ventures" transferee anywhere in the record. The LLC is the same operating business: it is the named plaintiff asserting the '437, and its RuleGATE and CleanINTERNET products were put in evidence as commercial embodiments. Note the counter-signal: an entity that ships products and sues competitors is not a shell regardless of the LLC suffix.

2. Known asserter in the chain — not present.
Neither recorded party (Centripetal Networks, Inc. / Centripetal Networks, LLC) nor either inventor matches any listed NPE (Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Spangenberg entities), and none appears in the Unified Patents or RPX asserter directories for this patent. Centripetal is the target of third-party IPR activity (Cisco's IPR2018-01443/-01444/-01505/-01506/-01513; Palo Alto's IPR2021-01153; Keysight's petitions), not a repeat petitioner-lender shop.

3. Repeat correspondent across the chain — unclear / not established.
Repeat correspondent requires the recorded correspondent name on each reel/frame. I could not retrieve the correspondents on either recording. The only recurring name I can verify is Banner & Witcoff, Ltd. as attorney of record on the face of the family patents. One firm handling prosecution across a straight continuation chain is normal continuity, and Banner & Witcoff is a large general-practice IP firm serving operating companies — a single firm appearing on a patent family is expressly not a finding under the stated criterion. I decline to elevate this to a signal.

4. Cascading transfers — not present.
Two recorded events across four years: one founder-assignment (recorded 2019-06-05) and one change of name (recorded 2023-01-20). No chain of LLCs, no shared registered-agent address across successive assignees, no sub-24-month multi-hop sequence.

5. Pre-litigation transfer — not present.
The first suit asserting the '437 is Centripetal Networks, LLC v. Palo Alto Networks, Inc., No. 2:21-cv-00137 (E.D. Va., filed 2021-03-12). The nearest prior recording, the founder assignment, was recorded 2019-06-05 — roughly 21 months earlier, which falls outside the 6-month window, and it was a routine confirmatory assignment of inventor interest, not a transfer arranged to manufacture standing or venue. The 2023 change of name post-dates the litigation.

6. Bankruptcy fire-sale — not present.
No Chapter 7/11 proceeding, trustee sale, or § 363 patent sale was located for Centripetal Networks. The company is identified as an operating, VC-backed business through at least 2023–2024 in the litigation record, and as the active record owner and appellant in 2026. (Caveat: I did not run a PACER bankruptcy-docket check; if Centripetal has undergone an insolvency event after my sources, this call would need revisiting.)

7. Privateering — not present.
Privateering means an operating company transfers to a third-party NPE that asserts on its behalf. Here the operating company asserts in its own name — Centripetal is the plaintiff in 2:21-cv-00137, the complainant in ITC Inv. No. 337-TA-1314, and the cross-appellant in CAFC 25-1167/25-1168. The patent never left the operating company.

8. Defensive aggregator — not present.
The chain terminates at Centripetal Networks, LLC, not at RPX, AST, LOT Network, Unified Patents, or OIN. (Unified Patents appears in the Google Patents metadata only as the litigation-data aggregator, and appears in the patent-art list as an entity that petitioned against family members — neither is a chain-of-title fact.)

Verdict

Operating-company assertion.

The chain is short and unremarkable: founders Rogers and Moore assigned to Centripetal Networks, Inc. (recorded 2019-06-05), and the entity later recorded a change of name to Centripetal Networks, LLC (recorded 2023-01-20) — same owners, same business, no third-party transferee. Centripetal ships products embodying the claims (RuleGATE Threat Intelligence Gateway, CleanINTERNET) and is suing actual competitors — Cisco, Palo Alto Networks, Keysight — in its own name, including the still-pending appeal 25-1167/25-1168. No NPE signal rises above "unclear," and none is supported by concrete reel/frame or documentary evidence.

Verification links.

• USPTO Assignment Center (search by patent number 10567437): https://assignmentcenter.uspto.gov/ — and the legacy index: https://assignment.uspto.gov/patent/index.html
• Google Patents legal events for the patent: https://patents.google.com/patent/US10567437/en

Open items I could not close (do not treat as resolved):

1. Reel/frame numbers and recorded correspondents for both entries — required for signal 3 and for exact citation; must be pulled from Assignment Center directly.
2. Whether the 2017-04-17 Douglas A. Smith security interest (EPAS ID PAT4374169) was ever extended to, or released as to, US 10,567,437 — its collateral list omits the '437 only because the patent did not yet exist.
3. Maintenance-fee status — Google Patents shows "Active," anticipated expiration 2032-10-22; not independently verified.
4. No PACER bankruptcy-docket check was run for Centripetal Networks; signal 6 rests on absence of evidence.

Cross-check against the earlier-generated sections: this ownership reconstruction is consistent with the "Patent summary" (original assignee Centripetal Networks, LLC, with 2019-06-05 assignment and 2023-01-20 change of name) and with the litigation section (Centripetal as plaintiff). No contradictions found. One refinement to flag: the litigation section's "Unknown/unverified" entry 1:21-cv-00313 is unrelated to title, and I found nothing in the assignment record that bears on it.

USPTO / Prior-Art Search Results for US Patent 10,567,437 (US10567437B2)

Method and sourcing caveats (read first)

I attempted to pull the examiner's "References Cited" section from USPTO PatentCenter/Public PAIR and the Google Patents "Citations" tab for US10567437B2. The full face-of-patent citation list was not fully retrievable through the available web-search tooling. What I did verify:

1. US-6826694-B1 ("High Resolution Access Control") is linked as a citation to US10567437B2 in the Unified Patents patent record (search result: US-6826694-B1 - High Resolution Access Control - US-10567437-B2).
2. US-20090262741-A1 ("Transparent Provisioning of Services Over a Network") appears on the US10567437B2 record (the Google Patents page for that reference lists US10567437B2 as citing it).
3. The Unified Patents portal page for US-10567437-B2 contains a "Patent Art (291)" list — this is the prior-art set assembled for the IPR2021-01153 petition (which the PTAB declined to institute — "Not Instituted – Merits") and/or the E.D. Va. litigation (2:21-cv-00137, Centripetal v. Palo Alto Networks). This is challenger-generated art, not the examiner's citation list.
4. I did not obtain the granted claim text of the B2 patent; claim mapping below uses the published-application claims (US20190230128A1) — independent method claim 2 and independent system claim 9 — which the earlier summary flagged as the best-available-but-not-authoritative version.

Per the operating rules, I flag this explicitly: I cannot assert with high confidence that the list below is the complete examiner-cited set. What follows is the verified relevant art from the searches, with confidence levels per item.

Verified prior-art references (with § 102 analysis)

Claim architecture used for mapping (published application):

• Claim 2 (method): provisioning a packet security gateway at a protected-network boundary with packet-filtering rules (each rule pairs a packet-matching criterion tied to malicious traffic with a packet-transformation function); configuring the gateway to receive packets through a communication interface having no network-layer address, to drop matching packets, and to modify a LAN switch's switching matrix so the switch also drops matching packets.
• Claim 9 (system): processor + memory configured to perform the same steps.
• Key dependents: claim 3 (spoofed-source-address rules), claim 4 (subscription-service threat intelligence), claim 5 (network-layer-addressed management interface), claim 6 (encapsulation/rerouting to a different address, e.g., a monitoring device), claim 7 (Layer-2 VLAN rerouting), claim 8 (dynamic rule creation/alteration on policy update); claims 10–15 mirror claims 3–8.

1. US 6,826,694 B1 — "High Resolution Access Control" (verified citation)

• Citation: US 6,826,694 B1; issue date January 12, 2004 (filed in the late-1990s/2000 timeframe per the record).
• Description: Directed to high-resolution, granular access control / packet filtering — precisely the "filter substantially all traffic at high resolution" problem the '437 patent's Background calls out as the scalability challenge. This appears to be an earlier-generation design in the same technical lineage (same inventor field, Steven Rogers), i.e., a likely self-citation.
• § 102 analysis: Potentially anticipates claims 2 and 9 (and the filtering-centric dependents 3 and 4) if it discloses: a gateway at a network boundary, criteria-based packet-filtering rules paired with transformation/deny functions, and high-resolution filtering. Gaps to check against granted text: whether it discloses the no-network-layer-address interface and the LAN-switch switching-matrix modification — I could not verify those disclosures from the available snippet.
• Confidence: Citation itself verified; element-by-element mapping not fully verified — treat as high-priority art requiring full-text review.

2. US 2009/0262741 A1 — "Transparent Provisioning of Services Over a Network" (verified citation; assignee per record: Lookingglass Cyber Solutions LLC; priority date listed: 2000-06-22)

• Citation: US 2009/0262741 A1; published October 22, 2009.
• Description: Discloses provisioning network services (including security services) via inline devices that operate transparently — i.e., not addressed at the network layer while still processing traffic at higher layers. This maps almost directly onto the '437 patent's "network layer transparent" gateway features (spec: interfaces 206/208 not addressed at the network layer; PSG still performs network-layer transformation functions).
• § 102 analysis: Strongest candidate against the transparency element of claims 2 and 9, and by extension the dependent-management-interface claims (claim 5). Whether it discloses the LAN-switch switching-matrix modification and the malicious-traffic-criteria limitation is unverified; if not, it may only partially anticipate or read as an obviousness reference rather than a full § 102 anticipation.
• Confidence: Citation verified; substantive disclosure details from the abstract/body not fully retrieved — requires full-text confirmation.

3. US 2006/0070122 A1 — "Method and Apparatus for a Distributed Firewall" (verified in Patent Art list; priority date listed: 1999-06-29; current assignee listed: RPX Corp.)

• Citation: US 2006/0070122 A1; published April 6, 2006.
• Description: Discloses a distributed firewall architecture with centrally managed security policy pushed to enforcement points — analogous to the '437 security-policy-management-server → packet-security-gateways model.
• § 102 analysis: Potentially anticipates claims 2 and 9 on the provisioning rules to a boundary gateway from a management server and drop-matching-packets elements. Unverified: the no-network-layer-address stealth interface and LAN-switch switching-matrix limitation.
• Confidence: Listed in the IPR/litigation art set (verified); disclosure details not fully retrieved.

4. US 2007/0083924 A1 — "System and Method for Multi-stage Packet Filtering on a Network-Enabled Device" (verified in Patent Art list; filed 2005-10-07)

• Citation: US 2007/0083924 A1; published April 12, 2007.
• Description: Discloses multi-stage packet filtering on a network device — conceptually similar to the '437's packet filter → multiple packet-transformation-function pipeline.
• § 102 analysis: Relevant to claims 2/9's rule-criteria → transformation-function pairing and to dependent claims specifying transformation functions (e.g., claims 6/7-style rerouting) if disclosed therein. Anticipation of the full independent claims is unlikely on its own unless it also discloses the stealth interface and LAN-switch programming — unverified.
• Confidence: Listed in IPR/litigation art set (verified); disclosure details not retrieved.

5. US 2006/0133377 A1 — Fortinet — "System and Method for Integrated Header, State, Rate and Content Anomaly Prevention with Policy Enforcement" (verified in Patent Art list; filed 2004-12-21)

• Citation: US 2006/0133377 A1; published June 22, 2006.
• Description: Discloses integrated anomaly detection and policy enforcement — i.e., criteria-based classification of traffic with enforcement actions.
• § 102 analysis: Relevant to the packet-matching criteria paired with transformation functions element of claims 2/9. Full anticipation depends on unverified disclosures (stealth interface, LAN-switch modification).
• Confidence: Listed in IPR/litigation art set (verified); disclosure details not retrieved.

6. US 7,792,775 B2 — NEC — "Filtering Rule Analysis Method and System" (verified in Patent Art list; filed 2005-02-23)

• Citation: US 7,792,775 B2; issued September 7, 2010.
• Description: Discloses analysis/merging/transformation of filtering rules — relevant to the '437 specification's discussion of merging overlapping rules into non-overlapping rule sets (FIG. 3 discussion).
• § 102 analysis: Not a standalone anticipator of claims 2/9 (it addresses rule analysis, not boundary-gateway provisioning with stealth interfaces); better suited to an obviousness combination or to dependent rule-structure claims.
• Confidence: Listed in IPR/litigation art set (verified).

7. US 2005/0071650 A1 — ETRI — "Method and Apparatus for Security Engine Management in Network Nodes" (verified in Patent Art list; filed 2003-09-28)

• Citation: US 2005/0071650 A1; published March 31, 2005.
• Description: Discloses management of security engines in network nodes — policy distribution/management to network enforcement points.
• § 102 analysis: Relevant to the management-server → gateway provisioning model of claims 2/9. Anticipation gap analysis as with items 3–5 (unverified stealth/switching-matrix elements).
• Confidence: Listed in IPR/litigation art set (verified).

8. US 7,853,996 B1 — "Methodology, Measurements and Analysis of Performance and Scalability of Stateful Border Gateways" (verified in Patent Art list; filed 2003-10-02)

• Citation: US 7,853,996 B1; issued December 14, 2010.
• Description: Addresses stateful border-gateway scalability/performance — the exact "scalability of high-resolution filtering" problem motivating the '437 invention.
• § 102 analysis: Diagnostic/analytical in nature; not a plausible single-reference anticipator of the structural method/system claims. Obviousness-combination relevance only.
• Confidence: Listed in IPR/litigation art set (verified).

9. US 2001/0039579 A1 — "Network Security and Surveillance System" (verified in Patent Art list; filed 1996-11-05)

• Citation: US 2001/0039579 A1; published November 8, 2001.
• Description: Discloses network security monitoring/surveillance — relevant to the '437's monitoring-service embodiment (copy/reroute packets to a monitoring device, FIG. 6).
• § 102 analysis: Relevant to dependent claims directed at rerouting/copying to a monitoring device (published claim 6 family). Not a standalone anticipator of claims 2/9.
• Confidence: Listed in IPR/litigation art set (verified).

10. US 2015/0237012 A1 — "Filtering Network Data Transfers" (verified in Patent Art list; filed 2013-03-11)

• Citation: US 2015/0237012 A1; published August 20, 2015. Note: filed after the '437's earliest priority date (Oct. 22, 2012), so it is not § 102 prior art against the '437 claims under pre-AIA § 102 (or under AIA § 102(a)(2) for the full claim scope) — relevant only for § 103 if at all.
• Confidence: Listed in IPR/litigation art set (verified); prior-art status as to date is a red flag.

11. US 7,080,581 B1 — "Security System Design Supporting Method" (verified in Patent Art list; filed 1999-11-29)

• Citation: US 7,080,581 B1; issued July 25, 2006.
• Description: Design-support methodology for security systems; secondary relevance to rule/policy engineering.
• § 102 analysis: Not a plausible single-reference anticipator of the independent claims.
• Confidence: Listed in IPR/litigation art set (verified).

12. Lower-relevance items from the "Patent Art (291)" set (verified as listed, minimal analysis)

• US 9,419,942 B1 (Palo Alto Networks, "Destination Domain Extraction for Secure Protocols," priority 2013-06-04) — post-priority-date filing; not § 102 prior art against the full claims; not claim-relevant on its face.
• US 2015/0347246 A1 (Hitachi, "Automatic-Fault-Handling Cache System…," priority 2012-12-04) — post-priority-date filing; cache-fault handling, low claim relevance.
• US 7,499,412 B2 (filed ~2005) — listed in the art set; I could not verify its title/disclosure from the available snippets; do not rely on my characterization — flagged as unverified.
• US 7,080,581 / US 7,792,775 / US 7,853,996 — see above.

Bottom-line ranking of "most relevant" prior art

Explicit uncertainty statement (per operating rules)

• The complete examiner "References Cited" list on the face of US10567437B2 was not retrieved. The analysis above draws on (a) two confirmed citation links (US 6,826,694 B1; US 2009/0262741 A1) and (b) the Unified Patents "Patent Art (291)" list, which is petitioner/litigation art from IPR2021-01153 (not instituted) and E.D. Va. 2:21-cv-00137, not examiner citations. Those are different populations; do not conflate them.
• Granted-claim text is unverified; mapping uses the published-application claims (claims 2 and 9 independent). If prosecution amendments narrowed the granted claims (e.g., added the LAN-switch limitation), the anticipation analysis shifts accordingly.
• Element-by-element disclosure verification for each reference requires full-text review, which the search tooling did not permit; where a mapping element is unverified, I say so rather than infer it.
• The fact that IPR2021-01153 was not instituted is relevant context: the PTAB did not find the petition's art sufficient to establish a reasonable likelihood of prevailing — a signal that none of the references above, as presented in that petition, cleanly reads on the claims.

Obviousness Analysis — U.S. Patent 10,567,437 B2 ("Methods and systems for protecting a secured network")

1. Scope note and claim reconstruction

Before analyzing § 103, I must flag an evidentiary limitation that affects precision: I could not retrieve the exact granted claim text of the B2 patent from any source in this session. The earlier sections of this analysis relied on the pre-grant publication US20190230128A1 (claims 2 and 9 independent, claim 1 canceled). The IPR and PTAB records, however, confirm the granted patent has 20 claims with three independent claims — claims 1, 8, and 15 — and that the preambles of all three are limiting (Patent Owner's Preliminary Sur-Reply, IPR2021-01153; see ptacts.uspto.gov document and the IPR2021-01153 institution decision, Paper 10, Jan. 24, 2022). The IPR record also confirms a key preamble/limitation phrase: "a plurality of packet security gateways that collectively provide an entire interface across a boundary of a network protected." The granted claims therefore differ in form from the published claims 2/9, and my mapping below is necessarily based on (a) the IPR-confirmed claim structure, (b) the published-application claims, and (c) the specification (provided in full), which supports the recited features. Any conclusion should be re-validated against the actual granted claim text (USPTO Patent Center) before reliance in litigation or IPR.

The core limitations at issue, drawn from the specification and the record, are:

• L1 (preamble): a plurality of packet security gateways that collectively provide an entire interface across a boundary of a protected network (per IPR record).
• L2: a dynamic security policy received from a security policy management server, comprising rules pairing packet-matching criteria with packet-transformation functions.
• L3: at least one packet-transformation function other than forwarding or dropping (e.g., rerouting/encapsulation to a monitoring device, IPsec-stack delivery, VLAN assignment, queueing).
• L4: network-layer-transparent operation — traffic received/sent on interfaces not addressed at the network layer, while transformation occurs at the network layer; separate management interface with a network-layer address, secured at the application level (e.g., SSH).
• L5: dropping packets matching criteria tied to malicious network traffic (blocklist/allowlist constructs).
• L6: modifying a LAN switch's switching matrix so the switch also drops matching packets.
• Dependent features: five-tuple rules; DSCP-based differentiated queueing; SIP-URI-based rerouting; VoIP-session-derived rules; subscription threat-intelligence feeds; IPsec security-association matching; phased restoration; series-configured gateways.

2. Prior art of record (the "Prior Art section" of the patent's family/records)

The examiner's References Cited (per PubChem's copy of the B2 bibliographic record) and the litigation/IPR record supply the following material references:

3. Primary § 103 combinations

Combination 1 — Jungck alone (claims 1–4, 6–11, 13–18, 20)

Asserted by the petitioner. Jungck discloses a high-speed, rule-based packet-processing system in which received packets are classified against rules and subjected to specified processing actions (forward, drop, and other transformations). For the preamble (L1), Jungck's architecture distributes processing across multiple packet-processing engines arranged to provide an interface between networks. For L2/L3, Jungck's rules pair matching criteria with configurable actions, including actions beyond simple pass/drop. For L5, block/drop rules tied to identified malicious traffic are routine. For the five-tuple, DSCP, queueing, and IPsec-stack dependent features, Jungck's programmable classification and action engine provides a natural home, combinable with the well-known DiffServ (RFC 2474/2475) and IPsec (RFC 4301) stacks.

Why the PTAB denied institution (important): the Board found the petitioner did not establish a reasonable likelihood on any independent claim (Paper 10). Based on the record, the likely gaps in Jungck alone are: the plurality-of-gateways-collectively-providing-an-entire-interface preamble (L1); the network-layer-transparent interface with an unaddressed data interface plus a separately addressed, application-secured management interface (L4); and the LAN-switch switching-matrix modification (L6). These are exactly the limitations a stronger combination must source from other references.

Combination 2 — Jungck + RFC 2003 (claims 5, 12, 19)

Asserted by the petitioner. Claims 5/12/19 (per the IPR) add rerouting of matched packets to a network address different from the destination — the monitoring-service embodiment of the specification (Fig. 6). RFC 2003 teaches precisely this: encapsulating an IP packet with an outer IP header so it is routed to an intermediate device, which can copy/inspect and then forward on. Motivation: the POSITA seeking to tap or mirror traffic at a security gateway — a routine lawful-intercept/IDS function — would use RFC 2003's encapsulation to redirect a copy to a monitoring device while preserving the original destination, exactly as the specification describes (col. 55–60 area: "encapsulate … with an IP header specifying a network address different from their respective destination addresses"). The combination is the textbook use of a standards-track encapsulation technique to implement a known function, with predictable results.

Combination 3 — Jungck + AT&T '694 (High Resolution Access Control)

AT&T '694 (claim 1 verbatim, from the record): "A method for filtering a packet … receiving a packet having at least one header parameter and a payload; selecting an access rule based upon the contents of the payload … implementing the access rule … selected based upon a combination of the contents of the packet … and the contents of at least one other packet." It also discloses rules "obtained from a node external to the firewall."

• Fills L2 (dynamic policy provisioning): '694's external-node rule delivery is the management-server-to-gateway policy channel; combined with Jungck's rule engine, the POSITA would configure the gateway to receive updated rule sets from a policy server — the "dynamic security policy received from the security policy management server" limitation.
• Fills the high-resolution-filtering problem the patent itself identifies as the reason proactive filtering was "deemed untenable": '694's payload/state-based rules and the Qiu/Fulp/Warkhede/Christiansen optimization literature (all of record) directly address the scalability concern that the specification concedes was the barrier.

Motivation: The patent's own Background frames high-resolution filtering as the core challenge; '694 is the canonical solution to that challenge and predates the '437 priority date by 14 years. A POSITA designing a scalable proactive filter would combine Jungck's high-speed engine with '694's high-resolution rule-selection and external provisioning as a matter of routine design.

Combination 4 — Jungck + Microsoft policy distribution (US 2006/0129808 / EP 1677484)

The Microsoft reference discloses a security-policy management server distributing policies to multiple enforcement engines at protected hosts, where each rule has a condition and an action and a rule type directing the rule to the correct enforcement engine.

• Fills L1/L2: the plurality of enforcement points, the central server, and the rule = (criterion → action) structure are all present. The POSITA would substitute Jungck's packet gateways for the host-based enforcement engines (or combine both) to scale the Microsoft distribution model to boundary devices — a predictable substitution of one known enforcement point for another in the same field.
• Motivation: centralized, dynamic policy management of distributed security devices was a well-documented trend by 2012 (also reflected in the Sourcefire and Palo Alto zero-trust literature of record). The POSITA would be motivated by operational efficiency (single-point policy updates, uniform enforcement at every boundary — the exact "uniform manner" goal of Figs. 7–8).

Combination 5 — Jungck + transparent/stealth-firewall prior art (L4)

The network-layer-transparent gateway (unaddressed link-layer data interfaces + addressed management interface secured at the application layer) was a well-known transparent firewall / "bump-in-the-wire" design long before 2012 (e.g., Cisco PIX/ASA transparent-mode and comparable products, which bridge at Layer 2 without IP addresses on data interfaces and are managed out-of-band/SSH).

• Fills L4: a POSITA implementing Jungck's rule engine as an inline boundary device would routinely place it in transparent/L2 mode to avoid renumbering the protected network and to make it unaddressable (and thus less attackable — the specification's own stated rationale at Fig. 2: "attack packets cannot be routed to the network interfaces"), with an out-of-band management interface secured via SSH/TLS.
• Motivation: non-disruptive deployment, stealth, and management security — all explicit in the specification — are standard design goals; the combination is the straightforward application of known transparent-firewall techniques to Jungck's engine.

Combination 6 — Jungck + switch-ACL/TCAM art (L6)

Programming a LAN switch's switching matrix/forwarding table (TCAM/ACL) to drop traffic matching a policy is standard switch security (port ACLs, VLAN ACLs, storm control, and switch-integrated security features).

• Fills L6: once the gateway derives drop criteria from its dynamic policy, programming the adjacent LAN switch to enforce the same drop set is an obvious defense-in-depth step — the specification itself says the switch is modified "so the switch also drops those matching packets," i.e., to extend enforcement to traffic that may not physically traverse the gateway (LAN-to-LAN flows, Fig. 9).
• Motivation: ensuring all traffic at the boundary is filtered (the patent's own scalability/full-coverage rationale); using the switch's hardware forwarding tables offloads enforcement from the gateway. The mechanism (writing filter entries into a switch's switching matrix) is a known, predictable use of switch ACL hardware.

Combination 7 — Jungck + subscription threat-intelligence feeds (dependent claim on malicious-address lists)

The "list of known network addresses associated with malicious network traffic … received from a subscription service" is precisely the DShield/Spamhaus/Emerging Threats model (aggregated malicious-IP feeds) that predates 2012.

• Fills the dependent limitation: a POSITA would feed such a list into Jungck's rule engine to auto-generate drop rules — the specification's Fig. 5 "malicious host tracker service 508" is a direct description of this known model.
• Motivation: automation of blocklist maintenance, reduction of human error, timeliness — all standard reasons to subscribe to threat feeds.

Combination 8 — Jungck + SIP-aware firewall/session-border-controller art + DiffServ (VoIP and DSCP dependents)

• VoIP-session-derived rules: SIP-aware firewalls and session border controllers that open/close pinholes based on SIP signaling were mature by 2012. The POSITA would combine Jungck with a softswitch/SBC signaling feed (as in Fig. 6) to create and tear down per-session rules — the patent's VoIP firewall service.
• DSCP-based differentiated queueing: RFC 2474/2475 DiffServ and DSCP-based scheduling/queuing in routers is foundational prior art; adding a DSCP selector to Jungck rules and mapping matched traffic to different forwarding queues with different rates is an obvious application of DiffServ to the gateway's egress queues (the specification's enqueueing service).

4. Why a POSITA would combine (general § 103 framework)

1. Same field, same problem. Every reference above is in network security/packet processing — the exact field of the '437 patent. The patent's Background concedes the problem (scalable, high-resolution, proactive filtering) was known; the Qiu/Fulp/Warkhede/Christiansen papers of record show the POSITA community was actively working on ordered, high-resolution rule sets.
2. Known elements, predictable combination. Each disputed limitation maps to a known technique: dynamic policy distribution (Microsoft; AT&T '694 external-node rules), transparent/stealth inline filtering (L2 firewall art), switch-ACL enforcement, RFC 2003 encapsulation, DiffServ queueing, SIP-pinhole firewalling, and threat feeds. § 103 does not require a reference to teach the combination; it requires the combination to be obvious, and here the specification itself describes each feature as a conventional "service" layered onto a rule engine.
3. Design incentives. The specification's own stated benefits — uniform enforcement at every boundary (Figs. 7–8), non-disruptive deployment (transparency), scalability via series-configured gateways (Fig. 4), and centralized management (Fig. 5) — are precisely the incentives that would drive a POSITA to combine Jungck with the policy-distribution, transparency, and switch-enforcement references.
4. Reasonable expectation of success. All components are off-the-shelf technologies with known interfaces; no new protocol or hardware is required. The series-gateway configuration (Fig. 4) is a straightforward application of the known "defense in depth"/chained-firewall model, and the two-policy decomposition (internal hosts / external hosts) is a standard rule-partitioning technique (cf. the rule-optimization NPL of record).

5. Countervailing factors and the record's cautionary signals

• The PTAB declined to institute IPR2021-01153 on the Jungck and Jungck + RFC 2003 grounds. That decision is not a merits finding of patentability, but it is a meaningful signal that Jungck alone (and with RFC 2003) was not shown to disclose the full independent claims — most plausibly the plurality-of-gateways preamble, the network-layer-transparent interface/management-interface combination, and the switch-matrix modification. A robust § 103 case therefore requires the additional references (Combinations 3–6), not just the IPR grounds.
• The "Law" reference relied on by the petitioner in the IPR reply for the plurality-of-gateways preamble could not be identified in this session; its strength is an open question.
• Secondary considerations are likely to be contested (long-felt need for scalable proactive filtering; industry praise; licensing/settlement activity in the Centripetal litigation family). The CAFC appeals 25-1167/25-1168 (reactivated Dec. 22, 2025) may produce a merits decision that bears on claim construction and validity, but none was located as of this analysis.
• § 101 exposure (raised in the Ixia/Keysight case, 2:17-cv-383) is outside the § 103 scope of this task but worth noting: the claims' emphasis on functional "services" could make eligibility an alternative invalidity path.

6. Bottom line

• Most vulnerable claims: the dependent claims adding encapsulation-based monitoring (claims 5/12/19 over Jungck + RFC 2003, as the IPR alleged), the subscription-feed malicious-address limitation, and the DSCP/queueing and VoIP-session limitations — all map to mature, well-documented prior art.
• Hardest limitations to reach under § 103: (i) the "plurality of packet security gateways that collectively provide an entire interface across a boundary" preamble; (ii) the combination of network-layer-transparent data interfaces plus a separately addressed, application-secured management interface; and (iii) the LAN-switch switching-matrix modification. Each requires a specific secondary reference (transparent-firewall art, switch-ACL art) rather than Jungck alone — which is consistent with the PTAB's institution denial.
• Strongest overall combination: Jungck (rule engine) + AT&T '694 (high-resolution rules, external provisioning) + Microsoft EP 1677484/US20060129808 (policy-server distribution to multiple enforcement points) + transparent-firewall art (L4) + switch-ACL/TCAM art (L6) + RFC 2003 (monitoring reroute), with the threat-feed, DiffServ, and SIP-pinhole references filling the dependent features. Under KSR-style reasoning (known elements, predictable combination, same field, express design incentives), this composite has a realistic chance of rendering the independent claims obvious — provided the granted claim text does not contain undisclosed limitations beyond those confirmed in the IPR record.

Key caveat restated: the granted claim text was not fully verified this session; the mapping relies on the IPR-confirmed claim structure (independent claims 1, 8, 15; plurality-of-gateways preamble) and the published-application claims. Before relying on this analysis, pull the granted claims from USPTO PatentCenter and re-validate each limitation mapping, and verify the identity and dates of the "Law," Wood, and Palo Alto zero-trust references.