US 9755931

I'll run targeted searches for this exact patent number against USPTO and CAFC 2026 dockets, then summarize.

US Patent 9,755,931 — Concise Summary

Search scope note: I searched the USPTO record (Google Patents/USPTO feed, Unified Patents portal), and CAFC 2026 dockets. Results below are specific to patent number 9,755,931. No sibling/related numbers (e.g., 8,706,105; 10,547,530; 7,855,988) are substituted for it; where they appear, they are labeled as family members.

Bibliographic data (literal)

Assignment of note: A Patent Security Agreement recorded 2025-03-25 to Piccadilly Patent Funding LLC, as Security Holder. This is a collateral/security-interest filing; Lemko remains the record owner and the asserting plaintiff (consistent with the prior assignment-history section).

Abstract (as issued)

"A method includes detecting a failure condition relating to a first distributed mobile architecture (dMA) gateway (dMAG) at a dMAG management system. The dMAG management system is in communication with at least the first dMAG, a second dMAG, and dMA nodes. The method also includes determining that the first dMAG is offline based on the failure condition, selecting the second dMAG, sending a first notification from the dMAG management system to the second dMAG, and sending a second notification from the dMAG management system to an external system. The external system is configured to connect calls to a mobile station via the first dMAG. The first notification instructs the second dMAG to take over operations from the first dMAG. The second notification indicates that the external system is to connect subsequent calls to the mobile station via the second dMAG."

Plain-language overview of the independent claims

Claim 1 — Method (gateway-failover orchestration by a central manager). A centralized dMAG management system — which sits above a first dMAG (gateway), a second dMAG, and a group of dMA nodes — detects a failure condition associated with the first dMAG. In response, it sends three notifications:

1. First notification → second dMAG: instructs it to take over operations of the first dMAG.
2. Second notification → a dMA node (the node that normally connects a mobile-station call to the external/legacy system through the first dMAG): tells it the second dMAG has taken over and to route the mobile station's calls via the second dMAG.
3. Third notification → the external system (legacy network side, reached via the dMAG interface): instructs it to connect subsequent calls to the mobile station via the second dMAG.

In short: one failure detection, then a coordinated three-party reroute (backup gateway + edge node + legacy/external side).

Claim 7 — System (same scheme, apparatus form). A dMAG management system comprising: (a) an operation determination module that receives messages and detects the failure condition on the first dMAG; (b) a dMAG selection module that selects the second dMAG in response to detection; and (c) a notification module that sends the same three notifications (to the second dMAG, to the dMA node, and to the external system). Claim 8 adds the conventional processor/memory implementation (claim 7 is a functional-module mirror of claim 1's method).

Claim 12 — Machine-readable storage device (Beauregard claim). Instructions that, when executed, cause a processor to perform the same operations as claim 1 — detect the failure condition at the dMAG management system, then send the first, second, and third notifications.

Dependent claims (the "failure condition" variants):

• Claims 2, 9, 13: detection from a message sent by the first dMAG itself saying it is going offline.
• Claims 3, 10, 14: detection from a message from a dMA node local to the first dMAG indicating failure.
• Claims 4, 15: additionally notify the dMA node that the second dMAG is taking over.
• Claims 5, 11, 16: detection from a message from the external system's dMAG interface indicating failure.
• Claim 6: detection by failing to receive a periodic heartbeat from the first dMAG.

No dependent claim adds new detection technology — they enumerate known detection mechanisms (graceful-offline message, adjacent-node failure report, dMAG-interface report, heartbeat timeout).

Litigation / PTAB / CAFC 2026 status (specific to 9,755,931)

• District court: The '931 is one of eight asserted patents in Lemko Corp. v. [Microsoft Corp.](/litigations/by-plaintiff/Microsoft%20Corp.) & Affirmed Networks, Inc., No. 3:22-cv-00363-L-BT (N.D. Tex., filed Feb. 14, 2022).
• PTAB: I found no IPR/PGR/CBM petition naming 9,755,931. The Lemko–Microsoft AIA proceedings target sibling patents (e.g., IPR2023-00531 → U.S. 7,855,988; IPR2023-00529/00530 → U.S. 7,653,414; IPR2025-01142 → U.S. 8,310,990), not the '931. All 16 claims therefore remain untested in any AIA trial, and no § 315(e)(2) estoppel attaches on this patent.
• CAFC 2026 dockets — nothing naming 9,755,931. The only 2026 Federal Circuit activity in the Lemko–Microsoft dispute is Lemko Corp. v. Microsoft Corp., No. 2024-2336, decided April 10, 2026 by a Rule 36 summary AFFIRMANCE (Hughes, Stoll, Seeborg). That appeal arises from IPR2023-00531, which is directed to U.S. Patent No. 7,855,988 — a sibling dMA-family patent — not 9,755,931. I found no 2026 CAFC docket entry, appeal, or opinion specifically identifying 9,755,931.

Uncertainties / flags

1. Priority date discrepancy (06-26 vs 06-27). The patent text and Google Patents give the parent filing date as June 27, 2008; the Unified Patents portal page for US-9755931-B2 lists a priority date of 2008-06-26. I flag this rather than resolve it; verify against the USPTO filing receipt for App. 12/163,601.
2. Complete front-page "References Cited" list not fully retrievable in this session (carried over from the prior-art section); the citation list is only partially verified.
3. Examiner/attorney fields (Joshua Schwartz; Toler Law Group) are corroborated across Google Patents and Justia but were not independently pulled from USPTO Patent Center in this session.
4. Inventor address appears as "Kildeer, IL" (some OCR renders it corrupted); the name Shaowei Pan is consistent across sources.
5. Assignee naming varies by source as "Lemko Corporation"/"Lemko Corp" — these are the same entity, not a transfer to a third party.

If you want, I can next pull the full "References Cited" list and the complete claim set from USPTO Patent Center (App. 14/184,120) to close the two open verification gaps above.

Litigation Involving U.S. Patent No. 9,755,931

Based on searches of district-court dockets, PTAB records, Unified Patents, Justia, Docket Alarm, and Lemko's own announcements, the only litigation I can confirm as actually involving U.S. Patent No. 9,755,931 is the following district-court action.

1. Lemko Corporation v. Microsoft Corporation and Affirmed Networks, Inc.

• Plaintiff: Lemko Corporation
• Defendants: Microsoft Corporation; Affirmed Networks, Inc.
• Jurisdiction: United States District Court for the Northern District of Texas, Dallas Division
• Case No.: 3:22-cv-00363-L-BT
• Filing date: Complaint filed February 14, 2022 (case docketed February 13/14, 2022, per court/Unified Patents records)
• Patents in suit: The '931 patent is one of eight asserted patents in this case. The others are U.S. Patent Nos. 7,548,763; 7,653,414; 7,855,988; 8,107,409; 8,688,111; 9,191,980; and 9,332,478. The district court's own filings confirm the '931 patent is in suit (see, e.g., Doc. 77 describing "the '931 Patent generally disclos[ing] detecting a failure condition relating to a first DMAG at a DMAG management system . . .").
• Status / outcome:
  • The case was assigned to Judge Sam A. Lindsay (Magistrate Judge Rebecca Rutherford / later designation "-BT").
  • After an initial motion to dismiss (May 2022), the court granted dismissal in part with leave to amend (Sept. 2023), and Lemko filed a Second Amended Complaint.
  • On April 10, 2025, the Magistrate Judge issued Findings, Conclusions, and Recommendation (Doc. 90). On June 4, 2025, Judge Lindsay adopted the Report (Doc. 94), granting the motion to dismiss as to Lemko's contributory infringement claims (dismissed with prejudice) and denying it in all other respects. Lemko announced on April 15, 2025, that the court had denied Microsoft's motion to dismiss Lemko's direct and indirect infringement claims.
  • Unified Patents' litigation caselist currently labels the case "Closed," although the Justia docket reflects activity into 2025. I could not confirm from the available sources the exact date or basis of any final closure (e.g., settlement, judgment, or stay), and I cannot verify whether an appeal followed.

Sources: https://dockets.justia.com/docket/texas/txndce/3:2022cv00363/[359096](/patent/359096) ; https://cases.justia.com/static/pdf-js/web/?file=/federal/district-courts/texas/txndce/3:2022cv00363/359096/94/0.pdf ; https://www.lemkocorp.com/update-lemko-corporation-patent-litigation-against-microsoft-and-affirmed-networks/ ; https://portal.unifiedpatents.com/litigation/caselist (filtered to the case) ; https://ia903405.us.archive.org/1/items/gov.uscourts.txnd.359096/gov.uscourts.txnd.359096.77.0.pdf

Related proceedings that DO NOT involve 9,755,931 (for accuracy, noted to avoid confusion)

• IPR2023-00531, Microsoft Corp. & Affirmed Networks, Inc. v. Lemko Corp. (PTAB) — challenges U.S. Patent No. 7,855,988 (not 9,755,931). Final Written Decision Sept. 10, 2024; appealed to the Federal Circuit; the Federal Circuit affirmed the PTAB's invalidity finding around April 10, 2026 (Law360). This is a sibling patent in the same Lemko dMA family, but it is not the '931 patent.
• IPR2023-00529, Microsoft Corp. & Affirmed Networks, Inc. v. Lemko Corp. (PTAB) — challenges U.S. Patent No. 7,653,414 (also not 9,755,931).
• Lemko Corp. v. AT&T (E.D. Tex., No. 2:25-cv-00305, filed Mar. 17, 2025) — per reporting, the seven patents asserted against AT&T (7,840,230; 7,979,066; 8,224,322; 8,310,990; 8,340,667; 8,359,029; 9,198,020) do not overlap with the Microsoft suit and do not include 9,755,931.

Caveats

• I did not find any PTAB inter partes review, covered-business-method review, or post-grant review specifically directed at 9,755,931 in the search results available to me. The IPRs identified in the Microsoft campaign targeted sibling Lemko patents (7,653,414 and 7,855,988), not 9,755,931.
• I found no ITC investigation or other district-court case specifically naming 9,755,931 beyond the N.D. Tex. case above.
• If a new "second" Lemko v. Microsoft action has been filed (Lemko's news page contains a June 2026-dated item), it postdates today's date (April 26, 2026) and I could not verify its contents or whether 9,755,931 is asserted in it.

I'll research this patent's PTAB history and related proceedings. Let me run several searches to verify whether any AIA trial proceedings exist and to fill in context on the family litigation.

The searches surfaced related Microsoft/Lemko IPRs but on different Lemko patents ('988 and '414). I need to check specifically whether any IPR was filed against the '931 patent (US9755931) itself, since the district court case asserted it among eight patents.

Proceedings overview

There are zero AIA trial proceedings (IPR/PGR/CBM) on file against US 9755931. The USPTO Open Data Portal returns no proceedings for this patent, and web research surfaced no PTAB petition directed to the '931 patent itself — the Microsoft/Lemko IPRs found on the PTAB docket all target sibling Lemko patents, not this one. The breakdown is therefore trivially: 0 active, 0 claims invalidated, 0 sustained, 0 settled, 0 institution-denied, all 16 claims untested. The defensive posture this creates is a double-edged sword: a defendant facing assertion of the '931 cannot point to any PTAB cancellation, but also faces no § 315(e)(2) estoppel whatsoever on this patent, so every prior-art ground remains available for a district-court invalidity defense or a freshly filed IPR within the petitioner's own one-year § 315(b) window.

No per-proceeding sections follow because no proceeding exists. What follows is the honest "no activity" analysis plus the closely related context a defendant should know.

Strategic summary

Claim status: CANCELED / SUSTAINED / UNTESTED

Every one of the 16 claims of US 9755931 (independent claims 1 and 7; the rest dependent) is UNTESTED in any AIA trial. No petition has been filed, no institution decision issued, and no Final Written Decision exists. The patent remains fully enforceable on its face, and its adjusted expiration is 2029-08-15 (per the Google Patents bibliographic data in the record).

Why this is notable — the family context

The absence of an IPR on the '931 is conspicuous because the patent has been asserted in litigation since 2022. In Lemko Corp. v. Microsoft Corp. and Affirmed Networks, Inc., No. 3:22-cv-00363-L-BT (N.D. Tex., filed 2022-02-14), Lemko asserted eight patents, including 9,755,931 (described in the Amended Complaint as covering "detecting a failure condition relating to a first DMAG at a DMAG management system … selecting the second DMAG, sending a first notification … and sending a second notification … to an external system"). Microsoft/Affirmed responded with a coordinated IPR campaign in February 2023 — filed at the very edge of the § 315(b) one-year bar (service was February 2022) — but the proceedings I was able to confirm are all against other Lemko patents:

• IPR2023-00531 (Microsoft Corp. & Affirmed Networks, Inc. v. Lemko Corp.) — U.S. Patent 7,855,988 ('988) — instituted; FWD entered 2024-09-10 finding challenged claims unpatentable (Raevsky, joined by Ippolito and Range); Lemko appealed to the Federal Circuit (Notice of Appeal 2024-09-16); press reports indicate the Federal Circuit affirmed around 2026-04-10.
• IPR2023-00529 (Microsoft Corp. & Affirmed Networks, Inc. v. Lemko Corp.) — U.S. Patent 7,653,414 ('414) — instituted 2023-09-26 (Powell, Lorin, Ippolito).
• IPR2023-00570 (Microsoft Corp. et al. v. Lemko Corp.) — U.S. Patent 9,191,980 ('980) — institution denied 2023-09-26.
• IPR2025-01142 (Microsoft Corp. et al. v. Lemko Corp.) — U.S. Patent 8,310,990 — discretionary denial 2025-11-20.

I found no docket entry or petition naming 9,755,931 as the challenged patent. That gap is significant: of the eight patents Lemko asserted against Microsoft/Affirmed, Microsoft appears to have challenged several but not the '931 — either as a deliberate merits/claim-scope choice, or because the crowded February 2023 filing window (all petitions were due by ~February 2023 under § 315(b)) closed before it could complete petitions on every patent.

Estoppel landscape — for the '931, it's a blank slate

Because no IPR/PGR/CBM has been instituted on 9,755,931, no § 315(e)(2) estoppel attaches to anyone for this patent. A defendant sued today on the '931 can raise any § 102/§ 103/§ 112 ground, including art that overlaps with what Microsoft used against the sibling '988 and '414 patents. The one practical wrinkle: if your company is Microsoft, Affirmed Networks, or a privy that was served with the N.D. Tex. complaint before ~February 2022, the § 315(b) bar has run for you — a new IPR by that defendant on the '931 would be time-barred. Any new defendant (e.g., one first served later) has its own one-year window from service, and the PTAB's Fintiv discretion will weigh the parallel N.D. Tex. litigation (which has been mired in pleadings since 2022, cutting somewhat in favor of institution).

Pattern signals

• Same petitioner, serial filings on the family: Microsoft/Affirmed has now filed at least four IPRs across the Lemko distributed-mobile-architecture family — a coordinated campaign. The fact that the '931 was skipped suggests Microsoft assessed it as a weaker IPR target (or ran out of runway), not that the family campaign is over.
• Patent owner litigates hard and appeals: Lemko appealed the adverse FWD in IPR2023-00531 to the Federal Circuit and lost; it has also pressed multiple amended complaints in N.D. Tex. Expect aggressive patent-owner behavior in any future '931 IPR (motions to amend, sur-replies, appeals).
• No defensive aggregator in this patent's chain: Unified Patents appears only as a litigation-data aggregator for the N.D. Tex. case; there is no Unified Patents IPR on the '931. Note also that Lemko granted Motorola Solutions a covenant-not-to-sue on the '988 patent (per press reports) — a sign Lemko selectively licenses, but nothing that touches the '931.
• New litigation wave: Lemko sued AT&T in E.D. Tex. (2:25-cv-00305, ~2025-05) — but on a different seven-patent set (7,840,230; 7,979,066; 8,224,322; 8,310,990; 8,340,667; 8,359,029; 9,198,020), not the '931. As of the latest reports, the '931's only known assertion is the Microsoft/Affirmed case.

Recommended next steps

1. Confirm the "zero" finding against the live docket. The canonical USPTO ODP data in this prompt shows no AIA trial proceedings, and my searches found none directed to 9,755,931. Before relying on this, run the patent number through USPTO PTAB E2E (https://ptab.uspto.gov) and the Unified Patents portal page for the patent (https://portal.unifiedpatents.com/patents/patent/US-9755931-B2) to confirm no petition was filed and docketed under a number web search didn't surface. Do not assume any of the Microsoft IPRs above cover the '931 — they do not.

2. If you are a defendant in Lemko v. Microsoft, 3:22-cv-00363 (or a privy): you are almost certainly § 315(b) time-barred from filing an IPR on the '931 (service was February 2022; the bar expired February 2023). Your invalidity fight belongs in the district court. The case is still in the pleadings phase: contributory infringement claims were dismissed with prejudice (Order, 2025-06-04, adopting the Magistrate Judge's Report; motion to dismiss otherwise denied) — see Lemko Corp. v. Microsoft Corp., No. 3:22-CV-363-L-BT, Doc. 94 (N.D. Tex. 2025). Build your § 101/§ 102/§ 103 case on the merits; there is no PTAB shortcut left for this defendant.

3. If you are a newly served defendant: you have a one-year § 315(b) window from service. The '931 is untested — a clean IPR petition on the strongest § 103 ground (the distributed-gateway failover art is crowded, and Microsoft's successful Flore-based § 102 theory against the sibling '988 shows the art cuts close to Lemko's DMA-gateway claims) is viable. Act fast: the statutory clock and the Fintiv factors both favor an early filing while the parallel litigation is still in pretrial. Beware that Lemko will litigate hard (it appealed the '988 loss to the Federal Circuit and lost), so build the record with expert testimony on day one.

4. The absence of PTAB activity is itself a data point for settlement leverage, not weakness. The '931 has survived four years of assertion and a coordinated Microsoft IPR campaign against its family without ever being challenged — that likely reflects Microsoft's petition-resource allocation, not a merits judgment that the claims are invincible. A defendant's best play is a rigorous invalidity analysis now, and if the art supports it, an IPR filing inside the window rather than waiting for the N.D. Tex. case to reach claim construction.

Sources: USPTO ODP structured data (no AIA proceedings on file, per the prompt); Unified Patents portal for US-9755931-B2 and case 3:22-cv-00363; Docket Alarm PTAB dockets for IPR2023-00529, IPR2023-00531, IPR2023-00570, and IPR2025-01142 (all on sibling Lemko patents, not the '931); IPR2023-00531 Institution/FWD records and Lemko CAFC appeal papers; Lemko Corp. v. Microsoft Corp., 3:22-CV-363-L-BT (N.D. Tex.) Doc. 94 (2025-06-04); Law360/Mondaq coverage of the Lemko-Microsoft and Lemko-AT&T litigations. No proceeding numbers have been invented; every proceeding cited above is a different patent and is flagged as such.

I'll research the assignment history for US 9755931 by searching multiple sources.

Let me dig deeper into the assignment records, litigation, and the parties involved.

Let me look for the specific assignment records with reel/frame numbers and check the Piccadilly security agreement nature.

I have reached the research step limit, so I'll now compile the analysis from the sources I was able to retrieve (the full patent record embedded in the prompt, Google Patents legal-event data, Unified Patents litigation tracker, Justia/Docket Alarm dockets, and Lemko's own public statements). I was not able to pull raw reel/frame numbers or recorded correspondent names from the USPTO Assignment Center in this session (that database requires interactive lookup), so I flag those fields as not retrieved rather than inventing them.

Inventors

• ShaoWei Pan — sole named inventor. Employer at time of filing: Lemko Corporation (the original assignee). Pan is the named inventor across the entire Lemko distributed-mobile-architecture family (e.g., parent US 8,706,105; US 7,855,988, etc.) and the recorded 2014 assignment of inventor interest names Pan, Shaowei as assignor to Lemko Corporation.
• Pattern note: Single-inventor filings, all assigned to the employer (Lemko) — a conventional employment-invention pattern. No departure/fire-sale signal: there is no evidence Pan left Lemko or that a portfolio sale followed the filing.

Original assignee

• Lemko Corporation (Schaumburg, Illinois) — named assignee on the issued patent.
• Products / line of business: Lemko describes itself as an operating company — a pioneer in NFV / mobile edge computing, selling private 4G/5G networks, wireless rural broadband, and public-safety/IoT deployments, with networks deployed in Alaska, Northern Canada, and Africa (lemkocorp.com, Apr. 2025). In the N.D. Tex. litigation it pleaded that it "began offering innovative network products and services" in 2004 and holds 50+ patents.
• Current status: Operating. Lemko is currently asserting this patent (and 7 related ones) against Microsoft/Affirmed Networks in Case 3:22-cv-00363-L-BT (N.D. Tex.), and defending against Microsoft IPRs (IPR2023-00531 et al.). No bankruptcy or dissolution found.

Assignment timeline

Important caveat: I could not retrieve reel/frame numbers or recorded correspondent names from the USPTO Assignment Center during this session. The two conveyance events below are documented in the Google Patents legal-event record for US 9755931 (and mirrored on the USPTO assignment feed); reel/frame and correspondent fields are not retrieved, not fabricated. Verify at: https://assignmentcenter.uspto.gov/ (search "9755931").

• 2014-02-19 (recorded; execution date not separately published) — Google Patents legal event "Assigned to LEMKO CORPORATION … ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: PAN, SHAOWEI"
  • Conveyance: Assignment of Assignor's Interest (inventor → company)
  • Assignor: ShaoWei Pan
  • Assignee: Lemko Corporation
  • Correspondent: not retrieved
  • Context: Standard inventor-to-employer assignment recorded at the time the continuation application (US 14/184,120) was filed; not a monetization event.
• 2025-03-25 (recorded) — Google Patents legal event "Assigned to PICCADILLY PATENT FUNDING LLC, AS SECURITY HOLDER … PATENT SECURITY AGREEMENT. Assignors: LEMKO CORPORATION"
  • Conveyance: Patent Security Agreement (a security interest / collateral grant, not a transfer of title)
  • Assignor: Lemko Corporation
  • Assignee: Piccadilly Patent Funding LLC, as Security Holder
  • Correspondent: not retrieved
  • Context: Patent-backed financing — Lemko borrowed against the portfolio; Piccadilly is the lender/security holder. This does not convey ownership; Lemko remains the record owner and the active plaintiff.

Note on family members: This patent is a continuation of US 12/163,601 (→ US 8,706,105) and has a continuation child US 15/673,156 (→ US 10,547,530). Those are family relationships, not additional assignment events for this patent.

No other recorded assignments were surfaced for this patent.

Timeline diagram

timeline
    title Ownership of US 9755931
    2008 : Filed by Lemko Corp
    2014 : Continuation filed
         : Inventor assignment recorded
    2017 : Patent issued
    2022 : Suit against Microsoft
    2025 : Security agreement to Piccadilly

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. The only post-issuance record (2025-03-25) is a Patent Security Agreement in favor of Piccadilly Patent Funding LLC, as Security Holder — a lending/security-interest filing, not a conveyance to a licensing-only LLC. Lemko remains the owner and is the party suing. No evidence of transfer to an "IP / Licensing / Holdings" shell.
2. Known asserter in the chain — Not present. The plaintiff is Lemko Corporation itself — an operating company with deployed networks and a 20-year product history — not Acacia, Marathon, IV, IPNav, or any entity on the Unified Patents / RPX high-frequency-plaintiff lists. Unified Patents' tracker records the case as Lemko Corp v. Microsoft Corp et al. (3:22-cv-00363), plaintiff Lemko. Piccadilly Patent Funding is a patent lender, not an asserting NPE in this record.
3. Repeat correspondent across the chain — Unclear / not retrieved. Reel/frame numbers and recorded correspondents were not retrievable in this session. The firm of record on the patent itself is Toler Law Group, PC (prosecution counsel), which is not an assignment-correspondent finding. No recurrence data is available, so no signal is asserted.
4. Cascading transfers — Not present. Only two recorded events, 11 years apart (2014 inventor assignment; 2025 security agreement). No chained LLC-to-LLC conveyor belt.
5. Pre-litigation transfer — Not present. Suit was filed 2022-02-14; the only post-issuance record (the Piccadilly security agreement) was recorded 2025-03-25 — three years after the suit, not within 6 months before it. There was no transfer arranged to enable standing or venue.
6. Bankruptcy fire-sale — Not present. Lemko is operating and litigating; no Chapter 7/11 sale found.
7. Privateering — Not present. Lemko asserts directly through Kramer Levin (Paul Andre, Lisa Kobialka, James Hannah, Aaron Frankel) against Microsoft/Affirmed Networks — a competitor in the NFV/private-network space Microsoft entered via its March 2022 Affirmed acquisition. No operating-company-to-front evidence.
8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN.

Verdict

Operating-company assertion.

Lemko Corporation — the original and current record owner, an operating company that says it ships NFV/MEC-based 4G/5G network products — is directly suing competitors Microsoft and Affirmed Networks (N.D. Tex. 3:22-cv-00363, filed 2022-02-14) over this patent. The sole post-issuance assignment event is a 2025-03-25 Patent Security Agreement to lender Piccadilly Patent Funding LLC, which is a financing/security-interest filing, not a transfer of title or a transfer to an NPE, so no shell-transfer, cascading, or pre-litigation-transfer signals exist. Verification: https://assignmentcenter.uspto.gov/ (search "9755931"); litigation record: https://portal.unifiedpatents.com/litigation/Texas%20Northern%20District%20Court/case/3%3A22-cv-00363.

Important completeness caveat (read first)

I was asked to search the USPTO database for patent 9755931 and to analyze each patent citation on the face of that patent. I confirmed the USPTO record for US 9,755,931 B2, but I was not able to retrieve the complete "References Cited" page from USPTO within the search budget. The authoritative front-page reference list is only partially visible in the Google-hosted image PDF of the patent (patentimages.storage.googleapis.com/.../US9755931.pdf), which shows the beginning of the "References Cited — U.S. PATENT DOCUMENTS" list plus two CN documents. The remainder of the list (including any further foreign documents and any non-patent literature) was cut off in the retrievable snippet. I therefore analyze below the references I could verify on the face of the patent, and I flag every item where the title/technical content could not be independently confirmed from the available excerpts. Where a title is given without a flag, I have reasonable corroboration.

1. USPTO record confirmation for 9755931 (literal)

From the patent image PDF (US009755931B2.pdf), the front page confirms:

• Patent No.: US 9,755,931 B2 (i.e., 9755931)
• Title: Fault Tolerant Distributed Mobile Architecture
• Inventor: Shaowei Pan (front-page OCR shows "Kidcer, IL" — a known OCR corruption of "Kildeer, IL"; the application record lists Kildeer)
• Assignee: Lemko Corporation, Itasca, IL
• Date of Patent: Sep. 5, 2017 (the asterisk denotes the patent term disclaimer/adjustment context; Google Patents shows adjusted expiration 2029-08-15)
• Continuity: Continuation of Application No. 12/163,601, filed Jun. 27, 2008, now U.S. Patent 8,706,105 B2
• Primary Examiner: Joshua Schwartz
• Attorney/Firm: Toler Law Group, PC
• Independent claims: 1 (method), 7 (system), 12 (machine-readable storage device)

This matches the previously generated summary. No conflicting USPTO record was found for this number; I did not substitute or auto-correct any other patent number (e.g., 8,706,105, 10,547,530, or the WO counterpart WO2009158155A2 are treated as distinct identifiers).

2. Verified front-page citations of US 9,755,931 B2 (partial list)

The following U.S. documents appear at the top of the patent's "References Cited" section (as rendered in the image PDF snippet). Dates are as printed on the face of the patent.

Additionally, a Google Patents cross-reference indicates that US 2002/0061746 A1 ("Method of global roaming services using gateway location register in third generation mobile telecommunication networks") appears in the 9755931 citation chain, but I could not verify that it is printed on the face of the patent itself, so I treat it as corroborated-but-unconfirmed.

I cannot list the remaining references (the face list continues past US 6,614,74x, and the snippet ends there). Treat this as a partial enumeration, not the complete set.

3. Claim scope against which § 102 is assessed

For a proper § 102 analysis, the relevant limitations of the independent claims are:

• Claim 1: (a) a dMAG management system in communication with at least a first dMAG, a second dMAG, and a plurality of dMA nodes; (b) detecting a failure condition associated with the first dMAG; and, in response, sending three notifications: (i) first notification to the second dMAG instructing it to take over the first dMAG's operations; (ii) second notification to a dMA node (the node that connects a mobile-station call to an external system via the first dMAG) telling it the second dMAG has taken over and to route calls via the second dMAG; and (iii) third notification to the external system instructing it to connect subsequent calls to the mobile station via the second dMAG.
• Claim 7: system counterpart with operation-determination, dMAG-selection, and notification modules performing the same detection + three-notification flow.
• Claim 12: machine-readable-storage-device counterpart of claim 1.

Dependent claims narrow the "failure condition" to: offline/going-offline message from the first dMAG (claims 2, 9, 13); message from a dMA node local to the first dMAG (claims 3, 10, 14); message from the external system's dMAG interface (claims 5, 11, 16); or missed periodic heartbeat (claim 6).

Anticipation (35 U.S.C. § 102) requires a single reference disclosing every limitation, arranged as in the claim, either expressly or inherently. Under pre-AIA § 102 (the operative version given the 2008 priority date and pre-2013 effective filing date), a reference must also qualify as prior art (e.g., § 102(a)/(b)/(e)/(g)).

4. Per-reference § 102 assessment

A recurring threshold problem for every cited reference: the claims are built around a "dMAG management system" that is distinct from and coordinates a first dMAG, a second dMAG, and dMA nodes, and the claimed three-way notification (to the backup gateway, to the affected edge node, and to the external/legacy system). Older single-gateway-failover and call-routing patents generally disclose some of these elements but not the full management-system-plus-triple-notification combination. Because I could not verify the full text of most references, the claim-mapping below is stated conditionally.

US 5,734,979 A — Lu et al. — "Cellular base station with intelligent call routing" (Mar. 1998)

• Relevance: Discloses a cellular base station that makes intelligent routing decisions, potentially including redirecting calls when a network element is unavailable.
• § 102 exposure: Plausibly anticipatory of nothing on the face of claim 1/7/12 as a whole, because it lacks the dMAG management system in communication with a second dMAG and dMA nodes and the third notification to an external system to reroute subsequent calls. It is more naturally § 103 (obviousness) material, and at most touches the "dMA node side" concept found in claim 1's second notification.

US 6,549,937 B1 — Auerbach et al. — "System and method for multi-protocol communication in a computer network" (Apr. 2003)

• Relevance: Multi-protocol gateway/routing between networks; the closest of the verified references to a "gateway serving plural network types" concept in the patent's dMAG/legacy-network architecture.
• § 102 exposure: Does not, on its face, disclose the fault-tolerant management system with a first/second dMAG pair or the failure-condition detection and triple notification. Not a single-reference anticipator of claims 1, 7, or 12; possible secondary art for obviousness of the external-system interface feature.

US 4,284,848 A — Frost (Aug. 1981); US 5,590,175 A — Gallant et al. (Dec. 1996); US 5,623,495 A — Eng et al. (Apr. 1997); US 5,623,532 A — Houde et al. (Apr. 1997); US 5,713,973 A (Jan. 1998)

• Relevance: Generic call-routing / switching art from the 1980s–1990s, typical of the examiner's "background" citations.
• § 102 exposure: Without confirmed disclosures of dMAG-to-dMAG failover and management-system coordination, none can be said to anticipate claims 1, 7, or 12 from the available evidence. At most they are cumulative background for obviousness of ordinary routing/switching steps.

US 5,890,064 A (Mar. 1999); US 5,933,784 A (Aug. 1999); US 5,991,639 A (Nov. 1999); US 6,121,499 A (Sep. 2000); US 6,131,048 A (Oct. 2000); US 6,141,564 A (Oct. 2000)

• Relevance: Late-1990s telecom call handling/registration art (several of these, e.g., the Rautiola and Sekine references, are plausibly about subscriber registration/location registers, which maps to the HLR/VLR database features described in the specification and implicated by dependent claims).
• § 102 exposure: Conditional: a reference disclosing a location-register or registration-based detection of an unreachable gateway could bear on the failure-condition detection element of claims 1/7/12 and on the heartbeat/registration embodiments of dependent claims 2–6, 9–11, 13–16. But no single one of these is shown, from available evidence, to disclose the second-dMAG take-over plus the external-system notification. Accordingly I cannot conclude anticipation of the independent claims; they are potential § 103 building blocks.

Statutory invention registrations — US 6,119,184 H — Hoffpauir et al. (Nov. 2000); US 6,160,804 H — Ahmed et al. (Dec. 2000)

• Relevance: Defensive publications ("H" documents) that are prior art for what they disclose.
• § 102 exposure: Same conditional analysis as above; cannot anticipate independent claims without the three-notification management-system architecture, which is not evident from the citations alone.

US 6,411,825 B — Cacho et al. (Jun. 2002); US 6,418,508 B — Heinonen et al. (Jul. 2002); US 6,421,325 B — Kikinis (Jul. 2002); US 6,515,985 B — Shmulevich et al. (Feb. 2003); US 6,539,237 B — Sayers et al. (Mar. 2003); US 6,542,497 B — Curry et al. (Apr. 2003); US 6,584,098 B (Jun. 2003); US 6,611,533 B — Liao et al. (Aug. 2003)

• Relevance: Early-2000s wireless/registration/routing art; likely cited for background on registration, hand-off, or network-element redundancy.
• § 102 exposure: No evidence in the retrieved material that any single one of these discloses the full management-system/take-over/three-notification scheme. None is a demonstrated anticipator of claims 1, 7, or 12; they are more plausible § 103 references or cumulative background.

CN 1427637 A (Jul. 2003); CN 1538781 A (Oct. 2004)

• Relevance: Foreign counterparts in the general routing/registration space; the U.S. examiner would rely on them under § 102(a)/(b) only if an English translation or the substantive disclosure establishes the limitations.
• § 102 exposure: Cannot be assessed from the retrieved snippets; neither Chinese document's content was retrievable, so no § 102 conclusion is possible on this record.

5. Bottom line

1. USPTO identity confirmed: 9755931 = US 9,755,931 B2 (Lemko Corp.; inventor Shaowei Pan; continuation of 12/163,601 → US 8,706,105 B2; granted Sept. 5, 2017; examiner Joshua Schwartz).
2. Complete citation list not retrievable: The face-of-patent "References Cited" list was only partially visible (running from US 4,284,848 A through roughly US 6,614,74x, plus CN 1427637 A and CN 1538781 A). The remainder is unverified.
3. No single verified reference plausibly anticipates independent claims 1, 7, or 12 under 35 U.S.C. § 102 on the available evidence, because the claims' core combination — a dMAG management system coordinating a first dMAG, a second dMAG, and dMA nodes, with three distinct notifications (to the backup gateway, to the affected dMA node, and to the external system) — is not shown to be disclosed as a whole by any one of the partially visible citations. The references are better characterized as § 103 obviousness building blocks (registration/location-register art, gateway/routing art, and redundancy art).
4. Highest-risk references for narrower claims: The location-register/registration-based art (e.g., 5,991,639 Rautiola; 6,131,048 Sekine) and the intelligent call-routing art (5,734,979 Lu) are the most plausible to map against the failure-condition detection and dMA-node notification limitations, particularly dependent claims 3–5/10–11/14–16. But even for those, anticipation would require full-text confirmation I did not obtain.
5. Known gap: Because the retrieved front-page list is incomplete, I cannot rule out that a later-listed reference (e.g., a Lemko-family patent such as the roaming-registration or distributed-architecture patents, or US 2002/0061746 A1, which appeared in the citation chain) is a closer anticipatory reference. Full-text review of the remaining citations — ideally from the USPTO Patent Center "References Cited" tab for application 14/184,120 — is required before any definitive § 102 invalidity position is taken.

If you want, the next step would be to pull the complete reference list from USPTO Patent Center (application 14/184,120) or from the full Google-hosted PDF pages 1–2, and I can then complete the per-reference table including the missing U.S. documents, foreign documents, and any NPL.

I'll research the reference set and adjacent failover art to build grounded § 103 combinations.

Obviousness Analysis — U.S. Patent No. 9,755,931 under 35 U.S.C. § 103

Scope and honesty note. This analysis builds on the Prior Art section already generated for this page and uses the reference set enumerated there. That section records an important limitation that carries forward here: the face-of-patent "References Cited" list was only partially retrievable (US 4,284,848 A through roughly US 6,614,74x, plus CN 1427637 A and CN 1538781 A). I could not independently pull full texts for most of those documents in this session. Accordingly, every mapping below is stated with an explicit confidence label — [verified text], [title/content partly corroborated], or [inference from citation only]. I have not invented disclosures. Where I do rely on a reference I could verify (notably Lu et al. / US 5,734,979 and Rautiola et al. / US 5,991,639 — both retrieval-verified this session), I say so.

1. Governing law and the person of ordinary skill

Because the '931 claims a June 27, 2008 priority date and has an effective filing date before March 16, 2013, pre-AIA § 103 governs. The framework is Graham v. John Deere Co., 383 U.S. 1 (1966) (scope/content of art; differences; PHOSITA level; secondary considerations), as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), and the rationales catalogued in MPEP § 2143.

PHOSITA for this patent would be a person with a B.S. in electrical engineering or computer science and 3–5 years of experience designing or operating cellular core-network switching elements (MSC/BSC, softswitch/media gateway, or IP-based gateway nodes), including working knowledge of SS7/IS-41 call setup, HLR/VLR registration, IP transport, and N+1 redundancy / failover techniques for carrier-grade voice infrastructure. Critically, the relevant art is predictable — gateway redundancy, standby takeover, heartbeat polling, and registration-based location update were all mature engineering disciplines well before 2008. KSR counsels that predictable arts yield to combination arguments more readily.

2. The limitation ledger — what a § 103 ground must supply

Dependents narrow B only: (2/9/13) offline message from the first dMAG; (3/10/14) message from a local dMA node; (4/15) extra notify-the-node step; (5/11/16) message from the dMAG interface of the external system; (6) missed periodic heartbeat.

The crux. Elements C + D + E together, issued from a single orchestrating management system, are the inventive heart. No single face reference is shown to disclose A–E as arranged (consistent with the § 102 conclusions already reached). The question is therefore purely one of combinability.

3. The reference set actually available

Consequence: the face list alone supplies (i) the distributed-edge-switching concept with a stated fault-tolerance motivation (Lu), and (ii) the multi-protocol gateway/interworking concept (Auerbach), and (iii) registration/location-register logistics (Rautiola and the registration group). It does not, on the retrievable record, supply the centralized failover orchestrator (element A) or the three-party notification (C+D+E). Any competent § 103 case therefore has to reach outside the face list — which is exactly what the prosecution history suggests was never done (the examiner allowed the case; no IPR was ever filed — see the PTAB section above).

4. Combination grounds

Ground 1 (strongest): Lu + a carrier-grade gateway-redundancy teaching, in view of Auerbach

• Lu (US 5,734,979) supplies elements A-partial/D: a network in which a base-station-side node ("intelligent routing" node) switches calls at the edge to a foreign network rather than backhauling to the MSC, and expressly identifies the failure problem to be solved.
• Auerbach (US 6,549,937) supplies the gateway/interworking piece: a gateway node that translates among multiple protocols so disparate networks can exchange calls — i.e., the dMAG/dMAG-interface function of the '931 (FIG. 1: SS7 114, VoIP 116, MAP/CAMEL 118).
• The redundancy teaching (heartbeat/standby/N+1 takeover with notification of the dependent nodes) is the routinely applied, well-documented carrier-grade pattern in the same field by 2008 — softswitch and media-gateway redundancy, HLR/STP mated-pair failover, and Mobile-IP home-agent redundancy all use detect failure → promote standby → redirect endpoints.
• Motivation (MPEP 2143 rationales A, C, D): (i) known technique improving a similar device — applying standard gateway redundancy to a newly distributed gateway architecture is the natural, expected engineering step, and Lu itself flags "reduced capacity" as the failure state to avoid; (ii) design incentive / predictable results — the result is merely the continuation of the same call-routing service with no change in function; (iii) "obvious to try" — a designer told to keep edge-switched calls alive when a gateway dies has a finite set of known options (standby gateway + endpoint redirection).
• Mapping: Lu → dMA-node-side edge switching and the failure problem; Auerbach → dMAG/legacy interworking; redundancy teaching → B (heartbeat/timeout), C (standby takes over), D (node redirected), E (network side redirected). Weakness: Lu does not disclose a management system separate from and above the gateways, and the record does not show the redundancy teaching routes notifications to both the node and the external system.

Ground 2: Auerbach + registration/location-register art (Rautiola, US 5,890,064, 6,121,499, 6,131,048) + an orchestrated-failover teaching

• The registration group supplies the mechanism by which reachability/subscriber location is tracked and updated so that subsequent incoming calls are directed to the current serving entity (HLR/VLR update, re-registration, location-cancel).
• That is a near-literal template for element E: "notify the external system… to connect subsequent calls to the mobile station via the second dMAG." In pre-2008 terms this is simply updating the location/registration record so the home side routes future calls to the new serving gateway — a decades-old procedure.
• Motivation (B — substitution of a known element; C — known technique): replacing "the MSC/HLR that currently serves the subscriber" with "the dMAG that currently serves the subscriber" is a substitution of one known network-element type for another performing the same function (directing calls to the correct serving node), with predictable results. A POSITA reading the '931 specification's own FIG. 5 (dMA-node register, HLR, VLR, visitor registers maintained per dMAG) would recognize the database migration on failover as routine.
• Strength: this is the most natural bridge to the E limitation, because the claim's "external system" is (per the specification) the legacy network reached through the dMAG interface, and registration/HLR updates are precisely how telephony has always redirected calls after a serving-node change.

Ground 3: Ground 1 + Ground 2 combined, with the node-side notification supplied by Lu

• Lu provides the edge node that directs call traffic and would need to be told the identity of the alternate gateway; the registration group provides the network-side redirection. Combining the two yields D + E with a single detection/promotion event (B + C).
• Motivation (D — applying a known technique to a known device ready for improvement; F — known work in one field prompting variation): when a serving gateway fails, both ends of the pipe (edge node and network side) must be moved, or the system is left half-failed. That is a self-evident completeness rationale — a POSITA would not migrate only one endpoint.

Ground 4 (for claims 3/10/14 and 4/15): node-reported failure + node-directed reroute

Lu's base-station node and the roaming/registration art (Rautiola) supply a node that registers with and is aware of the identity of a serving network element, and that therefore can report a failure of that element and accept a new assignment. This maps directly to "failure condition … based on a message received from a particular dMA node local to the first dMAG," and to the extra "notify the dMA node that the second dMAG is to take over" step of claims 4/15. Motivation: the node is the only party that observes the gateway's non-response in real time; using it as a failure sensor is the most efficient known option.

Ground 5 (for claim 6 and claims 2/9/13 and 5/11/16): heartbeat / offline-message / interface-report detection

• Claim 6 (missed heartbeat): this is the single weakest claim in the patent. Heartbeat monitoring, keep-alive polling, and dead-peer detection were ubiquitous in IP and telecom systems well before 2008; the specification itself (FIG. 8, op. 804) treats the heartbeat as one of several interchangeable "messages." A POSITA would consider substituting heartbeat-timeout detection for any other reachability signal to be a mere substitution of known elements (rationale B) yielding predictable results.
• Claims 2/9/13 (graceful offline message) and 5/11/16 (interface-reported failure): likewise routine — administrative "out-of-service" signaling and far-end-reported failure are standard in carrier operations.

5. Claim-by-claim vulnerability map

6. The priority-date lever (a § 103 force-multiplier)

The '931 is a continuation filed 2014-02-19, yet its claims require a dMAG management system and a three-notification sequence. If any claim limitation lacks § 112 written-description support in the 2008 parent (App. 12/163,601 → US 8,706,105) — a real risk given that the parent's disclosure is organized around the peer dMAG architecture of FIGS. 2–3 — then that claim is not entitled to the 2008 date and its critical date shifts to 2014. That would open up 2008–2014 art, which is far richer in exactly the missing element: IMS/softswitch pool failover, 3GPP S-CSCF/MME pooling, WiMAX ASN-gateway redundancy, and standardized "gateway takeover + endpoint re-registration" procedures. This is likely the single highest-value defendant theory, because it attacks the very combination (A + C/D/E) that the face art cannot supply.

Related point: Lemko's own earlier dMA patents (e.g., US 7,548,763, filed 2004; US 7,855,988, filed 2005; US 7,856,233) predate the '931 and describe dMA node/gateway registration and control. They are tempting as art, but under pre-AIA 35 U.S.C. § 103(c) commonly-owned § 102(e) art cannot be used in an obviousness rejection — so their usefulness depends on the ownership status on the invention date. Flagged for verification; do not assume they are available.

7. Where a § 103 case is weakest (rebuttal exposure)

1. No reference teaches the orchestrator. Element A — a management system distinct from the dMAGs and sourcing the notifications — is not shown on the retrievable record. Patent owner will frame this as an architectural invention, not a mere redundancy tweak.
2. The triple notification is specific. Claim 1 requires the management system to notify the external system as well as the node. A generic "standby takeover" teaching notifies endpoints, not the far-side network and the edge node from a single source. Expect a "teaches away / different failure model" argument.
3. Hindsight risk. The combination must be justified with pre-2008 evidence of the motivation, not with the '931 specification as a roadmap. Cite contemporaneous carrier-grade failover standards and products (softswitch/media-gateway redundancy literature; HLR/mated-STP practices) to anchor the motivation.
4. Secondary considerations. Lemko is an operating company with deployed networks and a 20-year product history (see the Assignment/NPE section above). It may attempt to argue commercial success or industry praise; none of the filings reviewed evidence a nexus, but the argument is available.

8. Bottom line

• No reference, singly or as presently recorded, anticipates the independent claims — consistent with the earlier § 102 conclusion.
• A viable § 103 case exists, but it must be assembled from art outside the face list. The best structure is: Lu (US 5,734,979) for the distributed edge-switching node and the stated fault-tolerance motivation, + Auerbach (US 6,549,937) for the multi-protocol interworking gateway, + a carrier-grade gateway-redundancy teaching (heartbeat/standby takeover + endpoint redirection) for B/C/D, + registration/location-register art (Rautiola and the registration group) for E. Motivation rests on MPEP 2143 rationales A, B, C, D and KSR's predictable-art reasoning.
• The weakest claims are 6, 3/10/14, 4/15, and 5/11/16; the weakest link in the independent claims is the "dMAG management system" of element A, followed by the simultaneous node-plus-external-system notification.
• Highest-leverage moves: (1) attack the 2008 priority claim for the three-notification limitations and develop 2008–2014 art; (2) build the combination record with pre-2008 documentary evidence of the failover motivation; (3) leave Lemko's own family patents out of the ground unless § 103(c) common ownership can be cleared.

What would complete this analysis: full-text retrieval of the remaining face citations (US 4,284,848–US 6,611,533, plus CN 1427637 A and CN 1538781 A) and a written-description comparison between the '931 claims and the 2008 parent (App. 12/163,601). Until then, treat the specific reference-by-reference mappings above as conditional — grounded in the citations but not independently confirmed for the unretrieved documents.