US 6789118

I'll search for authoritative records on this patent number, checking both USPTO-style sources and litigation dockets.

Let me check for litigation/PTAB activity and any Federal Circuit appeals specifically involving this patent.

Let me try to retrieve the actual claims and abstract.

US Patent 6,789,118 B1 — Analyst Summary

Source of record: The full patent text supplied in this session (Google Patents, fetched 2026‑09‑24 from https://patents.google.com/patent/[US6789118](/patent/US6789118)/en) is treated as authoritative and is preferred over search snippets. Supplemental searches were run against USPTO‑adjacent sources (uspto.report, ptacts.uspto.gov, patentimages/EPO, PCT/EP family documents).

1. Bibliographic data

Caveat on inventorship: The US face of '118 names only Ramakrishna Rao. The related PCT/EP "Multi-service network switch" filings list a larger group (Rao, Ghahremani, Young, Murthy, Tseng, Nagaraj, Farid, Schunk). I am reporting the US inventor literally as recorded; do not treat the PCT list as inventorship of '118.

2. Abstract

The abstract text of the US grant was not captured in the fetched full text (the fetch renders the description but the abstract block did not survive extraction). The abstract of the directly corresponding EPO family document (EP 1 225 730 A3, same title, same priority, same applicant/inventor) reads:

"A multi-service network switch capable of providing multiple network services from a single platform. The switch incorporates a distributed packet forwarding architecture where each of the various cards is capable of making independent forwarding decisions. The switch's supports policy based routing where specific routing paths are selected based a domain name, a telephone number, and the like."

This is family-consistent but not verbatim the US abstract — treat as a close proxy, not a quotation of '118.

3. Claim overview — with an explicit gap

I could not retrieve the issued claim text of US 6,789,118. The supplied full text terminates within the description (mid‑passage in the GFI/port‑addressing discussion, around the WKIM address material), before the claims. Searches for the claims returned unrelated patents. I therefore cannot state, with confidence, how many independent claims exist, their exact wording, or their precise recitation. The following is a plain‑language characterization of the invention as claimed appears to be directed, based on the specification's own statement of the invention — flagged as inference:

• Policy-based routing (the title feature). The specification states: "the switch identifies a characteristic of the connection request, selects a router based on the identified characteristic, and forwards a data block originated from the user to the selected router." Independent claim(s) most likely recite a method and/or apparatus performing this sequence: detect/inspect an incoming connection request → identify a characteristic (inlink/channel such as DS0, calling or called telephone number, domain name via login such as "user@isp1.com", source address, destination address, or combinations as search keys) → select a designated next-hop router → forward user data to that router. The specification emphasizes selection "without having to refer to a separate routing table," which would likely appear as a claim limitation if present.
• Characteristic→policy-record retrieval. The disclosure states the identified characteristic "may be used to retrieve a call policy database including call profile information that specifies how the call should be handled," keyed via search key 291 into call policy record 290 (fields include call type 294, service type 296, QoA 298, QoS 300, VR ID 302, VPN ID 304, authentication source 306, DNS 308/310, domain ID 311).
• Possible additional independent claims. Given the breadth of the disclosure, additional independent claims may exist directed to the broader multi-service switch platform (distributed forwarding across forwarding modules, generic forwarding interface, virtual routers/VPNs, QoA-tiered resource allocation, APS). I have no verified basis to enumerate these — this is speculation and should not be relied upon.

Do not quote any claim language of '118 from this summary. If claim text is required, obtain it from the USPTO PatentCenter/Patent Public Search full-text record for US09/511,145.

4. Litigation and CAFC 2026 check

• No Court of Appeals for the Federal Circuit 2026 docket involving patent 6,789,118 was located. Searches of Federal Circuit 2026 materials returned no case naming '118.
• The prominent 2026 WSOU/Federal Circuit matter that surfaced is WSOU Investments LLC v. Google LLC, W.D. Tex. No. 6:20‑cv‑00585, Fed. Cir. No. 24‑1499 — a location-based services patent case in which the Federal Circuit summarily affirmed JMOL of non-infringement in January 2026 (court holding the claimed method steps required a particular sequence). That case does not involve US 6,789,118 and should not be attributed to it.
• WSOU Investments (d/b/a Brazos Licensing and Development) has been active as an assignee-asserting entity across many Delaware and W.D. Tex. suits, and '118 appears in exhibit/reference lists in PTAB petition papers. I found no specific infringement suit, IPR, or appeal that I can confirm asserts or challenges US 6,789,118. Absence of a hit in these searches is not proof that none exists.

5. Plain-language technical overview (description-based, not claim-based)

The patent describes a chassis-based multi-service switch where each slot holds a forwarding module (FM) with its own RISC processors, and a system control module (SCM) hosts route server and management functions. It supports Ethernet, PRI/ISDN, digital modem server, and serial/frame-relay personality modules. Salient described features:

1. Policy-based routing / dial wholesaling — a wholesaling ISP's router can be selected per call based on inlink, phone number, domain name, or addresses, enabling one switch platform to serve multiple wholesale ISPs without separate POPs.
2. Domain database — maps domain name (e.g., ISP1) to a next-hop router address; user login "user@isp1.com" selects the ISP and its authentication (e.g., RADIUS) server.
3. Call policy records — per-profile records keyed by search key, carrying call type, service type, QoA, QoS, VR ID, VPN ID, DNS addresses, and authentication source.
4. QoA/QoS tiering — tiered access by resource-utilization thresholds plus prioritization of packet processing.
5. Dynamic resource allocation — modem/ISDN resources shared across cards via TDM/cell buses, not bound to ports.
6. Virtual routers and VPNs — partitioned per-VR routing tables and filter/session/rule tables.
7. APS / redundancy — 1:1 through 1:4 protection switching, redundant management/TDM/cell buses, RPL backup ports.
8. Generic forwarding interface (GFI) — protocol-transparent internal forwarding with dynamic runtime bonding of layer‑2 protocols to layer‑1 interfaces, functional port addressing (FPA/PPA).

Bottom line

US 6,789,118 is an Alcatel-origin, now WSOU Investments–held patent (1999 priority, filed 2000‑02‑23, issued 2004‑09‑07, expired 2020 for maintenance) titled Multi-service network switch with policy based routing, naming Ramakrishna Rao as the sole listed US inventor. Two data gaps are material and stated openly: (a) I could not obtain the verbatim US abstract or the issued claim set, so the claim overview above is inferred from the specification's statement of invention; and (b) I found no 2026 CAFC docket for this patent number. If you need claim-exact or docket-exact analysis, the authoritative sources are USPTO Patent Center/Patent Public Search for US09/511,145, PTAB P-TACTS, and PACER/CourtListener for district and appellate dockets.

I'll search for litigation involving US patent 6789118 specifically.

Let me search more specifically for litigation on this patent number.

Let me try more targeted queries on litigation databases.

Search Results: Litigation Involving U.S. Patent No. 6,789,118

Bottom line

I found no litigation specifically asserting U.S. Patent No. 6,789,118. My searches did not identify any district court action, ITC investigation, or Federal Circuit appeal in which this patent appears as an asserted patent.

I want to be explicit about the limits of this answer: I was unable to directly query the Unified Patents litigation portal (https://portal.unifiedpatents.com/litigation/caselist), PACER, or RPX Insight's case database in a way that returned patent-number-specific results for '118. The results I did obtain were keyword matches on "6789118" that turned out to be unrelated (e.g., an FDA adverse-event case number, a TSMC/GlobalFoundries patent designated "'118" that is a different patent — U.S. 7,169,677-class art issued 2007 on corrugated FET structures, and various citation lists). I did not find a complaint, docket, or case list naming US 6,789,118 as an asserted patent. This should be treated as "no litigation found in the sources searched," not as an affirmative verified negative.

Patent ownership context (relevant to litigation risk)

Per the Google Patents record, the chain of title is:

Two points matter here:

1. The current assignee is WSOU Investments LLC (also doing business as Brazos Licensing and Development), an NPE founded by Craig Etchegoyen. WSOU has been one of the most active patent plaintiffs in the U.S., with campaigns against Google, Cisco, HPE, Netgear, Canon, ZTE, Salesforce, F5 Networks, Juniper, Xilinx, and others.
2. The patent expired on 2020-02-23 (20 years from the 1999-02-23 priority date), and Google Patents lists its legal status as "Expired – Lifetime." Any infringement action would have had to be filed on or before that date to capture live damages; the six-year § 286 damages lookback also limits recovery.

WSOU campaigns I identified (none of which I could confirm asserted the '118 patent)

To be clear, these are not cases asserting US 6,789,118 — I list them only to document the search space I covered:

• WSOU Investments LLC v. Google LLC, No. 6:20-cv-00585-ADA (W.D. Tex.) — location-based services patent; JMOL of noninfringement granted, summarily affirmed by the Federal Circuit, No. 24-1499 (Fed. Cir. 2026).
• WSOU Investments LLC v. Google LLC, No. 6:20-cv-00580 (W.D. Tex.).
• WSOU Investments LLC v. Cisco Systems, Inc., No. 6:21-cv-00128-ADA (W.D. Tex.) — patents '014, '216, '859, '106; standing challenge under Rule 12(b)(1).
• WSOU Investments LLC v. Hewlett Packard Enterprise Co., No. 6:20-cv-00730 (W.D. Tex.) — voluntarily dismissed 2022-03-28.
• WSOU Investments, LLC v. Netgear, Inc., Nos. 1:21-cv-01117 and 1:21-cv-01119 (D. Del.) — stipulated dismissal entered 2022-10-28 in -01119; and No. 6:21-cv-00153 (W.D. Tex.).
• WSOU Investments LLC v. Canon, Inc., No. 6:20-cv-00980 (W.D. Tex.).
• WSOU Investments LLC v. Salesforce.com, Inc., No. 6:20-cv-01168 (W.D. Tex.).
• WSOU Investments, LLC v. F5 Networks, Inc. — Federal Circuit No. 25-1505; summary judgment of noninfringement affirmed (April 2025); patents '940, '945, '884, '330.
• WSOU Investments LLC (Brazos) v. Xilinx, Inc., No. 1:20-cv-01228-GBW-JLH (D. Del.).
• WSOU Investments, LLC v. ZTE Corp. et al., Nos. 2:20-cv-0487 through -0497 (W.D. Tex.); IPR2021-00695 through -00699 (PTAB); dismissed.

Recommended next steps to verify

Because I could not exhaustively search the litigation-specific databases, if you need a litigation-grade clearance opinion, I would recommend running:

1. RPX Insight / Unified Patents portal — patent-number search on 6789118 filtered to "Asserted Patents."
2. PACER / CourtListener RECAP — full-text search for "6,789,118" and "6789118" in complaint documents.
3. Lex Machina / Docket Navigator — patent-level case lookup (these are the databases that reliably map asserted patents to cases).
4. USPTO Patent Assignment / litigation records — the WSOU acquisition records (2017) and the Omega/BP Funding Trust security-interest filings may contain schedules listing which patents were involved in enforcement.

Caveat on identifiers

Per your instruction to interpret identifiers literally: note that the search results contained several near-miss identifiers that I did not treat as this patent — including U.S. Patent No. 7,169,677 (the "corrugated FET" patent referred to as "'118" in TSMC's complaint is a different number entirely), U.S. 6,789,110, and others. I have not auto-corrected any of these.

Conclusion: no litigation asserting U.S. Patent No. 6,789,118 was found in the sources searched. The patent is expired and is now held by WSOU Investments LLC, a prolific NPE, but I could not confirm that WSOU or any prior owner ever asserted this specific patent in court.

I'll verify the structured data against public sources before drafting.

Proceedings overview

Total AIA trial proceedings on US Patent 6,789,118: 0. Breakdown by status: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0.

I ran the structured "PTAB proceedings on file" block (canonical USPTO Open Data Portal source) as the primary source and supplemented with web searches for IPR/PGR/CBM filings, PTAB E2E and petition documents, CourtListener, and Federal Circuit coverage naming the patent. Nothing surfaced. The searches returned WSOU Investments enforcement activity and PTAB filings, but every hit concerned other WSOU/Nokia patents (e.g., U.S. 9,450,884; 8,155,012), not 6,789,118.

Bottom-line defensive posture: the absence of PTAB activity is not the same as a hardened patent. This patent is expired — Google Patents records an anticipated expiration of 2020-02-23 (20 years from the 2000-02-23 filing, off a 1999-02-23 priority date) and a legal status of "Expired - Lifetime." No IPR was ever filed because by the time the patent landed in a litigation-focused NPE's hands (WSOU Investments LLC, per the post-2017 assignment chain), its enforceable life was effectively over. For a defendant receiving a demand letter today, the PTAB record is not the main event — the expiration date is.

Per-proceeding detail

No proceedings to report. There are no IPR, PGR, or CBM numbers to list, no judge panels, no institution decisions, no Final Written Decisions, no settlements, and no Federal Circuit appeals arising from an AIA trial on this patent. I am deliberately not supplying proceeding numbers, because none exist in the records I could reach and inventing plausible-looking numbers would be worse than reporting the gap.

Strategic summary

Claim status: all claims UNTESTED in AIA proceedings. No claim of 6,789,118 has been canceled, confirmed, or otherwise adjudicated by the PTAB. There is no narrowing to map and no surviving-claim list to build around, because no trial ever reached a final written decision. (I also want to flag a data limit: the patent text supplied to me does not include the claims section, so I will not state how many claims the patent has or quote claim numbers — that would be fabrication.) The one adjudicative fact I can state with confidence from the record is the expiration: the patent's term ran out on 2020-02-23.

Estoppel landscape. § 315(e)(2) estoppel is a non-issue here — there was no IPR, so no petitioner is estopped and no prior-art ground has been "used up" before the Board. Conversely, there is also no PTAB record to leverage offensively: a defendant cannot point to canceled claims or an adverse FWD. Every invalidity theory remains available in district court (§§ 102, 103, 112, and § 101), unconstrained by § 325(d) or IPR estoppel. Note the timing interaction with § 315(b) is moot — the one-year bar runs from service of a complaint alleging infringement, and no live infringement exposure exists for conduct after the expiration date.

Pattern signals. No petitioner has ever filed against this patent — not a single, let alone multiple, IPRs. The patent owner of record has not pursued any PTAB appeal on it. There is no defensive aggregator (Unified Patents or similar) in the chain: the assignment history runs Alcatel Internetworking → Alcatel → Alcatel Lucent → WSOU Investments LLC (2017-09-25), with security interests recorded to Omega Credit Opportunities Master Fund (2017) and BP Funding Trust Series SPL-VI (2019), both later released. WSOU is an active NPE with a large Waco/E.D. Tex. campaign against Cisco, Huawei, TP-Link and others, and it does litigate aggressively and litigate PTAB appeals on other patents (e.g., WSOU Investments, LLC v. Google, LLC, Nos. 2022-1066, 2022-1067 (Fed. Cir. Sept. 25, 2023) (nonprecedential), affirming indefiniteness holdings on Nos. 7,304,563 and 8,238,681). But that pattern has not touched this patent, and the expiration date explains why.

Why no IPRs exist — the practical reading. IPRs cost six figures and are filed against patents that can still generate damages. A patent that expired in February 2020 offers, at most, a § 286 six-year lookback on pre-expiration conduct. Any complaint filed after roughly 2026-02-23 would capture a damages window of essentially zero. That economics — not claim strength — is the most likely reason no challenger ever petitioned.

Recommended next steps

1. If you have received a demand letter citing 6,789,118: the first check is not the PTAB docket, it is the calendar. Confirm the expiration date against the USPTO Patent Center record (application 09/511,145) and the maintenance-fee history. If the patent expired 2020-02-23 and the demand relates to post-expiration products or services, there is no direct infringement and no ongoing royalty base. Respond on that basis before engaging on claim construction.
2. If the demand relates to pre-expiration conduct: the § 286 six-year damages bar is your primary defense. Compute the accrual cutoff against the date any complaint was or would be filed, and note that a complaint filed on or after 2026-02-23 captures no recoverable damages period at all.
3. Because no IPR exists, you are not estopped — you may raise any § 102/§ 103 art, invalidity under §§ 112 and 101, prosecution-history estoppel, and any laches/§ 286 limitations defense without Board-imposed constraints. There is no FWD to link to because no FWD was ever issued; I am not able to provide a PTAB E2E or CourtListener link for a decision that does not exist.
4. Verify independently before relying on this. My conclusion rests on the ODP structured block plus web searches that returned no PTAB filings naming this patent. PTAB E2E searches by patent number, and the patent's "Legal Events"/PTAB tabs on Google Patents and Patent Center, are the fastest confirmatory checks. If a recently filed petition exists that neither ODP nor my searches surfaced, the analysis above changes materially — the expiration point would still be the lead defense, but a live proceeding would add a parallel track.
5. Do not characterize this patent as "hardened." That framing would be wrong and risky. Its practical immunity from PTAB challenge is a function of expiration, not of surviving attack. If anything, the untested claim set is a blank slate for district-court invalidity work — and the claims have never faced an adversarial obviousness challenge on the merits.

I'll research the assignment record before writing. Let me search for the specific records.

Good — I have concrete reel/frame leads. Let me dig for this patent's specific records and the security-interest chain.

Strong leads. Let me verify this patent's specific chain and the later security interests.

Research limitations up front

Google Patents' legal-events table for US 6,789,118 (supplied in the source text) gives dates and conveyance types but no reel/frame and no correspondent. I could not load the Assignment Center abstract of title for this patent number directly. I therefore reconstructed the chain from (a) the Google Patents legal events for this patent, (b) USPTO assignment abstracts of title reproduced verbatim in federal court filings (storage.courtlistener.com W.D. Tex. 6:21-cv-00128 and 6:20-cv-00980) and PTAB papers (ptacts.uspto.gov), and (c) the Alcatel Lucent → WSOU executed assignment documents filed in those cases. Reel/frame numbers below are attested at portfolio level; the mapping of US 6,789,118 into those reels is inferred from exact recording-date matches and flagged as such. Where I have no reel/frame, I say so rather than guess.

Inventors

• Single-inventor patent. Rao executed an assignment of assignors' interest to Alcatel Internetworking, Inc., recorded 2000-06-12 — roughly 3.5 months after the 2000-02-23 filing. That timing is a normal employee-assignment recordation, not a departure signal.
• No evidence of inventors leaving the original assignee within 12 months of filing. There is nothing in the record suggesting a pre-sale inventor exodus; the 2017 divestiture was a corporate portfolio sale, not a founder/inventor-driven one. I could not locate Rao's later employment history, so I make no claim either way.

Original assignee

• Named on the issued patent: Alcatel S.A. (Google Patents "Original Assignee"; the 2000 inventor assignment ran to Alcatel Internetworking, Inc., and rights were then assigned to Alcatel in 2002 — so "Alcatel S.A." on the face of the patent reflects the post-2002 owner).
• Product embodying the claims: Yes — this is a hardware/architecture patent, not a paper patent. The claims cover a physical multi-service switch chassis with interchangeable forwarding modules and personality modules (Ethernet switch PM, PRI PM, digital modem server PM with a 32/64-modem pool, serial data interface PM) on a redundant management/TDM/cell-bus backplane, plus policy-based routing of dial connections. That maps to Alcatel Internetworking's carrier multi-service access/aggregation switch line sold to ISPs. I could not confirm the exact marketing name of the product from sources reviewed (Alcatel Internetworking was the former Xylan Corp., acquired by Alcatel in 1999), so I am not naming a product number.
• Primary line of business: Carrier and enterprise data networking equipment (routers, LAN/WAN switches, remote-access concentration).
• Current status: Operating, but only as a subsidiary. Alcatel merged with Lucent in December 2006 to form Alcatel-Lucent; Nokia's acquisition of Alcatel-Lucent completed in 2016. Alcatel-Lucent S.A. is now inside Nokia and the brand is largely retired. Not in bankruptcy; not dissolved. Notably, the US 6,789,118 portfolio was sold out (see below) while the corporate entity kept operating — a divestiture, not a fire-sale.

Assignment timeline

Chronological. Where Google Patents shows the event but no reel/frame, that is stated explicitly.

1. 2000-02-23 (filing) — no assignment recorded; inventor's rights unrecorded at this point

   • Conveyance: n/a (application filed)
   • Context: original filing by Alcatel Internetworking, Inc.

2. 2000-06-12 (executed date not exposed) / recorded 2000-06-12 — Reel/frame not exposed by Google Patents legal events; not verified

   • Conveyance: Assignment of assignors' interest
   • Assignor: Rao, Ramakrishna
   • Assignee: Alcatel Internetworking, Inc.
   • Correspondent: not available (Google Patents does not publish correspondent for this event)
   • Context: standard employee invention assignment to the operating employer.

3. 2002-11-25 (executed date not exposed) / recorded 2002-11-25 — Reel/frame not exposed; not verified

   • Conveyance: Assignment of assignors' interest
   • Assignor: Alcatel Internetworking, Inc.
   • Assignee: Alcatel
   • Correspondent: not available
   • Context: intra-group reorg moving the patent up to the Alcatel parent.

4. 2006-12 (merger effective) — no assignment record located

   • Conveyance: none recorded for this patent
   • Context: Alcatel/Lucent merger creates Alcatel-Lucent. Chain gap: the 2017 assignor of record is "ALCATEL LUCENT," yet no Alcatel → Alcatel-Lucent change-of-name or merger record appears in Google Patents' event list for US 6,789,118. Either the change-of-name record was made at the entity level and not indexed per-patent, or the recordation is incomplete. Flag this as an unverified link, not as a fabricated one.

5. 2017-09-21 (executed 2017-08-22) / recorded 2017-09-21 — Reel 043966/0574 (portfolio-level record; US 6,789,118's inclusion inferred from the exact 2017-09-21 recording-date match in Google Patents)

   • Conveyance: Security Interest (see document for details)
   • Assignor: WSOU Investments, LLC
   • Assignee: Omega Credit Opportunities Master Fund, LP, c/o Omega Advisors, Inc., 810 7th Avenue, 33rd Floor, New York, NY 10019
   • Correspondent: Erick J. Palmer, P.O. Box 2828, Chicago, IL 60690-2828. Recurrence flag: Palmer is also the correspondent of record on the later OCO release in the same reel/frame family — a lender-side agent address (a bare Chicago P.O. box) used repeatedly across WSOU's secured-financing records.
   • Context: securitization — WSOU pledges the newly acquired portfolio as collateral. Note the grant was recorded four days before the underlying patent assignment was recorded (see item 6), i.e. WSOU encumbered assets it had not yet recorded as owning.

6. 2017-09-25 (executed 2017-07-22) / recorded 2017-09-25 — Reel 044000/0053 (portfolio-level record; US 6,789,118's inclusion inferred from the exact 2017-09-25 recording-date match)

   • Conveyance: Assignment of assignors' interest
   • Assignor: Alcatel Lucent (Alcatel Lucent SAS; co-sellers under the same transaction included Nokia Technologies Oy and Nokia Solutions and Networks B.V.)
   • Assignee: WSOU Investments, LLC, 11150 Santa Monica Blvd., Suite 1400, Los Angeles, CA 90025 — recorded as a Delaware LLC
   • Correspondent: Burdick Patents, 2526 W. State Street, Boise, ID 83702 — attorney Sean D. Burdick, USPTO Reg. No. 51,513, signing as "IP Counsel for WSOU Investments, LLC." Recurrence flag: Burdick is correspondent on this WSOU acquisition record and is the signatory/correspondent on multiple WSOU 37 CFR 3.73(c) ownership statements (e.g. for US 8,767,614, and a 2022 statement expressly citing "Reel 044000, Frame 0053"). Same lawyer, same reel, multiple filings — a repeat-player correspondent.
   • Context: transfer-to-asserter — bulk portfolio sale (~2,979 properties in 238 pages in a single recording), routed through "Wade and Company," which assigned its interest in the Patent Purchase Agreement to WSOU effective 2017-08-21.

7. 2019-05-20 (executed date not exposed) / recorded 2019-05-20 — Reel/frame not exposed by Google Patents; not verified

   • Conveyance: Security Interest
   • Assignor: WSOU Investments, LLC
   • Assignee: BP Funding Trust, Series SPL-VI
   • Correspondent: not available
   • Context: securitization / refinancing — the portfolio is pledged to a trust-series vehicle, replacing or layering over the Omega facility.

8. 2019-05-21 (executed date not exposed) / recorded 2019-05-21 — Reel/frame not exposed; not verified

   • Conveyance: Release by Secured Party
   • Assignor: OCO Opportunities Master Fund, L.P. (f/k/a Omega Credit Opportunities Master Fund LP)
   • Assignee: WSOU Investments, LLC
   • Correspondent: not available for this event; Erick J. Palmer is the correspondent on the OCO release documentation in the same reel/frame family per the W.D. Tex. abstract of title
   • Context: securitization lifecycle — Omega loans repaid/released the day after the BP Funding Trust lien was recorded; effectively a roll-over of the collateral facility.

9. 2020-02-23 — anticipated expiration (Google Patents legal status). Patent term ended; the asset is expired.

10. 2021-06-03 (executed date not exposed) / recorded 2021-06-03 — Reel/frame not exposed; not verified

    • Conveyance: Release by Secured Party
    • Assignor: Terrier SSC, LLC
    • Assignee: WSOU Investments, LLC
    • Correspondent: not available
    • Context: securitization lifecycle — final lien release; WSOU emerges as unencumbered record owner of an already-expired patent.

Bottom line: the Assignment Center is not empty for this patent. There is a full recorded chain, with two directly corroborated reel/frame numbers (043966/0574 and 044000/0053) and a name-level break in the chain around the 2006 Alcatel-Lucent merger that I could not verify.

Timeline diagram

timeline
    title Ownership of US 6789118
    1999 : Priority application filed
    2000 : Application filed by Alcatel Internetworking
         : Inventor Rao assigns rights to employer
    2002 : Rights assigned up to Alcatel
    2004 : US 6789118 issued
    2006 : Alcatel merges with Lucent
    2017 : WSOU grants security interest to Omega
         : Alcatel Lucent assigns portfolio to WSOU
    2019 : BP Funding Trust security interest recorded
         : Omega security interest released
    2020 : Patent term expires
    2021 : Terrier SSC release recorded

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT. Reel 044000/0053 (recorded 2017-09-25) moves ~2,979 properties from operating company Alcatel Lucent SAS to WSOU Investments, LLC, a Delaware licensing LLC at 11150 Santa Monica Blvd., Suite 1400, Los Angeles. This is not naming inference: W.D. Tex. briefing and PTAB filings state WSOU "has never employed any inventors at the time of the purported inventions … did not research or develop any of the technology claimed by its patents," and that "its activities to date have consisted of monetization of the patents through licensing and filing lawsuits." WSOU's own declarant describes it as "WSOU Investments, LLC d/b/a Brazos Licensing and Development," Waco, Texas.

2. Known asserter in the chain — PRESENT. WSOU Investments, LLC is the current record owner per Google Patents and per Reel 044000/0053. It is named as an NPE in defendant briefs and reported at 187 district-court cases in 2020 — ~5% of all US patent litigation and the most active single assertor that year, averaging roughly five new filings per week. It is a high-frequency plaintiff of the type tracked by Unified Patents/RPX. Defendant briefing further identifies WSOU as founded by Craig Etchegoyen (also founder of Uniloc, itself an NPE repeatedly found to lack standing for want of exclusionary rights). None of the enumerated lists in your prompt (Acacia, Marathon, IV, Wi-LAN, etc.) apply here; WSOU is a distinct high-frequency assertor rather than a match on those lists.

3. Repeat correspondent across the chain — PRESENT. The chain is small but the lawyer recurs exactly as your hypothesis predicts:

   • Sean D. Burdick, Burdick Patents, Boise, ID, Reg. No. 51,513 — correspondent of record on Reel 044000/0053 (the Alcatel Lucent → WSOU acquisition) and signatory "IP Counsel for WSOU Investments, LLC" on multiple 37 CFR 3.73(c) ownership statements, including one expressly reciting "Reel 044000, Frame 0053." Recurrence is documented, not inferred.
   • Erick J. Palmer, P.O. Box 2828, Chicago, IL 60690-2828 — correspondent on the WSOU → Omega security interest (Reel 043966/0574) and on the related OCO release record in the same reel family, i.e. the same Chicago agent box handling both the encumbrance and its discharge.
   • Caveat you asked for: Burdick Patents does ordinary prosecution work too, so a single appearance alone would not be a finding. Here he appears on the acquisition record and as WSOU's own chain-of-title signatory — that is recurrence.

4. Cascading transfers — PRESENT. Three portfolio-scale records land within nine days in September 2017 on adjacent reels: 043953/0822 (Nokia Technologies → WSOU), 043966/0574 (WSOU → Omega security interest, recorded 2017-09-21), 044000/0053 (Alcatel Lucent → WSOU, recorded 2017-09-25). The collateral grant was recorded before the ownership record — the exact sequencing defect defendants attacked in WSOU v. Cisco (6:21-cv-00128) and WSOU v. Canon (6:20-cv-00980). Further refinancing churn follows in 2019 and 2021 (BP Funding Trust / OCO / Terrier SSC), i.e. continued chained securitization over the same asset pool.

5. Pre-litigation transfer — UNCLEAR / not established for this patent. The 2017 assignment long predates WSOU's 2020-2021 filing wave against Cisco, Google, Huawei, ZTE, Canon, TP-Link, Arista, etc. But I found no litigation asserting US 6,789,118 itself; WSOU's asserted patents in the cases I reviewed are different numbers ('106, '014, '216, '859, '133, '346, '211, '961, '4180). So there is no 6-month pre-suit transfer to point to for this asset. Marking this signal present would be fabrication.

6. Bankruptcy fire-sale — NOT PRESENT. The seller side (Nokia / Alcatel Lucent SAS) was solvent and operating throughout; the 2017 transaction was a negotiated Patent Purchase Agreement, not a Chapter 7/11 § 363 sale. No bankruptcy proceeding appears anywhere in the chain.

7. Privateering — PRESENT (structural); coordination UNCLEAR. The structure is textbook: an operating vendor (Nokia, via Alcatel Lucent SAS and Nokia Solutions and Networks B.V.) divests patents to a licensing-only vehicle (WSOU), which then sues that vendor's competitors — Cisco, Huawei, ZTE, Arista, Canon, Google, TP-Link — the "Cisco's Motion" language explicitly frames WSOU's business as "threatening and filing lawsuits based on patents that it purportedly acquired from Nokia and related entities." However: I found no license-back, revenue-share, or retained-enforcement-interest evidence in the sources reviewed. So the divestiture-to-asserter structure is documented; Nokia's direction or participation in the assertions is not.

8. Defensive aggregator (anti-NPE) — NOT PRESENT. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. It terminates at WSOU Investments, LLC — an active asserter. The 2021 Terrier SSC "release" is a lien discharge, not an acquisition by a defensive aggregator; do not misread that event as neutralization.

Verdict

NPE — high confidence.

The patent left an operating vendor for a licensing-only Delaware LLC via a bulk Alcatel Lucent → WSOU Investments portfolio assignment (Reel 044000/0053, executed 2017-07-22, recorded 2017-09-25), the assignee is a publicly documented high-frequency NPE (≈187 suits in 2020 alone, ~5% of all US patent litigation, per W.D. Tex. briefing), and the transaction was immediately encumbered by a WSOU-granted security interest recorded days earlier (Reel 043966/0574, executed 2017-08-22, recorded 2017-09-21) with a further 2019 BP Funding Trust lien and 2019/2021 releases — i.e. collateralized patent monetization. Two independent strong signals (shell/asserter transfer + known high-frequency NPE with repeat correspondent Sean D. Burdick / Burdick Patents on the chain-of-title records) drive the call; the only caveats are that the 2006 Alcatel → Alcatel-Lucent link has no locatable per-patent record and that US 6,789,118 appears never to have been asserted before expiring on 2020-02-23.

Verify the records at: https://assignmentcenter.uspto.gov/ (search patent number 6789118) and https://patents.google.com/patent/US6789118/en — the two reel/frame numbers above, 043966/0574 and 044000/0053, are the entries to check first.

I'll run targeted searches to (a) pull the USPTO record for 6789118, (b) retrieve its front‑page patent citations, and (c) attempt the claim text that was missing from the earlier sections.

Let me pull the USPTO-side record and the Google Patents citation/timeline data specifically.

I have the family and litigation context. Now I need the actual front-page citations for US 6,789,118. Let me pull the EPO search report for the family (EP 1 225 730) and the Google Patents citation listing.

Scope note before the analysis

Two housekeeping items, stated rather than silently resolved:

1. Date discrepancy. The task header says "Current Date: April 26, 2026," while my operating context gives today as 2026‑10‑01. I've used the operating date for status statements and flag the conflict rather than picking one silently.
2. Coverage gap carried forward. The prior section of this analysis already flagged that the issued claim text and the US abstract of '118 were not captured. That gap has a direct consequence here: § 102 anticipation mapping is claim‑by‑claim work, and I still cannot cite verbatim claim numbers. I will therefore give the references with high confidence and the claim mapping as reasoned inference, labeled as such.

1. Confirming the record for US 6,789,118 (and excluding look‑alikes)

Searching "6789118" returns at least three distinct documents that are not this patent. Per your instruction not to auto‑correct identifiers, I note them explicitly so they are excluded from the prior‑art analysis:

The Google Patents record for the target (the authoritative corpus supplied earlier) does not render a "References Cited / Patent Citations" block — the fetch reproduced the classification tree, the definitions/description, and nothing in the claim or citation apparatus. So I could not read the US front page's citation list directly. What follows is the best verifiable evidence I could assemble, with its source and its limits stated.

2. What I could verify — the citation evidence that exists

2.1 The family search report (closest verifiable "patent citation" evidence)

The direct EP family member is EP 1 225 730 A2/A3 (App. 01128311.6, filed 2000‑02‑23, applicant Alcatel Internetworking, Inc., inventor Rao, same 31 US provisional priorities of 1999‑02‑23, title Multi-service network switch with policy based routing). Its A3 supplementary search report carries a Box C "Documents Considered to be Relevant" list. Source: EPO Global Patent Index record for EP 1225730 A3 (http://data.epo.org/gpi/EP1225730A3-Multi-service-network-switch-with-policy-based-routing.html) and the A3 front page (patentimages EP1225730A3.pdf).

Interpretation of the examiner categories (this is the § 102‑relevant signal):

• [X] = "particularly relevant taken alone" → the examiner considered US 5,550,816 and WO 98/39879 to individually anticipate the independent claims of the EP family counterpart. That is the closest thing to an official anticipation finding touching this disclosure.
• [E] = earlier application published on/after the filing date → a novelty‑only reference (EPC Art. 54(3)), the analogue of pre‑AIA § 102(e). It cannot be combined for obviousness.

Critical caveat: the EP claims are not the US claims. An [X] finding against an EP independent claim does not establish anticipation of any specific claim of US 6,789,118; it establishes that the same general subject matter was already in the art, per the EPO.

2.2 Forward citations of '118 — NOT prior art

These are patents that cite '118. They are not prior art against '118 (they post‑date it) and I list them only for completeness because "each patent citation for 6789118" can be read either direction.

2.3 '118 being used as prior art by later litigants (the most probative new finding)

'118 (or a Rao document with the identical specification) is being asserted as prior art in PTAB proceedings:

• Sable Networks, Inc. v. Splunk Inc., No. 21‑cv‑00040‑RWS (W.D. Tex.) — invalidity claim charts for U.S. Patent No. 7,630,358 quote a "Rao" reference (EX‑1004), at line cites 23:31‑34, 23:26‑38, 24:4‑8, 14:4‑5, 14:34, and 30:32‑36. Those passages map onto the specification text in our authoritative corpus ("each VR has an instance of an IP protocol stack and its own routing table 70 for routing protocols including RIP, OSPF, GBP4…"; "The forwarding table also includes a flag field 100 for describing the types of route"; the GFI definition). The same briefs reference an EX‑2006 "Rao '358 Patent Considered Reference."
• A Fortinet IPR exhibit index (PTAB, petition 1558605) lists "6789118" among its numbered reference set.
• Hendel (a "multi‑layer distributed network element / MLDNE") appears in the same IPR family as the distributed‑forwarding prior art (IPR re: U.S. 7,224,668).

⚠️ Flag for internal consistency: I cannot confirm that "EX‑1004 / Rao" is specifically US 6,789,118 rather than its specification‑identical sibling US 6,674,756 (Multi‑service network switch with multiple virtual routers, Rao et al.). The quoted text is common to both. Treat the line‑cite mapping as unverified. I raise it because the earlier section asserted '118 "appears in exhibit/reference lists in PTAB petition papers" — that is now confirmed for the Fortinet list, but the more substantive "Rao as anticipatory art" identification remains ambiguous.

2.4 Family members that are not prior art (important trap)

These share the specification and cannot be used as § 102 art against '118 on the same subject matter.

3. § 102 mapping — what can and cannot be said

What the authoritative corpus does now tell me about '118's claim vocabulary (new since the prior section): the Google Patents "Prior art keywords" extraction for '118 is "connection request · switch · characteristic · router · port." That field is machine‑derived from the claim language. It is the strongest internal evidence yet available that independent claim 1 recites, at minimum: a switch / receiving a connection request / identifying a characteristic / selecting a router / a port. This upgrades the earlier section's inference (which was drawn only from the specification's statement of invention) to something corroborated by claim‑derived metadata — though it is still not verbatim claim text.

Explicit non‑findings (stated, not glossed):

• I could not retrieve the US front‑page "References Cited" list of 6,789,118 from PatentCenter, Patent Public Search, the printed patent PDF, or uspto.report in this session. The US examiner‑cited set may differ from the EPO set above, and I am not representing the three items in §2.1 as the US front‑page citations.
• I therefore cannot assert that US 5,550,816 and WO 98/39879 are cited on the face of US 6,789,118, nor that no other US references were cited.
• I cannot name a single reference and a single claim number with the confidence a § 102 assertion requires, because the claim text is unavailable and no reference has been read in full.

4. Bottom line

• Target confirmed and de‑duplicated: US 6,789,118 B1, Rao, Alcatel‑origin/WSOU‑held, 1999 priority, expired 2020; the JP 6789118 and the Mirena "6789118" hits are unrelated and excluded.
• The best verifiable prior‑art signal touching this disclosure is the EP 1 225 730 A3 search report: [X] US 5,550,816 A (1996‑08‑27) and [X] WO 98/39879 A1 (1998‑09‑11), plus [E] WO 00/33204 A1 (2000‑06‑08). Those are EPO anticipation candidates against the EP counterpart's independent claims, which are the presumptive analog of '118 claim 1 — but they are not confirmed as the US front‑page citations.
• Forward citation set (US 7,050,430; 9,413,660; 7,735,114; 11,362,987) is a citing‑art set, not prior art.
• '118 is itself being used as prior art in later PTAB/IPR work (Sable v. Splunk charts; a Fortinet exhibit list) — but the "Rao" identification in the Sable charts is ambiguous as between 6,789,118 and 6,674,756.
• The § 102 claim‑by‑claim deliverable is not achievable from available sources. To close it, pull (a) the printed US 6,789,118 B1 front page for the exact "References Cited," (b) the full claim set from Patent Public Search / PatentCenter for App. 09/511,145, and (c) full texts of US 5,550,816 and WO 98/39879 to test the [X] findings against '118's actual independent claim.

I'll start by trying to identify the actual prior-art references listed on the Google Patents page and the issued claim text, since the supplied full-text extraction did not include those sections.

Important early finding: the EP family search report for EP 1 225 730 A3 contains the closest prior-art citations, and I also see a near-neighbor patent-number trap (US 6,779,118 is a different patent). Let me pin down the actual references and the claim text.

Scope note before the analysis

Two things must be stated up front, because they govern the confidence level of everything below.

1. The "Prior Art" panel of the Google Patents page was not captured in the supplied full text. The fetch I was given renders the bibliographic block, the description, the classification tree, and the CPC "Information" list — but no "Citations / Cited By / Similar Documents" table. What did survive, and what I have used, is the page's own prior-art metadata block: prior art keywords = "connection request, switch, characteristic, router, port." That string is itself probative of claim scope and I rely on it.
2. To fill the gap I substituted the search reports for this patent's own family, which are the closest thing to an authoritative prior-art record for the same disclosure: EP 1 225 730 A3 (same title, same applicant, inventor Rao) at https://patentimages.storage.googleapis.com/f2/5e/27/fc21854a77f3a8/EP1225730A3.pdf, and WO 00/51290 A3 (PCT/US00/04717, "Multi-service network switch") at https://patentimages.storage.googleapis.com/67/94/95/9762ce0bf7e5a7/WO2000051290A3.pdf. These are family documents, not the '118 US record itself, and I flag that transfer.

The material gap from the prior section is still open: I do not have the verbatim issued claims of US 6,789,118. The full text terminates mid-description (in the WKIM port-addressing passage). An obviousness analysis that cannot quote the claim is a framework, not an opinion of invalidity. I have built the framework below and marked every element-level assertion as verified / inferred / unverified.

Number trap — flagged explicitly. Search results in this space are saturated with US 6,779,118, a different patent (Willens; "network access control"; asserted by LinkSmart Wireless Technology; reexamined, Ex parte Linksmart Wireless Technology, LLC, Appeal No. 2011‑009566 — see https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1519835](/patent/1519835)/...), and with US 7,630,358 (Sable Networks v. Splunk, e.g. https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1548481](/patent/1548481)/...). Neither is US 6,789,118. Any chart, reexam, or invalidity exhibit labeled "'118" that discusses RADIUS/redirection of dial-up users or logical-router link advertisement belongs to those patents, not this one. I have not merged them.

1. Legal framework applied

US 6,789,118 has a 1999‑02‑23 priority and a 2000‑02‑23 filing, so pre‑AIA 35 U.S.C. §§ 102/103 govern, including pre‑AIA § 102(e) (published PCT applications designating the US and published in English are prior art as of their international filing date). The analysis uses Graham v. John Deere (scope and content of prior art; differences; PHOSITA level; secondary considerations) and KSR Int'l v. Teleflex (predictable combinations; known techniques; design incentives; "a court must ask whether the improvement is more than the predictable use of prior art elements according to their established functions").

Proposed PHOSITA (inference, dated to Feb 1999): a B.S. in EE/CS plus 2–4 years in data networking, or an M.S. plus 1–2 years, with working knowledge of IP routing (RIP/OSPF/BGP4), dial access aggregation (PPP, HDLC, ISDN PRI/T1), RADIUS (RFC 2058/2138), and ATM/frame relay encapsulation (RFC 1483/1490). This is a routine-implementation art, which lowers the inventiveness bar under KSR.

Claim scope (inferred, not verified). The EP search report for the family lists claims 1–14 as the relevant claim set, and the Google Patents prior-art keyword block for this record is "connection request / switch / characteristic / router / port." Taken together, the independent claim most plausibly recites the sequence the specification itself states as the invention: identify a characteristic of the connection request → select a router based on the identified characteristic → forward a data block originated from the user to the selected router (specification, Definitions block). I proceed on that construction and label every mapping accordingly. Do not quote claim language of '118 from this document.

2. The prior-art record actually available

Two honesty flags on this table. (a) The claim numbers 62/63/64/65 belong to the broader parent application WO 00/51290, which claimed the whole multi-service switch platform; the '118 continuation was examined on a narrower set (claims 1–14 per the EP record). I therefore use the WO 00/51290 citations only as evidence of what the platform claims were confronted with, not as citations against '118's claims. (b) I could not independently verify the internal disclosures of US 5,550,816, WO 98/39879, WO 99/00945, or US 5,740,175 in this session; the column above reproduces the examiner's designations and the examiner's own pinpoint passages. Where I describe content below, I say so only where a search hit independently corroborates it.

Independently corroborated content: a PTAB petition excerpt describes "Hardwick" as relating "to the operation of virtual switches within physical switching systems that direct the flow of protocol data units," involving a framer 164 and a database 192 that limits access so that a protocol data unit with a destination identifier not owned by a virtual closed user group (186, 188) "will not be delivered," and describing a "Closed User Group paradigm" that permits free communication within the group but restricts communication outside it. That is a logical/virtual router partitioned within one physical system with per-group routing restrictions — which is the same architectural genus as '118's virtual routers and VPN filtering. I treat this as corroborating the character of the Hardwick reference, while noting it appears in a brief adverse to a party relying on Hardwick, so its characterization is advocacy-colored.

3. The single most important observation about this record

The two X references and the E reference were cited by examiners working on this patent family, against this disclosure. That does not by itself prove the '118 claims obvious — an X designation in an EPO search report reflects a novelty/inventive-step-per-EPO judgment against EP claims, which may differ from the US claims, and the EPO's problem-and-solution approach is not § 103. But it means the obviousness case is not something I am constructing from a cold start. Two separate search authorities, on two family members, each independently assembled a small multi-reference combination against claims overlapping the '118 claim set. That is the strongest available signal, and it is a combination signal (note the explicit X+Y pairing at the PCT stage), i.e., the record itself evidences that PHOSITA-level skill was believed to supply the missing link.

4. Grounds of obviousness

Ground 1 — Hardwick in view of WO 98/39879 (claims 1–14)

• Hardwick supplies: a physical switching system partitioned into virtual/logical switches, each with its own set of destination identifiers and its own forwarding behavior, and a mechanism (the framer/database) that selects how a protocol data unit is handled based on attributes of that unit and the group it belongs to. This maps to the "switch that selects a routing path based on a characteristic of the connection request" preamble and to the selection/forwarding steps if the claim is apparatus-oriented.
• WO 98/39879 (Northern Telecom) supplies, per the examiner's citation to Fig. 20 and p. 25:22–32, a network architecture in which a call's attributes are used to determine the destination/route — the examiner treated this alone as X against claims 1–3 and 8–11, i.e., against the core routing and the multi-service-platform claims.
• Motivation (verified from the '118 background itself): the specification's own background states that "[t]he service provider traditionally purchases a separate POP for each outsourced network or reselling service provider," and that ISPs "add separate access servers, access routers, and/or stand alone LAN switches, generally resulting in an increase in cost and management complexities." That is an admitted problem in the field, and it supplies an express design incentive to consolidate per-subscriber routing into one chassis — which is precisely what Hardwick's virtual switches-in-one-physical-system provides. Under KSR, an express incentive in the field, plus a reference teaching the consolidation architecture, plus a reference teaching attribute-based destination selection, is a textbook predictable combination.
• Predictable result: substituting a connection-request attribute as the lookup key for route selection is a substitution of one known selection key for another with no change in the principle of operation of either reference.

Ground 2 — Hardwick in view of the E reference WO 00/33204 (claims 1–14)

• If WO 00/33204 qualifies as prior art under pre-AIA § 102(e) (English-language PCT designating the US, international filing date before 2000‑02‑23 — the examiner's "E" designation tells us only that it published after the EP filing date, which is consistent with an earlier international filing), then the combination of the Hardwick X reference with this E reference is the arrangement the EP examiner specifically flagged against claims 1–14.
• Motivation: the E reference is a Nortel Networks filing (Pillalamarri, Alles et al.) directed to the same access-network/policy problem space; same field of endeavor, same problem, contemporaneous. Common assignee-different-application references are the paradigm case of combinability under KSR ("if a technique has been used to improve one device…").
• Caveat: I have not verified WO 00/33204's international filing date, its designation of the US, or its English-language publication. That verification is required before this ground is usable. If its international filing date post-dates 2000‑02‑23 it drops out entirely as to '118 (it would remain relevant only as evidence of the state of the art generally, not as § 102(e) art).

Ground 3 — WO 99/00945 (Sun) in view of US 5,740,175 (Myers) — platform claims

• The PCT search report built an explicit X + Y combination against the parent's distributed-architecture and dependent claims (X: WO 99/00945 against claims 1, 3, 4, 62, 64, 65; Y: US 5,740,175 against claims 2, 63). An X+Y pairing means the examiner regarded the X reference as teaching the independent claim and the Y reference as supplying a dependent-claim feature that PHOSITA would add.
• Motivation, as recorded by the examiner's own citation practice: a Y citation against a dependent claim is, functionally, a statement that the added feature was a known, available option in the art. That is the KSR "known technique to a known device, ready for improvement" rationale.
• Caveat: I have not verified what WO 99/00945 or US 5,740,175 disclose. US 5,740,175 is cited only for its abstract, which is a thin citation and, standing alone, weak § 103 evidence.

Ground 4 (my own, KSR-based) — the admitted state of the art plus a prior-art access platform

Independent of the family search reports, the claim 1 sequence as characterized has a strong invalidity profile on the face of the '118 specification:

• Known technique 1 — policy-based route selection on ingress attributes. Route selection keyed on inbound interface, source/destination address, or a route-map is the definition of policy-based routing as practiced in IP routers of the period. Selecting a next hop based on the inlink is the exact analogue of matching an inbound interface; selecting based on source/destination address is the exact analogue of a route-map match on IP address; selecting based on domain name via login ("user@isp1.com") is the realm-routing practice of dial aggregation.
• Known technique 2 — retrieving a per-user service profile from an authentication server. The '118 specification itself describes service configuration information (compression, QoA, QoS, VPN ID) being returned by a RADIUS-type server and overriding defaults. RADIUS (RFC 2138, Apr 1997; predecessor RFC 2058, Jan 1997) is a printed publication predating the priority date and supplies, in a single reference, the "characteristic → profile → service handling" chain that the specification relies on.
• Known technique 3 — telephony call-attribute routing. Routing a call based on called/calling number (DNIS/ANI) and on the inbound channel (DS0) is the core competence of the PSTN/IVR switching art and long predates 1999.
• Rationale for combining: the problem the specification identifies (a wholesaler needing policy-driven routing of dial users to a designated ISP router without duplicating POPs) is a business-driven, well-understood objective in an established art, and each element had an established function. Combating "obvious to try": under KSR, where there is a design need or market pressure and a finite number of identified, predictable solutions, the combination is obvious. Here the "solutions" (route on inlink; route on DNIS/ANI; route on login realm) were a finite set of known selection keys.
• Caveat: I have not verified a specific dated Cisco or competitor document in this session for the policy-based-routing point. That document — a 1996–1998 router configuration guide or equivalent — must be produced before Ground 4 is asserted formally. RADIUS RFC 2138 I state from established knowledge and it is independently verifiable, but I did not retrieve it here.

5. The dependent claims (2–14) — where the § 103 case is easiest

The specification's own field inventory predicts what these recite, and each predicted feature has a documented prior-art source:

Dependent-claim obviousness here is stronger than the independent claim, because "apply a known selection key to a known lookup structure" is close to the definition of a predictable improvement. The proviso: dependent claims rise and fall with the independent claim they depend from, so the whole analysis still hinges on the unverified claim 1 language.

6. Secondary considerations and defenses

• No nexus evidence located. I found no evidence of licensing-nexus, unexpected-results, or industry-praise evidence tied to this patent. The 2017 – 2019 assignment chain (Alcatel Lucent → WSOU Investments; security interests to Omega Credit Opportunities and BP Funding Trust) reflects monetization of a portfolio, not commercial success of a product embodying the claims, and carries no nexus.
• Anticipated defense — "teaching away." The most credible rebuttal available to a patent owner is a straight § 103 argument that the combination changes the principle of operation of one reference. The Hardwick material about preventing internal-link disclosure outside a closed user group could be argued to teach away from the claim if the claim requires advertising or exposing per-subscriber routes. That is precisely the fight that was had over other patents in this family line, and it is the defense a patent owner would run. I flag it as the likely battleground rather than dismissing it.
• Rear-view-mirror caution. Because I cannot see the claim language, I cannot exclude that claim 1 contains a limitation (e.g., selection "without referring to a separate routing table," if that phrase was drafted into the claim) that materially narrows the combination. The specification states that effect at page-level, and if it is claimed, the analysis above weakens considerably, because all four grounds select a next hop from data that is functionally a routing table, and the patent owner would argue the claim requires otherwise.
• Expiry relevance. The patent expired for maintenance on 2020‑02‑23. Any § 103 opinion is therefore retrospective (validity as a defense, or portfolio-valuation diligence), not enforcement-facing.

7. What must be verified before this becomes an opinion

1. The issued claim text of US 6,789,118 — USPTO Patent Center / Patent Public Search for application 09/511,145. Everything in § 4 is conditional on this.
2. WO 00/33204's international filing date, US designation, and English publication — controls whether Ground 2 exists under pre‑AIA § 102(e).
3. The actual texts of US 5,550,816 and WO 98/39879 at the examiner's cited passages (cols. 10, 22–26, 34–35 and Figs. 6, 8; p. 25:22–32 and Fig. 20) — controls whether the element mapping in Grounds 1–3 can be written claim-by-claim rather than as a framework.
4. A dated policy-based-routing publication (1996–1998) and RFC 2138 / RFC 2058 — needed to convert Ground 4 from a KSR-framework into an evidenced ground.
5. The Google Patents "Citations / Cited By" table for US 6,789,118 — this is the literal "Prior Art" section the task points to and it was absent from the supplied page text; it should be pulled directly from https://patents.google.com/patent/US6789118/en to check for further US-format § 102(b) art not surfaced by the EPO/PCT reports.

Bottom line: On the available record, the policy-based-routing subject matter of US 6,789,118 sits squarely in the center of a crowded, well-developed 1998–1999 art, and two independent search authorities already assembled small combinations (Hardwick + WO 98/39879; Hardwick + WO 00/33204; WO 99/00945 + US 5,740,175) against overlapping claim sets in the same family. A strong prima facie § 103 case exists for the independent claim as inferred and a very strong one for the dependent claims — but I am not able to certify it, because the issued claims of US 6,789,118 were never retrieved, and the page's own prior-art citation table was not in the supplied text. The two gaps in the prior section therefore remain open, and this section has narrowed, not closed, them.