US 7154909

I'll verify against live sources before summarizing.

Let me run a couple more targeted checks on litigation/docket activity for this number.

US Patent 7,154,909 B2 — Verification Summary

I searched for the specific number 7154909 and cross-checked against the authoritative full-text record at https://patents.google.com/patent/[US7154909](/patent/US7154909)/en. Results confirm the Google Patents bibliographic data. No litigation, PTAB, or CAFC 2026 docket activity referencing US 7,154,909 was surfaced by my searches. I could not directly query PACER or the CAFC docket system, so I cannot affirmatively certify that no such proceeding exists — treat the "no litigation found" statement as a search-result observation, not a legal certainty.

⚠️ Note on search noise (literal ID rule): search hits for "7154909" also returned JP 7154909 B2 (a Japanese packaging-box patent, Rengo Co.) and a Brazilian JUcec corporate-registry entry. These are unrelated documents with a coincidentally matching number and are not the US patent.

Bibliographic Data (US 7,154,909 B2)

Abstract (as issued)

Apparatus, and an associated method, for providing a radio link layer control procedure to maintain the sensitivity of UDP-Lite-formatted data at an RLP (Radio Link Protocol), or other, layer. An indicator generator generates an indication of the sensitivity-type of data that is formatted at the RLP layer. An RLP packet is generated, including the sensitivity data indicator value, and sent to a receiving station. When received thereat, the value of the sensitivity data indicator is utilized and subsequent actions, whether to permit retransmission of the data in the event of erroneous-, or non-, delivery is made.

Plain-Language Overview of the Independent Claims

This patent has two independent claims — claim 1 (apparatus) and claim 17 (method). All other claims depend, directly or indirectly, on one of these two. (Claims 13–16 depend in whole or part on claim 3, which depends on claim 2, which depends on claim 1.)

Claim 1 (apparatus). Setting: a packet system with two formatting layers — a first layer/protocol that can split data into a "first part" and a "second part," and a second layer/protocol that formats the data afterward. The improvement is an indicator generator at the second layer. It produces an indicator whose value tells whether the data handed to the second layer was formatted into those two parts. That indicator is placed into the second-layer data packet. In plain terms: the lower layer tags its packets to flag whether the payload is the sensitive/insensitive split kind of data.

Claim 17 (method). The corresponding method steps: (1) provide the first-layer-formatted data to the second layer; (2) determine which of the first/second part the data belongs to; (3) generate an indicator value based on that determination; and (4) build a second-layer data packet, per the second protocol, that carries the indicator value.

Dependent claims worth flagging (they add the operationally significant limitations):

• Claims 2–6 narrow the first layer to UDP-Lite (sensitive vs. insensitive data) and the second layer to RLP, with the indicator as a Sensitive Data Indicator (SDI) carried in the RLP header.
• Claims 7–12 add a frame count value, a first-frame sequence number, and an SDI_MAP bit-map representation of per-frame sensitivity in the header.
• Claims 13 swaps in an RLC layer (W-CDMA context) instead of RLP.
• Claims 14–16 add a segmenter that divides data into separate second-layer packets based on which part (sensitive/insensitive) it belongs to.
• Claims 18–20 add receiving-side behavior: analyze the indicator and selectively permit retransmission — permitted for first-part (sensitive) data, prohibited where the packet contains only second-part (insensitive) data.

Drafting/Identifier Notes (literal reading)

Two lexical discrepancies exist in the record and should be reported literally rather than "corrected":

1. Claim 9 recites inserting the count value into a SEQ_MAP_COUNT field. The written description (and claim 7's context) consistently names this field SEG_MAP_COUNT. The claim text as issued says SEQ_MAP_COUNT; I am not auto-correcting this.
2. The specification text contains apparent transcription artifacts (e.g., "L ≦ V(R)" for L_V(R); "the data (name is stored in the re-sequencing buffer"). These appear in the fetched source and are flagged for accuracy, not corrected.

Uncertainty Statement

• The bibliographic data above is taken from the authoritative full patent text provided and is corroborated by search results naming US 7,154,909 B2 / Nokia / 2001-06-15 / 2006-12-26.
• Litigation/docket status is not authoritative. My searches returned no docket entry for this patent; that is not proof of absence. I recommend a direct PACER/Docket Navigator/CAFC-CM/ECF check if a litigation-free history must be certified.

I'll search for litigation involving US patent 7154909. Let me run several targeted searches.

Let me search more targeted litigation sources.

Bottom line

I could not find any U.S. litigation (district court, ITC, CAFC appeal, or PTAB proceeding) involving U.S. Patent No. 7,154,909 ("Apparatus, and associated method, for facilitating maintenance of sensitivity level of data communicated in a packet communication system"; Nokia; filed 2002‑06‑12; priority 2001‑06‑15; granted 2006‑12‑26).

No known litigation — no case to report. No plaintiff, defendant, jurisdiction, case number, filing date, or outcome can be listed because I found no case asserting or challenging this specific patent.

What I searched and what I found

Important caveat on the false positives: the number 7154909 appears in several unrelated contexts (a Japanese patent, a trademark docket, a Brazilian company registration, etc.). Per your instruction not to auto‑correct IDs, I am treating those as unrelated identifiers for different entities and not attributing them to U.S. 7,154,909.

Additional context bearing on litigation likelihood

• Prosecution/ownership: Original assignee Nokia Inc. / Nokia Corporation; reassigned to Nokia Technologies Oy on 2015‑05‑06 (effective 2015‑01‑16). Related family members include WO2002103482A2 and the co‑pending application that issued as US 7,593,428 / US 7,161,960 ("multiple‑checksum‑protected data packet," inventors Miller et al.) — those siblings are distinct patents and should not be conflated with the '909 patent.
• Status: Google Patents' legal‑events record shows a maintenance-fee reminder (2018‑08‑06), then lapse for failure to pay maintenance fees on 2019‑01‑28 (effective 2018‑12‑26). The front-page "expires 2025‑03‑06" line reflects the adjusted statutory term, but the patent lapsed earlier for non‑payment. A lapsed patent is rarely asserted, which is consistent with the absence of found litigation.

Confidence and limits

• Confidence that no litigation exists: moderate. I found no evidence, but the searches did not return a clean, authoritative "no cases" record from a litigation database keyed to this exact patent number, and I was unable to fully page through the Unified Patents / PTAB portals directly.
• To confirm definitively, a paid/subscription search (LexisNexis, Darts‑IP, RPX Insight, or a PACER full‑text query) on the exact string "7,154,909" and "7154909" would be the reliable next step; I cannot guarantee my search set captured every district‑court docket.

If you can point me to a specific docket or database export, I can map plaintiff/defendant/jurisdiction/case number/filing date/status precisely for any case that surfaces.

I'll verify the structured data with a few targeted searches before finalizing.

Let me run a couple more checks to be thorough.

One more check on the aggregator/docket side.

Proceedings overview

Total AIA trial proceedings on file for US 7,154,909: zero (0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied). The USPTO Open Data Portal returns no AIA trial proceedings for this patent, and my independent searches (Google Patents PTAB-style queries, PTAB E2E/CourtListener-style queries, and aggregator-defensive-filer checks) surfaced no IPR, PGR, or CBM docket naming 7,154,909 as the challenged patent. The defensive posture is therefore not "the patent was hardened by surviving IPRs" — it is that the patent was never PTAB-tested at all, and it is now expired. Separately, the structured data records the patent as Expired – Fee Related — lapse for failure to pay maintenance fees effective 2018-12-26 (lapse notice 2019-01-28), with an adjusted expiration date recorded as 2025-03-06. That expiry, not any PTAB outcome, is the dominant fact for a defendant.

Proceedings

None to report.

There are no proceedings to list, so the per-proceeding template cannot be populated for this patent. I am deliberately not inventing proceeding numbers, panels, or dispositions. For completeness, here is what the record actually shows and what I checked:

• USPTO structured source (canonical): no AIA trial proceedings. Default conclusion applies: no PTAB activity on file.
• Web search for IPR/PGR/CBM activity: no hits referencing 7,154,909 as a challenged patent. The searches returned unrelated PTAB matters (e.g., Centripetal v. Cisco, Keysight-related IPRs, oilfield-drilling IPRs) and a Japanese patent with the coincidentally identical number JP 7154909 B2 (a Rengo Co. packaging-box patent — not this patent; flagged so no one conflates the two).
• Federal Circuit: no appeal identified. With no PTAB FWD, there is nothing to have been appealed to the Federal Circuit from a PTAB proceeding. I found no CAFC docket on this patent.
• Defensive aggregator: no Unified Patents (or similar) filing identified against 7,154,909.

Corroborating non-PTAB facts from the structured record

• Original assignee: Nokia Inc. → Nokia Corporation (assignment recorded 2002-10-28; effective 2002-07-15) → Nokia Technologies Oy (recorded 2015-05-06; effective 2015-01-16).
• Inventors: Mark W. Cheng; Keith Miller.
• Prosecution: App. No. US 10/167,908, filed 2002-06-12; priority to provisional 60/298,748, filed 2001-06-15; published as US 2003/0035441 A1 on 2003-02-20; granted 2006-12-26.
• Maintenance-fee history: 4-year fee paid 2010-05-27; 8-year fee paid 2014-05-28; 12-year fee not paid → lapse.
• Examiner-cited art (6 references): US 6,057,943; US 6,111,866; US 6,208,620; US 6,259,684; US 2003/0185239 A1 (Miller); US 2004/0039833 A1. None of these were the basis of an AIA trial.
• Non-patent citation: International Search Report for PCT/US02/19134 (note: the WO publication appears in the "Similar Documents" list as WO 2002103482 A2).
• Related family member: US 7,163,960 B2 ("Apparatus, and associated method, for forming, and operating upon, multiple-checksum-protected data packet," Nokia) — same inventor group overlap (Keith Miller), not a PTAB matter.

Strategic summary

Claim status of 7,154,909: all 20 claims are UNTESTED before the PTAB. Claims 1–20 have neither been canceled nor affirmatively sustained in any AIA trial. The "improvement" claim 1 (indicator generator at the second formatting layer) and its dependents (2–16), plus method claims 17–20, remain in the exact form they issued in 2006. Practically, however, this is a moot point for validity purposes going forward: the patent lapsed for nonpayment effective 2018-12-26, so all 20 claims are unenforceable for failure to pay maintenance fees and the patent cannot be asserted for infringement occurring after that lapse.

Estoppel landscape — § 315(e)(2). Because no IPR/PGR was ever instituted, no § 315(e)(2) estoppel attaches to anyone. There are no petitioners, no privies, and no institution decisions to trigger estoppel. That cuts both ways: no one is barred, but there is also no PTAB record, no FWD, and no Federal Circuit affirmance to lean on. If the patent were somehow revived (it cannot be simply by paying a late fee after this long — see below) or if pre-lapse damages were being asserted, a defendant's only invalidity vehicle would be the district court, with the six-year § 286 damages window and the lapse date bracketing the exposure.

Pattern signals. No petitioner has filed once, let alone repeatedly, against this patent; there is no PATB trial-stage timeline to track; no patent-owner appeal activity exists; and no defensive aggregator appears in the chain. The chain of title (Nokia → Nokia Technologies Oy) is a large-entity operating company, not a litigating shell, and there is no visible assertion campaign. The single most telling signal is the absence of PTAB activity combined with the failure to pay the 12-year maintenance fee — a patent that was well-asserted typically attracts IPRs; this one attracted none and was allowed to lapse.

Recommended next steps

1. Confirm no PTAB activity directly. The authoritative check is the PTAB E2E / PTAB API trial-proceedings endpoint filtered on the patent number, cross-checked against the USPTO Patent Center "Status" and "Transactions" tabs for 7,154,909. If those corroborate the structured feed, treat "no PTAB activity" as final.
2. Lead with expiry, not validity. For any current or threatened assertion, the controlling point is the 2018-12-26 lapse for nonpayment of maintenance fees (37 C.F.R. § 1.362; recorded lapse notice 2019-01-28). Get the Patent Center fee/transaction record and the adjusted-expiration entry (2025-03-06) into the file. A patent that has lapsed cannot support a post-lapse infringement claim. The reanimation window under 35 U.S.C. § 41(c)(1) (petition to accept an unavoidably delayed payment, limited to the 2-year period after the fee became due) and the interim extension provisions of § 41(c)(2) have long since closed for the 12-year-fee lapse, so do not expect revival.
3. Scope any damages to the pre-lapse window. If a plaintiff is asserting the patent, map the accused conduct against the enforceable period ending 2018-12-26 and the § 286 six-year lookback from the complaint date. On today's date (2026-09-30), a six-year lookback reaches back only to 2020-09-30 — entirely after the lapse — so there should be no accruable damages at all.
4. Do not over-invest in an IPR strategy. There is no PTAB decision to distinguish, no estoppel to exploit, and no narrowing to rely on. The defense is a standing/enforceability and damages-window defense built on the maintenance-fee lapse. If you nonetheless need invalidity as a fallback, it goes in district court under §§ 102/103/112 using the six examiner-cited references above as the starting corpus.
5. Watch out for the number collision. JP 7154909 B2 (Rengo Co., packaging box) is a different patent; keep the two out of any docket or search memo.

Confidence and limits: I'm highly confident there is no PTAB activity on US 7,154,909 — the structured ODP block says so and searches corroborate it. I could not locate (and therefore do not assert) any PTAB E2E or CourtListener permalink, because no proceeding exists to link to; per the task constraints and the absence of a proceeding, I have not fabricated proceeding numbers, panels, grounds, or dispositions. If a proceeding surfaces later, the right primary sources to cite are the PTAB E2E docket (https://ptacts.uspto.gov/ptacts/) and CourtListener (https://www.courtlistener.com/), neither of which I can cite here for this patent.

I'll reconstruct the assignment chain. Let me first verify the recorded assignments and try to surface reel/frame and correspondent details from the assignment databases.

Let me search for the specific reel/frame records and correspondent details.

Let me probe the assignment databases further for the 2002 recording and press on correspondents.

Note on sourcing: The three follow-up probes for the 2002 recording's correspondent and for any PTAB/NPE activity returned no usable results before my search budget was exhausted. Everything below is grounded in the Google Patents legal-events record supplied in the patent text plus the one corroborating assignment document I did retrieve. Where I could not confirm a detail, I say so rather than fill it in.

---

Inventors

Unusual-pattern check: The classic red flag — all inventors departing the original assignee within ~12 months of filing, preceding a fire-sale — is not determinable from the records I could access. I found no employment-departure data for either inventor. What I can say:

• Only two inventors, both assigning to the same operating company on the same day (2002‑07‑15) — the ordinary employment-obligation pattern, not a distress signal.
• Keith Miller went on to accumulate a substantial Nokia portfolio (14 patents), and is a co-inventor on the sibling application that issued as US 7,161,960 / US 7,593,428 ("multiple-checksum-protected data packet"). A continuing prolific in-house inventor is the opposite of the "everyone bailed out" tell.
• No evidence that either inventor personally retained or later asserted rights. Both assigned at filing.

---

Original assignee

Two different names appear in the source, and I am flagging the discrepancy rather than reconciling it silently (per "do not auto-correct"):

• Google Patents front-page field: "Original Assignee — Nokia Inc."
• The actual recorded 2002 assignment (reel 013425/0447): assignee "NOKIA CORPORATION" (Finland).
• Current assignee per Google Patents: Nokia Technologies Oy.

These are the same corporate family (Nokia Inc. was the U.S. subsidiary/brand; Nokia Corporation is the Finnish parent that took the recorded assignment). I treat Nokia Corporation as the legally operative original assignee because that is what the recorded instrument names.

• Primary line of business: telecommunications — mobile handsets and network infrastructure, including cdma2000 infrastructure and devices (the patent's own exemplary implementation targets cdma2000 RLP).
• Did they ship a product embodying the claims? Not established from the record. The claims are directed to an RLP-layer control procedure that preserves UDP‑Lite sensitivity — an internal protocol-stack behavior. Nokia was an operating cdma2000 vendor in this era, so a practicing implementation is plausible, but I found no product mapping and will not assert one.
• Current status: Nokia Corporation remains an operating telecom/network-equipment company (handset division sold to Microsoft in 2014). Nokia Technologies Oy — a Nokia licensing/holding subsidiary (Karaportti 3 / Karakaari 7, Espoo, Finland) — is the current owner of record. Not acquired by a third party, not dissolved, not in bankruptcy for this asset.

---

Assignment timeline

Two recorded assignments. Both are intra‑corporate; there is no third‑party or NPE transfer anywhere in the chain.

• 2002‑07‑15 (executed) / recorded 2002‑10‑28 — Reel 013425/0447

  • Conveyance: Assignment (Assignment of inventors' interest)
  • Assignor: Mark W. Cheng; Keith Miller (joint inventors)
  • Assignee: Nokia Corporation (Finland)
  • Correspondent: Not retrievable from the sources I could access. The Google Patents legal-events entry does not expose the correspondent for this 2002 recording, and my targeted follow-up searches returned nothing. I will not guess a name.
  • Context: Original inventors' assignment to the operating-company employer — standard, mandatory employment assignment at filing.

• 2015‑01‑16 (executed) / recorded 2015‑05‑06 — Reel 035581/0384

  • Conveyance: Assignment
  • Assignor: Nokia Corporation
  • Assignee: Nokia Technologies Oy (Karaportti 3, FI‑02610 Espoo, Finland)
  • Correspondent: NOKIA USA, INC — 200 South Mathilda Ave, Attn: Intellectual Property, Sunnyvale, CA 94086 (phone 510‑303‑9002; email jacqueline.linebarger@nokia.com). Caveat: this correspondent string is drawn from a court‑filed copy of the Nokia→Nokia Technologies EPAS bulk submission on reel 035581 covering ~125 properties; the individual property shown there sits at frame 0654, i.e., the frame number varies per property on the same reel. I could not isolate the '909‑specific correspondent page, so treat the correspondent as highly consistent with, but not independently confirmed for, the '909 entry. Note it is in‑house Nokia IP staff, not an outside NPE lawyer.
  • Context: Internal corporate reorganization — Nokia's "housekeeping" transfer of a large patent batch to its licensing/holding subsidiary Nokia Technologies Oy. (The retrieved instrument states the assignment is effective 31 December 2014 and recites transfer of a described list of patents on Attachment I.)

Recurrence check on correspondents: I cannot run the recurrence test the task asks for. Only one correspondent is even partially visible (Nokia USA, in‑house), and it appears on a single link of the chain. A single in‑house appearance is not a finding of an NPE "repeat correspondent."

---

Timeline diagram

timeline
    title Ownership of US 7154909
    2001 : Provisional filed
    2002 : Utility app filed
         : Inventors assign to Nokia Corporation
    2006 : Patent granted
    2015 : Nokia to Nokia Technologies Oy
    2019 : Lapsed for unpaid maintenance fee

---

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. No licensing-only LLC ever appears. The only transferee after the inventors is Nokia Technologies Oy, a wholly owned Nokia subsidiary — an operating-company family member, not a single-purpose Delaware/Texas shell. (Reel 035581/0384.)

2. Known asserter in the chain — Not present. No assignee in the chain (Nokia Inc → Nokia Corporation → Nokia Technologies Oy) matches any public NPE list (Acacia, Marathon, IV, Wi‑LAN, Conversant/Mosaid, Vringo, Pendrell, etc.). Both entities are Nokia corporate vehicles.

3. Repeat correspondent across the chain — Not present / unclear. Only one partial correspondent is visible (NOKIA USA, INC, in‑house, on reel 035581), appearing once. No attorney or recording firm recurs, and none matches an NPE assertion list. Insufficient occurrences to score.

4. Cascading transfers — Not present. Exactly two recorded assignments, separated by ~13 years (2002 → 2015). No chained LLCs, no clustering in <24 months.

5. Pre-litigation transfer — Not present. No infringement suit naming this patent was found (consistent with the prior litigation section); the 2015 internal transfer has no adjacent litigation to enable.

6. Bankruptcy fire-sale — Not present. No Chapter 7/11 involving Nokia for this asset. The '909 patent's eventual lapse was for non‑payment of maintenance fees (2018‑12‑26 effective; lapse recorded 2019‑01‑28), not a distressed sale.

7. Privateering — Not present. No SEC filing, Patent Progress, or EFF coverage surfaced showing Nokia moving this patent to an NPE to assert on its behalf. Nokia Technologies Oy is Nokia's own licensing arm, which is normal corporate IP management, not privateering.

8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at Nokia Technologies Oy.

---

Verdict

Insufficient data (no NPE-pattern records).

The chain consists of exactly two recorded assignments — inventors → Nokia Corporation (reel 013425/0447, exec. 2002‑07‑15) and Nokia Corporation → Nokia Technologies Oy (reel 035581/0384, exec. 2015‑01‑16) — with zero third-party, LLC, or asserter transfers, so there is no NPE chain to evaluate and this bucket is the closest fit in the menu. Flagging a rubric mismatch explicitly: this case is affirmatively a clean operating‑company ownership line (all records intra‑Nokia), which none of the other four verdict categories cleanly captures — "Operating‑company assertion" requires the owner to be suing competitors (not observed), and "Defensive/non‑asserting" requires termination at a defensive aggregator (not observed). The patent lapsed for non‑payment of maintenance fees in 2018/2019, a state inconsistent with active monetization.

Verify at USPTO Assignment Center: https://assignmentcenter.uspto.gov/ (search by patent number 7154909), mirrored at https://assignment.uspto.gov/patent/index.html.

Confidence and limits: Moderate. The two reel/frame entries above come from the patent's authoritative Google Patents legal‑events record, and the 2015 transfer is corroborated by a retrieved copy of the reel 035581 EPAS submission. I could not retrieve the correspondent of record for the 2002 recording, and I could not run a full exact‑string sweep of the Assignment Center, so I cannot rule out an additional, unindexed nunc‑pro‑tunc or corrective recording. A direct Assignment Center query is the definitive next step; if you supply that export I can pin the 2002 correspondent and confirm the 2015 frame number precisely.

I'll verify the USPTO record and the cited references against live sources.

USPTO Record — US 7,154,909 B2 (verified)

I cross-checked the number against the authoritative full-text record at https://patents.google.com/patent/US7154909/en and corroborating sources. The number resolves cleanly and only to this patent — the earlier warning about "noise" hits (JP 7154909 B2, a Brazilian registry entry) still applies, and those are unrelated documents.

• Patent: US 7,154,909 B2 — "Apparatus, and associated method, for facilitating maintenance of sensitivity level of data communicated in a packet communication system"
• Application: US 10/167,908; filed 2002-06-12; priority 2001-06-15 (provisional 60/298,748); granted 2006-12-26
• Inventors: Mark W. Cheng; Keith Miller
• Assignee: Nokia (orig.) → Nokia Technologies Oy (current)
• Status: Lapsed for failure to pay maintenance fees (effective 2018-12-26); adjusted statutory expiry noted 2025-03-06

Note on method: I was able to retrieve the full Google Patents text and corroborating pages for four of the six cited references, plus the family/priority data. Two references (the Ericsson publication and the Miller/Nokia sibling publication) returned no independent page in my search budget; for those I rely on the citation data as printed on the face of US 7,154,909 and flag the reduced verification level. I could not pull USPTO PatentCenter/Public PAIR HTML directly, so the primary bibliographic grounding is the Google Patents full text (which mirrors the USPTO record) plus the sibling sources returned.

---

The six references cited on the face of US 7,154,909

These are the complete set of "Patent Citations (6)" listed for the '909 patent. Google Patents flags some as examiner-cited (*); I reproduce the set literally and do not re-characterize the citation provenance beyond what the record shows.

Also cited: Non-Patent Citation — International Search Report for PCT/US02/19134 (the PCT counterpart of this application is WO 2002/103482 A2).

---

Per-reference §102 analysis

Threshold point on the §102 exercise. Claims 1 and 17 (the only independents) require two cooperating formatting layers where a first layer splits data into a "first part / second part," and a second layer places an indicator into its own packet telling which part the data is. Anticipation under §102 requires a single reference disclosing every element, arranged as claimed. None of the six references below discloses a lower-layer sensitivity indicator for a split-sensitive/insensitive payload. They are, in the main, §103 / background art, not true anticipatory art. I say this explicitly rather than forcing a strained §102 mapping. Where a reference arguably touches one element, I identify the claim it most nearly reaches and flag that it is not a complete anticipation.

1. US 6,111,866 A — (Kweon et al.), ETRI

• Dates: filed 1998-03-31 (app. 09/050,978); granted 2000-08-29. The '909 face page lists "1997-04-01" (a likely foreign/priority listing); the underlying U.S. application date is 1998-03-31 per the Wanfang record.
• Description: A method of servicing asynchronous circuit/data calls and terminating-call signaling in a CDMA network built on TIA/EIA/IS-99. It defines a protocol stack (PPP, TCP/IP, IPCP/LCP, SNDCF) over an RLP link layer that "divides variable length data packet of the upper layer into CDMA traffic channel frames." It is about call setup and framing, not payload sensitivity. (https://patents.google.com/patent/[US6111866A](/patent/US6111866A)/en)
• §102 reach: Relevant only as teaching of an RLP layer that segments upper-layer packets — i.e., it touches the "second formatting layer" and "segmenter" concepts of claims 14–15. It does not disclose any UDP-Lite first layer, any sensitive/insensitive split, or any indicator field. It cannot anticipate claims 1–20.

2. US 6,259,684 B1 — Nortel Networks

• Dates: priority 1997-05-19; granted 2001-07-10.
• Description: Architecture/procedures for circuit-mode data over the A-interface using the Intersystem Link Protocol (ISLP) framing structure. Notably it recites an "indicator incorporated into said set" of call-processing/handoff messages, and RLP transport (per IS-707). (https://patents.google.com/patent/[US6259684B1](/patent/US6259684B1)/en)
• §102 reach: This is the closest of the group on surface language, because it uses the words "indicator" and "RLP." But its "indicator" is a call/handoff-message flag for transporting circuit-mode data — not a payload sensitivity-type indicator, and there is no first-layer split into sensitive/insensitive parts. It does not read on any claim element of the '909 indicator generator. No anticipation.

3. US 6,057,943 A — ETRI

• Dates: filed 1998-06-15 (app. 09/097,483); granted 2000-05-02.
• Description: G3 fax transmission over CDMA cellular networks; uses upper-layer protocols (PPP and above) in a data terminal and lower-layer (RLP and below) in the mobile terminal, with RLP segmentation of the encapsulated data. (https://patents.google.com/patent/US6057943.html)
• §102 reach: Confirms RLP segmentation of an upper-layer packet — again background for the "second layer/segmenter" concept (claims 14–15). No sensitive/insensitive data partitioning, no indicator field. No anticipation.

4. US 2004/0039833 A1 — Telefonaktiebolaget LM Ericsson

• Dates: filed 1998-07-15; published 2004-02-26 (abandoned-family application; published after the '909 filing because of a long prosecution, but its 1998 filing gives it §102(e) effect).
• Description: "Communication device and method." I was unable to retrieve the body of this publication within my search budget, so I describe it at the level the citation permits; its early-1998 filing puts it in the pre-AIA §102(e) category.
• §102 reach: Given the pre-1998 subject matter and the absence of any UDP-Lite (UDP-Lite post-dates it), it cannot disclose the claimed sensitivity indicator for a split payload. Best characterized as §102(e)/§103 art; cannot anticipate any claim on this record. Flagged: description is unverified beyond the citation.

5. US 6,208,620 B1 — Nortel Networks

• Dates: filed 1999-08-02; granted 2001-03-27.
• Description: TCP-aware agent sublayer (TAS) caching TCP packets and ACKs near the wireless link, monitoring the link, and retransmitting only the lost air-link frame rather than the whole TCP packet; explicitly discusses coupling link-layer ARQ (RLP/RLC/LLC) with TCP. (https://patents.google.com/patent/[US6208620B1](/patent/US6208620B1))
• §102 reach: This is the most substantively relevant of the six, because it addresses selective, frame-level retransmission recovery over a radio link. That concept underlies the receiving-side behavior of claims 18–20 (selectively permitting retransmission based on an indicator). But it does not disclose the indicator itself, nor a sensitive/insensitive payload split, nor retransmission decisions keyed to sensitivity type. It is §103 material against the retransmission-dependent claims; it does not anticipate them alone.

6. US 2003/0185239 A1 — Keith Miller (Nokia)

• Dates: filed 2002-03-26; published 2003-10-02. (Issued as US 7,593,428 / related to US 7,161,960.)
• Description: "Apparatus, and associated method, for forming, and operating upon, multiple-checksum-protected data packet" — a sibling application by a common inventor (Keith Miller) addressing multiple checksum protection of data packets. (Cited-listing data only; body not independently retrieved.)
• §102 reach: Because it was filed 2002-03-26 (before the '909 filing of 2002-06-12) it can qualify as §102(e) art. It is directed to checksum protection, an adjacent-but-distinct concept (protecting a partially covered payload), and shares an inventor; it does not disclose the claimed second-layer sensitivity indicator field. It is best treated as a same-family/§102(e) reference and as §103 context for the sensitivity-maintenance problem — not an anticipatory reference.

---

Bottom line for the §102 question

• No cited reference anticipates any of claims 1–20. Anticipation for claim 1 (and method claim 17) would require one reference showing: (a) a first formatting layer that selectably splits data into two parts; (b) a second, lower formatting layer that generates an indicator of which part the data is; and (c) that indicator embedded in the second-layer packet. None of US 6,111,866; 6,259,684; 6,057,943; 2004/0039833; 6,208,620; or 2003/0185239 shows elements (a) or (b).
• The references cluster into two §103/background roles:
• RLP-segmentation background (teaching a lower layer that segments upper-layer packets) → relevant to the "segmenter" claims 14–15: US 6,111,866; US 6,057,943.
• Link-layer recovery / selective retransmission background → relevant to the receiving-side retransmission claims 18–20: US 6,208,620 (and, at the call-signaling level, US 6,259,684's "indicator," though in an unrelated sense).
• US 2004/0039833 A1 and US 2003/0185239 A1 are §102(e)/§103 candidates by date, but on this record neither discloses the claimed sensitivity indicator.
• The PCT International Search Report for PCT/US02/19134 (non-patent citation) is the examiner's own search documentation for the PCT counterpart WO 2002/103482 A2, not prior art in itself.

Uncertainty / limits

• Direct USPTO PatentCenter/PAIR HTML could not be retrieved with my available searches; the bibliographic and citation data above rest on the Google Patents full text (authoritative for this number) plus corroborating pages (Justia/EveryPatent/FPO/Wanfang) for the individual references.
• For US 2004/0039833 A1 and US 2003/0185239 A1 I did not retrieve the specification bodies; their descriptions are drawn from the '909 citation listing and, for 2003/0185239, its sibling-family description. If a precise §102(e) element-by-element comparison against these two is required, the full texts should be pulled and mapped claim-by-claim.
• The "1997-04-01" date listed for US 6,111,866 on the '909 face page vs. the 1998-03-31 U.S. filing date is a discrepancy I am reporting literally rather than reconciling; it most likely reflects a foreign priority listing.

I'll ground this analysis in the actual prior-art references listed on the patent record. Let me pull their content and abstracts.

Obviousness Analysis — US 7,154,909 B2 under 35 U.S.C. § 103

Prepared 2026-04-26. Grounded in the "Prior Art" section of the Google Patents record (https://patents.google.com/patent/US7154909/en) plus corroborating source text retrieved by search. Builds on the previously generated bibliographic, claim-overview, and litigation sections; those are not repeated.

---

1. Scope of the prior art actually "on the page"

The Prior Art section of the record supplies exactly these references (I list them literally, including the quirky dates the record itself shows):

"Citations (6)" — the examiner's cited references

Non-Patent Citation (1): International Search Report for PCT/US02/19134 — this is the '909's own search report, so it is a procedural document identifying art, not substantive prior art itself.

Two important notes on the record's own framing:

• The Google page header "Prior art date 2001-06-15" and the "Prior art keywords" (data, layer, formatting, indicator, formatted) are indexing artifacts, not an admission.
• The written description of the '909 itself admits UDP-Lite as prior art (background: "A modified UDP formatting scheme, referred to as UDP-Lite, is sometimes utilized…"). That admission is usable as prior art under § 103. The UDP-Lite concept is also documented in the pre-2001 literature — the Larzon/Degermark/Pink paper "UDP Lite for Real Time Multimedia Applications" (cited as LDP99) and the UDP-Lite IETF drafts. ⚠️ Careful: RFC 3828 itself published July 2004, i.e., after the '909's 2001-06-15 priority, so RFC 3828 is not itself prior art by date; the pre-2001 paper/draft art and the '909's own admission are what carry the disclosure.

---

2. A threshold date/ownership problem with the Miller reference (flag before using it)

US 2003/0185239 A1 (Miller et al., "multiple-checksum-protected data packet"):

• Its earliest effective filing date is 2002-03-26, which is after the '909's 2001-06-15 priority.
• Therefore it is not § 102(a)/(b) art, and its § 102(e) date (2002-03-26) also post-dates the '909 priority, so it is not § 102(e) art against the '909.
• It is, moreover, commonly owned (both Nokia; co-inventor Keith Miller appears on both). Under pre-AIA 35 U.S.C. § 103(c) (the '909 was filed 2002, so pre-AIA applies), a 102(e) reference commonly owned at the time of invention is disqualified for use in a § 103 combination.

Conclusion: the Miller reference is essentially unusable as a § 103 combination reference against the '909. Any rejection built on it would be improper. I flag this explicitly rather than treating the examiner's citation as an endorsement of its availability. (If a challenger needs the "header fields identifying the checksum-covered parts" concept, it must come from elsewhere — e.g., the UDP-Lite checksum-coverage field, which is admitted prior art.)

The remaining five references all qualify by date (US 6,208,620 and US 2004/0039833 via § 102(b)/§ 102(e); the three CDMA references via § 102(b)/§ 102(e)).

---

3. Level of ordinary skill in the art (POSITA)

The '909 is directed to cross-layer signaling between a transport-formatting layer (UDP-Lite) and a radio link layer (RLP/RLC) in a cdma2000-type packet system. The relevant POSITA therefore has: a bachelor's degree in EE/CS (or equivalent) plus ~2–3 years of experience with wireless packet-data protocols — specifically the cdma2000 Radio Link Protocol (ARQ, NAK, resequencing, L_V(R)/L_V(N) variables), UDP/IP transport framing, and the UDP-Lite partial-coverage concept. This is a routine-engineering level of skill, which matters for the KSR "predictable results" analysis.

---

4. What the independent claims actually require

• Claim 1 (apparatus): at the second (lower) layer, an indicator generator that produces an indicator whose value represents whether the data handed up/down from the first layer was formatted into a "first part and second part"; the indicator forms part of the second-layer data packet.
• Claim 17 (method): the four steps — deliver first-layer-formatted data to the second layer; determine which of the first/second part it is; generate an indicator responsive to that determination; form a second-layer packet carrying the indicator.
• Dependent claims add the operationally meaningful substance: UDP-Lite⇒RLP (2–6); SDI in the RLP header (5–6); SEG_MAP_COUNT (7–9 — note the claim says SEQ_MAP_COUNT, already flagged); FIRST (10); SDI_MAP bitmap (11–12); RLC variant (13); a segmenter splitting frames by type (14–16); and receiver-side selective retransmission — permitted for sensitive, prohibited for insensitive-only (18–20).

The heart of the invention is narrow and conceptual: the lower layer learns and signals the sensitivity split created at the higher layer, and uses it to decide retransmission/resequencing.

---

5. What the qualifying references teach

(A) UDP-Lite (admitted prior art; LDP99 paper/drafts). Teaches splitting a datagram payload into a checksum-covered "sensitive" part and an uncovered "insensitive" part, with a header field (checksum-coverage) marking the boundary. Critically for motivation, the Larzon paper itself states: "For links that can deliver a significant ratio of damaged packets, it is important to reflect the UDP Lite policy of sensitive and insensitive data onto the link layer… Designs for new link layer protocols for such links should allow checksums to be partial; the insensitive part of the payload should not be covered by a checksum." That is an express teaching to push the sensitive/insensitive distinction down to the link layer — i.e., the very problem the '909 addresses.

(B) US 2004/0039833 A1 — Ericsson / Rathonyi (priority 1998; EP 0973302 A1 published 2000-01-19 → printed-publication art). Abstract (retrieved):

"…the implementation of a protocol that embeds packets is made sensitive to the type of data being embedded, so that adjustable parameters can be set automatically at the layer providing the parameters being adjusted in accordance with the packets being embedded… the data in packets being received from a higher layer… may be classified into different categories with respect to the adjustable parameters, where the parameters can automatically be adjusted for an individual packet in accordance with the category into which is classified."

This is a lower layer that (i) inspects a higher layer's traffic, (ii) classifies it per-packet, and (iii) changes its own per-packet parameters (e.g., reliability/retransmission behavior) accordingly. That maps almost element-for-element onto claim 1's "indicator… representative of whether the data… is formatted into the first part and the second part," and onto the receiving-side decision logic of claims 18–20 when combined with (C).

(C) US 6,208,620 B1 — Nortel "TCP-Aware Agent Sublayer (TAS)" (priority 1999; granted 2001-03-27 → § 102(b)). Teaches a link-layer sublayer that intercepts and parses higher-layer (TCP) packets, caches them, monitors the wireless link, and selectively retransmits immediately on air-link packet loss while suppressing inappropriate retransmissions. In other words: a lower layer that reads transport-layer semantics and performs selective retransmission based on them — the functional core of claims 1 and 18–20, in a wireless link context.

(D) The three CDMA references (US 6,111,866; US 6,259,684; US 6,057,943). Provide only background on cdma2000/circuit-mode data delivery and A-interface framing; they are weak teaching references but establish that RLP-style link layers over CDMA were well known.

---

6. Motivation to combine (KSR rationales)

1. Express teaching: Larzon explicitly calls for reflecting UDP-Lite's sensitive/insensitive policy onto the link layer. That is the problem statement, handed to the POSITA.
2. Known technique / predictable result: Rathonyi already teaches lower-layer per-packet parameter adjustment based on higher-layer data categories; applying that to the UDP-Lite sensitivity category is "using a known technique to improve a similar device in the same way," yielding the predictable result (protect sensitive data, don't stall on insensitive data).
3. Design incentive / market pressure: Real-time services (VoIP, video) over error-prone cdma2000 links create strong pressure to avoid retransmitting delay-sensitive but error-tolerant payload, and to avoid discarding irreplaceable sensitive payload — exactly the win the '909 claims.
4. Infrastructure already in place: RLP already performed ARQ/NAK and resequencing (described in the '909 background as known); adding a sensitivity flag to an existing RLP header is a minimal, obvious modification. TAS shows the same cross-layer parse-and-act pattern was routine.

---

7. Proposed § 103 grounds, claim by claim

Ground 1 — Claims 1–6 over UDP-Lite (A) in view of Rathonyi (B)

• Claim 1: (A) gives the first layer that "selectably formats into a first part and a second part" (sensitive/insensitive). (B) gives the second layer's per-packet classification of higher-layer data categories and adjustment of its own parameters. Making the lower layer carry a value indicating which of the two parts the data belongs to (so the peer lower layer can apply matching behavior) is the natural, predictable implementation of (B)'s classification. → Obvious.
• Claims 2–3 (UDP-Lite sensitive vs. insensitive ⇒ first/second indicator value): taught directly by (A).
• Claims 4–6 (RLP layer; SDI in RLP header): RLP is the admitted, well-known cdma2000 link layer; placing a one-bit sensitivity flag (SDI) in its header is routine header design. → Obvious.

Ground 2 — Claims 1–6 and 17–20 over UDP-Lite (A) in view of Nortel TAS (C)

• (C) supplies the lower-layer selective-retransmission engine keyed to higher-layer semantics, so claims 18–20 (permit retransmission for sensitive first-part data; forbid it for insensitive-only data) follow directly from (A)+(C): once the link layer knows a frame is insensitive, it does not need to retransmit it, and the delay-tolerance of insensitive data makes that the correct policy. The "semi-transparent procedure" of the specification is the obvious reconciliation of (A)'s tolerance for error with (C)'s retransmission machinery.
• Claims 14–16 (segmenter splitting the packet into separate frames by type): obvious given (A)'s part split plus the well-known practice of segmenting transport PDUs into link-layer frames; the only new choice is to keep sensitive and insensitive frames distinct, which (A)'s policy requires.

Ground 3 — Claims 7–12 (SEG_MAP_COUNT / FIRST / SDI_MAP) — weakest, most contested

• These claims require specific header fields: a frame count, a first-frame sequence number, and a bitmap indexing per-frame sensitivity. This is a design-detail layer. A challenger would argue these are routine implementation choices once Grounds 1–2 establish the concept (KSR: "a finite number of identified, predictable solutions"). But the qualifying references on the page do not clearly disclose a per-frame sensitivity bitmap in the lower-layer header, and the only reference that comes close (Miller, US 2003/0185239) is inadmissible under § 103(c) (see §2). Absent additional art (e.g., a link-layer bitmap/map header), claims 7–12 are the least obviously rendered and would likely withstand a § 103 attack built only on the page's qualifying references.

Ground 4 — Claim 13 (RLC variant)

• Mere substitution of one known link-layer protocol for another (RLP → RLC in the W-CDMA context, which the '909 itself acknowledges as an equivalent alternative). Obvious if Ground 1 or 2 succeeds.

---

8. Counter-considerations — why the claims might nonetheless survive

• Issuance over these very references. The examiner had US 6,208,620 and US 2004/0039833 (and Miller) before them and still allowed claim 1. That is not dispositive but is evidence the specific claimed indicator-in-the-lower-layer-header was viewed as not taught.
• Rathonyi's locus of action differs. (B) adjusts parameters at the layer itself; the claim requires an indicator embedded in the second-layer packet that communicates the higher-layer formatting to a remote entity. Whether the "classify locally and adjust locally" teaching fairly suggests "tag the packet so the distant receiver can classify" is the strongest available non-obviousness argument.
• A nexus / secondary-considerations gap. No evidence of unexpected results, industry praise, licensing, or copying was found in the record (consistent with the separate finding that the patent lapsed for non-payment effective 2018-12-26 and shows no litigation). Without a nexus, secondary considerations are unlikely to rebut a well-constructed prima facie case, but they also cannot help a challenger.
• The '909's own framing treats the advantage as "maintain the segregation between the parts" — a result the POSITA would expect from the Larzon teaching, which cuts toward obviousness (predictable result).

Net assessment: Claims 1–6, 13–16 (frame-splitting) and 17–20 are vulnerable under § 103 over UDP-Lite (Larzon) + Rathonyi (US 2004/0039833), and independently over UDP-Lite + Nortel TAS (US 6,208,620). Claims 7–12 (the count/first/map header fields) present the strongest non-obviousness position because the admissible art on the page does not clearly teach a per-frame sensitivity bitmap carried in the lower-layer header.

---

9. Evidentiary caveats (do not overstate)

• I could retrieve full abstract/description text for US 6,208,620 and US 2004/0039833 but not their complete claim sets; the mappings above are based on the retrieved abstracts and summaries. A rigorous invalidity contention would require pin-cite column/line mappings against the actual granted texts (e.g., via https://patents.google.com/patent/[US6208620B1](/patent/US6208620B1) and the Ericsson/EP 0973302 family).
• The record shows a date discrepancy for the Ericsson reference (page: 1998-07-15; Unified Patents: 1998-07-14). I have kept both rather than picking one (per the no-auto-correction rule).
• The Miller reference is flagged as inadmissible for § 103, not merely weak; any combination relying on it should be rejected on § 103(c) grounds.
• I have not performed a full prior-art search beyond the references in the page's Prior Art section plus the background UDP-Lite art; a broader search could materially strengthen or weaken the above.

---

10. Bottom line

Using only the references in the page's Prior Art section (and the UDP-Lite art the '909 itself admits), the strongest § 103 combinations are:

1. Larzon UDP-Lite + Ericsson/Rathonyi (US 2004/0039833) → renders claims 1–6, 13–16, 17–20 obvious; motivation is express (Larzon: "reflect the UDP Lite policy… onto the link layer") and reinforced by Rathonyi's lower-layer per-packet classification.
2. Larzon UDP-Lite + Nortel TAS (US 6,208,620) → renders claims 1–6 and 17–20 obvious (cross-layer parse + selective retransmission), with the strongest showing on the retransmission claims 18–20.
3. Claims 7–12 (SEG_MAP_COUNT/FIRST/SDI_MAP) are not clearly obvious from the admissible art on the page; the one reference that arguably teaches identifying parts in a UDP-Lite header (Miller, US 2003/0185239) is disqualified under pre-AIA § 103(c) and by date.

A valid rejection would therefore have to rest on combination 1 or 2 and would most plausibly reach claims 1–6 and 13–20, while stopping short of claims 7–12 without additional, independent link-layer bit-map art.