US 9578570

Let me search more specifically for CAFC activity involving this patent in 2026.

I have gathered what's available. Note up front: my searches did not surface any CAFC 2026 docket entries for this patent (the CAFC-specific queries returned no results), and I could not directly query the USPTO database itself. The bibliographic facts below are drawn from the authoritative full-text record (Google Patents) provided in the task plus corroborating search results (Justia, FreePatentsOnline, PatSnap/Unified Patents litigation data). Where a fact is search-derived rather than USPTO-verified, I flag it.

US Patent 9,578,570 B2 — Summary

• Title: Methods and systems for switching over a voice call
• Patent number: US9578570B2 (application no. US14/826,663)
• Inventors: Chi Fai Ho; Shin Cheung Simon Chiu
• Original assignee: TP Lab Inc. (California)
• Current assignee: Lab Technology LLC — recorded assignment from TP Lab, Inc., filed 2024-06-25, effective 2024-01-15 (per Google Patents legal-event record; assignment reel/frame 067830/0076)
• Filing date: 2015-08-14 (continuation of US14/457,643 → US13/598,133 → US12/835,402 → US11/615,636)
• Earliest priority date: 2006-12-22
• Issue date: 2017-02-21
• Status: Active (8th-year maintenance fee paid 2025-01-21, small entity; anticipated expiration 2026-12-22)
• Publication: Application pre-grant US20150358878A1 (2015-12-10)

Abstract (verbatim):

"Systems and methods for switching a voice call from an Instant Messaging (IM) based voice network to a second voice network. A telephone adapted to switch a voice call from an Instant Messaging (IM) based voice network to a second voice network."

Technical field: Telecommunication — switching a live voice call from an IM-based voice network (e.g., Skype, Yahoo! Messenger with Voice, Google Talk) to a cellular voice network when, e.g., a dual-mode phone leaves WiFi coverage.

Independent claims (plain-language overview)

The patent has 24 claims; four are independent (claims 1, 8, 16, 23), the remainder dependent.

• Claim 1 — Method. A method of switching an IM-based first voice call to a second voice call: an incoming call destined for a telephone number is received; a first call record identifying that call is created; an IM-based first voice call is set up over an IM voice network with an IM phone agent, and information about it is stored in the first record; the incoming call is connected to the IM call; at least part of the first call record is sent to a switchover agent; a second call record (tied to the phone number) is created containing that portion plus IM-call information; a second voice call is established, with its information added to the first record; the second call is associated with the IM call; a signal is sent to a phone agent telling it the second call is for switch-over purposes; and the second call is established.

• Claim 8 — Service gateway. A service gateway (processor + computer-readable medium with software) that receives an incoming voice call destined for a phone number and creates a first call record. The software directs the processor to: decide the circumstances under which the call should switch to a second voice network; establish an IM-based first voice call over an IM voice network; log IM-call information in the first record; send at least part of the first record to a switchover agent; create a second call record; establish a second voice call and log it in the first record; associate the second call with the IM call; and send a signal to the switchover agent indicating the second call is for switch-over purposes.

• Claim 16 — Computer-readable medium. A non-transient (sic; as printed), computer-readable storage medium whose program, when executed, performs essentially the same steps as claim 1 (receive incoming call → first call record → establish IM call → connect → send record portion to switchover agent → second call record → establish second call → associate → signal → establish).

• Claim 23 — Wireless phone. A wireless phone with a processor to switch an IM-based first voice call to a second voice call, comprising: (a) an IM phone agent to process the IM-based first call over an IM first voice network using an IM user identity; (b) an agent to process a second voice call from a second voice network, receive a call reference, and pass it to a switchover agent; and (c) a switchover agent that receives the call reference, creates a call record (call reference + IM-call information), retrieves the matching record, determines whether a voice call is associated with the IM-based first call, sends a response to the agent indicating the call is for switch-over purposes, stores IM-call information in the call record, and causes the agent to automatically establish the second voice call upon receiving that response.

Litigation / post-grant context (search-derived, not USPTO/CAFC-verified)

• Unified Patents and PatSnap litigation data show Lab Technology LLC has asserted US9578570B2 in 2024–2025 Texas and Florida district-court cases against defendants including Samsung, Qualcomm, U-blox, and others. Most reported dispositions were voluntary Rule 41 dismissals (several without prejudice, e.g., Qualcomm in 6:24-cv-00335; the U-blox matter in 4:25-cv-00610 was dismissed with prejudice). No merits invalidity ruling has been identified; the patent remains in force.

Uncertainty

• I found no CAFC (Federal Circuit) docket entries for 2026 involving this patent — my CAFC-specific searches returned empty. I cannot rule out that such a docket exists but is simply not indexed by the search tool; treat "no CAFC 2026 activity" as not found, not confirmed absent.
• USPTO database facts (assignment, maintenance fee, status) are corroborated by Google Patents legal-event data and search results, not by a direct USPTO query.

I'll research litigation involving US Patent 9,578,570 using multiple sources.

I found a substantial litigation record. Let me verify each case number, party, and status by checking multiple databases.

I have enough verified information to compile the litigation record. Note that I reached the search-step limit before I could verify every docket individually, so I flag confidence levels case by case below.

Litigation involving US Patent 9,578,570 (as of April 26, 2026)

All suits are brought by the same plaintiff: Lab Technology LLC (New Mexico; patent-holder since the 2024 assignment from TP Lab, Inc.), represented by Isaac Rabicoff / Rabicoff Law LLC. Every identified case is a patent-infringement suit under 35 U.S.C. § 271 asserting US9578570B2 ("Methods and systems for switching over a voice call"), typically alongside other Lab Technology portfolio patents.

June 2024 W.D. Tex. (Waco) batch — filed ~June 21, 2024, all quickly dismissed

Bloomberg Law (June 24, 2024) specifically identified US Patent No. 9,578,570 as the patent-in-suit against Nvidia, Intel, and Qualcomm.

May–July 2024 E.D. Tex. batch — filed ~May–June 2024

2025–2026 cases

Notable exclusions / caveats

• Google LLC (6:24-cv-00334, W.D. Tex., filed 2024-06-21) was sued by Lab Technology on the same day as the Intel/Qualcomm/Nvidia batch, but I found no confirmation that US9578570 is asserted in that case, and Google Patents does not link it to this patent — it likely involves a different portfolio patent (e.g., display-refresh or E911). Excluded pending confirmation.
• Lumen Technologies (7:25-cv-00172, W.D. Tex., filed 2025-04-17, voluntarily dismissed 2025-06-24/25) and a Yeastar Digital Technology case (2:26-cv-00709, E.D. Tex.) are other Lab Technology suits, but I could not confirm US9578570 is the patent-in-suit; they may involve sibling portfolio patents.
• The RPX article on the June 2024 E.D. Tex. wave notes Lab Technology also sued Amazon (2:24-cv-00409), AT&T (2:24-cv-00412), Bose (2:24-cv-00413), and Stryker (2:24-cv-00416) — but those cases are not on the Google Patents litigation list for US9578570 and likely assert other portfolio patents, so they are excluded.
• No CAFC (Federal Circuit) appeal, no IPR/PTAB proceeding, and no merits ruling (validity, claim construction, or infringement) on US9578570 was identified as of today's date. The patent remains in force (8th-year maintenance fee paid 2025-01-21; anticipated expiration 2026-12-22).

Bottom line

Every confirmed case shares the same pattern: Lab Technology LLC suing in Texas or Florida district courts (2024–2025), with nearly all resolving within weeks by plaintiff's voluntary Rule 41(a)(1)(A)(i) dismissal — mostly without prejudice (Samsung, Intel, Qualcomm) and at least one with prejudice (u-blox) — before any answer, claim construction, or merits determination. Sources: Stanford NPE Litigation Database, RPX Insight/Empower, Unified Patents Portal/PATROLL, PatSnap PACER-docket analyses, Justia/CourtListener dockets, Bloomberg Law, and the Google Patents litigation feed.

The Unified Patents angle is important — that contest finding prior art on the '570 patent often precedes a Unified-filed IPR. Let me check whether any such IPR was actually filed.

Let me verify whether Unified's prior-art finding matured into any PTAB filing and check USPTO/PTAB portals directly for any proceeding tied to this patent number.

Proceedings overview

There are zero AIA trial proceedings (IPR/PGR/CBM) on US Patent 9,578,570 — the USPTO Open Data Portal returns no proceedings on file as of its most recent ingest, and my independent searches of PTAB dockets, Unified Patents' portal, CourtListener, and USPTO sources surfaced no petition, no institution decision, and no Final Written Decision for this patent number. The status breakdown is therefore: 0 active · 0 claims invalidated · 0 claims sustained · 0 settled · 0 institution denied. Bottom line for a defendant: the patent is completely UNTESTED at the PTAB — every one of the 24 claims (independent claims 1, 8, 16, 23 and all dependents) remains legally presumed valid, and no estoppel of any kind binds any would-be challenger; but the absence of PTAB activity is itself notable given Lab Technology LLC's aggressive multi-district assertion campaign and Unified Patents' public prior-art hunt on this exact patent.

Because no proceedings exist, the per-proceeding sections below do not apply. What follows is the strategic picture that the absence creates.

Strategic summary

Claims status: CANCELED / SUSTAINED / UNTESTED

The four independent claims cover the method (claim 1), the service gateway (claim 8), the computer-readable medium (claim 16), and the wireless phone (claim 23). None has been challenged anywhere. Every confirmed 2024–2025 district-court case — Samsung (2:24-cv-00324), Intel (6:24-cv-00332), Qualcomm (6:24-cv-00335), NVIDIA (6:24-cv-00337), Ericsson (2:24-cv-00414), Anritsu (2:24-cv-00411), Quectel (2:24-cv-00415), u-blox (4:25-cv-00610), Mitel (1:25-cv-22712) — ended by Lab Technology's Rule 41(a)(1)(A)(i) voluntary dismissal (mostly without prejudice, u-blox with prejudice), all before a merits adjudication. There is no claim-construction order, no § 101/§ 102/§ 103 holding, and no finding of infringement anywhere in the record.

Estoppel landscape (§ 315(e))

Because no IPR has ever been instituted, § 315(e) estoppel is a non-issue today. No petitioner exists, so no defendant — and no defendant's privy — is barred from raising any ground. Every prior-art theory is fully available to any accused infringer in district court (subject only to the ordinary § 282 presumption of validity and the prosecution-history art already of record). Two cautions:

1. If Unified Patents (or another third party) files an IPR and you are a Unified member or otherwise could be deemed a real party in interest/privy, institution would trigger § 315(e)(2) estoppel against you for any ground raised or reasonably available in that IPR. Members should track whether Unified's challenge (see below) matures into a petition naming them as an RPI.
2. § 315(b) one-year bar: any defendant served with a complaint more than one year ago can no longer file its own IPR on this patent; those defendants must rely on district-court invalidity or a third-party petition.

Pattern signals

• Defensive-aggregator interest is confirmed, but no filing has followed. Unified Patents ran a public PATROLL crowdsourcing contest on US 9,578,570 from 2024-06-06 to 2024-08-30, expressly "seeking prior art on at least claim 1," citing the assertions against Samsung, Anritsu, Quectel, and Ericsson (https://patroll.unifiedpatents.com/contests/PPTa62DzDJqMcxQjd). On 2024-12-17 Unified announced "[p]rior art has been found on U.S. Patent 9,578,570" (https://www.unifiedpatents.com/insights/2024/12/17/lab-technology-location-tracking-patent-prior-art-found). Unified's standard playbook after a winning contest submission is a PTAB challenge or ex parte reexamination — as of 2026-09-08, no such proceeding appears on any docket I could access, but this is the single most important item to monitor.
• Same plaintiff, same pattern, no merits fight: Lab Technology LLC (Rabicoff Law LLC) has filed and then voluntarily dismissed suits in W.D. Tex., E.D. Tex., N.D. Tex., and S.D. Fla. — a classic "demand-and-dismiss" campaign. No defendant has yet forced a merits test of the '570 patent, which may explain why no IPR has been needed by the patent owner to defend (and why no defendant has yet paid for one).
• Patent expiration is imminent — anticipated 2026-12-22, roughly 3.5 months from today. After expiration, forward-looking relief disappears and only past damages remain, which typically collapses settlement value and may further deter a costly IPR investment. Note also that the PTAB retains jurisdiction to consider IPR on an expired patent (estoppel and remedy calculus change, but the proceeding remains available).
• No Federal Circuit activity exists because there is no PTAB decision to appeal and no district-court judgment.

Recommended next steps

1. If you are a defendant facing assertion today, do not assume PTAB relief is coming — and act before the expiry date.

• There is no FWD to cite and no canceled claim to leverage. Any demand letter citing claims 1–24 of the '570 patent is legally unimpaired on its face — the invalidity case has simply never been made.
• The patent expires 2026-12-22. If you were served recently, your § 315(b) one-year IPR window is open but short and overlaps the expiry; an IPR filed now would have limited forward-looking value but could still defeat past-damages exposure. Weigh cost against the rapidly shrinking damages window.
• The cleanest near-term path for most defendants is district-court invalidity with no estoppel constraints — all § 102/§ 103 grounds (including the art Unified's contest identified) are fully available in litigation.

2. Monitor Unified Patents' portal (https://portal.unifiedpatents.com/patents/patent/9578570) for a petition or reexamination request derived from its December 2024 winning prior-art submission. If Unified files, check whether your company is an RPI/privy for estoppel purposes before the petition is filed, and coordinate invalidity strategy accordingly.

3. If you are a Unified member, confirm with Unified whether the '570 patent is an active target and whether any petition is in preparation — membership status can determine both your access to the prior art and your estoppel exposure if institution occurs.

4. Watch the district-court refile risk. Nearly all dismissals were without prejudice (only u-blox obtained a with-prejudice dismissal). Lab Technology can and may re-assert the '570 patent against the same or new defendants before expiry. A new complaint against you starts a fresh § 315(b) clock and gives you a clean one-year IPR window if you act promptly.

Bottom line: this is a patent that has never been stress-tested. All 24 claims stand. The defendant who first forces a merits test — in district court or via a promptly filed IPR before the 2026-12-22 expiry — writes the playbook for everyone else.

Sources: USPTO Open Data Portal (canonical structured data — no proceedings on file); Unified Patents PATROLL contest page and 2024-12-17 insight announcement; PatSnap PACER-docket litigation analyses (cases 6:24-cv-00335, 2:24-cv-00324, 4:25-cv-00610); Bloomberg Law (2024-06-24); Google Patents litigation feed. No proceeding numbers are cited because none exist — I did not and will not invent any.

I'll research the ownership chain for US 9,578,570. Let me pull from the assignment record, entity history, and asserter directories.

Excellent — the Mondaq/RPX coverage reveals a critical timing detail. Let me dig into the recorded assignment specifics and the entity history.

I've hit the search-step limit. Here's the ownership-chain reconstruction based on what I could verify. Two process notes up front: I could not directly query the USPTO Assignment Center (no direct database access this session), so reel/frame and correspondent data below are drawn from the Google Patents legal-events record supplied in the task plus corroborating search results, and I flag every search-derived item. I also did not fabricate any assignment — where a link is missing from the record, I say so.

Inventors

Chi Fai Ho (Palo Alto, CA) — founder, CEO, and sole principal of the original assignee, TP Lab, Inc., at the time of filing. He was therefore effectively self-employed at his own invention-licensing lab. Background (from TP Lab's own bio): Senior Director of Business Strategy at Nokia; VP of Engineering at Amber Networks (acquired by Nokia, 2001); CTO of the Ethernet Division of Vitesse Semiconductor; VP of Engineering at Tidal Networks; Ph.D. Mathematics, Caltech. He also founded IPLearn LLC in 1997 — described by RPX as "an entity that has asserted online education patents against several dozen defendants," and in 2004 founded TP Lab. (Source: tplab.com/aboutus.htm; Mondaq/RPX, 2024-05-31.)

Shin Cheung Simon Chiu (Palo Alto, CA) — prolific TP Lab co-inventor; named on numerous TP Lab filings (e.g., US 8,219,061; US 9,282,372; the universal-phone-number family). He appears as a repeat collaborator with Ho across the TP Lab portfolio. No public employment history beyond TP Lab located.

Pattern note: No "inventors depart within 12 months" signal here — the opposite. Both inventors remained at TP Lab, and TP Lab held and re-asserted this patent family for ~18 years (2006 → 2024) before transferring it. The unusual feature is not inventor flight but that the founder-inventor is also the principal of the original assignee, and that assignee ran an explicit invention-and-license model rather than a product business (see next section).

Original assignee

TP Lab, Inc. (Palo Alto, CA) — named on the issued patent. Founded 2004 by Chi Fai Ho. TP Lab's stated business model is monetization, not manufacturing: per its own "About" page, Ho "founded TP Lab with a focus on invention with licensing to monetize the Lab's ideas." Main areas: "Internet-telecommunication convergence, Web-oriented personal communication, and consumer digital media."

• Did it ship a product embodying the claims? No evidence found. The '570 claims a service gateway plus a dual-mode wireless phone; TP Lab is not a carrier, handset OEM, or VoIP operator. No commercial product was identified in any source.
• Primary line of business: patent development and licensing (~80–100+ applications filed; USPTO PatentsView shows 84 grants 2015–2024, concentrated in H04L).
• Current status: Appears to be an ongoing (non-bankrupt) concern — it is a serial transferor, having moved patents to Dolby; Knapp Investment Company Limited (Dec 2015); InnoBrilliance, LLC (Jul 2017); Fiver LLC (Sep 2017); the emergency-caller-location family to Dominion's Local Intelligence (Aug 2017), which returned those assets to TP Lab (Nov 2019); and ZAMA Innovations LLC (Aug 2021, after a Nov 2019 transfer to Allied Security Trust). (Sources: Mondaq/RPX 2024; PlainPatent/PatentsView.)

Assignment timeline

Important limitation: the authoritative record I can reach shows one post-issuance recorded assignment for US 9,578,570 — the TP Lab → Lab Technology transfer. The pre-issuance inventor → TP Lab assignment that must underlie the "Original Assignee: TP Lab Inc." designation is not present in any source I could retrieve, and I will not invent a reel/frame for it.

• 2006-12-22 (approx.) / recorded not retrieved — Reel unknown

  • Conveyance: Assignment (implied by patent face naming TP Lab as original assignee)
  • Assignor: Chi Fai Ho; Shin Cheung Simon Chiu
  • Assignee: TP Lab, Inc.
  • Correspondent: not retrieved
  • Context: Original prosecution assignment to the founder's own lab. Flag: reel/frame not verified — do not cite.

• 2024-01-15 (executed) / 2024-06-25 (recorded) — Reel 067830/0076

  • Conveyance: Assignment
  • Assignor: TP Lab, Inc.
  • Assignee: Lab Technology LLC (New Mexico)
  • Correspondent: Not disclosed in the sources I could access. The only attorney-of-record fact I can verify is litigation counsel: Isaac Rabicoff, Rabicoff Law LLC, 4311 N Ravenswood Ave, Suite 315, Chicago, IL 60613, per Lab Technology's Rule 7.1 disclosure in Lab Technology LLC v. Zoom Communications, Inc., No. 1:24-cv-01711 (D. Colo., filed 2024-06-20). I cannot confirm Rabicoff was the assignment-recording correspondent — flag as unverified.
  • Context: Transfer-to-asserter. Executed seven days after Lab Technology LLC was formed (2024-01-08) and ~3.5 months before the first suit naming this patent (Samsung, 2:24-cv-00324, filed May 2024).
  • Timing tell: As of 2024-05-31, RPX/Mondaq reported that "Currently available USPTO records do not reflect the assignment of any of these three patents away from TP Lab … but Lab Technology pleads ownership." The recording (2024-06-25) therefore post-dates the first infringement suits — the classic "file first, paper the chain later" sequence.

If the pre-issuance link is genuinely unrecorded/not retrievable, the only verifiable conveyance is reel 067830/0076.

Cross-reference note: no separate Local Intelligence → Lab Technology record appears in the '570 chain (the 2017/2019 Local Intelligence round-trip involved the emergency-location family, not this patent).

Timeline diagram

timeline
    title Ownership of US 9578570
    2004 : TP Lab founded by Chi Fai Ho
    2006 : Filed and assigned to TP Lab Inc
    2017 : Issued as US 7778630 B2
    2024 : Lab Technology LLC formed in New Mexico
         : TP Lab assigns patent to Lab Technology
         : Samsung suit filed 2 24 cv 00324
         : Assignment recorded at reel 067830 0076
    2024 : Intel Qualcomm Nvidia suits filed
    2025 : u-blox suit dismissed with prejudice

(The "2017" issue line reflects US 9,578,570's own grant date; parent US 7,778,630 issued 2010.)

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT. The patent moved from TP Lab (reel 067830/0076, executed 2024-01-15) to Lab Technology LLC, a New Mexico LLC formed 2024-01-08 — seven days before the assignment executed. Concrete corroboration beyond the name: (a) Lab Technology's Rule 7.1 disclosure states it "does not have a parent corporation and … no publicly held corporation owning ten percent or more of its stock" (D. Colo. 1:24-cv-01711); (b) no product, service, or website tied to the entity was found (the similarly-named labtechnologygroup.com is unverified and may be unrelated); (c) RPX describes it as "of uncertain management or personnel." The name suffix is "Technology" rather than "IP/Holdings," so the name alone is not the finding — the finding is the single-purpose, just-formed LLC acquiring a patent for assertion.

2. Known asserter in the chain — PRESENT. Lab Technology LLC is classified "Non Practicing Entity" by the Stanford NPE Litigation Database (npe.law.stanford.edu/party/lab-technology-llc) and listed as an NPE by ex parte's asserter directory (43 cases; 29 patents; E.D. Tex., W.D. Tex., S.D. Fla.). It is a high-frequency plaintiff surfaced by Unified Patents and RPX, the two directories the task names. Separately, inventor Ho's earlier entity IPLearn LLC asserted education patents against dozens of defendants — a monetization pedigree, though IPLearn is not in this patent's chain.

3. Repeat correspondent across the chain — UNCLEAR. Only one verifiable link exists in this chain (reel 067830/0076), so recurrence within the chain cannot be established — a single appearance is not the signal. The recording correspondent is not disclosed in the sources I reached. Isaac Rabicoff / Rabicoff Law LLC is confirmed as Lab Technology's litigation counsel (Zoom Communications Rule 7.1, June 2024; Lumen complaint, Apr. 2025), and Mondaq separately links "Rabicoff Law LLC" to "plaintiffs with ties to … monetization figure Jeffrey M. Gross" — but tying Rabicoff to the recorded assignment would be inference. Flagged, not asserted.

4. Cascading transfers — NOT PRESENT (for this patent). A single hop: TP Lab → Lab Technology. TP Lab has cascaded other families (Local Intelligence out and back; AST → ZAMA), but this patent did not traverse a chain of LLCs.

5. Pre-litigation transfer — PRESENT. Assignment executed 2024-01-15; first suit naming the '570 (Samsung, 2:24-cv-00324) filed May 2024 — ~3.5 months, inside the 6-month window. Recorded 2024-06-25, i.e., after the first complaints, per RPX's May 2024 observation that USPTO records then showed no assignment. Arranged before assertion; papered after.

6. Bankruptcy fire-sale — NOT PRESENT. No bankruptcy of TP Lab found; TP Lab is an active licensor. (Contrast the classic Kodak/Nortel pattern.)

7. Privateering — UNCLEAR / largely NOT PRESENT. The traditional definition (operating company → NPE asserting against its competitors) does not fit cleanly, because TP Lab is not a product competitor of Samsung/Intel/Qualcomm — it is an invention-and-licensing house. This is closer to pure monetization transfer than classic privateering, though the practical effect (assertion against operating companies) is the same.

8. Defensive aggregator — NOT PRESENT. The chain terminates at an asserter, not at RPX/AST/LOT/Unified/OIN. Note the inverse for a sibling: Unified Patents filed an ex parte reexamination against US 8,498,388 (another TP Lab→Lab Technology patent) on 2024-11-27 — evidence of defensive aggregation aimed at this portfolio, but it does not neutralize the '570.

Verdict

NPE — high confidence.

The chain shows two strong signals: (i) a shell-entity transfer — patent moved per reel 067830/0076 (executed 2024-01-15) from TP Lab to Lab Technology LLC, a New Mexico LLC formed 2024-01-08, seven days before execution, with a Rule 7.1 disclosure confirming no parent and no 10% public holder; and (ii) a known asserter — Lab Technology is listed as a Non Practicing Entity by the Stanford NPE Litigation Database and by ex parte, and is a repeat filer in E.D. Tex./W.D. Tex./S.D. Fla. These are reinforced by a pre-litigation transfer (assignment executed ~3.5 months before the Samsung suit, recorded ~1 month after the first suits per RPX's May 2024 snapshot). The original assignee, TP Lab, is itself an invention-licensing entity under the same founder-inventor, so no link in this chain ever shipped the claimed subject matter.

Verify at: https://assignmentcenter.uspto.gov/ (search patent number 9,578,570; look for reel 067830/0076 and any earlier inventor→TP Lab record) and https://assignment.uspto.gov/patent/index.html.

Caveats: (1) I could not query the Assignment Center directly; the reel/frame and dates come from the Google Patents legal-events record plus RPX/Mondaq corroboration. (2) The inventor→TP Lab assignment reel/frame was not retrieved and is not cited. (3) The recording correspondent on reel 067830/0076 is unconfirmed; Rabicoff is verified only as litigation counsel. (4) labtechnologygroup.com is not confirmed as this entity's site. (5) Addition vs. the prior Litigation section: RPX/Mondaq confirms the '570 was asserted against Samsung only among the May-2024 E.D. Tex. wave — Verizon (2:24-cv-00323) received different portfolio patents, so no contradiction, but the Verizon case is now on the record and the '570 was not in it.

I've reached the search limit, so I'll now compile the prior-art analysis from the verified record I've assembled. First, two process notes:

• USPTO database query limitation: I could not directly query USPTO databases (PatFT/AppFT/Patent Center) in this session. The citation list below is taken from the authoritative full text of US 9578570 B2 (Google Patents record provided in the task), which reproduces the examiner-considered "Patent Citations (8)." Where reference content is search-verified, I say so; where I could not verify content, I flag it explicitly rather than invent details.
• Number literalism: I treated every identifier literally. Notably, a search for KR20050001924A surfaced a different Korean family (an aluminum-recycling patent, KR100648386B1, application KR20050001924A filed 2005-01-08). I have not conflated that document with the Bizmodeline citation on the '570 record, which Google Patents lists as priority 2003-06-28 / published 2005-01-07, "System and Method for Transforming Channel."

Prior-art framework applicable to US 9578570 B2

The '570 patent is a continuation chain whose earliest effective filing date is 2006-12-22 (parent US 11/615,636). Because the chain began before 2013-03-16, pre-AIA § 102 governs. Under that regime:

• § 102(a) – reference must have been known/used/patented/described in a printed publication before the invention date (presumed 2006-12-22).
• § 102(b) – statutory bar for patents/publications dated more than one year before the earliest US filing date, i.e., before 2005-12-22.
• § 102(e) – a US patent or US published application "by another" filed before 2006-12-22 is prior art as of its filing date, even if published later.

Three of the eight face-of-the-patent citations published in 2007–2008 (after the priority date) are therefore only available under § 102(e), not § 102(a)/(b).

The eight citations on the face of US 9578570 B2

1. KR20050001924A — "System and Method for Transforming Channel" (Bizmodeline Co., Ltd.)

• Full citation: KR 2005-0001924 A; applicant 주식회사 비즈모델라인 (Bizmodeline Co., Ltd.); priority/filing 2003-06-28; published 2005-01-07.
• § 102 basis: § 102(a) and § 102(b) (published 2005-01-07, well before both 2005-12-22 and 2006-12-22).
• Brief description: Per the record, a "channel-transforming" system/method. I could not verify the technical content of this Korean-language document in this session (searches returned an unrelated Korean family sharing the numeric string — see note above).
• Potential § 102 anticipation: Cannot responsibly be assessed on the verified record. Screening-level: channel-transformation concepts circa 2003–2004 for Bizmodeline (a mobile-commerce/communications services firm) rarely map to the IM-to-cellular switchover agent architecture of claims 1, 8, 16, or 23, but without the text this is speculation. Flag: content unverified.

2. US 2006/0023695 A1 — "Predictive communication routing scheme using instant messaging preferences and/or settings for routing packet traffic for voice, video and other multimedia communications" (Yahoo! Inc.; Yarlagadda et al.)

• Full citation: US 2006/0023695 A1; filed 2004-02-13 (provisional priority Feb. 2004); published 2006-02-02.
• § 102 basis: § 102(a), § 102(b) (published before 2005-12-22), and § 102(e).
• Brief description (search-verified from the published PDF): Routes PSTN calls addressed to a telephone number to IM clients: a call for the number is received from a PSTN device; routing preferences for the number indicate whether the call should go to one or more IM usernames associated with the number; presence of those IM usernames is checked; and the online IM usernames are rung for the call. This is the classic "call a phone number, ring the associated IM identity" paradigm.
• Potential § 102 anticipation: Discloses the IM-routing leg of claim 1 — i.e., "receiving an incoming voice call destined for a telephone number," "establishing over an IM voice network an IM-based first voice call with an IM phone agent," and the concept that the IM user identity is distinct from the telephone number (relevant to dependent claim 7, and claim 6's distinct-identity concept). It does not disclose: the first/second call-record structure with a call reference, a switchover agent, establishment of a second (cellular) voice call, the signal to a phone agent "indicating that the second voice call is for switch over purpose," or the switchover itself. As a single reference it cannot fully anticipate claims 1, 8, 16, or 23, but it is strong § 103 base art for the IM-establishment steps and reads directly on claims 6–7 in combination.

3. US 2006/0286984 A1 — "Multi-mode handset services" (Cingular Wireless LLC; Bonner) — cited by the examiner

• Full citation: US 2006/0286984 A1; filed 2005-06-16; published 2006-12-21 (one day before the '570 priority date); published-family members EP 1884087, WO 2006/138019, CA 2612372.
• § 102 basis: § 102(a) (printed publication 2006-12-21, before 2006-12-22) and § 102(e) (filed 2005-06-16). Not § 102(b) (published after 2005-12-22).
• Brief description (search-verified): Dual-mode GSM/WiFi voice service delivery with transparent handover between WiFi/VoIP and GSM, including mid-call redirection between networks. The subscriber's mobile number is the main number — mobile-terminated calls are routed to the HPLMN and forwarded over broadband to a VoIP/SIP client at an indoor network; as a user leaves the building the redirection component redirects the active call from the indoor VoIP network back to the cellular network, transparently to the user. It manages "voice call handovers between packet-switched networks and circuit-switched networks" and establishes a circuit-switched bearer using SIP signaling over WiFi. This is conceptually the closest cited art to the '570's problem statement (active-call switchover from a VoIP/Internet leg to a cellular leg for a dual-mode phone).
• Potential § 102 anticipation: Reads on many elements of claims 1, 8, 16, and 23 — incoming call destined for a phone number; a network-side element routing the call over a packet/Internet leg; mid-call handover to a cellular leg; terminal-side handover control. Element gaps remain that defeat strict § 102 anticipation on the verified content: (i) the packet leg is SIP/VoWLAN/UMA, not an "IM voice network" with an "IM phone agent" using an "IM user identity" as the claims require; (ii) no disclosure of the dual call-record structure (gateway record + phone record keyed by a call reference); and (iii) no teaching of establishing the second (cellular) call specifically for switchover purposes with a signal to the phone agent to that effect. Because Cingular stresses operation "transparently to the subscriber," it also teaches against the user-acceptance embodiment of claims 5/20. *Ranking: closest conceptual art and the only -cited reference; primary § 103 anchor; realistic anticipation case is weak but must be examined element-by-element on the full specification.

4. US 2007/0094374 A1 — "Enterprise-managed wireless communication" (Divitas Networks, Inc.; Karia, Mittal, Rao, et al.)

• Full citation: US 2007/0094374 A1; filed 2005-10-03; published 2007-04-26; PCT counterpart WO 2007/041663.
• § 102 basis: § 102(e) only (filed before 2006-12-22; published after). Not § 102(a)/(b).
• Brief description (search-verified): A client device operable in multiple networks (WiFi + cellular) with an enterprise-managed server that stores the client's connectivity information and routes incoming media addressed to the client using that connectivity information, plus client-side modules: a mobility manager monitoring connectivity to the plurality of networks and a call control module that "determines whether to trigger a handoff … from a first radio station to a second radio station based on the connectivity." The spec details fast media handoff between WiFi and cellular, rendezvous calling, and an explicit call-roaming scenario of an active call moving from WiFi to cellular (FIGS. 12–13) and back (FIGS. 14A–B).
• Potential § 102 anticipation: The client-side architecture (mobility manager + call control module deciding handoff based on connectivity) is structurally reminiscent of the '570's switchover agent + IM/cellular phone agents of claim 23; the server-side routing of incoming media mirrors the service gateway of claim 8. Gaps versus the verified content: the networks are WiFi/enterprise-VoIP and cellular — again not an IM voice network/IM phone agent; no call-record/call-reference pairing between gateway and phone; no "signal indicating the second call is for switch over purpose." Ranking: strong secondary § 103 anchor for claims 1, 8, and 23; strict § 102 anticipation doubtful.

5. US 2007/0167167 A1 — "Network-based system for rerouting phone calls from phone networks to VoIP clients for roamers and subscribers who do not answer" (Roamware, Inc.; Jiang)

• Full citation: US 2007/0167167 A1; filed 2003-02-18; published 2007-07-19.
• § 102 basis: § 102(e) only (filed 2003, long before 2006-12-22; published after).
• Brief description (search-verified): A gateway 120 (expressly including "a VoIP gateway, a Skype gateway, a Vonage-like gateway, a SIP/IMS gateway, or an IM-VoIP gateway") couples the cellular network (HPMN/VPMN) to a subscriber's VoIP/IM client 122 ("Skype, Yahoo!, Google, GizmoProject, MSN, Vonage, or any SIP, VoIP or messaging client"). Calls intended for the subscriber's ordinary mobile number are automatically routed over IP to the VoIP client when the client is online, conditioned on the subscriber roaming or not answering. FIGS. 6–7 show switching traffic from a second communication network to a first communication network and vice versa — i.e., two-way network switching.
• Potential § 102 anticipation: Of the cited art, this comes closest on the "gateway reroutes a call placed to a cellular phone number to an IM-based voice client" element — and unlike the Yahoo reference, it also teaches two-way switching of traffic between the networks. It still does not disclose, on the verified content: the two call records (first record at the gateway, second record at the phone) keyed by a call reference; a switchover agent at the phone that decides whether to switch based on the two calls; or the specific act of establishing the second voice call with a signal to the phone agent that it is for switchover purposes. Ranking: strong § 103 base art for claims 1, 8, 16; strict § 102 anticipation doubtful for the same reasons as references 3–4.

6. US 2007/0280445 A1 — "Method for Interacting Via an Internet Accessible Address-Book Using a Visual Interface Phone Device" (Roy Shkedi)

• Full citation: US 2007/0280445 A1; filed 2006-06-05; published 2007-12-06.
• § 102 basis: § 102(e) only (filed before 2006-12-22; published after).
• Brief description: Per the title/record, it concerns interacting through an Internet-accessible address book using a visual-interface phone device. Content not search-verified in this session.
• Potential § 102 anticipation: On the verified record this appears tangential to IM-to-cellular call switchover; it was likely cited for background on phone-device/IM interaction. I will not assert anticipation of any claim without reviewing the text. Flag: low relevance, content unverified.

7. US 2008/0133580 A1 — "Method and system for providing automated real-time contact information" (James Andrew Wanless)

• Full citation: US 2008/0133580 A1; filed 2006-11-30; published 2008-06-05.
• § 102 basis: § 102(e) only (filed 22 days before the '570 priority date; published after).
• Brief description: Per the title/record, automated real-time contact-information provision. Content not search-verified in this session.
• Potential § 102 anticipation: Appears tangential to the call-switchover claims. Filed only 22 days before the '570 priority date, it is unlikely to have been considered substantive to the switchover concept. Flag: low relevance, content unverified; do not rely on it without reading the full disclosure.

8. US 7,778,630 B2 — "System for switching over a voice call" (TP Lab, Inc.; Ho & Chiu)

• Full citation: US 7,778,630 B2; filed 2006-12-22; issued 2010-08-17 — the grandparent application in the same family as the '570 patent (same two inventors, same assignee).
• § 102 basis: Not prior art at all. It is not "by another" (same inventive entity), is commonly owned, and is in fact the priority document from which the '570 claims priority. It cannot be used under any subsection of § 102 against the '570 claims. It appears on the citation list only because the examiner/continuation practice carries the family's own ancestor forward.
• Potential § 102 anticipation: None. (It is substantive evidence that the claims are entitled to the 2006-12-22 priority date, which is what makes the pre-AIA framework above applicable.)

Bottom line and ranking

Analyst's assessment: No single face-of-the-patent reference appears, on the verified content, to fully anticipate any of the four independent claims (1, 8, 16, 23) — the recurring gaps are (i) the "IM voice network / IM phone agent / IM user identity" limitation, (ii) the dual call-record structure keyed by a call reference created both at the gateway and at the phone, and (iii) the distinctive act of establishing the second (cellular) call with a signal to the phone agent indicating it is for switch-over purposes, followed by automatic establishment. The Cingular, Roamware, and Divitas references together make a formidable § 103 obviousness combination against claims 1, 8, 16, and 23, and the Yahoo reference supplies the IM-specific routing leg. The strongest prima facie § 102 candidates to press are Cingular (claim 1/16 method elements) and Divitas (claim 23 apparatus elements), each of which would need a meticulous element-by-element comparison against the full specifications — which I could not complete within this session's search budget.

Caveats restated: (1) I could not query USPTO databases directly; citation data is from the authoritative Google Patents full-text record. (2) The content of KR 2005-0001924 A, US 2007/0280445 A1, and US 2008/0133580 A1 is unverified — do not rely on my characterization of those three beyond their bibliographic entries. (3) Anticipation opinions above are screening-level, based primarily on abstracts/summaries; a formal § 102 analysis requires reading each full disclosure against each claim limitation.

The searches confirmed the technical content of the three most important references (Cingular '984, Divitas '374, Roamware '167) directly from the published documents. I now have enough verified detail — combined with the Prior Art section's analysis of Yahoo '695 — to construct the § 103 analysis.

Obviousness Analysis — US Patent 9,578,570 B2 Under 35 U.S.C. § 103

1. Legal framework applied

• Pre-AIA § 103 governs (earliest effective filing date 2006-12-22), but the analysis is conducted under the Graham v. John Deere factors as elaborated by KSR Int'l Co. v. Teleflex, Inc., 550 U.S. 398 (2007) — decided the year after the priority date and the controlling law for any present-day validity challenge:
  • (1) scope and content of the prior art;
  • (2) differences between the prior art and the claimed invention;
  • (3) level of ordinary skill in the art;
  • (4) secondary indicia.
• Under KSR, a § 103 case does not require an express teaching-suggestion-motivation in the references. Obviousness may be shown by design need, market pressure, known interchangeability of elements, and the "ordinary creativity" of the skilled artisan — points that matter here because the asserted combination (generic VoIP-to-cellular handover + IM voice routing) is precisely the convergence the marketplace was driving in 2005–2006.

2. Level of ordinary skill in the art

A PHOSITA circa December 2006: a telecommunications engineer (B.S./M.S. in EE/CS or equivalent experience, 3–5+ years) with working knowledge of cellular telephony (GSM/3G/SS7/ISUP), VoIP and SIP signaling, WiFi/unlicensed mobile access (UMA/GAN), and instant-messaging voice services (Skype, Yahoo! Messenger with Voice, Google Talk, AIM) — i.e., precisely the technologies each reference discloses. The '570 patent itself is addressed to this person; its background describes dual-mode phone usage of IM voice over WiFi hotspots as a known, pre-existing phenomenon at filing.

3. The claimed invention in one paragraph (claim scope recap)

Independent claims 1 (method), 8 (service gateway), 16 (computer-readable medium), 23 (wireless phone) describe: an incoming call destined for a telephone number → a first call record at a service gateway → an IM-based first voice call set up over an IM voice network with an IM phone agent (using an IM user identity distinct from the phone number — claims 6–7) → the gateway connects the incoming call to the IM leg and sends at least part of the first record to a switchover agent on the phone → a second call record on the phone → a second (cellular) voice call established and logged in the first record → association of the two legs → a signal to a phone agent that the second call is "for switch over purpose" → automatic establishment of the second call (claim 23) → later, connection/disconnection of voice paths and a switchover command carrying the call reference to the phone (spec ¶¶ [0046]–[0052]).

The recurring claim features most likely to carry validity are therefore: (i) the IM voice network / IM phone agent / IM user identity leg, (ii) the dual call-record structure keyed by a call reference (one at gateway, one at phone), and (iii) the "signal … for switch over purpose" + automatic-establishment coordination between the network and the phone's agents.

4. The prior-art combination set

Two further references are combinable but weaker: KR 2005-0001924 A and US 2007/0280445 A1 / US 2008/0133580 A1 are content-unverified or tangential on the record and add nothing affirmative to a § 103 case; they are not needed. US 7,778,630 B2 is the family ancestor and is not prior art (same inventive entity / not "by another"), though its disclosure confirms the claims' own architecture is a continuation of what TP Lab filed in 2006.

5. Primary combination — Cingular '984 + Yahoo '695 (+ Roamware '167 as express interchangeability teaching)

This is the strongest case and tracks the '570 patent's own problem statement almost exactly.

5.1 Why the combination is natural (motivation)

• Same problem, complementary halves. Cingular '984 exists to solve the dropped-call problem when a dual-mode user on an active call loses WiFi coverage — its flow (¶ [0055]-style description, sheets 2–3) is: call on cellular → enter building → hand over to indoor VoIP → user exits → "handover from the home network to the GSM cellular network is triggered due to a loss of signal … the call is maintained on the GSM network automatically and transparently to the user." That is the identical use case motivating the '570 patent (the dog-walker / café-patron leaving WiFi coverage). But Cingular's indoor leg is generic SIP/VoWLAN — it does not route the call to an IM voice network.
• IM voice was the known, commercially salient instance of the VoIP leg. Roamware '167, filed three years earlier, teaches that the gateway rerouting calls to a subscriber's mobile number can be "a VoIP gateway, a Skype gateway … a SIP/IMS gateway, or an IM-VoIP gateway," and that the client may be "Skype™, Yahoo®, Google®, GizmoProject, MSN®… or any SIP, VoIP or messaging client." Yahoo '695 independently teaches routing a PSTN call addressed to a phone number to IM usernames associated with that number, gated on presence. A PHOSITA reading Cingular in late 2006 would immediately recognize that the "indoor VoIP network" leg it hands calls over to is, in the consumer market of the day, most often Skype/Yahoo! Messenger with Voice/Google Talk — the references themselves (Roamware ¶ describing "GSM WiFi phone … configured with a Skype client") say so.
• Known interchangeability. KSR's "known interchangeability" rationale applies directly: Roamware lists IM-VoIP, Skype, Vonage, SIP/IMS gateways as interchangeable, and Yahoo/Google/MSN clients as interchangeable with SIP clients. Substituting Yahoo '695's IM-routing leg (presence-gated, keyed to IM identities) for Cingular's SIP leg requires no more than selecting among known interchangeable elements.
• No incompatibility. Yahoo '695's routing and Cingular's redirection both anchor on the same telephone number and both terminate on the same dual-mode handset. Yahoo supplies the "call destined for a telephone number → IM agent using an IM identity distinct from that number" relationship (claims 6–7); Cingular supplies the "active call on the non-cellular leg → establish the cellular leg mid-call → switch voice paths" mechanism (claims 1, 3). Nothing in either reference teaches away from the other; they operate in the same protocol stack (SIP/SIMPLE, RTP).

5.2 Element mapping for claim 1

Dependent claims fed by the combination: claim 2 (trivial — phone associated with the number); claim 3 (Cingular's redirection component is the "service gateway" that switches the call); claim 6–7 (Yahoo '695 and Roamware '167 both use identities — IM screen names / Skype IDs — distinct from the cellular telephone number; Yahoo's routing maps number→IM username, which is the exact "user identity different from the telephone number" limitation).

5.3 Why a PHOSITA would have had a reasonable expectation of success

All three references predate or bracket the priority date, operate on the same dual-mode handset and the same call model (one telephone number, two access networks, mid-call continuity), and use standard protocols (SIP/RTP for the IP leg; GSM/ISUP for the cellular leg). Implementing Yahoo-style IM-leg routing inside Cingular's redirection architecture is an integration task — selecting the IM signaling protocol (SIMPLE/XMPP, which the '570 spec itself lists as conventional) and mapping the IM identity to the subscriber record — well within the routine competence of the PHOSITA. The '570 patent's own spec concedes these protocols and networks were all known.

6. Secondary combination — Divitas '374 as the client-side/claim 23 anchor

Claim 23 is the wireless-phone apparatus: IM phone agent + an agent processing a second voice call that receives a call reference and forwards it to a switchover agent, which creates/retrieves a call record, determines the second call is associated with the IM call, and causes automatic establishment.

• Divitas '374 is the closest client-side art. Its client device has (verified from the published application) a "mobility manager application module for monitoring connectivity … to a plurality of networks" and "a call control application module for determining whether to trigger a handoff of the client device from a first radio station to a second radio station based on the connectivity" — structurally the functional equivalent of the '570's switchover agent deciding between the IM and cellular legs. Its FIGS. 12–13 are an express WiFi→cellular active-call handoff flow; FIGS. 14A–B the reverse. Its server routes incoming media to the client using stored connectivity information (the gateway-side mirror of claim 8). Divitas even frames the design choice in KSR terms — it criticizes carrier-controlled UMA and moves the handoff decision to the client, which is exactly the "switchover agent on the phone" design of claim 23.
• Adding Yahoo '695 (or Roamware '167) supplies what Divitas lacks: the IM voice network / IM phone agent / IM user identity leg. Divitas's handoff is between enterprise VoIP (WiFi) and cellular; making the packet leg an IM-voice leg is the same known-interchangeability substitution as in § 5.
• Claim 23's "receive a call reference and send it to a switchover agent" and "signal … for switch over purpose" are the coordination details any client-side handoff controller must implement to distinguish an incoming call that is the continuation of the active call from an ordinary new call — see § 8.

7. Tertiary combination — Roamware '167 + Cingular '984 + Divitas '374 as the claim 8 (gateway) anchor

Claim 8 requires the gateway-side software to "determine the circumstances under which the incoming voice call is switched to a second voice network," establish the IM leg, maintain the records, establish the second call, associate it, and signal the switchover agent.

• Roamware '167 supplies the gateway concept most on point: a gateway coupled to the cellular network that reroutes calls placed to the subscriber's mobile number to an IM/VoIP client, with FIGS. 6–7 expressly teaching switching traffic from a second communication network to a first and vice versa — two-way network switching at the gateway.
• Cingular '984 supplies the mid-call, active-call redirection trigger logic ("trigger component … redirection component") that decides when (on loss of WiFi signal) to re-establish the call on the cellular leg — the "circumstances under which the voice call is switched."
• Divitas '374 supplies the server that stores connectivity and routes incoming media — corroborating the gateway's record-keeping role.
• A PHOSITA combining Roamware's IM-rerouting gateway with Cingular's active-call redirection trigger would have had every claim-8 element except the literal dual-record/call-reference protocol — see § 8.

8. The genuine battleground — and why it likely does not save the claims

Three feature clusters recur as the apparent differentiators. Each needs an honest assessment:

(a) "IM voice network / IM phone agent / IM user identity." This is the weakest differentiator. Yahoo '695 and Roamware '167 both disclose routing a call placed to a phone number to an IM client/identity, and Roamware expressly declares IM-VoIP gateways and Skype/Yahoo/Google clients interchangeable with SIP/VoIP ones. Under KSR's known-interchangeability principle, substituting an IM voice leg into Cingular's/Divitas's generic VoIP handover architecture is an obvious design choice — indeed it is the commercially dominant instance of the VoIP leg in the 2005–2006 market the references themselves describe.

(b) Dual call records keyed by a call reference. No face-of-the-patent reference literally discloses "first call record … at the gateway" plus "second call record … at the phone … keyed by a call reference." But the function — correlating the two legs of the same call so the handover is seamless — is inherent in every reference: Cingular's redirection component must know which redirected call corresponds to which active cellular call; Divitas's server stores the client's connectivity/call state and its client tracks the active session across handoff (FIGS. 12–15); Roamware's gateway maintains subscriber routing state. Maintaining per-call state and passing a call identifier between network and terminal during a handover was textbook telephony practice (ISUP call-reference / transaction identifiers, SIP dialog identifiers, GSM call references) — precisely the kind of implementation detail KSR says "ordinary creativity" supplies. A challenger should press the examiner's file history here, but on the record as assembled this is a routine-record-keeping gap, not an inventive concept.

(c) "Sending a signal to a phone agent indicating that the second voice call is for switch over purpose" + automatic establishment. This is the strongest surviving limitation. The specific protocol act — the network telling the phone "this incoming cellular call is not a new call; it is the switchover leg," so the phone answers it without connecting it to the audio module until the switchover command arrives — is not literally disclosed in Cingular (which emphasizes the handover is "transparent to the user" but does not detail the in-band call-setup signaling distinguishing a handover leg from a fresh call), Divitas (whose call-control module decides handoff but whose disclosure, per the verified abstract-level content, does not detail this exact "for switch over purpose" indicator), or Roamware (whose FIGS. 6–7 switching flows are forwarding-oriented, not active-call-leg coordination). Whether this limitation is obvious depends on art outside the face-of-the-patent record — e.g., UMA/GAN and 3GPP VCC procedures for establishing and correlating a CS bearer for an ongoing VoIP session (Divitas's own spec references the UMA standard; Cingular describes SIP registration and IMS CSCF elements). If the challenger introduces UMA/GAN or VCC prior art showing the standard procedure of establishing the target leg with an indicator that it is for continuity purposes, this limitation falls too. On the face-of-the-patent record alone, a careful district court could find this one limitation non-obvious.

(d) User-acceptance embodiments (claims 5, 20). Cingular teaches against user prompting ("automatically and transparently to the user"), and Divitas likewise automates handoff. But the '570 spec describes the user-acceptance embodiment as optional, and handset prompts to confirm a network/cost-affecting action (e.g., accepting a callback or a roaming call) were conventional on GSM handsets of the era. Claims 5 and 20 add little; dependent claims elsewhere (4, 19 — quality-based switchover) are supported by the universal practice of signal/quality-triggered handoff (Cingular: loss-of-signal trigger; Divitas: connectivity-based trigger; voice-quality monitoring was standard in the UMA/VoIP art).

9. Secondary considerations (Graham factor 4)

None favor the patentee on the verified record: no commercial product practicing the claims has been identified; no long-felt-need or industry-acclaim evidence exists; no licenses are of record; and the patent has never survived a merits challenge (all 2024–2025 district-court assertions ended in Lab Technology's voluntary dismissals; no IPR has ever been filed). The absence of any objective indicia of non-obviousness materially weakens the patentee's position if a defendant forces a § 103 fight. The sole counterweight of note: the claims were examined and issued with Cingular '984 (the closest art) -cited by the examiner on the face of the patent — but that cuts both ways, since it confirms the examiner saw Cingular and still allowed the claims, which a patentee will argue shows the combination (Cingular + IM-leg art) was not before the examiner.

10. Bottom-line assessment

Most credible invalidity theory: A § 103 challenge combining Cingular '984 (active-call, dual-mode WiFi→cellular handover with network-side redirection) with Yahoo '695 and/or Roamware '167 (routing a call placed to a cellular number to an IM voice network using an IM identity distinct from that number), optionally adding Divitas '374 for the client-side switchover-agent architecture. The motivation is the textbook KSR showing: the same dropped-call problem, converging on the same dual-mode handsets, in a market where the references themselves document IM-voice (Skype/Yahoo/Google Talk) as the salient instance of the WiFi voice leg and expressly teach the interchangeability of IM-VoIP and SIP/VoIP elements.

Honest caveats: (1) The "signal … indicating that the second voice call is for switch over purpose," with the phone holding the second leg off the audio module until the switchover command, is the one limitation the face-of-the-patent combination does not cleanly cover; winning on claim 1/16/8/23 requires supplementing with UMA/GAN or VCC prior art showing the standard "establish the continuity leg and flag it as such" procedure — which my searches did not independently verify, so that leg of the case is unproven on this record. (2) Content of KR 2005-0001924 A, Shkedi '445, and Wanless '580 remains unverified and contributed nothing to this analysis. (3) This is a screening-level opinion based on abstracts, claims, and verified excerpts — a formal § 103 opinion requires reading each reference in full against each limitation and reviewing the '570 prosecution history (Office Actions in the parent chain, listed in the non-patent citations, are the first place to check what the examiner already considered).