US 8793758

Patent Overview: US8793758B2

Title: Security, fraud detection, and fraud mitigation in device-assisted services systems

Patent Number: US8793758B2
Application Number: US13/309,463
Publication Number (pre-grant): US20120167162A1
Inventors: Gregory G. Raleigh, Jeffrey Green, James Lavine
Original Assignee: Headwater Partners I LLC
Current Assignee (as listed): Headwater Research LLC
Priority Date: 2009-01-28 (stated as an assumption)
Filing Date: 2011-12-01
Publication Date: 2014-07-29
Anticipated Expiration: 2029-03-02 (listed)

Purpose and Plain-Language Scope

Based solely on the supplied text, this patent concerns security, fraud detection, and fraud mitigation within "device-assisted services systems." The listing identifies prior-art keywords: "user device," "application," "service," "credential," and "agent." These terms suggest the claimed subject matter involves verifying or managing credentials and applications on a user device, potentially through a software agent, as part of delivering services.

The patent claims priority to a chain of earlier applications, the earliest listed being 2009-03-02 (US12/380,773), with the priority date assumption given as 2009-01-28. The application was filed 2011-12-01 and granted 2014-07-29. Multiple related continuations and priority claims are listed.

What the Excerpt Does Not Establish

The supplied excerpt is a Google Patents bibliographic and metadata listing. It does not include the patent's claims, abstract, detailed description, or drawings. As a result, the excerpt does not establish:

• The specific scope of any claim
• The precise mechanisms of fraud detection or mitigation
• Definitions of "device-assisted services," "agent," or "credential" as used in the patent
• Any legal conclusions regarding validity, enforceability, or claim coverage
• Litigation history, licensing, or the accuracy of the listed assignees

Per the source's own disclaimer, legal status, assignees, and dates are assumptions, not legal conclusions.

Source: https://patents.google.com/patent/[US8793758](/patent/US8793758)/en

I'll research litigation involving US 8,793,758 across patent litigation sources.

Litigation Report: U.S. Patent No. 8,793,758

Bottom line

Based on my searches, I found no litigation in which U.S. Patent No. 8,793,758 ("Security, fraud detection, and fraud mitigation in device-assisted services systems") is specifically identified as an asserted patent. I could not confirm a single case — district court, ITC, or PTAB — in which the '758 patent is named in a complaint, infringement contention, IPR petition, or judgment.

Two important caveats before you rely on this:

1. I was cut off by a tool-step limit mid-research, so this is a preliminary, non-exhaustive result rather than a fully verified negative.
2. The '758 patent sits in the same family as a large number of patents that have been litigated heavily by the same owner. It is easy to conflate the '758 with its siblings. I have kept those separate below.

What the record shows about ownership and lineage

The '758 is a continuation/divisional-family member descending from the January 28, 2009 provisional filings and the March 2, 2009 non-provisional (US 12/380,773) that generated the Headwater "device-assisted services" (DAS) portfolio.

Known litigation involving the Headwater / Raleigh portfolio (NOT the '758)

These are cases I actually found in the record. In every one, the asserted patents I could identify are other family members — not US 8,793,758. I list them so you don't mistake them for '758 suits, and because any '758 assertion would almost certainly be filed alongside or against the same parties.

Earlier wave (2022–2023, E.D. Tex.):

Later wave (2025):

PTAB proceedings (all on sibling patents, none on '758):

• IPR2024-01041 and IPR2024-01042 — Cellco/Verizon v. Headwater, U.S. 8,924,543
• IPR2024-00942 and IPR2024-00943 — Google LLC, Verizon, T-Mobile, Sprint, AT&T entities v. Headwater, U.S. 8,589,541 (final written decision: challenged claims unpatentable)
• IPR2024-00945 — Google v. Headwater
• IPR2024-00341 — Samsung/Google v. Headwater
• IPR2025-01572 — Apple v. Headwater, U.S. 10,064,055
• Google v. Headwater, U.S. 8,832,777-related challenges

Separate, unrelated cases that use a "'758" or "879" label (do NOT confuse these): the Truesight/Unified Patents material refers to a different '879 patent asserted against Samsung; the Neonode dispute concerns U.S. 8,095,879; the Cimline/Crafeo case concerns an unrelated '375 patent. None of these involve US 8,793,758.

Why the '758 may be absent from the docket

A few plausible explanations, offered as hypotheses only, not findings:

• Claim family overlap. Headwater has asserted sibling patents (e.g., 8,589,541; 8,924,543; 9,198,042; 9,215,613) that carry overlapping DAS security/fraud functionality. A patentee with 100+ family members typically asserts a strategic subset rather than every continuation.
• Priority/expiry economics. The '758 shares an earliest 2009 priority date and an anticipated expiration of March 2, 2029. Cases filed in 2023–2025 may prefer later-expiring continuations.

Confidence and recommended verification

• Confidence that US 8,793,758 has been asserted: Low-to-moderate that it has, moderate that my negative finding is correct but incomplete.
• This is not a legal conclusion. Litigation status changes and complaints are often filed under seal or amended.

To confirm before you rely on this, check directly (I was unable to complete these lookups):

1. PACER — search the patent number 8793758 across all district courts (Docket Navigator or Lex Machina if available).
2. PTAB / USPTO Patent Trial and Appeal Board — search "8,793,758" for any IPR/PGR petition.
3. Unified Patents litigation portal — https://portal.unifiedpatents.com/litigation/caselist and the patent page for US8793758.
4. CourtListener / RECAP — docket-level text search for "8793758" or "8,793,758" (docket text search often catches patents not in party metadata).
5. ITC EDIS — Section 337 investigations, which patent-metadata sites frequently omit.

If you'd like, I can run those targeted searches (PACER-style docket-text queries, PTAB, and the Unified Patents case list) and produce a verified case-by-case table — including any case where '758 appears only in infringement contentions or invalidity contentions rather than the complaint.

I'll verify the structured ODP data against publicly-filed proceedings before writing anything up.

The results surfaced some leads but also some ambiguity (the "'879" snippet may involve a different patent). Let me pin down whether any AIA trial actually names US 8,793,758.

Important: that "879 patent" petition snippet is actually about US 8,949,879 (Truesight Communications), not US 8,793,758 — the certificate of service confirms it. Let me rule out any proceeding that genuinely names this patent.

Proceedings overview

Zero (0) AIA trial proceedings are on file for US 8,793,758 — the USPTO Open Data Portal structured block returns no IPR, PGR, or CBM naming this patent, and my web searches surfaced no petition, institution decision, or Final Written Decision directed to U.S. Patent No. 8,793,758. Breakdown by status: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0.

Bottom line for a defendant: none of 8,793,758's claims have been canceled or even tested at the PTAB. There is no FWD to hand a district court, no § 315(e)(2) estoppel to exploit, and no PTAB-driven narrowing. That cuts both ways — the patent is not "hardened" by a win, but it is also completely un-vetted, with no institution decision, no claim construction, and no expert record you can borrow. Any IPR defense on this patent must be built from scratch.

---

Proceedings

None — no AIA trial names US 8,793,758

There is no proceeding to enumerate. I am reporting that as a finding rather than filling the section with near-miss cases.

Verification performed (2026-09-29): serialized searches on the patent number, the title ("Security, fraud detection, and fraud mitigation in device-assisted services systems"), the owner (Headwater Research LLC / Headwater Partners I LLC), and the asserted-patent dockets in the Headwater campaigns. Headwater/Headwater Research is extremely PTAB-active — Samsung alone has filed on the order of 26 IPR petitions against Headwater patents across five suits, and Verizon, AT&T, T-Mobile/Sprint, and Google have all filed as petitioners or co-petitioners — so the absence of a proceeding here is a real datapoint, not an artifact of a quiet patent owner.

Caveat: ODP ingest lags. If a petition on 8,793,758 was filed in the last few months, it may not appear in ODP yet. The related-matters tables I reviewed in other Headwater IPRs did not list any 8,793,758 proceeding, which is meaningful because 37 C.F.R. § 42.8(b)(2) compels petitioners to disclose them.

---

⚠️ Misattribution watchlist — the "'879 patent" trap

Do not confuse this patent with U.S. Patent No. 8,949,879. Search results (and, in my experience, docket-monitoring feeds) routinely collapse "8,793,758" and "8,949,879" because both are colloquially called "the '879 patent." They are different patents with different owners:

The Google/Samsung petition's certificate of service reads "Petition for Inter Partes Review of U.S. Patent No. 8,949,879." Its § VII Fintiv argument ("Patent Owner asserted the '879 patent against Samsung… in the Eastern District of Texas") refers to Truesight, not Headwater. If a demand letter or an IPR-monitoring alert cites "the '879 patent" against you, verify the seven-digit number before you rely on — or fear — anything in that record.

---

Adjacent Headwater IPRs (informative only — none reaches 8,793,758)

These do not bind or estop anyone with respect to 8,793,758, but they define the litigation ecosystem and supply reusable prior art and claim-construction history:

Representative citations: PTAB E2E, https://ptacts.uspto.gov; Unified Patents PTAB portal, https://portal.unifiedpatents.com/ptab/case/IPR2025-00481; CourtListener docket Headwater Research LLC v. Verizon Communications Inc., https://www.courtlistener.com/docket/67648165/440/1/headwater-research-llc-v-verizon-communications-inc/.

Two observations that matter strategically even though they don't touch '758: (1) Headwater settles instituted IPRs and drops the corresponding patent from the litigation rather than fighting to FWD — that is its revealed preference; (2) Headwater has also shown willingness to disclaim claims to defeat institution (48 of 174 asserted claims of the '541 patent in its POPR to IPR2024-00944), which is itself a form of de facto narrowing you can exploit if it happens on your patent.

---

Strategic summary

Claim-level status of 8,793,758: CANCELED — none. SUSTAINED — none. UNTESTED — all. Every claim of 8,793,758 stands exactly as issued on 2014-07-29, with no PTAB or (as far as I can confirm) reexamination-driven amendment or disclaimer on the record. If you are mapping the patent, you are working against the full original claim set. I could not confirm whether this patent has ever been asserted in the Headwater campaigns or in the Truesight suits; treat any assertion of it as a fresh assertion.

Estoppel landscape: there is none to work with — and none against you. Because no IPR was instituted on this patent, no petitioner is barred by § 315(e)(2) from raising any ground, and no patent owner benefit accrues from surviving a trial. Practically, this means (a) you have the complete universe of § 102/§ 103 prior art available, including art that others may already have developed against sibling members of the Burgunder/Raleigh family, and (b) you face no Sotera-style stipulation, no adverse FWD findings on claim construction, and no § 325(d) "same or substantially the same art" risk from a prior Board record on this patent. The flip side: Headwater has no estoppel against it either, and it has a documented habit of settling instituted IPRs on confidential terms — so a favorable outcome may arrive as a settlement rather than a cancelation.

Pattern signals. (1) Same-petitioner repetition: Samsung and the carrier/Google coalition are serial filers against Headwater, and Samsung has filed 26 IPR petitions across the portfolio — a fresh petition on 8,793,758 by an existing petitioner is entirely plausible if the patent is asserted. (2) Patent-owner posture: Headwater pursues discretionary-denial briefing aggressively and asks the Director to deny institution on Fintiv/§ 314(a)/§ 325(d) grounds far more often than it litigates validity to a merits FWD; it has also responded to institution threats by disclaiming claims and dropping patents from litigation. Expect that playbook, not a merits fight. (3) Defensive aggregators: I found no Unified Patents (or similar defensive-aggregator) petition in the chain for this patent — the counter-evidence is only a Unified Patents patent profile page listing US-8793758-B2 as a Headwater asset. Do not infer an IPR from that page.

---

Recommended next steps

1. Do not rely on any PTAB invalidation of this patent — there isn't one. If your invalidity contentions or an IPR draft currently cite a decision on "the '879 patent," stop and confirm the number is 8,793,758 and not 8,949,879. That single digit-pair error is the most likely failure mode on this file; a merits argument built on Truesight's record would be a Rule 11 problem, not a defense.
2. Build the challenge yourself, and start with the family. Because there is no institution decision or FWD on 8,793,758, there is no PTAB claim-construction guidance for it. Pull the prosecution history and the sibling patents and the art already deployed against them: Freund (U.S. 5,987,611), Rao (U.S. Pub. 2006/0039354), Cole (U.S. Pub. 2008/0080458), Kelz (EP 1 484 871 A1), Van Camp (U.S. 8,914,783), Oestvall (U.S. Pub. 2007/0038763), Montemurro (U.S. Pub. 2009/0207817), Fadell (U.S. Pub. 2010/0017506), Smith (VM/sandboxing), Owen (U.S. Pub. 2005/0213763), Bartlett (U.S. Pub. 2003/0177396), and Wyld (U.S. 8,028,060). These are the combinations that have produced institutions and at least one all-claims-invalidated FWD (IPR2024-00942/-00943) in the same portfolio; they are the highest-yield starting point for § 103 and for the § 112 indefiniteness theories Verizon raised against sibling Headwater claims.
3. Watch the ODP and E2E docket, and calendar the statutory clock. If a petition on 8,793,758 is filed, the § 314(b) institution deadline is six months from the preliminary-response due date, and trial, if instituted, closes with an FWD within one year of institution (extendable to 18 months for good cause). Because your patent is 12+ years past priority (2009-01-28) and issued 2014-07-29, expect the patent owner to fight institution on Fintiv/§ 314(a)/§ 325(d) and on the Stewart Memorandum's "settled expectations" factor (length of time in force) — which, notably, Samsung has successfully argued against in the neighboring proceedings.
4. Note the reexamination alternative. Ex parte reexamination is not an AIA trial and would not appear in the ODP AIA-trial data — a reexam (e.g., 90/019,892 concerning the '541 patent family) is a separate track. I found no reexam naming 8,793,758, but if your goal is a cheaper, lower-profile validity challenge, an ex parte reexam grounded on printed publications is available here precisely because no prior PTO challenge has consumed the art (§ 325(d) exposure is low).

Confidence and limits: High confidence that the ODP record showing zero AIA trials matches the public record as of 2026-09-29, and high confidence in the 8,793,758 / 8,949,879 distinction. I could not confirm whether 8,793,758 has ever been asserted in litigation, whether any Headwater-defendant has pre-filing diligence or § 315(b) constraints with respect to it, or whether a petition has been filed within ODP's ingest lag window. Those gaps should be closed by pulling the PTAB E2E docket directly and by checking the related-matters (§ 42.8(b)(2)) tables in any newly filed Headwater petition.

I'll research the assignment record for US 8,793,758, starting with targeted searches for the reel/frame data and correspondent information.

Good finds. Let me dig deeper into the specific assignment records and the corporate/ownership history.

Excellent — this is yielding concrete reel/frame and correspondent data. Let me pin down the '758-specific assignments and the correspondent's role.

Assignment & Ownership Reconstruction: US 8,793,758 B2

Verification note up front: I hit the tool-step ceiling before I could pull the USPTO Assignment Center's per-patent result screen directly. What follows distinguishes (a) reel/frame values I actually retrieved, (b) assignment events recorded on the patent's Google Patents legal events table (which mirrors USPTO Assignment records but does not display reel/frame), and (c) items I could not verify. I have not fabricated any reel/frame. Where I only have the family-level analogue, I say so.

---

Inventors

Per the issued patent's front page: Gregory G. Raleigh; Jeffrey Green; James Lavine.

Unusual patterns to flag:

• Inventor-set mismatch with the litigation record. The '758 names Raleigh, Green, Lavine. The Headwater patents actually put at issue in the 2022–2025 campaigns name Raleigh, Lavine, and Alireza Raissinia — a different inventor set. This is consistent with the previous section's finding that '758 was not itself asserted, and it means the Qualcomm co-ownership dispute that dominates the Samsung/Verizon/AT&T record does not, on the face of the patent, attach to '758 via Raissinia.
• Founder-departure timing. Raleigh filed the root provisional ('354, Jan 28, 2009 — the '758's earliest priority) within roughly four months of leaving Qualcomm. Headwater's own adversaries built their standing challenge on exactly this: the one-year post-departure presumption of employer ownership. The '758 descends from that same priority chain, so the same fact pattern is latent for '758 even though it is not (so far as I found) in suit.
• I found no evidence that Green or Lavine departed Headwater within 12 months of filing. I cannot confirm their employer, so the "all inventors bail out" tell is unclear, not present.

---

Original assignee

Headwater Partners I LLC, a Delaware limited liability company, formed ~Nov 2008 by Gregory Raleigh. Original business address on the early family assignments: 2995 Woodside Road, Ste. 400, Woodside, CA 94062-2048, later 350 Marine Parkway, Suite 300, Redwood Shores, CA 94065 and 555 Twin Dolphin Drive, Suite 310, Redwood Shores, CA 94065.

• Primary line of business: mobile "device-assisted services" (DAS) — an operating system layer that classifies app-level traffic and applies per-application charging/control/notification policy on wireless devices.
• Did it ship a product embodying the claims? Indirectly. Headwater's own litigation says the technology was distributed to carriers (Sprint) and OEMs (Samsung) and that "by the end of 2015, millions of Sprint devices…were running the ItsOn application, which included Headwater's intellectual property." ItsOn is a separate company Raleigh also founded (incorporated Sept 17, 2008); Headwater itself was an IP/R&D entity that partnered with ItsOn rather than selling hardware or software under its own brand.
• Current status: not operating; dissolved by merger. On December 29, 2016, Headwater Partners I LLC merged into Headwater Management LLC, and the surviving LLC was renamed Headwater Research LLC (Delaware Certificate of Merger, File No. 4966025, SR 20167326785). The successor, Headwater Research LLC, is active and is the current assignee of record.

---

Assignment timeline

Every recorded link I could retrieve is an internal transfer — inventors → operating company, then a corporate name-change/merger. No assignment to a third-party NPE appears in the recorded chain. Dates are recording dates where the source is the Google Patents legal-events table, execution/effective dates where the source is the assignment abstract itself.

• 2012-03-15 (recorded) — Reel/frame not retrieved

  • Conveyance: Assignment (inventor → company)
  • Assignor: Raleigh, Gregory G.
  • Assignee: Headwater Partners I, LLC
  • Correspondent: not shown on the legal-events table. (Family analogue, below, shows Krista S. Jacobsen, Reg. No. 59,374, in-house Head Counsel, Headwater Partners I LLC, patent@headwaterllc.com / krista@jacobseniplaw.com.)
  • Context: original inventor assignment of the Raleigh interest.

• 2012-04-03 (recorded) — Reel/frame not retrieved

  • Conveyance: Assignment (inventor → company)
  • Assignors: Green, Jeffrey; Lavine, James; Raleigh, Gregory G.
  • Assignee: Headwater Partners I, LLC
  • Correspondent: not shown. (Same family correspondent pattern as above — Jacobsen.)
  • Context: original inventor assignment completing company ownership of the three-inventor '758 family.

  Family baseline (not confirmed for '758 specifically): the earliest Headwater "Raleigh → Headwater Partners I, LLC" conveyance in this portfolio is recorded at Reel 024248, Frame 0209 (cited in the SB/80 for application 12/694,455, executed July 16, 2012). The '758 is a later continuation, so its own inventor assignments were likely recorded at a different reel/frame than 024248/0209 — I could not confirm.

• 2016-12-29 (executed/effective) / recorded 2017-01-04 (per legal-events table; the abstract's own recording stamp was not captured) — Reel 041248 / 0777

  • Conveyance: Merger and Change of Name
  • Assignors: Headwater Partners I LLC and Headwater Management LLC
  • Assignee / receiving party: Headwater Research LLC
  • Correspondent: Ann Taylor, 1011 Pruitt Place, Tyler, TX 75703; tel. 903-201-6211; patent@headwaterllc.com. Signed "/Ann Taylor/". Flag: this is the only post-2013 correspondent in the chain, and the sole recorded address for the surviving entity is a Tyler, Texas street address — the seat of the E.D. Tex. Tyler Division where Headwater's later campaigns landed.
  • Context: internal corporate reorganization (same legal successor; change of name only, no change in beneficial ownership). Delaware Certificate of Merger confirms Headwater Partners I LLC merged into Headwater Management LLC under the new name Headwater Research LLC.

Caveat on the 2016 reel/frame: I retrieved Reel 041248 / Frame 0777 from the assignment abstract attached to a related Headwater application (14/667,516, exhibit in E.D. Tex. 2:23-cv-00641 and 2:23-cv-00379), not from the '758-specific page. The '758's own legal-events entry displays the same merger ("Assigned to HEADWATER RESEARCH LLC," recorded 2017-01-04, assignors Headwater Management LLC / Headwater Partners I LLC), so '758 is covered by the merger recording, but I have not independently confirmed that '758's specific reel/frame is 041248/0777. Treat the reel/frame as family-level corroboration, not a '758-specific citation.

No security agreement, license, release, correction, or third-party transfer recorded against '758 was found. If any exist, they did not surface in my searches.

---

Timeline diagram

timeline
    title Ownership of US 8793758
    2008 : Raleigh founds Headwater
    2009 : Earliest priority provisionals filed
    2011 : Application 13 309 463 filed
    2012 : Inventors assign to Headwater Partners I
    2014 : Patent 8793758 issues
    2016 : Merger creates Headwater Research
    2017 : Merger change of name recorded
    2022 : Headwater sues Samsung
    2023 : Suits vs Verizon ATT and T-Mobile

Parser note: I replaced the application number's comma/slash with spaces ("13 309 463") to keep the event text punctuation-free; the number is 13/309,463.

---

NPE / troll-pattern signals

1. Shell-entity transfer — not present (recorded chain), partial on conduct. The patent did not move from an operating assignee to a new licensing-only third party. It moved by merger and change of name only (Reel 041248/0777, eff. 2016-12-29): legal successor, same beneficial owner. The classic tells are unmet — no "IP/Patents/Licensing" suffix, no separately formed NPE acquirer, no registered-agent service address proven. The weak supporting fact is that the surviving entity Headwater Research LLC holds and licenses IP with no product sales of its own; but that is conduct evidence, not an assignment-record finding.

2. Known asserter in the chain — present. Current assignee Headwater Research LLC is a documented high-frequency patent plaintiff: suits against Samsung (2:22-cv-00422; 2:23-cv-00103 — ~$279M jury verdict, Apr. 2025), Verizon (2:23-cv-00352), AT&T (2:23-cv-00397), T-Mobile (2:23-cv-00379), plus the 2025 wave against Apple, Amazon, Google, Comcast, Charter, and Dish. It does not appear on the legacy lists named in the brief (Acacia, Marathon, IV, Wi-LAN, Pendrell, Round Rock, etc.); it is a new-generation prolific asserter surfaced by litigation coverage.

3. Repeat correspondent across the chain — unclear / weak. Two different correspondents appear: Krista S. Jacobsen (Reg. No. 59,374), Head Counsel, Headwater Partners I LLC, on the 2012–2013 filings, and Ann Taylor, 1011 Pruitt Place, Tyler, TX, on the 2016 merger recording. A single repeat-player attorney running multiple shell LLCs is not the pattern here — both correspondents appear to be in-house, tied to the same company, not outside NPE counsel cycling assets through a stable of LLCs.

4. Cascading transfers — not present. The chain has one substantive link beyond the original inventor assignments: the 2016 merger. No <24-month chain of successive LLCs, no shared-principal sequence.

5. Pre-litigation transfer — not present. The only entity-level recorded event (the 2016 merger) predates the first headwater assertion campaign by ~6 years. The chain was not rearranged within 6 months of suit. The Tyler, TX address is adjacent to the venue story but is not itself a transfer-date fact.

6. Bankruptcy fire-sale — not present. No Chapter 7/11 record, no court-supervised sale found.

7. Privateering — unclear. Headwater's technology was commercialized through ItsOn (separate company) and shipped on Sprint/Samsung devices, and Headwater now asserts against device makers and carriers. That is directionally suggestive of an operating-technology origin, but I found no evidenced transfer from a practicing operating company to an NPE asserting on that company's behalf (no SEC filing, no stated license-back/privateering relationship).

8. Defensive aggregator — not present. The chain terminates at a litigating owner, not at RPX / AST / LOT / Unified / OIN.

---

Verdict

NPE — moderate confidence. The recorded chain is not a classic shell transfer: it is inventors → Headwater Partners I LLC (2012, Reel-family 024248/0209 analogue) → Headwater Research LLC by merger/change of name only (Reel 041248/0777, eff. 2016-12-29) — the same legal successor throughout, with no third-party NPE acquisition. One signal is nevertheless strong and record-grounded: the current assignee (Headwater Research LLC) is a documented high-frequency asserter/non-practicing monetizer (the 2022–2025 campaigns and the ~$279M Samsung verdict), and it holds IP without selling a product of its own (the practicing product, ItsOn, is a separate company). With one strong signal plus weak supporting facts, the honest call is moderate, escalating to high if you independently confirm (i) Headwater Research LLC has no product revenue and (ii) the Reel 041248/0777 merger schedule expressly lists US 8,793,758.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ · Assignment Search — https://assignment.uspto.gov/patent/index.html (query patent 8,793,758 and also application 13/309,463; the merger recording is at Reel 041248 Frame 0777).

---

Two flagged uncertainties / potential cross-reference notes:

• The inventor set on '758 (Raleigh, Green, Lavine) differs from the Raleigh/Lavine/Raissinia set behind the Qualcomm co-ownership dispute discussed in the litigation record. Nothing in the '758 assignment chain shows Raissinia, so the Qualcomm standing theory does not visibly reach '758. No contradiction with the prior litigation section, which likewise could not tie '758 to an asserted case.
• The prior litigation section listed the current assignee as Headwater Research LLC and priority as 2009-01-28 — both are confirmed, not contradicted, by the assignment record above.

I'll research the actual prior-art citations for US 8,793,758. First, let me note a critical gap: the patent text provided to me is the description only — it does not include the "References Cited" / "Patent Citations" section. So I need to retrieve the citation record directly. Let me search.

The searches returned the patent's description but not its citation tables. Let me try to retrieve the front-page references directly.

I'm finding the family's citation lists. Let me get the '758-specific front page and citation list.

Prior-Art Analysis — US 8,793,758 B2

Important disclosure up front: I hit the tool-step limit mid-research. More importantly, the patent text supplied to me is the description only — it does not include the front page ("References Cited") or the claims. I therefore could not produce a fully verified, per-reference §102 anticipation table. Below I give (a) what I could confirm from live sources, (b) the reference landscape I did retrieve, and (c) a clearly-labeled conditional analysis. I have not invented any citation, date, or title to fill gaps.

---

1. USPTO / bibliographic confirmation for 8793758 (literal ID, no auto-correction)

Statutory note on the critical date. Because application 13/309,463 was filed before the March 16, 2013 AIA changeover, pre-AIA §102 governs. The operative prior-art cutoff is the 2009-01-28 priority date. This matters: only references with an effective date before 2009-01-28 can qualify as §102(a)/(b) art; later-dated documents could only qualify under §102(e) (a U.S. patent/application with an earlier effective filing). Many references surfaced below are 1990s–2000s and therefore eligible; any 2009–2014-dated sibling would need a §102(e) analysis.

---

2. Limitation on this analysis (flagged explicitly)

• The provided '758 text contains no claims section and no front-page citation list. The document ends mid-sentence in the FIG. 4 discussion ("…must pass through firewall 18 to reach file server 126, and through").
• I could not, within budget, open the USPTO "References Cited" page (uspto.report/patent/grant/8793758 or the patentimages PDF front page) to confirm the examiner-cited list specific to the '758.
• Consequently, the reference list below is drawn from sibling family members' front pages and from the patent's Unified Patents page, not from the '758's own front page. Treat it as a strong proxy, not a verified list.

---

3. Reference landscape retrieved

3a. U.S. Patent Documents appearing on sibling Headwater front pages (proxy for the '758 list)

These were visible in the "References Cited" lists of co-family patents (US 8,893,009; US 9,578,182; US 9,609,544; US 10,841,839). All are pre-2009 and thus §102-eligible. Titles were not visible in the retrieved snippets and are marked "title unverified."

Additional numbers visible in the same lists (title unverified): 6,263,055 Garland; 6,292,828 Williams; 6,381,316 Joyce; 6,418,147 Wiedeman; 6,505,114 Luciani; 6,532,235 Benson; 6,532,579 Sato; 6,539,082 Lowe; 6,542,992 Peirce; 6,563,806 Yano; 6,574,321 Cox; 6,574,465 Marsh; 6,581,092 Motoyama; 6,119,933 Wong; 6,115,823 Velasco; 6,061,571 Tamura; 6,058,434 Wilt; 5,974,439 Bollella.

3b. Documents surfaced on the '758's own patent page (Unified Patents portal) — verification status uncertain

I could not confirm whether these three are cited-by, citing, or merely similar documents on that portal. Do not treat them as examiner-cited art without front-page confirmation.

---

4. Conditional §102 assessment against the '758's claimed subject matter

The '758's disclosed independent-claim concepts (from the "Definitions" summary in the supplied text) are:

• (i) end-user device with modem(s) + stored application-specific network access policy enforced by device agents, including flow-tagging;
• (ii) an API to arrange an application setting to implement the policy;
• (iii) credential-based integrity — an application/agent may be modified only if the update's credential matches the stored credential; agent checks (hosts file, service-processor authentication, rooting);
• (iv) a network system obtaining a trusted measure of device usage to confirm policy compliance; and
• (v) a multi-tiered policy-verification process comparing trusted vs. device-generated usage measures, with pass/fail or quantized criteria and defined error actions (flag, re-rate, quarantine, suspend).

Anticipation (§102) vs. obviousness (§103). On the record available, no single retrieved reference appears to disclose every element of any independent claim. The references split across subject-matter clusters:

Bottom line on §102: On the evidence I could retrieve, the '758 claims look like a combination of known elements (device-side policy enforcement + credential gating + multi-tier reconciliation), i.e., an obviousness (§103) posture rather than a clean anticipation (§102) posture. I cannot responsibly name a single anticipatory reference.

---

5. Contradiction / consistency check with the earlier litigation section

• The prior Litigation Summary concluded no case was found asserting the '758. Nothing in this prior-art research contradicts that.
• One caution: the family's front-page reference lists are near-identical across siblings (e.g., US 8,893,009, 9,578,182, 9,609,544 all begin "5,283,904 Carson… 5,577,100 McGregor…"). It is easy to attribute a sibling's citation list to the '758. I have labeled them as proxies for exactly that reason.

---

6. Confidence and required verification

To finish this properly (I was unable to complete these):

1. Open the '758 front page at uspto.report/patent/grant/8793758 or the patentimages PDF to capture the exact "References Cited" (U.S. patents, U.S. publications, foreign, and NPL).
2. Retrieve the full claim set of the '758 (app. 13/309,463) from USPTO Patent Center/Public Search.
3. Confirm the citation role (cited-by vs. citing vs. similar) of US 2004/0168052 A1, US 7,609,700 B1, and US 2009/0180391 A1.
4. Then produce a true element-by-element §102 chart against independent claim 1 and the dependent claims.

If you'd like, I can run those four lookups next and return a verified, claim-charted prior-art table.

I'll research the prior art landscape for the '758 patent before analyzing.

Let me dig into the specific prior-art citations and related IPR records.

Let me try to retrieve the patent's own cited-references list and the '758-specific record.

Obviousness Analysis — U.S. Patent No. 8,793,758

("Security, fraud detection, and fraud mitigation in device-assisted services systems") under 35 U.S.C. § 103

Analyst note on scope and evidentiary basis. The instruction is to use "the results from the Prior Art section of this page." Two limitations must be flagged up front, consistent with the earlier sections of this analysis:

1. The authoritative full text supplied for the '758 does not contain a rendered "Prior Art"/"Citations" section. The fetched patents.google.com text terminates mid-Description (in the FIG. 4 discussion of security zones). I therefore could not quote the '758's own front-page "References Cited" list. Rather than invent one, I have reconstructed the operative prior-art set from prior art actually deployed against the '758's own child patent and against the '758's siblings in co-pending IPRs and district-court invalidity contentions. That is a defensible proxy, and I label it as a proxy throughout.
2. The '758 claim set is still unavailable to me (unchanged from the Patent Summary section). I therefore analyze at the level of claim-adjacent limitation sets that appear in the supplied '758 specification, cross-checked against the '055's claim 1 (the '055 being a division of application 13/309,463, i.e., of the '758 itself) and against the sibling '541/'613 claim families. I do not state or impute claim numbers for the '758.

Date inconsistency (already flagged once, repeated for the record): the task prompt says April 26, 2026; my system date is September 29, 2026; prior sections retrieved documents dated May–August 2026. I report what the sources show.

---

1. Governing legal framework

Critical threshold point — the effective filing date. The '758 is a continuation-in-part layered over the Jan. 28, 2009 provisional/non-provisional priority chain, but the security- and fraud-specific disclosure traces to later provisionals. The '758's own priority statement (as reproduced in the family) claims, inter alia:

• 61/418,509, filed Dec. 1, 2010 — "Service Usage Reporting Reconciliation and Fraud Detection for Device Assisted Services"
• 61/422,574, filed Dec. 13, 2010 — "Security and Fraud Detection for Device Assisted Services"
• plus 61/418,507, 61/420,727, 61/422,565, 61/422,572, 61/435,564, etc.

Source: https://www.freepatentsonline.com/[9491199](/patent/9491199).html (reproducing the 13/309,463 priority paragraph).

Why this matters for § 103: if any '758 claim limitation lacks written-description support in the Jan. 28, 2009 priority document and is supported only by the Dec. 2010 provisionals, that limitation's critical date slides forward by nearly two years. Headwater itself has contended a May 25, 2010 priority date for the sibling '541 and '613 patents (Verizon expert report of Kevin Jeffay, https://archive.org/download/gov.uscourts.txed.[224022](/patent/224022)/gov.uscourts.txed.224022.200.15.pdf, ¶ 119). A later critical date broadens the § 102/§ 103 art window and pulls in references such as U.S. 2010/0017506, U.S. 2011/0019557, and — if the date slides past their § 102(e) dates — U.S. 2012/0102504 and U.S. 2012/0265897, which AT&T listed in its contentions. This is the single most consequential variable in any '758 obviousness case, and it is a priority/§ 112 fight dressed as a § 103 fight.

---

2. The limitation sets being analyzed

I organize the '758's claim-adjacent subject matter into four limitation sets, each drawn from language in the supplied '758 text (marked "’758 spec") or from a closely related claim as labeled:

Set A — Device-side application-specific network access policy (’758 spec, claim-adjacent summary text)

"one or more modems to allow communications over a wireless access network, memory configured to store an application-specific network access policy… and one or more device agents configured to detect attempted or successful activity by the particular application program and to apply the application-specific network access policy."

includes: flow-tagging a data flow, associating the flow tag with an application identifier, applying the policy to the flow-tagged flow; agents using an API "to arrange an application setting" to implement the policy; policy types = control / charging / notification.

Set B — Credentialing and secure-link architecture (proxy: '055 claim 1 — a division of 13/309,463)

receiving a request for a secure device credential; verifying device identifiers; generating/associating/securely sending the credential; "based at least in part on the secure device credential, negotiating a secure message link" between a network message-link server and a device link agent; receiving over that link a request for a secured application credential comprising a "general application credential"; generating a device-unique secured application credential and sending it back.
Source: https://litigation.rpxcorp.com/patent/[US10064055B2](/patent/US10064055B2) and the '055 petition at ptacts.uspto.gov (IPR2025-01572).

Set C — Application/agent integrity and update control (’758 spec)

agents prevented from modifying/updating/replacing an application unless the update software carries a credential consistent with the application credential; agent credential = proper signature/certificate/hash of kernel, framework, or application agent software; checks for a hosts file, successful service-controller authentication, and rooting.

Set D — Multi-tiered policy verification and fraud detection (’758 spec)

a network system that "obtain[s] a trusted measure of access network usage… and… use[s] the trusted measure to confirm that the end-user device is properly implementing the access network policy," with a multi-tiered verification process including (a) missing/absent device reports vs. received trusted reports; (b) trusted measure vs. expected range; (c) trusted vs. non-secure measure within a tolerance; (d) non-secure measure vs. expected range; (e) usage classification vs. expected range; and an error action set (flag, re-rate, notify, quarantine, suspend).

---

3. Prior-art reference inventory

3.1 Art deployed against the '758's own child (U.S. 10,064,055) — the most probative set

Apple's IPR2025-01572 petition challenges the '055 (division of 13/309,463) on six grounds, all built from the same three-reference backbone plus objective-evidence-dependent swaps:

Source: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1558526](/patent/1558526)/download-documents?artifactId=1bjmteI_j_EZ13vY05YLrReWNiNkxyvoMcLYmHgTLnNy7uWnmHcdeZU

The petition's own analogy statement: the art is "from the same field of endeavor as the ['055] Patent, namely, security of devices and/or software communicating data via a network," citing the '055 abstract and cols. 1:53-55, 56:54-56, 57:40-42. That field-of-endeavor finding transfers directly to the '758, since the '758 and the '055 share one specification and one priority chain.

Analytical significance. Because the '055 is a division of the '758 — not a different invention — any ground that renders a '055 claim obvious on this specification is presumptively available against the corresponding '758 disclosure. The only escape for Headwater is that the '758's claims recite different limitations than the '055's. That is precisely the material I could not retrieve, and it is the gap a competent '758 obviousness case must close with the actual claim text.

3.2 Art deployed against the '758's siblings in the Verizon / AT&T / Samsung campaigns

These are the "same-family, same-specification" contentions, and they are the natural secondary-art pool for a '758 challenge.

(a) Verizon / PTAB — '541 patent (closely related policy-verification claims): the Board's Final Written Decision in the '541 IPR records the ground "Rao, Fadell, Freund" under § 103(a) for claims 97, 98, 100, 101, 139, and found "the challenged claims [unpatentable]" by a preponderance. Source: https://storage.courtlistener.com/recap/gov.uscourts.txed.224022/gov.uscourts.txed.224022.440.1.pdf

(b) Verizon P.R. 3-3 invalidity contentions — asserted against '541 / '543 / '042 / '613, with per-patent "Appendix of Secondary References & Obviousness Combinations" (Appendices A-A through D-D) and a dedicated "Motivations to Combine Prior Art" section. Source: https://www.docketalarm.com/cases/PTAB/IPR2024-00945/Google_LLC_v._Headwater_Research_LLC/docs/09-18-2024-Patent-Owner/Exhibit-2018-Exhibit_2018.pdf

(c) AT&T invalidity contentions — a representative reference list for the DAS family:

Source: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1555958](/patent/1555958)/download-documents?artifactId=NRCQgHfkY-yCWBx3HQ1DCZ2jbSBOWqF4n5xciwJna4IjtMIE8ayhiVo

Note that U.S. 2012/0102504, 2012/0265897, 8,397,087, 8,667,513, and 9,098,333 postdate the '758's Dec. 1, 2011 filing. Their appearance in a contentions list is a tell: AT&T expected to rely on them as § 102(e) art (US patents/publications with earlier effective filing dates), which only works against a target whose claims are pushed to a later critical date than Jan. 28, 2009. Again: priority is the pivot.

(d) Samsung invalidity contentions — reference list including U.S. 2004/0267897 (Hill), U.S. 2005/0273850 (Freund), U.S. 2003/0204596 (Yadav), U.S. 8,549,173 (Wu), U.S. 2008/0183996 (Field), WO 2009/112044 (Malomsoky), WO 2007/023467 (Tonsing '467), WO 2007/118638 (Oester), U.S. 2006/0136679 (O'Connor). Source: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1557850](/patent/1557850)/download-documents?artifactId=Ntg4HsuFgAUB7YiDjPvrjnlg9WJHVkhyf3JUdUurdweFduZa6hVb0bc

Freund is the bridge reference — it appears both in Samsung's district-court list and in the instituted Verizon PTAB ground ("Rao, Fadell, Freund"). A reference that the Board has already accepted as part of a successful § 103 combination against the same specification is the strongest building block available for a '758 challenge.

(e) Non-patent literature — the WAP standard. Samsung charted the WAP Push Architecture Overview (both the Nov. 1999 and July 2001 versions) as a complete § 103 ground, and survived Headwater's motion to strike on the theory-disclosure point. Source: https://storage.courtlistener.com/recap/gov.uscourts.txed.220857/gov.uscourts.txed.220857.311.0.pdf

Why this matters to the '758 specifically: the WAP Push material is asserted art for the "service control link"/"device messaging agent"/"device identifier" concepts — i.e., exactly the secure message link + device link agent + device identifier limitations that sit inside the '758's Set B, and that Apple maps to Lundblade/Jobst/Hardjono for the '055. Two independent routes to the same limitation.

3.3 What I could not obtain

• The '758's literal front-page References Cited list (the "Prior Art" section as such).
• The '758's literal claim set.
• Whether any IPR/PGR has ever been filed on 8,793,758 specifically (consistent with the Litigation section's negative finding; no search returned one).

I state these as gaps, not as findings of absence.

---

4. Obviousness combinations

Below I set out four grounds. For each: what the references supply, the motivation to combine, the KSR rationale, and the residual gap that Headwater would defend.

---

GROUND 1 — Sets B and C (credentialing / secure link / update integrity)

Lundblade + Jobst + Hardjono (primary), optionally + Zuccherato, Chen, Lundblade '330, Fosback, or Scarlata

• Lundblade supplies the device-credential provisioning and secure-message-link architecture: request → verification → generate → associate → send; the secure link then carries a second, application-scoped credential request. Apple charts this against the '055's claim 1 steps ("receiving a first request… for a secure device credential"; "negotiating a secure message link"; "a second request… for a secured application credential… comprising a general application credential").
• Jobst supplies the application/software credential-verification and trusted-execution aspects — the "consistent credential" gate on modification/update/replacement that the '758 recites for both application software and agent software (Set C), and the signature/certificate/hash-of-kernel-framework-application agent credential.
• Hardjono supplies the cryptographic credential-issuing-token and general→device-unique credential-derivation model (generic credential for a class, specialized credential unique to the device).
• The dependent-claim references map to specific narrow features: Zuccherato (certificate/CRL handling), Chen (secure execution/memory partitioning), Lundblade '330 (secure provisioning of a second/alternate credential type), Fosback (secure software download/update), Scarlata (trusted platform / attestation).

Motivation to combine (as Apple frames it, and as KSR supports): all three references address the same problem — how a network authenticates an untrusted client and how that client proves the provenance of code it runs — from complementary layers (link/transport credentialing; application-level credential verification; issuer-side credential generation). Combining them "would have been the combination of prior art elements according to their established functions, within the skill of an ordinarily skilled artisan, and yielding predictable results" (KSR, 550 U.S. at 416–18, as quoted in the Board's own framing in the analogous SightSound CBM institution). Field-of-endeavor identity makes the combination presumptively motivated; there is no apparent teaching away, because each reference performs its function independent of the others' function.

Residual gap for Headwater: the specific architecture of a distinct network-side "message link server" negotiating a link with a distinct on-device "device link agent," and the device-unique derivation of the application credential from the device credential. Expect Headwater to argue that the petitions' art individualizes credentials at the application level but not at the device-credential → application-credential chained level. The '055 IPR is the live test of exactly that argument.

---

GROUND 2 — Set D (trusted-measure reconciliation and multi-tiered verification)

Rao + Fadell + Freund (the instituted, adjudicated Verizon ground), optionally + AT&T's network-mediation art (U.S. 7,260,635; U.S. 2002/0166117; U.S. 2006/0059556; U.S. 5,987,611)

• Rao supplies network-side service-usage monitoring/mediation and the reconciliation of network-observed usage against a second usage record — the structural core of the '758's "trusted measure… to confirm that the end-user device is properly implementing the access network policy."
• Fadell supplies device-side service-state and connection-state information used to shape service behavior across heterogeneous access networks — the '758's connection-manager/network-type/roaming dimensions ("device connection state information can comprise… type of network information").
• Freund supplies per-application traffic handling and service-plan/policy attribution in a mobile data network — the '758's application-classification breakdown and sponsored-service accounting.
• Secondary network art (AT&T list): U.S. 5,987,611 (per-application network-access control/firewall on a client — directly on-point for Set A's "application-specific network access policy"), U.S. 2002/0166117 and U.S. 7,260,635 (policy-based charging/metering in a data network), U.S. 2006/0059556 (usage monitoring and notification).

Motivation to combine — and the '758's own admissions. This ground is unusually strong because the '758 specification supplies the motivation itself. The supplied text (see FIGS. 24–27 and the accompanying description) treats SGSN notification, GGSN notification, AAA/RADIUS start-stop accounting, and OCS start-stop accounting as existing, standard network infrastructure. Once the patentee concedes that network elements already generate trusted start/stop usage records, the recited step of comparing a trusted network measure to a device-generated measure requires no inventive faculty — it requires only the ordinary operator's desire to detect under-reporting. KSR's "design incentive" and "market pressure" rationales do the rest: carriers deploying sponsored/limited service plans had a direct financial incentive to detect devices that mis-apply policy.

The tolerance-band limitation ("comparing… to determine if the difference between the two usage measures is within a specified tolerance") is a quintessential obvious design parameter — a numerical tolerance chosen by routine experimentation to balance false positives against fraud leakage. In re Aller; In re Boesch.

Residual gap for Headwater: the multi-tiered / quantized aggregation — a "pass/fail" value per tier combined into an overall criterion, or a "quantized value associated with an error likelihood." This is the limitation with the best non-obviousness story, because it is a specific combination rule, not a single comparison. Whether it survives depends on whether the art shows a layered or scoring-based decision (Verizon's expert and secondary-reference appendices are aimed at exactly this).

---

GROUND 3 — Set A (per-application policy with flow tagging and API-based configuration)

U.S. 5,987,611 (per-application network access control) + Yadav or Hill (policy/usage management) + Freund (mobile per-application attribution); alternative route via Malomsoky (WO 2009/112044, traffic classification) + Oester (WO 2007/118638, policy control) + Tonsing (WO 2007/023467, policy-based charging)

• The application-to-policy binding, the classification of a traffic flow to an application, and the enforcement of allow/block/throttle by application are all long-predating the '758's 2009–2011 window. U.S. 5,987,611 (1999) is the classic client-side per-application access-policy reference.
• The "arrange an application setting via an API" limitation is, on the supplied '758 text, merely the integration mechanism: the agent "post[s], send[s], or otherwise communicate[s] a message comprising a setting configuration." Implementing a policy agent's configuration through a programmatic interface was, by 2009–2011, a universal engineering convention in mobile OS design — the kind of "improvement… consistent with the broad purpose of the prior art" that KSR treats as obvious.

Motivation to combine: the references are all directed to the same problem the '758 itself identifies — a device sending "large quantities of information," with the operator needing granular, per-application control for capacity protection and sponsored billing. Combining a client-side per-application policy engine with a mobile per-application flow-classification mechanism yields, predictably, a device that enforces policy per application. No new technical principle is required.

Residual gap for Headwater: the flow-tagging mechanics — specifically the virtual tagging book-keeping through the stack, and literal tagging in unused packet-frame bit fields removed before egress. This is the most distinctive technical disclosure in the '758's Set A. If the '758's claims reach tagging mechanics (rather than merely "associating a flow with an application"), this limitation requires art showing in-stack flow-to-application correlation, and the ordinary § 103 case would need DPI/flow-correlation art. This is exactly why the missing claim text is dispositive. The supplied '758 text is candid that "some standard communication stacks provide for unused, unspecified or otherwise available bit fields in a packet frame or flow" — an admission that the substrate for literal tagging was standard. The residual question is only whether the use of that substrate for policy attribution was known.

---

GROUND 4 — Set C's environment-integrity sub-limitations

Tkacik-type secure-processing art (separate security engine from processor core) + Lundblade/Jobst + Fosback — cf. the PTAB petition using Tkacik's secure processing system 14 (host processor 16 + secure memory 20, on-chip components 15) for a "security engine separate from a processor core" limitation (https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1556859](/patent/1556859)/download-documents).

• The '758's recited checks — "determining if a hosts file is present and properly configured," "determining if the end-user device has been rooted," "determining if a service processor… successfully completed an authentication procedure" — are the standard jailbreak/root/tamper-detection checks ubiquitous in mobile security by 2010. Note the petitions argue as common knowledge that "a security engine should be separate from a processor core," citing Weise (HSM/crypto accelerator operating "without the intervention of the host operating system").
• Motivation: a device-enforced policy engine that can be rooted is worthless; the entire point of the '758's DAS architecture creates the incentive to detect rooting and hosts-file tampering. That is a textbook KSR "design incentive" combination.

Residual gap: essentially none at the level of individual checks; the vulnerability is aggregation and claim scope, not novelty of the individual checks.

---

5. Cross-cutting KSR rationales (usable across all grounds)

1. Same field of endeavor. Every reference above is in wireless data service policy, device software security, or usage accounting/metering. Apple's filed petition states the field-of-endeavor finding for the same specification; the Board has already accepted Rao/Fadell/Freund against the same specification.
2. Known problem, known solution. The '758's own background frames the problem as mobile devices "sending and receiving increasing amounts of data" and supporting many applications. Per-application policy, sponsored services, and network-side verification are the recognized responses.
3. Predictable results / no new principle. Credentialing, policy enforcement, trusted-vs-claimed measurement reconciliation, and tolerance thresholds are each individually established techniques.
4. Design incentive + market pressure. Sponsored data, tiered plans, and roaming upsell create direct financial incentive to detect and mitigate device-side policy evasion. The '758's own FIG. 17–18 (fraud server, reconciliation server) and FIG. 41–51 (layered fraud, rules/statistics/time-series) are engineering implementations of that incentive, not departures from the art.
5. Obvious design parameters. Tolerances, thresholds, reporting intervals, and quantization levels.
6. Standard-stack admission. The specification's own statements about standard bit fields, and its treatment of AAA/RADIUS/OCS/SGSN/GGSN as pre-existing, undercut any argument that the substrate for the claimed mechanisms was inventive.

---

6. Counterweights Headwater will (and should) press

• No verified '758 claim construction. Indefiniteness has already been a live and partially successful theme: the Board and district courts have grappled with "device agent" / "device messaging agent" / "software components" as indefinite (courtlistener txed.220857.94.7, ¶¶ 42–53). A patent whose claims are indefinite under § 112 ¶ 2 is also hard to invalidate cleanly under § 103, because the claim scope needed to read on the art is contestable.
• Objective indicia. A $175M willful-infringement verdict on Claim 1 of U.S. 9,215,613 and Claim 83 of U.S. 8,589,541 (Verizon, E.D. Tex. 2:23-cv-00352) is a real, court-tested data point — but its nexus to the '758's claims (as opposed to the siblings' claims) is unproven and would have to be established claim-by-claim.
• The multi-tiered/quantized fraud-decision rule (Set D) and the device-credential→application-credential chaining (Set B) are the two limitations with the best chance of surviving. Any '758 § 103 case lives or dies on these.
• Priority-date defense. If Headwater can establish Jan. 28, 2009 support for all limitations, the art window narrows sharply and several of the references above (e.g., U.S. 2010/0017506, U.S. 2011/0019557, WO 2009/112044) drop out or move to § 102(e) only.

---

7. Bottom line

Confidence: Moderate as to the reference sets (these are art that the patent's own family adversaries have actually assembled for the same specification). Low as to claim-by-claim application, because I do not have the '758's literal claims — and I decline to construct it. The prior sections of this analysis established that absence; it remains the binding constraint.

Highest-value next steps to convert this into a defensible opinion:

1. Pull the granted '758 claim set from the USPTO PatentCenter full-text for app. 13/309,463.
2. Pull the '758's own "References Cited" from the front page / IFW (the section I could not retrieve).
3. Chart Lundblade / Jobst / Hardjono (per IPR2025-01572 exhibits EX1004, EX1005 et al.) and Rao / Fadell / Freund (per the '541 FWD) against each '758 independent claim.
4. Resolve the effective filing date claim-by-claim — this is the threshold § 103 issue.
5. Confirm whether the '613 Claim 1 / '541 Claim 83 willfulness verdict record contains any § 103 evidence specific to '758-type limitations (Set B/D), since those trial records are the only place the family's validity was actually tested on a full record.

---

Sources used in this section

• '758 full text (authoritative to me; truncated before claims and before the Citations section): https://patents.google.com/patent/[US8793758B2](/patent/US8793758B2)/en
• 13/309,463 priority/provisional listing: https://www.freepatentsonline.com/9491199.html
• Apple IPR2025-01572 petition on U.S. 10,064,055 (division of 13/309,463), grounds and analogous-art section: https://ptacts.uspto.gov/ptacts/public-informations/petitions/1558526/download-documents?artifactId=1bjmteI_j_EZ13vY05YLrReWNiNkxyvoMcLYmHgTLnNy7uWnmHcdeZU
• PTAB filing concerning the '055 and the family's IPR history: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1557287](/patent/1557287)/download-documents?artifactId=Mx-8SluhoP_cO_SOsZ8viX014So8o9pSS3_XipsLFvTA_h3B2lYHgnQ
• '541 Final Written Decision (Rao, Fadell, Freund; claims unpatentable): https://storage.courtlistener.com/recap/gov.uscourts.txed.224022/gov.uscourts.txed.224022.440.1.pdf
• Verizon P.R. 3-3 invalidity contentions cover pleading and obviousness appendices: https://www.docketalarm.com/cases/PTAB/IPR2024-00945/Google_LLC_v._Headwater_Research_LLC/docs/09-18-2024-Patent-Owner/Exhibit-2018-Exhibit_2018.pdf
• AT&T invalidity contentions reference list: https://ptacts.uspto.gov/ptacts/public-informations/petitions/1555958/download-documents?artifactId=NRCQgHfkY-yCWBx3HQ1DCZ2jbSBOWqF4n5xciwJna4IjtMIE8ayhiVo
• Samsung invalidity contentions reference list: https://ptacts.uspto.gov/ptacts/public-informations/petitions/1557850/download-documents?artifactId=Ntg4HsuFgAUB7YiDjPvrjnlg9WJHVkhyf3JUdUurdweFduZa6hVb0bc
• Samsung WAP-standard § 103 ground and motion-to-strike opposition: https://storage.courtlistener.com/recap/gov.uscourts.txed.220857/gov.uscourts.txed.220857.311.0.pdf
• POSITA definition / indefiniteness of "device agent" terms: https://storage.courtlistener.com/recap/gov.uscourts.txed.220857/gov.uscourts.txed.220857.94.7.pdf
• Jeffay invalidity expert report ('541/'613), priority-date contention: https://archive.org/download/gov.uscourts.txed.224022/gov.uscourts.txed.224022.200.15.pdf
• Tkacik secure-processing art (security engine separate from processor core): https://ptacts.uspto.gov/ptacts/public-informations/petitions/1556859/download-documents
• '055 claim 1 text and family abstract: https://litigation.rpxcorp.com/patent/US10064055B2

Explicitly not relied on / not auto-corrected: "8,793,758," "13/309,463," "10,064,055," "8,589,541," "9,215,613," "8,924,543," and "9,198,042" are kept as distinct literal identifiers and are not merged or substituted. The '055 claim 1 is labeled throughout as the child/division's claim, not the '758's. I have not invented any '758 claim number or claim language.