US 8004994

I'll research this patent number across the sources you specified.

Let me try more targeted searches.

Research note on sources and method

I ran targeted web searches for the exact string "8004994" / "8,004,994" in USPTO and Federal Circuit contexts. The searches returned no litigation, PTAB, or CAFC‑2026 docket entries for this patent. The only hits containing "8004994" were an unrelated Centripetal Networks IPR file‑history exhibit (IPR2018‑01512, Ex. 1002 for U.S. 9,565,213), where "8004994" appears only as a search string inside an examiner's EAST search history — it is not a citation to, or proceeding involving, this patent. I also reached the tool‑call step limit before I could query USPTO PatentCenter / CourtListener directly, so I cannot certify the litigation status of this patent. Everything below is taken from the authoritative full patent text supplied in the prompt (Google Patents, US8004994B1).

Bottom line on the docket search: I found no CAFC 2026 docket and no district‑court or PTAB proceeding involving US 8,004,994. Treat that as "no evidence found," not as proof none exists.

Bibliographic data (US 8,004,994 B1)

Uncertainty flags:

• The Google Patents header lists an "adjusted expiration 2028‑09‑30," which is the nominal 20‑year term. The legal‑events record shows the patent actually lapsed in 2019 for non‑payment of maintenance fees. These two statements are inconsistent in the source; the lapse entry is the operative one.
• Entity‑status events show a small‑entity claim (2015‑12‑07) later withdrawn in favor of large‑entity/undiscounted status (2016‑12‑03).
• I cannot confirm current ownership/chain of title beyond the original 2006 assignment (e.g., no acquisition by a successor entity is shown in the supplied record).

Abstract (as issued)

A method and apparatus is provided that enables accurate measurement of drop rate and delay in a System Under Test (SUT) by one or more monitoring devices even when the frame error rate of the monitoring devices may be imperfect. During a packet drop measurement process, ancillary information is identified and analyzed to determine if the ancillary information can be used to infer receipt of packets when explicit information regarding receipt is not present. A delay measurement process incorporates the time required to re‑transmit packets into the delay measurement to more accurately reflect SUT operation.

Plain‑language overview of the independent claims

Claim 1 — Method of measuring a quality‑of‑service (QoS) statistic of an SUT. The steps are: (a) monitor traffic between a Source DUT and a Destination DUT and store the packets in a source list; (b) monitor traffic received by the Destination DUT and store packets in a destination list; (c) compare the two lists to build a dropped packet list containing packets present in the source list but missing from the destination list; (d) validate that list using "ancillary information" to decide whether a packet marked as dropped was in fact received by the Destination DUT; (e) compute the QoS statistic from the validated list; and (f) output the statistic in tangible form. The inventive core is step (d): the monitor's own missed frames are not automatically treated as device‑under‑test drops.

Claim 8 — Analyzer apparatus. Same two‑phase idea expressed as an apparatus: a processor with a computer‑readable medium carrying program code that (i) generates the dropped packet list by comparing the source and destination lists, (ii) validates the list using ancillary information from the source list to identify packets that were actually received by the Destination DUT, (iii) calculates a QoS statistic from the validated list, and (iv) outputs it in tangible form.

Claim 11 — Method of measuring delay of an SUT. Monitor traffic and keep a source list with per‑packet timestamps for packets sent/received by the Source DUT, and a destination list with timestamps for the Destination DUT; select a packet from the source list and record its packet transmission timestamp (PTT); search the source list for a valid acknowledgement associated with that packet — expressly including an acknowledgement to a re‑transmitted version of the packet; find the matching entry in the destination list (which may be the retransmission) and retrieve its timestamp (D‑MATCH); compute the delay from PTT and D‑MATCH; output the delay in tangible form. In short: the delay clock starts at first transmission and stops only when a valid ACK arrives, folding retransmission delay into the measurement instead of under‑reporting it.

Claim 13 — Analyzer apparatus for delay. Program code that searches a list of packets exchanged between Source and Destination DUTs to measure SUT delay, specifically by comparing the timestamp of a transmitted packet to the timestamp of a matching destination‑list entry that corresponds to a retransmission of that packet, to determine the delay; plus program code to output a QoS statistic based on the determined delay in tangible form.

Dependent claims, briefly

• 2 — The ancillary information is acknowledgements recorded by the source monitor; 4 — receipt is inferred when an ACK arrives within a Protocol Acknowledgement Window (PAW); 5 — such a packet is removed from the dropped packet list; 6/7 — a gap list of packet ranges found on the source list but not the destination list is generated and used to locate relevant ancillary information; 3 — 802.11 WiFi ACK packets specifically; 9 — the gap list is a set of gap entries each identifying ancillary information usable to infer receipt; 10 — the ancillary information is an ACK received within a PAW; 12 — the valid ACK is one received within the PAW for the retransmitted packet.

Context worth noting

• The specification frames the problem as monitoring equipment sharing the same radio/receiver technology as the devices under test, so monitors lose frames too and can mischaracterize a received packet as dropped. The fix is to use protocol acknowledgements (e.g., IEEE 802.11 ACKs) as inferential evidence of receipt when the monitor has no explicit record.
• The patent lists six cited prior‑art references (e.g., US 2006/0012388 A1 to Acradyan Technology, US 2006/0120283 A1 to Northrop Grumman, US 2006/0274657 A1 to Olgaard) and shows a large "cited by" set, but those citations do not indicate any assertion of this patent against another party.
• Applicant described a commercial embodiment: an Azimuth Systems "W‑Series" test platform running a Voice‑over‑IP WiFi Test Suite (VoWiFi). Note the specification text has minor typos (e.g., "packet 4 on the source list corresponds to packet 9 ('YOU')" immediately after an earlier "packet 3 ('YOU')" reference) — I am quoting the source as given, not correcting it.

If you need the litigation/ownership picture confirmed, the next step would be a direct query to USPTO PatentCenter (application 11/555,324) and CourtListener/PACER for the exact number — I was unable to complete those lookups within my search budget, so I am flagging that gap rather than asserting a negative.

I'll research litigation involving US patent 8004994. Let me search multiple sources.

Let me search more targeted litigation databases.

Summary of Findings: Litigation Involving US Patent No. 8,004,994

Based on my searches, I found no evidence of any litigation involving US Patent No. 8,004,994.

What the searches returned

I searched for the specific number "8004994" and "8,004,994" across patent-litigation aggregators (including the Unified Patents portal litigation caselist, Docket Alarm, Justia, and general web sources). Every substantive hit was one of the following, none of which constitutes litigation:

1. The patent itself — US 8,004,994 B1, "System and method for intelligently analyzing performance of a device under test," assigned to Azimuth Systems, Inc. (Acton, MA). Inventors: Prashant Darisi, Jeremy Spilman, Jose Graziani. Filed 2006-11-01; granted 2011-08-23. (Google Patents; wiki.golden.com)

2. Prior-art citations in other patents' file histories — The string "8004994" appearing in Docket Alarm documents (e.g., the file history of US 9,565,213, Cisco Exhibit 1002, in IPR2018-01512) is a citation of US 8,004,994 as prior art, not a party to that proceeding. Likewise, the Unified Patents portal pages (e.g., for US 10,659,573; US 10,511,572) list US-8004994-B1 only as a related/cited reference.

3. Co-inventor attribution — A patent-leaderboard entry noting Jose Graziani as an inventor (now associated with Sonos).

Litigation table

No district court case, ITC action, PTAB proceeding (IPR/PGR/CBM), or CAFC appeal asserting or challenging US 8,004,994 was identified.

Important caveats

• I could not confirm this is an exhaustive negative. My tool access was limited and I was unable to complete every planned query. The absence of results is meaningful but should be verified against PACER, the Unified Patents litigation caselist, and Docket Alarm directly before being relied upon conclusively.
• This is consistent with the patent's commercial posture. US 8,004,994 remained with its original assignee, Azimuth Systems, and lapsed for failure to pay maintenance fees, effective 2019-08-23 (expired 2019-09-30). A patent that expired for non-payment and was never transferred to a monetization entity is unlikely to have been asserted in litigation. The patent's re MAIN(tenance) lapse, combined with the lack of any assignment to a patent-assertion entity, supports the conclusion that no known litigation exists.
• Do not auto-correct. I searched only for the literal identifier 8,004,994 / 8004994. I did not include results for similar numbers, per your instruction.

Bottom line: No known litigation involves US Patent No. 8,004,994. If litigation exists, it is not reflected in the sources reachable through my searches, and I would recommend a direct PACER docket search to conclusively rule it out.

Proceedings overview

Zero AIA trial proceedings (IPR, PGR, or CBM) have ever been filed against US 8,004,994 — no institutions, no denials, no Final Written Decisions, no settlements, no appeals: the structured USPTO Open Data Portal block reports no AIA trial proceedings on file, and every web query I ran corroborates that negative. The bottom-line defensive posture is therefore not "the patent is hardened" and not "the claims are dead" — it is "the patent was never stress-tested at the Board, and it does not need to be, because it expired for non-payment of maintenance fees in 2019 and is unenforceable by lapse." A defendant receiving a demand letter citing this patent should lead with the maintenance-fee lapse (effective 2019-08-23, per the legal-events record), not with an IPR strategy.

Individual proceedings

None. There is no proceeding to report for US 8,004,994.

I am deliberately not fabricating a proceeding number, petitioner, panel, or disposition. The canonical structured block states there are no AIA trial proceedings on file, and I found no counter-evidence.

Disambiguation — the "8004994" hits that are not proceedings on this patent

Three search hits contain the literal string 8004994, and none is a PTAB proceeding involving this patent. Flagging them so they aren't mistaken for docket entries:

1. IPR2018-01512 — Centripetal Networks / Cisco, Ex. 1002 (File History of U.S. 9,565,213). In that exhibit, "8004994" appears adjacent to "20100242098" in what is plainly an examiner's EAST/SEARCH string list — it is a search query recorded during prosecution of a different patent, not a citation to, or party in, an AIA trial on the '994 patent.
   • Source: Docket Alarm — IPR2018-01512, Exhibit 1002
2. Unified Patents portal — US-10511572-B2 and US-10659573-B2 list US-8004994-B1 (Azimuth Systems Inc, priority 2006-10-31) in a "related references" panel. That is a citation relationship, not a party or a challenge.
   • Source: portal.unifiedpatents.com/patents/patent/US-10511572-B2
3. Unrelated Azimuth-named hits (an Azimuth Motor/Worm Gear claim chart in IPR2023-00827, and azimuthinc.com biometrics material) — different entities entirely.

⚠️ Cross-reference flag (priority date). The previously generated summary states priority/filing 2006-11-01 (from Google Patents). The Unified Patents portal entries render the same patent's priority date as 2006-10-31. This is a one-day discrepancy in third-party data, not a contradiction in the patent record itself — Google Patents' 2006-11-01 remains authoritative for this analysis, but note it if you cite priority externally.

Strategic summary

Claim status: everything is UNTESTED. All 13 claims — independent claims 1, 8, 11, and 13 and dependents 2–7, 9, 10, 12 — are in their as-issued form. No claim has been canceled, disclaimed, or narrowed by a PTAB certificate. There is no IPR certificate of cancellation, no adverse judgment, no statutory disclaimer on file in the record supplied. So any statement that "claims 1–5 are canceled" or that "the patent is hardened by surviving two IPRs" is false for this patent in both directions. The claims are simply unadjudicated, and now expired.

Estoppel landscape: empty. Because no IPR/PGR was ever instituted, 35 U.S.C. § 315(e)(2) estoppel never attached to anyone. There are no petitioners, no privies, and no grounds that are "off the table" against this patent. Practically, though, the estoppel question is moot: a patent that lapsed under 37 C.F.R. § 1.362 cannot be enforced for acts occurring after lapse, so invalidity grounds matter far less than the enforceability/lapse defense. If any pre-lapse infringement is at issue (the patent's enforceability window ran from grant 2011-08-23 until lapse effective 2019-08-23), all prior-art grounds from the six cited references — US 2002/0095636 (Tatsumi), US 2003/0133458 (Sato), US 2005/0068900 (Intel), US 2006/0012388 (Acradyan Technology), US 2006/0120283 (Northrop Grumman), US 2006/0274657 (Olgaard) — remain available to any challenger, subject only to § 315(b)/§ 325(a) timing under the patent's post-issuance window.

Pattern signals: none. No petitioner has filed even one IPR against this patent, let alone multiple. No defensive aggregator (Unified Patents, RPX, etc.) appears in the chain for the '994 patent — the Unified Patents hits above are merely citation pages. The patent owner never filed a Notice of Appeal to the Federal Circuit, because there was never a FWD to appeal. The patent never left Azimuth Systems' hands (only the original 2006 assignment, Reel/Frame 018462/0641, appears of record), and it was never transferred to a monetization entity. That combination — original assignee, no enforcement entity, maintenance fees abandoned in 2019, zero IPRs — is the classic profile of a patent that was never asserted and never needed a defense; well-asserted patents reliably attract IPRs, and this one attracted none.

Recommended next steps

• If you are a defendant facing assertion of US 8,004,994: the dispositive point is not an IPR — it is the lapse. Per the legal-events record, LAPS/STCH entries dated 2019-09-30 record expiration "due to nonpayment of maintenance fees under 37 C.F.R. 1.362," with an effective lapse date of 2019-08-23. Confirm this directly at USPTO PatentCenter, application 11/555,324 (patentcenter.uspto.gov) and at the USPTO Patent Fee/Status portal, then move to dismiss or seek summary judgment on unenforceability for any post-2019-08-23 conduct. Note the small-entity/undiscounted entity-status shuffle on 2015-12-07 / 2016-12-03 if fee-payment compliance ever becomes contested.
• If some pre-lapse window is genuinely at issue: there is no PTAB record to rely on, so you would be litigating invalidity from scratch in district court under §§ 102/103/112. The six cited references in the "Citations" table are the natural starting set; I have not analyzed their substantive effect on any claim, so treat that as a research task, not a conclusion.
• If you were hoping for an FWD to quote: there isn't one. Do not cite an IPR number, panel, or disposition for this patent — none exists. Any document that quotes an FWD on US 8,004,994 should be treated as fabricated or as referring to a different patent number.
• Verify before relying on the negative. I confirmed the absence of PTAB activity against the structured ODP block and multiple web sources, but my tool access was limited. To certify the negative, query PTAB E2E (ptacts.uspto.gov/ptabweb/) by patent number and CourtListener's RECAP docket search (courtlistener.com) for both "8,004,994" and "8004994". Treat my result as "no evidence found," not as a certified null.

Bottom line: There is no PTAB story for US 8,004,994 — zero AIA trials, zero FWDs, zero appeals, zero estoppel. The real story is a lapsed maintenance fee in 2019 and an unasserted, unmonetized patent that never warranted an IPR in the first place.

I'll research the assignment chain and the assignee's corporate history before building the record.

Inventors

Unusual-pattern check: Not present. There is no evidence that all three inventors departed Azimuth within 12 months of the 2006-11-01 filing. Two of the three are demonstrably still tied to Azimuth work for years afterward (Spilman's Azimuth patents run through 2011; Azimuth itself continued filing into 2016, ending with ~16–17 patents per CB Insights). The inventors' later careers (Spilman → TapLink, Graziani → Sonos) read as normal engineer mobility, not a portfolio wind-down.

Caveat / contradiction flag: The patent text in the supplied record lists "packet 1 from source 12 corresponds, packet 3 ('YOU') on the destination list" and then immediately "packet 4 on the source list corresponds to packet 9 ('YOU')" — an internal inconsistency in the source document, noted but not corrected.

Original assignee

Azimuth Systems, Inc., 31 Nagog Park, Acton, Massachusetts 01720 — the entity named on the issued patent and the only recorded assignee in the chain.

• Line of business: Wireless/RF test and measurement equipment. Founded 2002; primary products were wireless channel emulators (ACE™ MIMO portfolio, Spider™ controlled RF environments) and a Wi-Fi/VoWiFi test platform (the "W-Series" and "Director II" test executive referenced in the specification). Customers were network operators, device/module OEMs, chipset vendors, and infrastructure providers.
• Did it ship a product embodying the claims? Yes — on the record. The specification itself identifies the commercial embodiment: "a Voice over IP WiFi Test Suite (VoWiFi) executing on a W-Series test platform provided by Azimuth Systems, Inc. of Acton Mass." The claims recite QoS (drop rate, delay, jitter) measurement of a DUT — precisely the function of that platform. Azimuth was also the first to market a Wi-Fi test solution and first MIMO test solution per its own corporate materials. This is a genuine operating-company patent, not a paper asset.
• Funding / ownership posture: VC-backed, ~$35.8M raised across Series A (2002) through Series D (Dec 2006 — one month after this filing), with North Bridge Venture Partners and Kodiak Venture Partners.
• Current status: Acquired. Anritsu Corporation (Atsugi, Kanagawa, Japan) signed the purchase agreement 2016-09-28 and announced the acquisition 2016-09-29, making Azimuth a wholly owned subsidiary. AGC Partners was Azimuth's sole financial advisor. Azimuth had ~85 employees at the time. No bankruptcy, no Chapter 7/11, no fire-sale.

Assignment timeline

The Assignment Center returned only one recorded assignment chain for this patent in the record available to me. I could not complete a live interactive query of https://assignmentcenter.uspto.gov/ within my tool budget, so the entries below are reconstructed from the USPTO legal-events block reproduced in the authoritative patent record, which carries the same reel/frame data Assignment Center surfaces.

• 2006-10-30 / 2006-10-31 (executed, per signing dates) / recorded 2006-11-01 — Reel 018462 / Frame 0641

  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST" (USPTO event code AS)
  • Assignor: Prashant Darisi; Jeremy Spilman; Jose Graziani (all individual inventors)
  • Assignee: AZIMUTH SYSTEMS, INC., a Massachusetts corporation, Acton, MA
  • Correspondent: Not available in the source I could reach. The Google Patents legal-events entry for Reel 018462/0641 does not reproduce a correspondent/attorney-of-record field, and I was unable to run the Assignment Center detail query that exposes it. I am flagging this as a gap rather than guessing — the correspondent field is the single most valuable datum for the NPE test requested, and it is the one item I could not verify.
  • Context: Pre-filing inventor-to-company assignment. Executed two days before the 2006-11-01 filing date and recorded the same day the application was filed. This is the standard employment/obligation-to-assign instrument, not an acquisition, sale, securitization, or reorg.

• No post-issuance assignment is recorded for US 8,004,994 in any source I reached. Specifically, there is no recorded assignment to Anritsu Corporation notwithstanding the 2016 acquisition. The Google Patents "Current Assignee" field still reads Azimuth Systems, Inc., one decade after the Anritsu deal.

  Why this is coherent, not a data gap: the Anritsu transaction was structured as a share purchase ("acquires shares of Azimuth Systems, Inc. … to make it a wholly owned subsidiary"). In an equity acquisition, patent title stays with the acquired corporation as the surviving legal owner — no 37 CFR 3.11 recordation is required, and none appears. The patent therefore sits inside the Anritsu subsidiary without an assignment link in the public record.

  Corroborating evidence that ownership/control did in fact pass to a large entity: the fee-payment record shows the patent holder claiming small-entity status on 2015-12-07, then filing PAT HOLDER NO LONGER CLAIMS SMALL ENTITY STATUS / entity status set to UNDISCOUNTED on 2016-12-03 — roughly nine weeks after the Anritsu closing. That entity-status flip is the clearest fingerprint of the Anritsu acquisition in this patent's prosecution history, even in the absence of an assignment.

Timeline diagram

timeline
    title Ownership of US 8004994
    2006 : Inventors assign rights to Azimuth
         : Application filed same day
    2011 : Patent issues to Azimuth Systems
    2015 : Small entity fee status claimed
    2016 : Anritsu acquires Azimuth as subsidiary
         : Entity status flipped to undiscounted
    2019 : Patent lapses for unpaid fees

(Event text kept to alphanumerics and spaces; no parentheses, slashes, or commas.)

NPE / troll-pattern signals

Verdict

Insufficient data — in the specific sense the rubric contemplates: only the original assignment exists.

The chain is a single link, Reel 018462/0641 (executed 2006-10-30/31, recorded 2006-11-01), from the three inventors to Azimuth Systems, Inc. — a VC-backed operating company that shipped the very product described in the specification (the VoWiFi test suite on the Azimuth W-Series platform) and was later acquired as a going concern by Anritsu Corporation on 2016-09-28/29 via share purchase, leaving title with the Azimuth subsidiary and no assignment on record. Zero of the eight NPE signals is present; the closest thing to a monetization event is the Anritsu equity acquisition, which is the profile of a strategic operating-company buyer rather than an asserter. The patent then lapsed for non-payment of maintenance fees effective 2019-08-23 (LAPS/STCH recorded 2019-09-30) — an economically irrational thing for an NPE holding company to do, and further inconsistent with any assertion-driven chain.

Verification links:

• Assignment Center (search by patent number 8004994): https://assignmentcenter.uspto.gov/
• Legacy assignment search index: https://assignment.uspto.gov/patent/index.html
• Authoritative patent record and legal events: https://patents.google.com/patent/US8004994/en
• Anritsu acquisition release: http://dl.cdn-anritsu.com/ja-jp/about-anritsu/news/Anritsu_Azimuth_newsrelease_E.pdf

Two unverified items I am explicitly not asserting either way:

1. The correspondent of record on Reel 018462/0641 — not exposed in the record I could reach. Recurrence analysis (Signal 3) cannot be completed without it.
2. Whether any Azimuth/Anritsu assignment recorded after 2016 covers this patent under a different application-number indexing — I found no such record, and the persistence of "Azimuth Systems, Inc." as current assignee plus the entity-status flip argues the equity-acquisition explanation is correct, but a direct Assignment Center query would settle it.

Do not read the negative findings above as proof that no assignment exists — read them as "no assignment beyond Reel 018462/0641 appears in any source reachable here."

I'll search for the patent and its cited references to build a grounded prior-art analysis.

Let me look up the specific cited references to characterize each one.

Prior-Art Analysis — US 8,004,994 B1

Building on the prior sections: the bibliographic record (App. 11/555,324; filed/priority 2006‑11‑01; granted 2011‑08‑23; Azimuth Systems; 13 claims, independent 1, 8, 11, 13; expired fee‑related) and the plain‑language claim map already provided are treated as authoritative and are not repeated here. This section addresses your specific request: the prior art cited on the face of US 8,004,994 and its potential § 102 relevance.

Scope and source note

Google Patents lists six references under "Patent Citations (6)" — i.e., the references the examiner cited against this application. Those six, and only those six, are the prior art for this analysis. The much larger "Cited By (93)" list is subsequent art (patents that cite this patent) and cannot be § 102 prior art against it; I exclude it.

Important grounding caveats:

• I verified the full text of three cited references by search (Acradyan US 2006/0012388 A1; Northrop Grumman US 2006/0120283 A1; Olgaard US 2006/0274657 A1).
• For US 2005/0068900 A1 (Intel) I confirmed the citation, title, and dates only.
• For US 2002/0095636 A1 (Tatsumi) and US 2003/0133458 A1 (Sato) my search budget ran out before I could pull full text; I characterize them from the patent's own citation table and title, and I flag where the substance is unverified. Do not treat those two descriptions as confirmed.
• Statutory framing: because this application was filed 2006‑11‑01 (pre‑AIA), § 102 includes the pre‑AIA subsections (a), (b), and (e) (published applications/patents as of their effective U.S. filing date). I note the date issue for the Olgaard reference below.

The six cited references

Date trap on reference #6 (Olgaard): it published 2006‑12‑07, after the 2006‑11‑01 filing date, so it is not § 102(a) or § 102(b) art. It can only be § 102(e) art, available as of its 2005‑06‑01 U.S. filing date. Any anticipation case built on Olgaard must be pleaded as § 102(e).

Per-reference analysis

1. US 2002/0095636 A1 — Tatsumi (transmitter/receiver/broadcast data delivery)

• Citation: US 2002/0095636 A1; pub. 2002‑07‑18; priority 2000‑11‑06; Hidenori Tatsumi.
• Description (title‑level, unverified): concerns a transmitter, a receiver, and a method of delivering broadcast data — i.e., error handling/reception reporting in a broadcast transport. (I could not pull the full text; treat as provisional.)
• Potential § 102 target: at most the retransmission / acknowledgement‑reporting background relevant to independent claim 11 (delay incorporating retransmission) and dependent claim 2 (recording acknowledgements). It does not appear to disclose the source‑list/destination‑list comparison or the ACK‑based validation of a dropped‑packet list, so on the record available it does not anticipate any claim. Verify against full text before relying on it.

2. US 2003/0133458 A1 — Sato (unicast‑to‑multicast converter and monitoring system)

• Citation: US 2003/0133458 A1; pub. 2003‑07‑17; priority 2002‑01‑17; Masaaki Sato.
• Description (title‑level, unverified): a unicast‑to‑multicast conversion apparatus plus a "monitoring system" in which it is used. (Full text not retrieved; description provisional.)
• Potential § 102 target: the "monitoring system" element touches the generic monitoring step of claim 1 / claim 8 ("monitoring traffic… storing packets"), but monitoring alone is not the inventive feature. No evidence it discloses the dropped‑packet‑list comparison or ancillary‑information validation. No anticipation of any claim on the available record.

3. US 2005/0068900 A1 — Intel (data burst transmission in WLAN devices/systems)

• Citation: US 2005/0068900 A1; pub. 2005‑03‑31; priority 2003‑09‑30; Intel Corporation.
• Description: WLAN data‑burst transmission methods, including framed burst/acknowledgement exchanges in devices and systems. This is the 802.11 / WLAN‑acknowledgement contextual art that the examiner of an 802.11‑test patent would naturally cite.
• Potential § 102 target: background supporting the "802.11 … ACK packet" limitation of dependent claim 3 and the general acknowledgement concept in claim 2. It does not, on the title/date record, disclose comparing source and destination captures or validating a dropped‑packet list, so it does not anticipate claims 1, 8, 11, or 13, and — being about transmission of bursts, not about measuring DUT drop rate — does not fully anticipate claim 3 either. Best viewed as § 103 context.

4. US 2006/0012388 A1 — Acradyan Technology (System and method for testing wireless devices) — closest in subject matter to claims 1/8

• Citation: US 2006/0012388 A1; pub. 2006‑01‑19; priority 2004‑07‑13; Acradyan Technology Corporation.
• Description (verified): a wireless test system with two DUTs, a channel emulator, and a signal monitor (power meter + spectrum analyzer) coupled through 3‑port power dividers/couplers, enabling simultaneous Tx/Rx testing. The specification expressly addresses loss measurement by counting received vs. transmitted packets — e.g., "if the golden radio sends 1000 packets … and 900 packets are recorded in the register as having been received, 10% of the packets obviously have been lost" — and obtains "minimal input power and PER" and uplink/downlink throughput.
• Potential § 102 target: the closest cited art to the drop‑rate / PER measurement thrust of claim 1 (a "quality of service statistic") and claim 8 (analyzer computing a QoS statistic). It discloses counting transmitted vs. received packets to quantify loss.
• Anticipation verdict: No. It lacks the two elements that define the invention — (i) building a dropped packet list by comparing a source list to a destination list, and (ii) validating that list with ancillary information (ACKs) to determine whether a "dropped" packet was actually received by the Destination DUT. It is properly a § 103 reference for the general "measure packet loss in wireless test" environment, not a § 102 anticipatory reference.

5. US 2006/0120283 A1 — Northrop Grumman (Real‑time packet processing system and method)

• Citation: US 2006/0120283 A1; pub. 2006‑06‑08; priority 2004‑11‑19 (prov. 60/629,331), filed 2005‑11‑18; Northrop Grumman Corp.; later granted as US 8,213,413 B2.
• Description (verified): real‑time packet detection/filtering/routing in packet‑switching systems for audio/video/multimedia. A header‑test module routes real‑time traffic to a real‑time packet handler; the handler (a) determines the packet's source address, (b) checks it against a list of source addresses, and (c) checks whether the source address is associated with an active channel (table‑based), modifying and queuing the packet. Explicitly frames the goal as delivering "the quality of service expected in mission critical systems."
• Potential § 102 target: the source‑address list / channel‑table and QoS language could be argued against the bare "source list" and "quality of service statistic" wording in claims 1 / 8, and its real‑time multimedia (VoIP‑like) context parallels the specification's VoIP focus.
• Anticipation verdict: No. It is a packet routing/processing system, not a test/monitoring system; it has no destination capture list, no dropped‑packet list, no ACK‑inference validation, and no delay‑from‑retransmission computation. Relevant, if at all, as § 103 background showing list‑based packet handling and QoS in packet networks.

6. US 2006/0274657 A1 — Olgaard (Method for measuring sensitivity of data packet signal receiver) — closest on the "drop/PER" concept; § 102(e) only

• Citation: US 2006/0274657 A1; pub. 2006‑12‑07; filed 2005‑06‑01; Christian Volf Olgaard (and Carsten Andersen), assigned to LitePoint Corporation; later granted as US 8,189,483 B2.
• Description (verified): measures receiver sensitivity by varying power level and/or modulation across a controlled sequence of data packets; computes packet error rate (PER) as "packets transmitted less packets correctly received, divided by packets transmitted," and derives the expected PER‑vs‑power curve. Explicitly about counting correctly received packets versus transmitted packets to characterize receiver loss.
• Potential § 102 target: the PER / correctly‑received‑packet‑counting concept bears on the "quality of service statistic" of claims 1 / 8 (packet drop rate) and on the QoS‑statistic‑based‑on‑delay output of claim 13.
• Anticipation verdict: No — and note the § 102(e)‑only date status. It discloses PER measurement of a receiver under test in a controlled sensitivity sweep; it does not disclose monitoring a Source DUT→Destination DUT exchange, source/destination list comparison, ACK‑based validation of a dropped‑packet list, or retransmission‑aware delay. Best cited as § 102(e)/§ 103 art for the packet‑loss‑statistic background.

Coverage judgment: what each reference can and cannot reach

Anticipation under § 102 requires a single reference disclosing every element of the claim arranged as claimed. Against that standard:

Bottom line: On the record I can verify, none of the six examiner‑cited references anticipates any claim of US 8,004,994 under § 102. The independent claims 1, 8, 11 and 13 all require at least one element absent from every cited reference:

• Claims 1 & 8 require the combination of a source‑list‑vs‑destination‑list comparison producing a dropped‑packet list and validation of that list via ancillary information (ACKs) to determine actual receipt by the Destination DUT. Acradyan and Olgaard address packet loss/PER but not this two‑list + ACK‑inference structure; Northrop addresses source lists but in a routing (non‑test) context.
• Claims 11 & 13 require computing delay from a first‑transmission PTT against a destination‑list D‑MATCH tied to a retransmission, keyed to a valid ACK (including an ACK to the retransmitted version). None of the six references discloses retransmission‑aware delay on this basis; Tatsumi/Intel are the nearest in touching acknowledgements/retransmission but show nothing resembling the PTT‑vs‑D‑MATCH computation.

Accordingly, the citations are best characterized as § 103 obviousness context and claim‑construction aids, not anticipatory art — which is consistent with the fact that the examiner allowed the application over them and with the prior-art "keywords" listed on the face of the patent (packet / list / destination / source / dut).

Flags and gaps (explicit, not fabricated)

1. Two references unverified in full text. US 2002/0095636 A1 (Tatsumi) and US 2003/0133458 A1 (Sato) are described here from the patent's citation table and titles only; I did not retrieve their specifications. Their descriptions must be confirmed before any § 102 argument rests on them.
2. Intel reference partially verified. US 2005/0068900 A1 is confirmed as to citation, title, dates, and assignee; claim‑level content was not pulled.
3. No contradiction found with the previously generated sections; this prior‑art analysis is consistent with them. Nothing here changes the bibliographic or litigation findings already stated.
4. Standard caution on "most relevant." "Most relevant" depends on the claim asserted. For the drop‑rate/validation concept, Acradyan (US 2006/0012388 A1) is the closest cited reference; for the statistic itself, Olgaard (US 2006/0274657 A1) is closest — but only as § 102(e) art and still non‑anticipatory. I state this as my assessment on the available record, and I flag that a full‑text review of all six (plus non‑cited art such as US 6,577,648 and CN 101926154B, which appear in "Similar Documents") would be needed for a definitive invalidity opinion.

I'll research the cited prior-art references to ground the § 103 analysis in their actual disclosures.

Obviousness Analysis — U.S. Patent No. 8,004,994 B1

Scope, sources, and a threshold caution

Per instructions, I limited this § 103 analysis to the prior art appearing in the Prior Art / Citations (6) section of the Google Patents page for US 8,004,994 B1. All six are pre-2006 publications, so all qualify as prior art under § 102(a)/(b) as of the 2006‑11‑01 priority date:

Two cautions before the merits:

1. Ref. B (Sato) is unusable here. My retrieval of its specification hit the tool‑call limit. I will not build a combination on a reference whose disclosure I have not read — doing so would be fabrication.
2. All six references were already before the examiner, and the claims issued over them. Under 35 U.S.C. § 282 the patent carries a presumption of validity. A § 103 attack therefore has to do more than re‑argue the art of record; realistically it needs (a) a combination the examiner did not make, or (b) an express motivation the examiner overlooked. My analysis below is candid about where the record is strong and where it is thin.

Legal framework and level of ordinary skill

Under Graham v. John Deere, I assess (1) scope and content of the prior art, (2) differences between the art and the claims, (3) the level of ordinary skill, and (4) objective indicia. Under KSR v. Teleflex, a claim is obvious where the elements were known and there existed a "reason to combine" — including where the combination is "a predictable variation," "obvious to try," or the product of "ordinary creativity," and the art is a predictable one. This field is highly predictable: packet‑based network testing is a mature, rules‑based engineering discipline, which weighs in favor of obviousness for many of the dependent claims.

POSITA (my construction): a Bachelor's in EE/CS (or equivalent) plus 2–4 years' experience in wireless LAN protocol testing / network performance measurement, with working familiarity with the IEEE 802.11 MAC (DCF, ACK, retry limits, IFS/SIFS/ACK‑timeout), packet capture tooling, and test‑bench hardware. This is the level at which the objective evidence (Refs. A, C, D, F) is written.

Core differences between the art of record and the claims

The claimed subject matter divides into two families:

• Claims 1–10 (drop‑rate / QoS with ACK‑based validation): compare a source list to a destination list → build a dropped packet list → validate it with "ancillary information" (recorded ACKs, within a PAW) → compute and output a QoS statistic.
• Claims 11–13 (delay including retransmission): timestamp a first transmission (PTT), find a valid ACK (including an ACK to a retransmission), locate the matching destination‑list entry (D‑MATCH), compute delay from PTT→D‑MATCH, output in tangible form.

The single most important difference from the art of record is this: the patent uses an ACK observed by the monitor as inferential proof of receipt when the monitor's own capture shows no packet — i.e., it distinguishes the Destination DUT's drops from the monitor's frame errors. None of Refs. A, C, D, E, or F states that problem expressly, and none states that solution. That is where any nonobviousness argument must be fought.

Combination I — Claims 1, 2, 4–10 (and 3)

Primary reference: F (Olgaard). Ref. F teaches the essential measurement skeleton: transmit a known sequence of packets to a DUT, count the correctly received packets, and derive an error/loss statistic from the difference. Its disclosure that "the PER is the number of packets transmitted less the number of correctly received packets… divided by the number of packets transmitted" supplies the compare‑sent‑against‑received → loss metric → output structure that reads on claim 1 steps (a)–(c), (e), (f), and on the analyzer of claim 8.

Secondary reference: C (Intel). Ref. C supplies the element the patent actually leans on: the 802.11 acknowledgment. Its specification states verbatim that "the minimal MAC protocol consists of two frames: 1) a frame sent from a transmitter to a receiver; and 2) an acknowledgment (ACK) from the receiver that the frame was received correctly," and that under 802.11 "the transmitter [must] wait for an ACK after transmission of each packet." That is precisely the ancillary information of claims 2, 4, 9, and 10 — an ACK is, by definition, evidence of correct receipt. It also directly supplies claim 3's "802.11 ACK packet."

Secondary reference: D (Acradyan). Ref. D supplies the test‑bench architecture of claim 1's two monitoring steps: two DUTs, a channel emulator, and a signal monitor coupled to both DUTs. Its background section even frames loss measurement as "if the golden radio sends 1000 packets… and 900 packets are recorded… 10% of the packets obviously have been lost."

Secondary reference: E (Northrop Grumman). Ref. E supplies the list mechanics: a packet handler that "determine[s] the source address included in the packet data," checks it "against a list of source addresses," and stores the filtered/processed packets in a queue — i.e., a capture store keyed by source, reading on the "source list" data structure.

Motivation to combine (KSR rationales):

• Same field, same problem: A, C, D, E, F are all directed to wireless/packet test and measurement, or to the 802.11 ACK mechanism those tests must observe. KSR sanctions combining references "designed to address the same problem."
• Known technique, predictable result: counting received frames against transmitted frames (F, D) is the standard PER methodology; adding the protocol's own ACK as a second, independent receipt indicator (C) is the kind of "predictable variation" KSR treats as obvious, especially where measurement accuracy is the recognized goal (F expressly worries about test‑result stability).
• Design incentive: the whole point of a test platform is to characterize the DUT, not the instrument; a POSITA building a WLAN test set (D) would be motivated to avoid attributing the monitor's own RF frame errors to the DUT — exactly the objective problem the specification identifies — and the ACK (C) is the ready‑made, protocol‑native corroborating signal.

On claims 4, 5, 9, 10 (PAW / removal from list): Claim 4's PAW finds support in C's teaching that the transmitter waits for an ACK and re‑transmits if none arrives (and in the 802.11 standard's ACK‑timeout/retry behavior), and claim 5's "remove the packet from the dropped packet list" is a routine bookkeeping step once the ACK‑within‑window inference is made. Claims 6, 7, 9 (gap list) are a data‑structure choice: once you perform a sequential match between source and destination lists (F, E) and find a miss, recording the surrounding matched sequence numbers to bound the "gap" is a predictable implementation detail — a weak basis for patentability.

Assessment: Prima facie obvious as to claims 2, 3, 4, 5, 6, 7, 9, 10, and arguable as to claims 1 and 8. The vulnerability in the attack is that the combination must be shown to yield the specific inference "ACK observed by the monitor ⇒ packet was received by the DUT even though absent from the destination list." That inference is nowhere stated in the record; it must be supplied by the POSITA's reasoning. Expect the patentee to argue the examiner already had F, C, D, E and still allowed the claims.

Combination II — Claims 11, 12, 13 (delay incorporating retransmission)

Primary reference: A (Tatsumi). Ref. A is the closest art for the delay family: it detects dropped/erroneous packets, coordinates a retransmission, and provides a "data recovering unit… [that] recovers the data using packets corresponding to a retransmission request… and normally received packets." That is the claimed concept of associating a first transmission with a later valid ACK/retransmission (claim 11's "valid acknowledgement… includes an acknowledgement to a re‑transmitted version of the packet"; claim 12's PAW‑qualified retransmission ACK).

Secondary reference: C (Intel). Ref. C supplies the 802.11 retransmission‑timing framework — ACK required per frame, IFS spacing, and (via Block ACK) acknowledgment of previously transmitted packets — i.e., the timing substrate on which PTT→ACK interval measurement (claim 11) rests.

Secondary reference: F (Olgaard) / D (Acradyan). Both measure packet‑signal/test performance over transmitted sequences and provide the source‑side "known number of packets transmitted / timestamps" foundation for PTT and D‑MATCH.

Motivation to combine: A POSITA measuring delay/QoS in a retransmission‑capable WLAN would want the delay actually experienced by a delivered packet. It is a matter of ordinary engineering knowledge — and squarely within KSR's "predictable variation" — that an effective delay measured only from a first transmission can be extended by retransmissions; measuring from first transmission to valid ACK is the straightforward fix. Ref. A supplies retransmission‑aware recovery; Ref. C supplies the 802.11 ACK timing. Claims 12 and 13 are then conventional details (applying the PAW to the retransmitted packet; expressing the result as a QoS statistic).

Assessment: Obviousness here is plausible but materially weaker than for Combination I. The record contains no reference that frames the problem (under‑reporting delay because retransmission time is ignored) or the solution (start the clock at first transmission, stop at the ACK). The specification presents that recognition as the invention. A court or examiner could accept the KSR "obvious to try / predictable variation" rationale, or could find genuine ingenuity in selecting the first transmission as the delay origin. I would rate claims 11–13 as the most defensible of the patent.

Where the patentee can resist, and where the record is thin

• Express motivation is missing for the core validation step. No reference of record states the monitor‑frame‑error problem or the ACK‑as‑inferential‑evidence solution. The combination relies on KSR "common sense" reasoning about instrument accuracy, which is arguable rather than documentary.
• Different purposes. A concerns broadcast data delivery efficiency; E concerns routing/queueing of real‑time packets; F concerns receiver sensitivity, not a two‑point monitor architecture. The patentee will stress that none measures a monitor's frame error rate, and that F's "received packet" count comes from the DUT itself, not from an independent monitor.
• All six references are of record, so the presumption of validity (§ 282) applies, and the strongest attacks will likely require art outside this list (e.g., a network‑monitoring/QoS patent that expressly distinguishes instrument loss from device loss).
• Objective indicia: I found none in the record — no evidence of commercial success tied to these claims, no licensing, no copying, no praise. The patent lapsed for non‑payment of maintenance fees effective 2019‑08‑23, and the prior litigation section found no assertion of the patent. There is therefore no objective‑indicia counterweight to obviousness.

Procedural note / contradiction flag: the task states the current date is April 26, 2026, whereas the previously generated sections are dated 2026‑09‑29. This is immaterial to the § 103 analysis (validity is assessed as of the 2006‑11‑01 priority date), but I am flagging the inconsistency rather than silently harmonizing it.

Bottom line

• Claims 1–10: prima facie obvious over Olgaard (F) + Intel (C) + Acradyan (D), optionally with Northrop Grumman (E) for the capture‑list mechanics. Claims 2–7, 9, and 10 are the weakest (the ACK, PAW, removal, and gap‑list elements are all supplied or rendered predictable by C, F, and A). The inference step in claims 1 and 8 is the only genuinely contestable element.
• Claims 11–13: substantially harder. Tatsumi (A) + Intel (C) frames the retransmission/ACK relationship, but the record does not expressly teach measuring delay from the first transmission to the acknowledgment, so these claims retain the best nonobviousness footing.
• Highest‑value gap to close: cite prior art off the examiner's record that expressly recognizes the network monitor's own frame‑loss problem and uses protocol ACKs to correct for it. That is the pivot on which a successful § 103 challenge to claims 1 and 8 would turn.