US 6198920

I'll search for authoritative information on US 6198920 and any 2026 CAFC activity.

Let me verify claim structure and any 2026 litigation/docket activity.

Let me try to confirm the exact independent claim language and check docket sources.

US Patent 6,198,920 — Summary

Bottom line on the searches: I found no 2026 CAFC docket activity for U.S. 6,198,920, and none is plausible on the face of the record — the patent is recorded as expired. I also could not retrieve the patent's verbatim U.S. claim set from the authoritative full text (the fetched document text ends mid-sentence in the description, before the claims). Below I flag exactly where I am certain versus inferring.

Bibliographic data

Literal-ID caution: the third inventor's name appears as "Flex Houvig" on the 6,198,920 record and as "Flex Hourig" in the related US 7,602,782 B2 patent. I am not auto-correcting either spelling; treat both spellings as they appear in their respective records.

Assignee chain (as recorded): Padcom, Inc. → Padcom Holdings, Inc. (2006) → NetMotion Wireless Holdings, Inc. (2013) → Mobile Sonic Intermediate, Inc. (2022) → Mobile Sonic, Inc. (2022). Security interests were recorded to Liberty Partners Lenders L.L.C. (2001), Wells Fargo (2012) and Consortium Finance, LLC (2014), all later released (2016).

Abstract (as published)

"An apparatus and method is provided for transparent communication between a remote or mobile device and a fixed communication host network. The apparatus and method may include a remote network controller that logically resides between the host network and the existing infrastructure(s) that are used to provide communications network contact with one or more remote devices. The remote network controller is connected to the host communication network as a protocol-appropriate communications controller so that remote devices are indistinguishable to the host network from the locally-attached devices. Each remote device may be provided with an asynchronous serial data interface to communicate with a mobile data controller. The mobile data controller, in combination with the remote network controller, provides end-to-end data communication such that incompatible protocols are transparent to the remote device and host communication network. A router may be provided which selects a communications network in accordance with user configured parameters. The router communicates over a plurality of incompatible networks and is capable of using a variety of different protocols. Switching between the plurality of incompatible networks is transparent to the remote device and host communication network."

Independent claims — plain-language overview

The specification's own "aspects of the invention" summary maps onto the independent claims. Four independent aspects are described:

1. Apparatus for transparent data transport (first independent claim). An apparatus transports data between a remote device and a host communication network over a wireless link. It has two halves:

• a mobile data controller at the remote end that converts data between the remote device's own format and the wireless link's transmission format; and
• a network interface means at the host end that converts between the wireless link format and a network-interface format, and then again between that format and the host network's format.
  The wireless link format and the host network format are expressly incompatible, and data is carried over the wireless link in the wireless link format. The point is that neither the remote device nor the host network has to know a wireless hop exists.

1. Apparatus in which the network interface means is a remote network controller (further independent aspect). Same architecture, but the host-side network interface means is specifically a remote network controller that logically resides on the host communication network and performs the functions of a network communications controller — i.e., the remote devices are made indistinguishable from locally attached devices.

2. Apparatus with plural network interface means plus synchronization (further independent aspect). Adds multiple network interface means joined by a local network, plus a synchronization means that synchronizes routing tables and health-and-status information between them (the redundancy/failover embodiment of FIGS. 27–28, with dual-ported host data controllers).

3. Method of transporting data (independent method claim). A method counterpart of aspect 1: the mobile data controller converts data between the remote device format and the wireless link format; the data is transported over the wireless communications link; the network interface device receives it, converts between the wireless link format and its own network interface format (the wireless format and host network format being incompatible), converts again to the host network format, and forwards it to the host network. The remote device and wireless link are connected via the mobile data controller, and the host network and wireless link are interfaced by the network interface device.

4. Apparatus for transporting data over a plurality of incompatible networks (further independent aspect). A system determines network selection criteria; a selection system dynamically picks one of the plurality of incompatible networks per those criteria; and a switching system switches to the selected network for data transport. A further aspect adds a system for interfacing protocolized data into the plurality of incompatible networks using different protocols. This is the "Router 200 / Decision process / Network Availability / Switch" embodiment (FIGS. 29–35), including RSSI-based availability checks on a CDPD network interface.

Uncertainty flag: I could not verify the exact U.S. claim numbers, dependency structure, or verbatim claim language for these independent claims from the authoritative full text, because the fetched document text terminates in the description (at "...This informat"). Separately, the claims visible in EP 1 042 931 A1 (a family member, not US 6,198,920) recite a related but broader/different claim set (apparatus with link-selection criteria and dynamic switching; a method claim; an apparatus with a "remote device communication interface module"; and an apparatus for transporting data over a plurality of incompatible networks). Those EP claims are not US 6,198,920's claims and should not be attributed to it.

Litigation / 2026 CAFC check

• No 2026 CAFC appeal or docket entry involving U.S. 6,198,920 was found. Searches surfaced no Federal Circuit docket, opinion, or briefing referencing this patent number. The only 2026 CAFC material returned was unrelated (e.g., Receivership Estate of AudienceScience v. Google, §101, different patents entirely).
• Reason to expect none: the record lists the patent as Expired – Lifetime, with an anticipated expiration of 2015-06-01 (20 years from the 1995-06-01 priority filing). Even under a pre-URAA measure, 17 years from the 2001-03-06 issue date would have run in 2018. Either way the patent is long expired, so an infringement appeal in 2026 is implausible (an IPR-derived appeal remains theoretically possible, but none was found).
• Limits of this search: I queried the open web and patent databases (Google Patents, FreePatentsOnline, Justia, uspto.report, Espacenet, Unified Patents portal). I did not have direct access to PACER or the CAFC CM/ECF docket system, so I cannot rule out a sealed or very obscure 2026 filing. Google Patents does flag that the family "has litigation" and links a Darts-IP family litigation record (family 25462463), which I could not open directly.

Date discrepancy note: the session metadata shows today's date as 2026-09-27, while your instruction states April 26, 2026. I have not resolved this; it does not affect the patent's bibliographic data, only the recency window of the docket check.

I'll research litigation involving US patent 6198920 across patent litigation databases.

Let me dig deeper into specific cases and parties.

Let me search for additional details, including any PTAB proceedings and other suits.

Litigation Involving U.S. Patent No. 6,198,920

Important scope note first: I was able to run only a limited number of searches before hitting my tool-call limit, so this answer is based on the sources retrieved (Google Patents' family/litigation data, the D. Del. court website, Casetext/Cetient opinions, and trade-press coverage) plus my own background knowledge. Where I could not verify a detail with high confidence, I say so explicitly rather than guessing. I also want to flag that U.S. Patent No. 6,198,920 ("Apparatus and method for intelligent routing of data between a remote device and a host system") has a priority date of 1995-06-01 (continuation of Ser. No. 08/456,860, Doviak et al., which issued as U.S. 5,717,737), was filed 2000-03-16, issued 2001-03-06, and expired 2015-06-01 (anticipated expiration per Google Patents). The patent is now Expired – Lifetime, so no new infringement suits asserting it are possible.

1. Padcom, Inc. v. NetMotion Wireless, Inc. and Database Solutions, Inc. — the core case

Grounded sources:

• Padcom's filing announcement (Oct. 27, 2003) explicitly names both patents: "At issue are Padcom's U.S. Patent Nos. 6,418,324 and 6,198,920."
• A later Padcom press piece confirms the '920 patent was part of Civil Action No. 03-983: "...where another action is pending... in which Padcom alleges NetMotion infringes three of its patents: U.S. Patent Nos. 6,418,324, 6,198,920 and 6,826,405 (Civil Action No. 03-983)."
• The district court's summary-judgment opinion (Padcom, Inc. v. NetMotion Wireless, Inc., D. Del., Feb. 22, 2006), denying NetMotion's § 102 invalidity motion, addresses the asserted claims of the '324, '920, and '405 patents: cetient.com case page; D. Del. opinion page (03-983, D.I. 286 / 296)
• Post-trial record listing (including "Stipulation and Order regarding U.S. Patent No. 6,198,920 (DI 475), dated Mar. 15, 2006" and "Stipulated Entry of Final Judgment in favor of Padcom Inc. and against NetMotion Wireless Inc.—Case Closed (DI 512), Jul. 12, 2006"): freepatentsonline.com/7602782

Two nuances you should be aware of before relying on this as "'920 infringement verdict'

1. The '920 patent appears to have been resolved by stipulation before the March 2006 verdict. The docket index shows a "Stipulation and Order regarding U.S. Patent No. 6,198,920 (D.I. 475), dated Mar. 15, 2006" — i.e., on the eve of trial. Trade coverage of the actual verdict describes it as a finding that the '324 and '405 patents were valid and infringed (e.g., the NGB report: "同特許の有効性および NetMotion の 'Mobility' ソフトによる侵害を認定する評決" — "the validity of those patents (USP 6,418,324; 6,826,405) and NetMotion's infringement"). I could not retrieve the text of the D.I. 475 stipulation, so I cannot state with high confidence what its effect on the '920 patent was (dismissal of '920 claims, entry of judgment, or something else). Treat the '920 patent as having been part of the 03-983 case but likely resolved by that March 15, 2006 stipulation rather than by the jury verdict.

2. I could not confirm any Federal Circuit appeal outcome from the sources retrieved. Given Padcom prevailed at trial and a stipulated final judgment was entered, an appeal by NetMotion is plausible but I have not verified it — do not assume one occurred.

2. Related cases in the same dispute (for context; not '920 assertions)

• Padcom, Inc. v. NetMotion Wireless, Inc. (declaratory judgment action) — Padcom filed a DJ action in the District of Delaware on December 28, 2005, seeking a declaration that a then-newly-issued NetMotion patent was invalid, unenforceable and not infringed. Source: Light Reading, Dec. 28, 2005. I could not verify the separate case number for this DJ action; do not treat a number as confirmed.

• NetMotion Wireless, Inc. v. Padcom, Inc. — NetMotion filed in Washington state court (King County Superior Court) alleging Padcom tortiously interfered with NetMotion's contracts/business expectancies; Padcom removed it to the U.S. District Court for the Western District of Washington on March 24, 2004. The Casetext opinion identifies it as 04-cv-622-JCC (W.D. Wash. 2004). This is a tort case, not a '920 patent assertion. Source: Padcom, Inc. v. NetMotion Wireless, Inc. (D. Del. venue/jurisdiction opinion)

• Personal-jurisdiction / venue rulings in 03-983 (Delaware long-arm analysis re: NetMotion's Delaware marketing to police, fire and hospital customers; denial of transfer motions) are reported in the same Casetext opinion and are part of the 03-983 docket rather than separate '920 suits.

3. Later cases involving successor entities (NOT '920 assertions)

• Sectra Communications AB and Columbitech, Inc. v. Absolute Software, Inc. and Mobile Sonic, Inc., W.D. Wash., Case No. C22-353RSM — antitrust/unfair-competition and related claims concerning mobile VPN competition between Columbitech and NetMotion (NetMotion/Mobile Sonic being the current owner of the '920 patent per Google Patents' reassignment chain). This is not a patent-infringement case on the '920 patent. Source: Sectra v. Absolute Software, W.D. Wash. order

4. Ownership chain (relevant to who could have sued)

Google Patents' reassignment record (from the patent page you supplied) shows the '920 patent moving: Padcom, Inc. → (2001 security agreement to Liberty Partners Lenders LLC) → Padcom Holdings, Inc. (2006) → NetMotion Wireless Holdings, Inc. (2013 name change) → Mobile Sonic Intermediate, Inc. / Mobile Sonic, Inc. (2022 mergers). Google Patents also carries a "Family has litigation" flag with a Darts-IP litigation-family link for this family. Source: patents.google.com/patent/US6198920B1

5. Bottom line

• *The one identified U.S. litigation asserting U.S. 6,198,920 is Padcom, Inc. v. NetMotion Wireless, Inc. and Database Solutions, Inc., D. Del. No. 03-983, filed Oct. 27, 2003*, which proceeded to a March 28, 2006 plaintiff's jury verdict and a July 12, 2006 stipulated final judgment for Padcom (although the '920 patent itself appears to have been the subject of a March 15, 2006 stipulation, with the verdict press coverage emphasizing the '324 and '405 patents).
• Related 2003–2005 filings in the same dispute (the Dec. 2005 Delaware DJ action and the W.D. Wash. 04-cv-622-JCC tort action) are part of the controversy but are not '920 assertions.
• I did not find any other litigation asserting U.S. 6,198,920, and I found no PTAB IPR/PGR identified by name against the '920 patent in the results retrieved. Because the patent expired June 1, 2015, no further infringement litigation is possible.
• Caveats: I could not complete searches for (a) the exact text of the D.I. 475 '920 stipulation, (b) any Federal Circuit appeal in 03-983, (c) the DJ action's case number, or (d) any PTAB proceeding specifically against the '920 patent. Those items remain unverified rather than negative, and I have not guessed at them.

Proceedings overview

Zero AIA trial proceedings (IPR / PGR / CBM) have been filed against US 6,198,920. The structured ODP block ("no AIA trial proceedings on file") is the canonical list and it is empty; my independent searches of PTAB-related sources, practitioner reporting and the Google Patents "family has litigation" record surfaced no petition, no institution decision, no FWD, and no Federal Circuit appeal from a PTAB decision on this patent. Bottom line for a defendant: there is no PTAB estoppel, no canceled claim, and no PTAB win to inherit — but also no PTAB loss to fear. Any defense must be built in district court (or in a fresh IPR, subject to the 2025–26 restrictions discussed below) from scratch, against a patent that expires/has expired and whose asserted siblings survived a 2006 jury verdict.

Proceedings on file

None. No AIA proceeding exists to summarize. To avoid fabrication, I am not supplying proceeding numbers, panels, grounds or dispositions — there are none to supply for this patent.

What I verified, and what I found instead:

FLAGGED — pre-AIA inter partes reexamination (95/001629) — UNVERIFIED

Search results surfaced, inside the reference list of an unrelated later patent (US 8,902,081), citations to a Decision Denying Petition under 37 CFR 1.181 and 37 CFR 1.515(c), issued 2011-09-30, and a "Denying Request for Inter Partes Reexamination," issued 2011-06-24, in connection with Inter Partes Reexamination 95/001629, in a document cluster that also cites Padcom's '920 and '324 patents. I could not confirm from the available sources that 95/001629 names US 6,198,920, and inter partes reexamination is a pre-AIA, CRU-administered proceeding — it is not an AIA trial (IPR/PGR/CBM), does not carry § 315(e)(2) estoppel, and would not appear in the ODP AIA-trials list. Do not rely on this without pulling the reexamination file wrapper directly (USPTO Patent Center / reexam file history). Treat it as a lead, not a fact.

Strategic summary

Claim status of 6,198,920. Because there is no AIA proceeding, no claim of the '920 patent has been canceled, amended, or held unpatentable by the PTAB. Every claim — including the independent apparatus claim directed to the mobile data controller / network interface combination and the independent method claim, plus the dependent claims drawn to multiple interconnected remote network controllers and synchronized routing/health tables — is UNTESTED at the PTAB. (Note the specification text in the record supports claim sets in the 1–33 range and a Router-based set; I am not going to assign specific claim numbers to "canceled" or "sustained" status, because no tribunal has done so.) The only adjudicated validity event in the family is the *2006 Delaware jury verdict in Padcom, Inc. v. NetMotion Wireless, Inc., D. Del. No. 03-983*, which reportedly found US 6,418,324 and US 6,826,405 valid and infringed (see Judge Robinson's 2006-02-22 opinion, https://www.ded.uscourts.gov/opinion/padcom-inc-v-netmotion-wireless-inc-et-al-di-286-di-296). I could not confirm that the '920 patent was part of that verdict — contemporaneous reporting lists only the '324 and '405 patents. Assume the '920 patent's validity has never been adjudicated.

Estoppel landscape. With no FWD, § 315(e)(2) estoppel is zero. A current defendant can raise any § 102/§ 103 ground, including art the Patent Office already considered and art it never saw, both in district court and (subject to the timing/eligibility rules below) in a new IPR. There is also no § 325(e) estoppel and no General Plastic "same petitioner" problem for a first-time petitioner. The absence of prior IPR also means no prior petitioner's expert discovery, claim-construction record, or institution-stage admissions exist to borrow — the whole case has to be built fresh.

Pattern signals and a timing trap. There is no pattern of PTAB filings against this patent: no serial petitioner, no defensive aggregator (no evidence of Unified Patents or a similar entity challenging the '920 patent), and no PTAB appeal activity by the patent owner because there has been nothing to appeal. What does exist is a litigation/ownership chain that matters more than any PTAB stat: the patent was asserted-era Padcom property (Padcom v. NetMotion, D. Del. 03-983), went through Liberty Partners / Padcom Holdings / NetMotion Wireless Holdings, and now sits with Mobile Sonic, Inc. as current assignee (2022-10-18 merger record), with NetMotion Wireless Holdings Inc / Mobile Sonic Intermediate Inc / Mobile Sonic Inc listed as current assignees. Two practical consequences: (1) the patent expired on 2015-06-01 (20 years from the 1995-06-01 priority date; Google Patents lists "Anticipated expiration 2015-06-01" — expired status does not by itself bar IPR, but it caps recoverable damages to past infringement); and (2) the 2025–26 PTAB policy environment now cuts against filing. Director Squires' 2025-10-16 memorandum centralizes institution decisions with the Director (https://www.willkie.com/publications/2025/10/ptab-announces-new-institution-policy), and the 2025-10-15 proposed rules would reportedly bar an IPR against a patent that has already survived a validity challenge in district court or at the PTAB and require a § 102/§ 103 stipulation. A 1995-priority, 2015-expired patent is also precisely the profile the Board has used for "settled expectations" discretionary denial. Budget accordingly: an IPR here is high-risk of denial regardless of merit.

Recommended next steps

• If you are a defendant and the demand letter cites 6,198,920: there is no FWD to link to and no canceled claim to point at. The honest position is: "The '920 patent has never been challenged at the PTAB; we are the first." Do not tell a court or opposing counsel that any claim has been invalidated — that would be sanction-bait in the other direction.
• Prior-art search scope is wide open. Because no IPR was filed, no art was selected, briefed, or conceded, and no § 325(d) "same or substantially the same art" argument was ever made. Pull the '920 file wrapper and the Padcom v. NetMotion D. Del. No. 03-983 record (including the 2006 trial exhibits and any invalidity contentions against the '324/'405 siblings) — that corpus is the cheapest source of ready-made § 102/§ 103 grounds.
• Before paying for a petition, check the threshold gates. (i) Confirm expiry and any PTA/terminal-disclaimer effects on damages; (ii) check whether the proposed 2025 rules have been finalized and whether they would bar an IPR on a patent whose sibling claims survived the 2006 jury verdict; (iii) expect a § 314(a) / settled-expectations discretionary-denial fight given the 1995 priority date. If the demand is for a modest past-damages sum on an expired patent, district-court invalidity plus laches/§ 286 damages limits may be the better spend than an IPR petition.
• Verify the 95/001629 lead at https://patentcenter.uspto.gov (reexamination file history) before relying on or ignoring it. If it is the '920 patent, note that a denied inter partes reexamination request is both (a) useful § 325(d) ammunition for a patent owner and (b) a sign the art considered in 2011 was cumulative of prosecution — which narrows, but does not close, your art options.
• No appeal to monitor. There is no PTAB FWD, so there is no CAFC docket number, no oral-argument date, and no CourtListener entry to track for this patent. If you want the canonical docket anchor for the only AIA-relevant activity, check https://ptacts.uspto.gov/ and the USPTO Patent Center record for 09/527,014.

Caveat on completeness. My conclusion rests on (1) the structured ODP block stating no AIA proceedings on file and (2) web searches that produced no contrary evidence. Older, settled, or terminated proceedings that were never indexed by the ODP and never reported in practitioner literature could in principle exist; if absolute certainty matters, run the patent number through PTAB E2E / PTAB Center (https://ptacts.uspto.gov/) and Docket Navigator or Lex Machina's PTAB module as a final confirmation step.

I'll research the assignment record for this patent before writing. Let me run targeted searches.

Let me search for the specific assignment reel/frame records and the successor-entity mergers.

I hit my tool-call limit mid-research, so I could not execute the Assignment Center query that would return reel/frame numbers and correspondent (recording attorney) fields. Rather than fabricate those, I reconstruct the chain from the USPTO-derived legal-events data on the patent record itself (which does carry conveyance type, assignor, assignee and recording date) plus press/press-release and PitchBook corroboration, and I flag every field I could not verify.

Sourcing caveat up front: The entries below use the reassignment events listed on the patent's own record (patents.google.com/patent/US6198920B1, "Reassignment" legal events, which mirror USPTO assignment data) and the filing/priority events. I do not have the reel/frame numbers or the correspondent-of-record names for any link in this chain — those require a live Assignment Center query I was unable to complete. Do not treat any reel/frame or attorney name as present; it is simply absent from this report.

Inventors

• All three are named on the '920 patent (Application No. 09/527,014, filed 2000-03-16), which claims priority to Ser. No. 08/456,860, filed 1995-06-01 (issued as U.S. 5,717,737, also Doviak et al.).
• Unusual-pattern check — no departure signal found. The opposite pattern appears: all three inventors were still reachable and were deposed as fact witnesses in the 2006 Padcom v. NetMotion litigation ("Doviak Deposition," "Houvig Deposition," "Whitmore Deposition" transcripts listed on the post-trial record). That is evidence they remained associated with the portfolio and the company through at least 2006 — i.e., roughly 11 years after the 1995 priority filing — not a "all inventors gone within 12 months" fire-sale tell.
• I found no evidence of the inventors assigning away personal rights or departing Padcom within 12 months of filing. The inventor→Padcom assignment of record exists (the patent issued to Padcom at grant on 2001-03-06), but I do not have its reel/frame.

Original assignee

• Entity on the issued patent: Padcom, Inc., Bethlehem, PA.
• Primary line of business: Wireless/mobile networking software and hardware — "mobile VPN" and wireless remote-access connectivity products (PitchBook lists PADCOM as a 1989-founded "Provider of wireless networking solutions… wireless remote access connectivity solutions"). Public-safety sector focus.
• Did it ship a product embodying the claims? Yes. Padcom marketed the TotalRoam mobile-VPN product line and Remote Access Router 6100 / 8100 family; these are catalogued in the 2006 trial record ("Padcom Inc., 'TotalRoam; Technical Overview'… '6100 Remote Access Router'… '8100 Remote Access Router'"). Its largest customer was Cox Communications. This is a practicing entity, not a paper owner.
• Current status: acquired/merged. Padcom merged with rival NetMotion Wireless in June 2006 (deal announced 2006-06-15; PitchBook shows M&A completed 15-Jun-2006, acquirer "NetMotion Software"). The combined company operated as NetMotion Wireless → later NetMotion Software; NetMotion was acquired by Clearlake Capital (2012), The Carlyle Group (2016), and Absolute Software (~US$340M, 2021).
• Note a corporate-naming nuance: the 2006 recorded assignment ran to "Padcom Holdings, Inc.," not "Padcom, Inc." That is the holding entity created around the merger, not a third-party buyer.

Assignment timeline

Field-completeness warning: The recorded-date, conveyance-type, assignor and assignee below are taken from the patent record's reassignment events. Reel/frame and correspondent-of-record are unavailable in this session and are listed as "not retrieved." Execution dates are frequently not the same as recording dates; where the record shows only one date I label it as recorded.

• Recorded 2001-07-16 — Reel not retrieved

  • Conveyance: SECURITY AGREEMENT
  • Assignor: Padcom, Inc.
  • Assignee: Liberty Partners Lenders L.L.C.
  • Correspondent: not retrieved
  • Context: Securitization — Padcom pledges its patents as collateral for financing (Liberty Partners is separately listed by PitchBook as a Padcom PE/VC investor).

• Recorded 2006-09-06 — Reel not retrieved

  • Conveyance: ASSIGNMENT OF ASSIGNORS INTEREST (see document for details)
  • Assignor: Padcom Inc.
  • Assignee: Padcom Holdings, Inc.
  • Correspondent: not retrieved
  • Context: Internal reorg — transfer to the Padcom holding entity created around the June 2006 NetMotion merger.

• Recorded 2012-08-31 — Reel not retrieved

  • Conveyance: RELEASE BY SECURED PARTY
  • Assignor: Liberty Capital Partners, Inc.
  • Assignee: Padcom, Inc.
  • Correspondent: not retrieved
  • Context: Lien release — release of the 2001 Liberty security interest.

• Recorded 2012-08-31 — Reel not retrieved

  • Conveyance: RELEASE BY SECURED PARTY
  • Assignor: Liberty Partners Lenders L.L.C.
  • Assignee: Padcom, Inc.
  • Correspondent: not retrieved
  • Context: Lien release — closing out the remaining Liberty security interest.
  • Discrepancy to flag: the two same-day releases name two different Liberty legal entities — "Liberty Capital Partners, Inc." and "Liberty Partners Lenders L.L.C." — while the 2001 grant named only "Liberty Partners Lenders L.L.C." Either two related lenders held interests, or the secured party was renamed. I cannot resolve which from the available records.

• Recorded 2012-09-14 — Reel not retrieved

  • Conveyance: PATENT SECURITY AGREEMENT
  • Assignor: NetMotion Wireless Holdings, Inc.; NetMotion Wireless, Inc.
  • Assignee: Wells Fargo Bank, National Association
  • Correspondent: not retrieved
  • Context: Securitization — new secured lender taken on by the NetMotion entity (same period as the Clearlake acquisition).

• Recorded 2013-04-03 — Reel not retrieved

  • Conveyance: CHANGE OF NAME
  • Assignor: Padcom Holdings, Inc.
  • Assignee: NetMotion Wireless Holdings, Inc.
  • Correspondent: not retrieved
  • Context: Change of name only — "Padcom Holdings, Inc." formally becomes "NetMotion Wireless Holdings, Inc." This is the link that moves the '920 formally into the NetMotion naming family.

• Recorded 2014-07-23 — Reel not retrieved

  • Conveyance: PATENT SECURITY AGREEMENT (SECOND LIEN)
  • Assignor: Lumension Security, Inc.; NetMotion Wireless Holdings, Inc.; NetMotion Wireless, Inc.
  • Assignee: Consortium Finance, LLC
  • Correspondent: not retrieved
  • Context: Securitization — second-lien collateral pledge.

• Recorded 2016-10-20 — Reel not retrieved

  • Conveyance: RELEASE OF SECURITY INTERESTS IN PATENTS
  • Assignor: Wells Fargo Bank, N.A.
  • Assignee: NetMotion Wireless, Inc.
  • Correspondent: not retrieved
  • Context: Lien release — Wells Fargo interest discharged around the Carlyle acquisition.

• Recorded 2016-10-25 — Reel not retrieved

  • Conveyance: RELEASE BY SECURED PARTY
  • Assignor: Consortium Finance, LLC
  • Assignee: NetMotion Wireless, Inc.; Lumension Security, Inc.
  • Correspondent: not retrieved
  • Context: Lien release — clearing the 2014 second-lien pledge.

• Recorded 2016-10-27 — Reel not retrieved

  • Conveyance: RELEASE BY SECURED PARTY
  • Assignor: Wells Fargo Bank, National Association
  • Assignee: NetMotion Wireless, Inc.
  • Correspondent: not retrieved
  • Context: Lien release — final Wells Fargo discharge.

• Recorded 2022-10-18 — Reel not retrieved

  • Conveyance: MERGER
  • Assignor: NetMotion Software, Inc.
  • Assignee: NetMotion Wireless Holdings, Inc.
  • Correspondent: not retrieved
  • Context: Internal reorg / merger step.

• Recorded 2022-10-18 — Reel not retrieved

  • Conveyance: MERGER
  • Assignor: NetMotion Wireless Holdings, Inc.
  • Assignee: Mobile Sonic Intermediate, Inc.
  • Correspondent: not retrieved
  • Context: Internal reorg / merger step (same-day cluster).

• Recorded 2022-10-18 — Reel not retrieved

  • Conveyance: MERGER
  • Assignor: Mobile Sonic Intermediate, Inc.
  • Assignee: Mobile Sonic, Inc.
  • Correspondent: not retrieved
  • Context: Internal reorg / merger step — the chain terminates at Mobile Sonic, Inc., the current assignee of record (successor in the Absolute Software/NetMotion corporate family).

Timeline diagram

timeline
    title Ownership of US 6198920
    1995 : Priority app filed by the inventors
    2000 : Continuation filed by Padcom Inc
    2001 : Issued to Padcom Inc
    2001 : Security agreement to Liberty Partners
    2003 : Padcom sues NetMotion for infringement
    2006 : Padcom merges into NetMotion
    2006 : Patent assigned to Padcom Holdings Inc
    2012 : Liberty releases restore Padcom Inc
    2012 : Wells Fargo security agreement
    2013 : Renamed NetMotion Wireless Holdings
    2014 : Consortium Finance second lien
    2016 : Security interests released
    2022 : NetMotion merged into Mobile Sonic

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. The only holding-company hop is 2006-09-06 Padcom Inc. → Padcom Holdings, Inc., which is the standard holdco created around the June 2006 Padcom/NetMotion merger, not a licensing-only vehicle. Every downstream assignee is an operating mobile-VPN vendor (NetMotion entities; Mobile Sonic, Inc., which sits in the Absolute Software family). I have no product-absence or registered-agent-address evidence pointing to a single-purpose licensing LLC, and I will not infer shell status from naming alone.

2. Known asserter in the chain — NOT PRESENT. No link matches Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg entity. The one party that did assert this patent — Padcom, Inc. — is an operating vendor that shipped TotalRoam/RAR products and sued a direct competitor (NetMotion). That is the operating-company pattern, not the NPE pattern.

3. Repeat correspondent across the chain — UNKNOWN / cannot be assessed. This signal turns entirely on the correspondent-of-record field, which I could not retrieve for any recording. I therefore make no finding either way. To close this out, query the Assignment Center by patent number and pull the correspondent on each reel/frame, then compare against RPX/Unified NPE-counsel lists.

4. Cascading transfers — NOT PRESENT (as an NPE cascade). There is no run of chained LLC assignments inside 24 months. The recordings are spaced years apart and are dominated by security agreements and lien releases. The one same-day cluster (2022-10-18, three MERGER recordings) is a single corporate restructuring executed and recorded together (NetMotion Software → NetMotion Wireless Holdings → Mobile Sonic Intermediate → Mobile Sonic), not a series of NPE hand-offs. The 2012-08-31 pair is likewise two same-day lien releases between the same parties.

5. Pre-litigation transfer — NOT PRESENT. The first suit on this patent was filed 2003-10-27 (D. Del. 03-983). The only recorded event predating it is the 2001-07-16 Liberty security agreement — a securitization roughly two years before suit, not a within-6-months standing/venue arrangement. No assignment was recorded in the months immediately before the 2003 complaint. (The patent did change hands after trial, in the 2006 merger that also settled the case — but that is post-litigation, not pre-litigation.)

6. Bankruptcy fire-sale — NOT PRESENT / no evidence found. I found no Chapter 7/11 proceeding for Padcom. The company's documented timeline (VC financing round Feb 2005; merger with NetMotion completed June 2006) is inconsistent with a bankruptcy liquidation. I flag this as "no evidence," since I could not run a docket/PACER check.

7. Privateering — NOT PRESENT. Padcom enforced the '920 itself, as the patent owner and product maker, against a head-to-head competitor — the opposite of a hidden third-party NPE asserting on a sponsor's behalf. No SEC filing or Patent Progress/EFF coverage surfaced showing a transfer to an entity litigating for Padcom's benefit.

8. Defensive aggregator — NOT PRESENT. The chain does not terminate at RPX, AST, LOT, Unified Patents or OIN. It terminates at Mobile Sonic, Inc., an operating successor in the Absolute Software/NetMotion family. The litigation involving successors that I did surface — Sectra Communications/Columbitech v. Absolute Software & Mobile Sonic (W.D. Wash., C22-353RSM) — is an antitrust/unfair-competition matter about mobile-VPN competition and is not a defensive-aggregation or '920 event.

Verdict

Operating-company assertion.

The patent was enforced by its practicing owner: Padcom, Inc. of Bethlehem, PA shipped the TotalRoam and Remote Access Router product lines and asserted the '920 against a direct rival (NetMotion) in D. Del. 03-983, filed 2003-10-27. Every subsequent recorded transfer is either a security agreement / lien release (2001-07-16 Liberty; 2012-09-14 Wells Fargo; 2014-07-23 Consortium Finance, all later released) or an internal reorg/name change/merger (2006 Padcom Inc.→Padcom Holdings; 2013 change of name to NetMotion Wireless Holdings; 2022-10-18 merger steps into Mobile Sonic, Inc.), never a transfer to a licensing-only NPE. No shell-entity, known-asserter, cascading-transfer, pre-litigation-transfer, bankruptcy or privateering tell is present; the only open item is signal #3 (repeat correspondent), which I could not test because reel/frame and correspondent fields were not retrievable here.

Verify at: USPTO Assignment Center — search by patent number 6198920 (https://assignmentcenter.uspto.gov/ ; mirrored at https://assignment.uspto.gov/patent/index.html). Pull the reel/frame and correspondent on each of the 13 post-issuance recordings above to complete signal #3 and to confirm the exact execution vs. recording dates.

I'll research the prior art references cited on the face of US 6,198,920 and related prosecution history.

Let me get the complete cited-reference lists from the authoritative patent pages.

Let me pull the complete cited-reference lists (U.S. patents, foreign references, and non-patent literature) from the authoritative records.

Prior Art Analysis — U.S. Patent No. 6,198,920 (Doviak et al.)

0. Scope, method, and honesty flags (read first)

What I searched. I queried the open web for the record of U.S. Patent No. 6,198,920 specifically (Google Patents, FreePatentsOnline's 6198920.html, Justia Patents /patent/6198920, EveryPatent's /comp/pat6198920.html, uspto.report /patent/grant/6198920, Espacenet) and for prosecution/family records naming it. I retrieved only a limited number of search results before the tool limit stopped me, so this is built from those sources plus the authoritative full text you supplied. I did not query USPTO PatentCenter / Patent Application Information Retrieval (PAIR) directly, so I have not read the actual office actions.

Three flags you must carry into anything built on this:

1. The authoritative full text you supplied does not contain the front-page "(56) References Cited" list, nor the printed claims. It truncates mid-description at "...This informat". I therefore reconstructed the cited-reference list from three independent mirrors — EveryPatent, FreePatentsOnline, and Justia (the latter also carrying the non-patent-literature list). Those three agree on the U.S. patent list, which gives me reasonable confidence, but this is a reconstruction, not the primary document. If exact sub-list membership matters (e.g., for an IDS or a validity chart), verify against the printed patent.
2. Verbatim claim language is still unverified (as flagged in the earlier summary section). I therefore express the § 102 mappings against functional claim groups corresponding to the independent aspects the specification itself enumerates, not against confirmed claim numbers. I have not invented claim numbers. Treat every "potentially anticipates" statement as an analytical hypothesis, not a prosecution fact.
3. I did not obtain the actual § 102 rejections from prosecution. The '920 issued, so the examiner did not ultimately allow any of these references to defeat the claims as filed. Citing a reference ≠ applying it. I cannot tell you from the retrieved sources which of these were applied under § 102 and which were cited as background, or under which subsection.

A chronology caveat that materially narrows the § 102 field. The '920 claims priority to 1995-06-01 (continuation of Ser. No. 08/456,860) and is a pre-AIA patent. That means:

• Only references published or patented before 1995-06-01 can be § 102(a)/(b)/(e)-type art on their face for the invention date; and
• References dated after 1995-06-01 (a large fraction of the list — 1996 through 2000) can generally only bite as pre-AIA § 102(e) art, i.e., only if their underlying U.S. filing date is earlier than the applicant's invention date. Several entries below (e.g., the 1998–2000 items) are almost certainly background/two-year-window citations, not anticipation art.
• US 5,717,737 is a "cited" family member, not § 102 art (see § 3.a) — it is the parent that issued from Ser. No. 08/456,860, from which the '920 claims benefit; it is the same inventive entity. It cannot anticipate a continuation claiming its priority.

1. The claim groups I am mapping against

Because I cannot cite verified claim numbers, here is the mapping scheme, tied to the independent aspects the specification itself recites (per the earlier summary section, do not re-derive them):

2. Complete examiner-cited U.S. patent list (reconstructed)

All items below are from the mirrored "(56) References Cited / U.S. Patent Documents" list for 6,198,920. "§ 102 target" is my hypothesis against the groups in § 1.

(Note: the FreePatentsOnline and Justia mirrors list these in a slightly different subset/order than EveryPatent; the EveryPatent "(56) References Cited U.S." enumeration above is the fullest of the three. US 5,412,375, 5,420,574, 5,452,471, 5,537,220, 5,602,843 and 5,717,737 appear consistently across all mirrors.)

3. Foreign patent documents cited

Caveat: these two are the foreign references in the U.S. list per the FreePatentsOnline mirror. The family's EP 1 042 931 B1 description additionally cites EP 0 766 490 A, EP 0 793 395 A and WO 97/15160 A, and US 5,452,471 (LEOPOLD) — but those are citations in the EP document's description, which is a different document from US 6,198,920. Do not attribute the EP description's citation list to the U.S. patent. (I flag this because the earlier litigation section already distinguishes EP 1 042 931 from the '920; the same discipline applies here.)

4. Non-patent literature cited (from the Justia "Referenced Cited" record)

This is, in my assessment, the most probative block of prior art in the record, because it describes real-world systems that integrate mobile data terminals with multiple RF networks and a host — i.e., squarely the "transparent routing between mobile device and host over dissimilar media" problem.

Non-patent caution: the family's EP description also cites Martin R. Arick, The TCP/IP Companion, Wiley-QED, pp. 18–19. I list it for completeness but cannot confirm it appears in the U.S. '920 reference list — same EP-vs-US attribution caution as above.

5. References discussed in the '920 specification itself (highest interpretive weight)

These are the references the applicants chose to characterize and distinguish in the specification. They carry the most weight for understanding what the '920 is not, and they are the natural starting points for any § 102/§ 103 analysis:

1. US 5,412,375 — Wood, Method of selecting an air interface for communication in a communication system (1995-05-02). Spec's own distinction: the Wood routing decision is based on capabilities of the endpoints, not preference metrics of the transporting networks; endpoints must be aware of wireless-environment peculiarities. → Directly aimed at Group E; the '920's Decision process/User Configured parameters are the stated point of novelty over Wood.
2. US 5,420,574 — Erickson et al., Channel allocation mechanism (1995-05-30). Spec's distinction: switches among incompatible message formats but provides only a single data path; incompatibility is format-level, not independent-incompatible-network-level. → Group E and Group A.
3. US 5,602,843 — Gray, Integrated wired and wireless telecommunications system (1997-02-11). Spec's distinction: does not transport data between the same two endpoints and provides no alternate-path routing for a terminal across wired/wireless handsets. → Groups B and C (the redundancy/synchronization aspect).
4. US 5,452,471 — Leopold et al. (1995-09-19). Spec's characterization: primary/secondary systems on orthogonal channels; subscriber picks the smallest-coverage system. → Group E.
5. Ser. No. 08/456,860 — Doviak et al. (which issued as US 5,717,737, 1998-02-10). Spec's characterization: transports data over a plurality of dissimilar wireless networks, but does not automatically transmit through differing networks per preference metrics and does not automatically provide redundant/alternate pathways. → This is the parent/family document. Its characterization is the express antecedent basis for Groups C and E (i.e., the '920's stated advance over it). It is not § 102 prior art.

6. Where I would rank the "most relevant" art

If the question is which references a challenger would lead with against the groups in § 1:

• Group E (multi-network selection/switching) — strongest cluster: Wood '375, Leopold '471, Erickson '574, Tayloe '188, plus the Racotek/RAM Mobile/ARDIS NPL. Wood and Leopold are the closest on "select among multiple networks/interfaces," and the specification's own distinctions are the intended rebuttals.
• Group A/D (transparent transport, mobile data controller + network interface) — strongest cluster: the MAVRIC 2000 brochure (Metric Systems, 1991) and the Racotek "Adaptive Network Modem" / "Mobile System Data Terminal" collateral, plus US 4,697,281 and US 4,969,184 on terminal-to-network data over radio.
• Groups B/C (remote network controller transparency; plural interfaces + sync) — strongest cluster: Gray '843 (the express target of the spec's failover and endpoint-identity distinctions) plus Chin '775 (integrated router/hub) and Murthy '905.
• Not usable as § 102 art: US 5,717,737 (family parent).

7. One genuinely useful cross-record datapoint (reverse direction)

Searching for the patent number turned up a third-party PCT search report — WO 2005/094008 A1 — in which US 6,198,920 (rendered there as "DOUTAK WILLIAM ET AL. 6 March 2001") is cited in category "X" (alone defeats novelty/inventive step) against that application's claims 1, 3, 5–10, 12–18, with pinpoint cites to the '920's column 9 line 55–col. 10 line 65 (fig. 2), col. 24 line 37–col. 26 line 5 (fig. 25), col. 28 line 46–col. 29 line 11, and col. 29 line 45–col. 36 line 33 (figs. 29–32).

Two uses of this: (a) it shows the '920's figs. 29–35 Router/Network Availability disclosure is broad enough that a later examiner used it as anticipatory art against a "smart connection management / multiple network connection" application — a useful proxy for how the Group E disclosure reads on later selection art; and (b) it confirms real column/figure boundaries for the Router material (cols. 29–36, figs. 29–32), which I could not get from the truncated full text.

Literal-ID note (no auto-correction applied): the search report writes the inventor as "DOUTAK" where the '920 record shows "Doviak"; this is a transposition/OCR artifact in the search report, and I have quoted it as it appears rather than silently correcting it. Similarly, the FreePatentsOnline record shows the third inventor as "Houvig, Flex" while Google Patents shows "Flex Houvig" — both are reproduced above as found, not normalized.

8. Bottom line and residual uncertainty

• US 6,198,920's cited-art list is a large, mostly-background cellular-data/dual-mode/roaming list (≈55 U.S. patents, 2 Japanese published applications, and a substantial Racotek/MAVRIC/RAM Mobile/ARDIS non-patent corpus). Only five references are affirmatively discussed and distinguished in the specification: Wood '375, Erickson '574, Gray '843, Leopold '471, and Doviak ('737 / Ser. No. 08/456,860).
• The most relevant prior art on the face of the record is the multi-network selection cluster (Wood, Leopold, Erickson, Tayloe) plus the integrated mobile-data-systems NPL (MAVRIC 2000; Racotek; RAM Mobile; ARDIS).
• Three things remain unverified and should not be assumed: (i) the verbatim claim set and therefore any specific claim-number § 102 mapping — my mappings are to functional groups A–E; (ii) which references were actually applied under § 102 during prosecution, and under which pre-AIA subsection; and (iii) the exact membership of the "(56) References Cited" list as printed, since I reconstructed it from three mirrors rather than the primary front page.
• Do not treat US 5,717,737 as anticipatory art against the '920 — it is the parent arising from the same priority application.

If you need this hardened for a chart, the next steps would be: (1) pull the printed front page and full claim set from USPTO PatentCenter for 6,198,920; (2) pull the file wrapper office actions to identify actual § 102 rejections and the applied subsections; and (3) obtain the 1990–1993 Racotek/MAVRIC documents' actual publication evidence, since those NPL items predate the critical date and are the most dangerous art in the list.

I'll gather the cited prior art and verify what I can before building the § 103 analysis.

Let me confirm the dates/identities of the key background references, since availability-as-prior-art turns on them.

Obviousness Analysis — U.S. Patent 6,198,920 under 35 U.S.C. § 103

Scope note / carry-forward: Per the earlier sections, the authoritative full text retrieved for this patent terminates mid-description ("…This informat"), before the claims. I therefore still do not have verbatim claim language, claim numbering, or dependency structure. The analysis below is built at the level of the five independent aspects the specification itself recites (as mapped in the Patent Summary section) and uses only the references the patent's Background/Prior Art discussion identifies, as instructed. Where a claim-level mapping is required, I flag it as inference. The "Flex Houvig"/"Flex Hourig" spelling caution and the April 26 vs. September 27, 2026 date discrepancy from the earlier sections remain unresolved and are carried forward.

I. Threshold issues that determine whether each reference is even available

1. The statute is pre-AIA

Application 09/527,014 was filed 2000-03-16, before the AIA's 2013-03-16 cutoff, so pre-AIA § 103 governs, with the pre-AIA § 102(a)/(b)/(e) categories. Priority is claimed from US 08/456,860 (1995-06-01), which issued as US 5,717,737. (US6198920B1; US5717737A)

2. The single most important variable: what is the effective filing date of each claim?

This decides the case more than any technical argument. The '920 was filed ~4.8 years after its claimed 1995-06-01 parent. The Background discussion identifies and discusses Gray, U.S. 5,602,843, which did not issue until 1997-02-11 — i.e., after the claimed 1995-06-01 priority date. (US5602843 PDF; FPO/5602843)

Two consequences:

• If the claims are entitled to 1995-06-01, then Gray (issued 1997) is available only under pre-AIA § 102(e) if its underlying application was filed before 1995-06-01 — a fact I could not verify before exhausting my search budget. Wood (issued 1995-05-02, filed 1993-09-27) is available under § 102(a)/(e) but not § 102(b), because 1995-05-02 is only about 30 days before 1995-06-01. (US5412375) I likewise could not verify the filing/issue dates of Erickson U.S. 5,420,574 or of the Leopold et al. reference.
• If the more advanced subject matter — the Router 200 of FIGS. 29–35 and the subsystem-synchronization/redundancy of FIGS. 27–28 — was added by amendment or intervening continuation (i.e., new matter), those claims get only 2000-03-16. Gray (1997) then becomes § 102(b) art outright, as would any 1996–1999 art (CDPD packet data, cellular RSSI-based handoff, router-redundancy protocols). This is the classic Padcom-family vulnerability: the specification text of the '920 discusses post-1995 references, which is itself evidence that not all of the disclosure was in the 1995 parent.

Practical takeaway: Grounds I and III below are robust either way (they rest on Wood/Erickson/Leopold plus admitted general knowledge). Ground II (redundancy/synchronization) is materially stronger if the 2000 date applies, because Gray's own claim 7 ("a plurality of said group controllers dividing the task between them") and claim 6 (a LAN subsystem with a LAN controller connected through the central switch) would then be squarely § 102(b) art.

3. The specification's own characterizations are admissions

The Background section pre-commits the patent owner to what each reference does and does not disclose. That is useful under § 103 because it narrows the disputed delta to: (a) using preference metrics of the transporting networks rather than endpoint capabilities; (b) providing redundant/alternate pathways between the same two endpoints; and (c) automatic, transparent switching among dissimilar networks. Notably, the specification does not claim that protocol/format conversion, transparent encapsulation, or controller-mediated network selection was novel — only these three differentiators.

4. Level of ordinary skill

For a 1995 priority date: a bachelor's degree in electrical engineering or computer science (or equivalent), plus roughly 2–3 years of experience in wireless data communications and/or data networking, including RF link protocols and host-network protocols (e.g., Ethernet/Token-Ring/TCP-IP). A POSITA would be familiar with trunked-radio and cellular data systems, gateway/protocol-converter design, and routing-table management.

5. § 112 ¶ 6 affects how the art must map

The aspects as recited use "network interface means," "remote data conversion means," "wireless link conversion means," "host network conversion means," and "synchronization means," plus the functional "selection system"/"switching system" of the router aspect. Under § 112 ¶ 6 the art need only disclose structure performing the recited function or an equivalent — not identical structure. That lowers the § 103 hurdle for the means-plus-function limitations considerably.

II. The references as § 103 tools

Important caveat: U.S. 5,717,737 is the '920's own parent. It is not § 103 prior art against claims entitled to its benefit — but it is the disclosure baseline the '920's claims must add something patentable over, and the specification's own "objects" (transparent data path, devices indistinguishable from locally-attached devices) are already recited there. That is a strong § 103 posture: the '920's claims are largely the parent's disclosure plus the admitted arts of network selection, format switching, and controller redundancy.

(Also appearing in the family's cited-art lists alongside Wood is U.S. 5,448,619, "Apparatus and a method of allowing private cellular operation within an existing public cellular system"; I did not verify its date or content and do not rely on it. See FPO/7602782 reference list.)

III. Grounds of rejection

Ground I — Transparency + end-to-end format conversion (independent aspects 1, 2, and the method of aspect 4): Gray in view of Erickson, further in view of Wood

Proposed combination: Gray '843 as the primary reference; Erickson '574 and Wood '375 as secondary.

Element-by-element logic:

• "Mobile data controller connected to the remote device and the wireless link; converting between a remote device transmission format and a wireless link transmission format." — Gray's base stations "act as multiplexers forming a transparent encapsulation bridge connecting said data link layer between the handsets and the controller," and its claim 1 recites a wireless subsystem communicating over wireless links to mobile terminals. Erickson supplies the data input on the subscriber unit and the setting of communication parameters according to the type of information being carried — precisely "converting… between a remote device transmission format and a wireless link transmission format."
• "Network interface means… logically resides on the host communication network… performs the functions of a network communications controller," and remote devices "indistinguishable from locally-attached devices." — Gray discloses this almost verbatim in substance: "The PBX controller regards the wireless devices as ordinary station sets. It can control these sets by using the mobility controller as a virtual peripheral device controller."
• "Wireless link conversion means… host network conversion means… the wireless link transmission format and the host network format being incompatible." — Gray's common pool of transcoders performs the equivalent two-stage conversion (RF-compressed format ↔ network format) centrally; Erickson supplies the express teaching of switching between incompatible message formats without endpoint intervention. A POSITA would recognize that any gateway bridging a trunked-radio/RF link to a wired LAN must by definition translate between formats that are not interoperable.
• Automatic, transparent operation (aspect 4's method). — Gray's group controller performs message-based mobility management over the data-link layer independently of the switching controller; Erickson's switching is "without the intervention of the endpoint devices."
• Controller-mediated selection / capability matching (for any dependent claim keying on selection). — Wood's controller comparison of capability lists.

Motivation to combine (KSR / MPEP 2143):

1. Same field, same problem. All three are wireless/wireline integration systems addressing how to carry traffic between a mobile unit and fixed infrastructure across a wireless hop.
2. Express problem-recognition in the art. Wood frames the need to select among multiple air-interface types; Erickson frames the need to pass voice and data and to change formats automatically; Gray frames the need to make wireless devices behave as ordinary station sets. Combining them is a "known technique improved" / "use of a known technique to improve a similar device in the same way" (MPEP 2143(A)(4)).
3. Predictable results. A person adding per-link format conversion at each end of a wireless hop obtains the expected result (a transparent data path) with no change in the function of any part. KSR, 550 U.S. 398, 416–17 ("a combination of familiar elements according to known methods… does no more than yield predictable results").
4. Only a finite, identified set of solutions to the transparency problem at the time (central gateway conversion, per-terminal conversion, or hybrid) — KSR's "finite number of identified, predictable solutions."

Caveat on this ground: As drafted, the aspect-1/2/4 subject matter is very close to Doviak '737 itself. Ground I is therefore best understood as establishing that the '920's independent claims add nothing over the parent plus the admitted art — i.e., it is a § 103 rejection aimed at the independence of these claims rather than at their detailed limitations. If the claims in fact contain structural limitations not captured in the specification's "aspects" summary (e.g., specific queuing, session-table, or acknowledge/retry machinery of FIGS. 3–12), this ground must be re-mapped claim-by-claim — which is why the missing verbatim claim text matters.

Ground II — Plural network interface means + synchronization (aspect 3, FIGS. 27–28): Gray in view of Wood, further in view of the admitted routing/failover art

Proposed combination: Gray '843 (claims 1, 6, 7, 8) + Wood '375 + the admitted state of the router art described in the '920's own Background ("Networks may be interconnected by routers which… make logical decisions about the pathway… based upon a variety of preference metrics," including the "virtual router" formed by two routers loosely-coupled through a protocol-neutral data-link).

Element-by-element logic:

• "A plurality of network interface means connected by a local network." — Gray claim 6 recites a LAN subsystem comprising a plurality of LANs interconnectable through the central switch, with a LAN controller establishing connections between the LANs; Gray's FIG. 3 shows the base stations/controllers on a shared virtual LAN. That is a plurality of coordinating interface elements joined by a local network.
• "Synchronization means for synchronizing… routing tables and health and status information." — Gray claim 7 recites "a plurality of said group controllers dividing the task between them" for mobility management; the '920's spec describes exactly the corresponding function ("keeping track of all routing tables and health and status information associated with both… remote network controllers" and periodically sharing them). Gray additionally discloses monitoring the operation of each channel… responsive to received control packets and switching to the secondary channel when the active channel degrades — i.e., a health/status-driven failover.
• Dual-ported host controllers and port switching (dependent claims). — Gray's central-switch architecture and channel-monitoring/hand-off mechanism, combined with the admitted routing art, make port-level failover an obvious design choice; the '920's own text concedes the purpose is merely "greater reliability."

Motivation to combine: The specification states the object is redundancy/reliability ("This configuration provides a redundant system which insures greater reliability"). Redundancy by duplicating a controller and periodically exchanging tables/health data is one of the most conventional engineering responses in networking — a textbook "obvious to try" / "design incentive" combination (MPEP 2143(A)(2) and (C)). Gray supplies the multi-controller divide-the-task teaching; Wood supplies the controller-based decision architecture; the admitted router art supplies routing tables and preference metrics. The result is a predictable aggregation of known elements (KSR at 416–17).

Strength/weakness: This ground is strong if Gray is § 102(b) art (i.e., effective date 2000). If all claims get 1995-06-01 and Gray is available only under § 102(e) — or not at all — the ground survives but with a thinner primary reference, and the opponent should substitute the parent Doviak disclosure plus art pre-dating 1995-06-01 for the redundancy aspects.

Ground III — The Router: dynamic selection among a plurality of incompatible networks (aspect 5, FIGS. 29–35): Wood in view of Leopold, further in view of Gray, and further in view of Erickson

This is the ground that most directly attacks the patent's own stated point of novelty, because the specification identifies the novelty as metric-driven selection among the transporting networks rather than endpoint capability matching.

Element-by-element logic:

Motivation to combine:

1. The references themselves supply the design incentive. Wood's stated need is to choose among multiple air interfaces; Leopold's stated rationale is capacity utilization (prefer the most subordinate/highest-capacity system); Gray's rationale is maintaining a usable link. Combining them answers the POSITA's ordinary objectives — maximize throughput, avoid degraded links, keep the session alive.
2. Extremely strong "finite number of identified, predictable solutions." A POSITA seeking to pick a network from several available RF systems has essentially a closed menu of selection inputs: endpoint capability (Wood), system capacity/coverage (Leopold), or link quality (Gray). Selecting among them per user-set priority yields exactly the predictable result claimed — this is the KSR "obvious to try" paradigm at 550 U.S. at 421.
3. Interoperability pressure. Once an application needs to run over more than one RF infrastructure, adding a decision layer plus a switch is the natural (and, per the '920's own architecture section, conventional) engineering step.
4. Networking arts are combinable with wireless arts here because the problem (route selection based on link metrics) is common to both, and the '920's Background itself invokes the generic router and "virtual router" as known.

Ground IV — The fallback ground if priority is broken

If any claim is held not entitled to 1995-06-01 (e.g., because the Router/failover subject matter was added later), then:

• Doviak '737 / 08/456,860 is no longer shielded by § 103(c)'s common-ownership carve-out for the later-added claims to the extent the shield, and more importantly it ceases to be a benefit-conferring parent for that new matter; it then functions as § 102(b) art against those claims (issued 1998-02-10, more than one year before 2000-03-16).
• Combined with Gray (1997), Wood (1995), Erickson (1995) and the intervening 1996–1999 art, a single-reference or two-reference § 102/§ 103 case becomes available against the Router claims.

IV. Rebuttal arguments the patent owner would raise, and how they fare

V. Bottom line

1. Strongest single ground: Wood + Leopold + Gray (+ Erickson) against the Router claims (aspect 5). Leopold supplies the exact element the specification brands as novel (preference metric of the transporting systems rather than endpoint capability), Gray supplies quality-driven reselection and the "ordinary station set" transparency, and Erickson supplies format switching without endpoint intervention. Motivation: capacity utilization (Leopold), link maintenance (Gray), transparent service continuity (Erickson); result predictable; solution set finite — squarely KSR.
2. Strong ground against the transparency/method claims (aspects 1, 2, 4): Gray + Erickson + Wood, with the further observation that Doviak '737 (the parent) already discloses most of this, so the independent claims add only what the admitted art supplies.
3. Ground against the redundancy/synchronization claims (aspect 3): Gray (claims 1, 6, 7, 8) + Wood + the admitted routing/virtual-router art, motivated by the stated reliability objective; its force depends heavily on the effective filing date, because Gray is only cleanly § 102(b) art if those claims are not entitled to 1995-06-01.
4. The decisive issue is not technical — it is the priority date. The '920 issued 2001 on a 2000 filing claiming 1995-06-01, and its Background discusses references (Gray, 1997; and Leopold) that post-date or are ambiguous relative to 1995-06-01. Resolving the continuation chain behind 09/527,014 — and whether FIGS. 27–35 added new matter — would determine whether Gray/Leopold are full § 102(b) art or restricted (or unavailable) under § 102(a)/(e).

VI. Confidence and open items (stated rather than guessed)

• High confidence: the identities/content of Wood '375 and Gray '843 (verified against the primary documents above), the pre-AIA framework, and the specification's own characterizations of each reference.
• Unverified: the verbatim claim language, numbering and dependencies of US 6,198,920 (fetched text truncated before the claims) — all claim mapping above is at the "aspect" level, carried forward from the earlier Patent Summary; the filing/issue dates of Erickson '574 and the Leopold et al. reference and the number of the Leopold patent; Gray's underlying filing date (which controls whether it is § 102(e) art as of 1995); the continuation chain and any new-matter history for 09/527,014; and whether the Router/failover disclosure in FIGS. 27–35 was present in the 1995 parent. I exhausted my search budget before closing these gaps and have not filled them by inference.
• Carried-forward flags: the "Flex Houvig" vs. "Flex Hourig" spelling variance and the April 26 vs. September 27, 2026 date discrepancy from the earlier sections remain unresolved.
• Practical relevance: the patent is recorded as Expired – Lifetime (anticipated expiration 2015-06-01), so this § 103 analysis is of historical/validity-review significance only; no prospective infringement exposure turns on it.