US 6233073

I'll search for the specific patent number in the specified databases.

Let me run additional targeted searches for litigation and docket information.

Search Results — US Patent 6,233,073

Important scope note: I searched USPTO/Google Patents, FreePatentsOnline, Justia, and general web sources for patent number 6233073 and for any Court of Appeals for the Federal Circuit (CAFC) 2026 docket activity. I found no CAFC docket, IPR, or litigation naming this patent. One search hit mentions a "'307 patent" in StratosAudio, Inc. v. Hyundai (W.D. Tex.) and related IPRs — but that is a different patent (a StratosAudio media/broadcast patent), not US 6,233,073. It should not be conflated with this IBM fiber-optic patent. If you are looking for CAFC 2026 activity, I have no authoritative evidence that US 6,233,073 appears on any 2026 CAFC docket; treat that as an open/uncertain point rather than a confirmed "no."

Bibliographic Data (from the authoritative full text)

Abstract (as issued)

Disclosed are a subsystem and a method for diagnosing a fiber optic network, the network comprising a fiber carrying an optic stream adhering to a defined protocol and a target for receiving/responding to the stream. The diagnosis comprises receiving the fiber optic stream and converting it to an electrical input stream with a FOSA (fiber optic subassembly), and recognizing, from the electrical input stream, a selected event (e.g., frame or command) related to the protocol. Then, in response to recognition of the selected event, recognizing a selected pattern of the electrical input stream (e.g., via mask logic), injecting a transmission error into the optic stream, and transmitting the error-bearing stream to the target via a FOSA. A FIFO buffer buffers the electrical input stream so the error can be injected at a selected point.

Plain-Language Overview of the Independent Claims

The patent has three independent claims — 1, 12, and 17 (claims 2–11 depend from claim 1; claims 13–16 depend from claim 12; claims 18–20 depend from claim 17).

Claim 1 — Method (optical-domain injection). A method for diagnosing a fiber optic network where the fiber carries an optic stream under a defined protocol and feeds a target. Steps: (a) receive the fiber stream and convert it to an electrical input stream; (b) recognize a selected event in the electrical stream that relates to the protocol; (c) in response to that event, recognize a selected pattern in the electrical stream; (d) in response to that pattern, inject a transmission error into the optic stream; and (e) transmit the optic stream (with the injected error) to the target.

Claim 12 — Method (electrical-domain injection). Same overall framework as claim 1, but the error is injected into the electrical input stream, and the method then explicitly recites (f) converting the electrical input stream to an optic stream and (g) transmitting that optic stream to the target. In other words, claim 12 is the same diagnostic concept framed with the inject-then-convert order spelled out in the claim body.

Claim 17 — Diagnostic subsystem (apparatus). A diagnostic subsystem inserted into the fiber optic network, comprising: (a) a first fiber optic subassembly converting the fiber's optic stream to an electrical input stream; (b) a comparator coupled to that subassembly that recognizes the selected event related to the protocol; (c) mask logic coupled to the comparator that recognizes the selected pattern (after the event) and injects a transmission error in response; and (d) a second fiber optic subassembly coupled to the mask logic that transmits the error-bearing optic stream to the target.

Representative dependent-claim concepts (context only): retransmit unmodified when no pattern is recognized (claims 2/13); abort injection if the incoming stream already has an error (claims 5/8/16); FIFO-buffer the input so the error can be placed at or before the recognized pattern (claim 6); compare buffered data to a preset value gated by a mask (claim 9); change a command and update the CRC (claim 10); substitute idle characters (claim 11); alter phase or timing of the optic stream (claims 14/15); and 8B/10B codec/serializer–deserializer details in the FOSAs (claims 19/20).

Key Points / Uncertainty Flags

• Filing-to-issue: filed 1998-07-30, issued 2001-05-15, ~2.8-year pendency; sole assignee IBM from filing.
• Status: Listed "Expired – Lifetime"; the anticipated expiration date is 2018-07-30 (20 years from filing). An expired patent generally cannot be asserted for ongoing infringement, which is consistent with the absence of any 2026 CAFC docket.
• Prior art cited within the patent: Glista (US 5,229,875 and 5,228,105), US 5,321,813 (Teradata), US 5,396,357 (Honeywell), US 5,524,218 and 5,602,667 (Unisys), US 5,559,963 (IBM), US 5,706,278 (Raytheon), US 5,774,242 (Northern Telecom).
• Notable downstream citation: EP 2,158,727 B1 (network in-line tester) cites US 6,233,073 as a system for testing a fiber optic network by injecting isolated errors — useful evidence of the patent's technical relevance even though it is expired.
• Uncertainty: I could not locate any 2026 CAFC docket entry, USPTO post-grant proceeding, or litigation naming US 6,233,073. I therefore cannot confirm the absence of such activity with certainty — I simply found no authoritative record of it. All bibliographic and claim details above are taken from the authoritative full text you supplied (source: patents.google.com/patent/US6233073/en), which I treated as controlling over conflicting web snippets.

I'll search for litigation involving US patent 6233073.

Let me search litigation-specific databases.

Let me try more targeted litigation databases.

Findings: Litigation involving US Patent 6,233,073

No litigation involving US Patent 6,233,073 was found in the sources searched.

What I searched

I ran targeted queries against patent litigation resources (including the Stanford NPE Litigation Database, Unified Patents materials, Docket Alarm/PTAB records, and general web/court-document sources), using both the plain number and the comma-formatted number ("6233073" and "6,233,073"), plus combinations with the title and assignee. None of these returned any district court, Federal Circuit, or PTAB proceeding in which US 6,233,073 is identified as an asserted or challenged patent. Search results that did mention "073" numbers related to clearly different patents — e.g., US 10,311,073 (MasterObjects v. Facebook/Meta), US 9,336,307 (Zaxcom), US RE50,307 (Kiwi Intellectual Assets), none of which is the patent at issue. I did not auto-correct or conflate any of these; they are distinct patents and are excluded.

Supporting evidence from the patent record itself

The Google Patents record for US 6,233,073 (fetched 2026‑10‑01) contains no litigation section. Its tabular data shows only:

• Patent Citations (9) — prior art cited (e.g., US 5,229,875 Glista; US 5,396,357 Goosen et al.)
• Cited By (54/97) — later patents citing the '073 patent (e.g., US 6,560,720, IBM "Error injection apparatus and method"; US 7,669,095, IBM "Methods and apparatus for error injection")
• Similar Documents and
• Legal Events limited to assignment and maintenance‑fee activity, ending with anticipated expiration on 2018‑07‑30 and status "Expired – Lifetime."

Google Patents includes litigation/reexamination panels for patents that have them; the absence of any such panel is consistent with there being no recorded litigation for this patent.

Case table

Important caveats

• Confidence level: moderate-to-high that no litigation exists, but not absolute. My conclusion rests on web-search results and the patent's public bibliographic record. I was unable to directly load a per-patent litigation page (e.g., a Unified Patents portal case list or a PACER docket search) for US 6,233,073 within the searches performed, and reached the tool-step limit before exhausting every database.
• Databases such as PACER, Docket Alarm, Lex Machina, or Unified Patents' portal would provide authoritative confirmation. A direct query for litigation where "6,233,073" appears as the asserted patent returned nothing.
• This patent (filed 1998‑07‑30, granted 2001‑05‑15, expired 2018‑07‑30) would have been enforceable roughly 2001–2018; if it was ever asserted, the case would fall in that window.

Bottom line: Based on the sources available to me, there is no known litigation involving US Patent 6,233,073. If you need a definitive negative confirmation, I recommend a direct PACER/RECAP or Lex Machina docket search on the exact patent number before relying on this for a legal opinion.

Proceedings overview

Total AIA trial proceedings on file for US 6,233,073: zero (0 IPRs, 0 PGRs, 0 CBMs). The USPTO Open Data Portal structured block returned no AIA trial proceedings for this patent, and independent web searches for a PTAB proceeding naming US 6,233,073 — by patent number, by title, and by inventor names (Bowers / Denning / Emberty / Klein) — surfaced nothing. There is no breakdown to give because there is no proceeding: no active trial, no FWD invalidating claims, no settlement, no institution denial, no Federal Circuit appeal. Bottom-line defensive posture: there is no IPR-based shortcut here and none is needed — this is a 1998-filed IBM patent that went Expired – Lifetime on 2018-07-30, so a defendant's leverage comes from the expiration date and the § 286 damages lookback, not from a PTAB record.

False positive to flag explicitly (do not conflate): Web searches repeatedly return hits for a "'307 patent." None of them are this patent. Two distinct distractors appeared:

1. StratosAudio, Inc. v. Hyundai Motor America (W.D. Tex.) and related IPRs — that "'307 patent" is US 9,336,307 B2 (StratosAudio media/broadcast patent, inventor Kelly Christensen). Its IPR2018-00972 FWD and the auto-OEM Fintiv briefing are not about US 6,233,073.
2. Universal Connectivity Technologies Inc. v. Lenovo / HP / Dell and IPR2024-00425 — that "'307 patent" concerns wireless devices with unique IDs and DAB receivers; it is likewise not US 6,233,073.

If any of those appear in a docket research summary alongside "6,233,073," they are cross-references or search noise, not proceedings against this patent.

Proceedings (none)

No proceeding block is provided because the canonical structured source lists none.

Strategic summary

Claim status of US 6,233,073. All 20 claims stand as issued: independent claims 1, 12, and 17, with dependents 2–11, 13–16, and 18–20. Because no AIA trial was ever instituted, no claim has been canceled, disclaimed, or amended in a PTAB proceeding, and there is no FWD, no certificate, and no substitute claim to point at. There is also no PTAB narrowing that a defendant could weaponize or that a patent owner would have to live with. The claims are "untested" before the Board, not "sustained" — an important distinction: a lack of adverse PTAB rulings here reflects disinterest, not demonstrated strength.

Why the absence of PTAB activity here is weak evidence of anything. The normal heuristic ("well-asserted patents eventually attract IPRs") does not apply cleanly to this patent. The patent issued 2001-05-15, was IBM-assigned throughout, and expired 2018-07-30 (20 years from the 1998-07-30 filing). CBM review was never available (not a financial-services patent), and PGR was long time-barred after the nine-month post-grant window. IPR remained technically available even post-expiration, but there is little economic incentive: no injunction exposure, and past damages are bounded by 35 U.S.C. § 286's six-year lookback. The last day on which infringement of this patent could occur was 2018-07-30, meaning a suit filed after 2024-07-30 would, on its face, recover nothing. I found no 2026 CAFC docket entry or litigation naming US 6,233,073 either, which is consistent with that timeline.

Estoppel landscape. Not applicable. 35 U.S.C. § 315(e)(2) estops only petitioners and their privies in an instituted IPR/PGR. With zero proceedings, no party is estopped, and any prior art is still fully available under §§ 102/103 in district court (subject only to the ordinary limits of § 282 and estoppel from other litigation). Conversely, there is no petitioner-side record to borrow and no IPR-file-history disclaimer to mine for a narrower construction — the only intrinsic narrowing evidence is the 1998–2001 prosecution history.

Pattern signals. No repeat petitioner (there is no petitioner). No defensive aggregator (Unified Patents or similar) in the chain. No patent-owner PTAB appeal activity. None of the "troll-hardened by surviving IPRs" markers are present — and equally, none of the "claims killed in IPR" markers are present.

Recommended next steps

• If no PTAB activity exists, say so plainly — it does. There is no FWD to quote, no proceeding number to link, and no claim-level disposition to report. Do not accept a citation to any "IPR on the '307 patent" without verifying the full patent number: the StratosAudio (US 9,336,307) and Universal Connectivity (IPR2024-00425) matters are different patents and different owners.
• Reframe the defense around expiration, not PTAB. Confirm the legal status verbatim ("Expired – Lifetime," anticipated expiration 2018-07-30) at Google Patents — https://patents.google.com/patent/US6233073/en — and check the USPTO Patent Center maintenance-fee/expiration record for the same event. Then evaluate the § 286 six-year damages window against the date any demand letter or complaint was served.
• Verify the absence yourself at the canonical sources rather than relying on this summary:
  • USPTO PTAB E2E / PTAB Center patent-number search — https://ptab.uspto.gov (and the PTAB petition/decision portal at https://ptacts.uspto.gov)
  • CourtListener full-text search for "6233073" — https://www.courtlistener.com (also spot-check for "6,233,073" and the patent title)
  • USPTO Patent Center for the application US 09/124,837 — https://patentcenter.uspto.gov
  • Federal Circuit docket/opinion search on CourtListener for the patent number
• If an active proceeding surfaces later (I found none), the milestones to track would be: the § 314(b) institution deadline (3 months from the patent owner's preliminary response, or its due date), the oral hearing, and the § 316(a)(11) one-year FWD deadline from institution — plus any § 315(b) one-year bar tied to service of a complaint.
• Caveat on confidence: my finding of zero PTAB proceedings rests on (a) the structured ODP block in this prompt and (b) targeted web searches. I cannot prove a negative, and a very recently filed petition might not yet be indexed. Treat "no PTAB activity" as highly likely but not formally certified; the E2E patent-number search above is the definitive check.

I'll research the assignment chain for US patent 6,233,073. Let me run targeted searches.

Let me dig deeper for the specific assignment reel/frame and any post-issuance transfers.

Let me verify whether there were any post-issuance transfers and confirm the inventors' employer.

I've reached the search-tool limit, so the following synthesizes the authoritative full text you supplied (Google Patents legal-events block for US 6,233,073) plus the assignment/docket searches already run. Where a datum could not be confirmed from a live USPTO record, I say so explicitly rather than fill it in.

Inventors

All four are listed as assignors on the single recorded assignment (Reel/Frame 009353/0987), which conveys "ASSIGNMENT OF ASSIGNORS INTEREST" from each inventor to IBM. That places all four as IBM personnel on the filing date (1998-07-30); their specific IBM site is not determinable from the assignment record I could access. The subject matter (Fibre Channel / ESCON FOSA diagnostics) is consistent with IBM's storage-subsystem development, but I will not assert a specific lab as fact.

Unusual-pattern note: No "all inventors depart within 12 months" signal is visible. There are no inventor-to-third-party re-assignments, no quitclaims, and no subsequent inventor-side recordings in the chain — consistent with a large, centralized corporate filer rather than a start-up whose team fragmented. This is a not present finding, not an absence of data.

Original assignee

International Business Machines Corporation (IBM) — named on the issued patent (original and current assignee per the record).

• Primary line of business: diversified enterprise computing — hardware, software, storage, and networking. This patent sits in IBM's storage/fibre-channel interface work.
• Product embodying the claims: The claimed subject matter is a diagnostic error-injection subsystem used to test fault tolerance of a fiber-optic link. It is plausibly an internal test harness / engineering tool rather than a standalone sold product. I have no evidence that IBM commercialized the claimed subsystem as a product; I therefore do not assert it "shipped a product embodying the claims." IBM unquestionably shipped the surrounding Fibre Channel/ESCON adapter and storage products, but that is not the same as shipping the claimed diagnostic subsystem.
• Current status: Operating. IBM has never been dissolved, acquired, or placed in bankruptcy. It remains one of the world's largest patent holders.
• Legal status of the patent: Expired – Lifetime; anticipated expiration 2018-07-30.

Assignment timeline

The Assignment Center grid could not be rendered directly in this session, so I rely on the Google Patents legal-events block embedded in the authoritative full text you provided. That block shows exactly one conveyance and a series of non-assignment fee/maintenance events. There are no merger, change-of-name, security-agreement, license, or release recordings.

• 1998-07-21 to 1998-07-24 (executed) / recorded 1998-07-30 — Reel 009353/0987
  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: Bowers, Michael Dana; Denning, Donald Eugene; Emberty, Robert George; Klein, Craig Anthony
  • Assignee: International Business Machines Corporation, New York (n/k/a Armonk, NY)
  • Correspondent: Not disclosed in the source. The Google Patents legal-events text records the reel/frame and signing dates but does not surface a correspondent attorney/firm for this internal-IBM recording. I will not fabricate one.
  • Context: Ordinary inventor-to-employer assignment executed days before the 1998-07-30 filing — the standard pre-filing corporate capture, not a fire-sale, reorg, securitization, or transfer-to-asserter.

Post-issuance events recorded (non-assignment, for completeness): fee-payment entries on 2003-12-09, 2004-09-22, 2008-09-23, and 2013-03-06 (with a 2012-12-24 late-payment reminder and a 2013-03-06 surcharge). These are maintenance-fee bookkeeping only and carry no ownership change.

Bottom line: There are no post-issuance recorded assignments. The original assignee, IBM, appears to have retained title throughout the patent's life.

Timeline diagram

timeline
    title Ownership of US 6233073
    1998 : Inventors assign to IBM
         : Recorded reel 009353 frame 0987
    2001 : Patent issued
    2018 : Anticipated expiration

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The only assignee on record is IBM (Reel/Frame 009353/0987, 1998-07-30). No "IP / Patents / Licensing / Holdings / Ventures" LLC ever appears.

2. Known asserter in the chain — not present. No assignee matches Acacia, Marathon, Intellectual Ventures, Wi-LAN, Conversant/Mosaid, Vringo, Pendrell, Round Rock, etc. IBM is an operating company, not an NPE. (Note: the "'307 patent" hit in StratosAudio v. Hyundai is a different patent — see the bookkeeping flag below.)

3. Repeat correspondent across the chain — not present / unclear. There is only a single recording, so no recurrence is possible. The correspondent of record is not disclosed in the source, so I cannot evaluate the identity of the recording attorney. Marked unclear on the raw data, not present as a signal.

4. Cascading transfers — not present. One assignment total; no chained LLC hops.

5. Pre-litigation transfer — not present. No transfer exists, and I found no litigation naming US 6,233,073. (Consistent with the earlier report of no 2026 CAFC docket and no IPR.)

6. Bankruptcy fire-sale — not present. IBM has no Chapter 7/11 history.

7. Privateering — not present. No operating-company-to-NPE conveyance exists.

8. Defensive aggregator — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN; it simply remains with IBM. This is original-assignee retention, which is the inverse of both privateering and defensive-aggregation.

Verdict

Insufficient data — specifically, only the original assignment exists.

Justification: The sole recorded conveyance is Reel/Frame 009353/0987 (executed 1998-07-21/24, recorded 1998-07-30), inventors → IBM; the entire remaining legal-event history is maintenance-fee bookkeeping (2003–2013). There is no post-issuance transfer, no asserter, no litigation, and no defensive-aggregator acquisition, so the file does not meet the threshold for any NPE verdict and does not fit "operating-company assertion" (no suit) or "defensive/non-asserting" (no aggregator). Ownership effectively terminated with IBM, and the patent expired 2018-07-30, removing any assertion value — which is consistent with the absence of any 2026 CAFC activity.

Bookkeeping flag (cross-reference to prior section): The prior section correctly warned that the StratosAudio "'307 patent" is not US 6,233,073. My searches likewise returned no assignment, IPR, or litigation tying US 6,233,073 to any NPE. No contradiction with the earlier section was found.

Verification link: USPTO Assignment Center search — https://assignmentcenter.uspto.gov/ (search "6233073"); mirror at https://assignment.uspto.gov/patent/index.html. I was unable to render the live Assignment Center result grid in this session, so the reel/frame and dates above are cited from the USPTO-sourced legal-events data embedded in the authoritative Google Patents full text (https://patents.google.com/patent/US6233073/en) rather than from a fresh Assignment Center query. Re-query the Assignment Center directly to confirm the correspondent of record, which is the one field the authoritative text does not expose.

Caveat on the deadline: I could not confirm the absence of 2026 CAFC activity with certainty — I simply found no authoritative record of any. Treat that as an open point, not a confirmed negative.

I'll analyze the prior art cited in US 6,233,073. Let me first verify the patent record and then research each cited reference.

Let me research the remaining cited references.

Prior-Art Analysis — US 6,233,073 B1

Scope and method

Per your instruction I searched the USPTO/Google Patents record for the exact number 6,233,073 and did not substitute any similar number. The record confirms a single, real patent — US 6,233,073 B1, "Diagnostic injection of transmission errors in fiber optic networks," filed 1998‑07‑30, granted 2001‑05‑15, IBM, application 09/124,837 — consistent with the authoritative full text and the earlier summary sections. I did not auto‑correct any identifier.

This section builds on the bibliographic and claim‑overview sections already generated; it does not repeat them. It analyzes the nine references cited on the face of US 6,233,073 ("Patent Citations (9)" in the full text you supplied as controlling), plus the one non‑patent publication the patent itself cites.

A threshold legal point up front: every one of the nine references has a filing and/or grant date before the 1998‑07‑30 filing date of the '073 patent, so all qualify as prior art under pre‑AIA § 102 (at minimum as § 102(e) art as of their US filing dates). However, anticipation under § 102 requires that a single reference disclose every element of a claim, arranged as in the claim. As analyzed below, none of the nine references discloses the core feature of the '073 patent — the deliberate injection of a transmission error into the optic stream in response to recognizing a selected pattern following a selected event. These are background/contextual references, not anticipatory art. I note which claims each is conceptually closest to, but I do not assert anticipation where the disclosure does not support it.

Summary table of cited references (all § 102‑qualifying)

Detailed per‑reference analysis

1. US 5,228,105 — Glista, "Programmable electro‑optic packaging and interconnect system"

• Citation: US 5,228,105 A; filed 1987‑05‑04; granted 1993‑07‑13; inventor Andrew S. Glista. (Related to Glista's earlier US 4,837,856 and US 5,020,152.)
• Description: A programmable electro‑optic packaging/interconnect system — a monolithic optical backplane for computers/parallel processors. Each terminal has multiple photodetectors and logic circuitry that determines optical power level/ratio in two or more input lines, compares it to a preset or programmable range, and selects the "best" signal while rejecting others; a built‑in test circuit signals faults. Notably, it expressly states it "does not require conversion of an optical signal to an electrical signal to achieve fault tolerance."
• Anticipation assessment: None. No element of claims 1/12/17 (error injection, FOSA opto‑electronic conversion, pattern/event recognition) is disclosed. Its express avoidance of O/E conversion is the opposite of the '073 claim‑17 architecture (first FOSA converts optic→electrical). If anything, this is § 103 background only.

2. US 5,229,875 — Glista, "Fault‑tolerant fiber optic coupler/repeater…"

• Citation: US 5,229,875 A; filed 1989‑05‑30; granted 1993‑07‑20; inventor Andrew S. Glista. (Expressly cited in the '073 Background section.)
• Description: A fault‑tolerant fiber‑optic coupler/repeater for high‑speed data buses. It "analyzes the power of the optical signals from upstream terminals and compares the signal power to a preset value range and rejects those out of the range, activating a built‑in 'test circuit' alarm telling downstream terminals of the failure" (as the '073 patent itself summarizes it). Detection/reporting of faults, not generation of them.
• Anticipation assessment: None. It detects and reports existing faults; it does not inject an error, does not recognize a protocol "selected event" then a "selected pattern," and does not transmit an error‑bearing stream to a target. Closest conceptual overlap is "fault handling in a fiber network" — § 103 territory only.

3. US 5,321,813 — Teradata, "Reconfigurable, fault tolerant, multistage interconnect network and protocol"

• Citation: US 5,321,813 A; filed 1991‑05‑01; granted 1994‑06‑14; Teradata Corporation.
• Description: A reconfigurable, fault‑tolerant multistage interconnect network with an associated protocol (fault avoidance/reconfiguration around failed paths in a multistage fabric). I was unable to retrieve the full specification text; the description is based on the title, assignee, and citation record — flagged as lower‑confidence.
• Anticipation assessment: None on the available record. Network reconfiguration/protocol art; no disclosure of selective error injection in an in‑service optical stream, no FOSA/mask‑logic combination. The single claim element it could conceivably touch is the generic "fiber optic network" environment of claims 1/12/17 — insufficient for § 102.

4. US 5,559,963 — IBM, "Suspending, resuming, and interleaving frame‑groups"

• Citation: US 5,559,963 A; filed 1992‑02‑20; granted 1996‑09‑24; International Business Machines Corp. (same assignee as '073 — note this is IBM's own earlier work, so § 102(a)/(e)/(b) but also potentially § 103 "same‑field" context.)
• Description: Management of frame‑groups in a serial protocol (suspend/resume/interleave). Full text not retrieved in this session; description from title/assignee — lower‑confidence.
• Anticipation assessment: None on the available record. This is frame‑stream bookkeeping (relevant only tangentially to the "frame" vocabulary of claims 1/12), with no teaching of error injection or event/pattern‑triggered modification. § 103 background only.

5. US 5,602,667 — Unisys, "Extended distance fiber optic interface"

• Citation: US 5,602,667 A; filed 1992‑10‑07; granted 1997‑02‑11; Unisys Corporation.
• Description: A fiber optic interface that extends usable link distance (typically a repeat/regenerate/retime function involving clock recovery and buffering). Full text not retrieved; description from title/assignee — lower‑confidence.
• Anticipation assessment: None on the available record. Its closest conceptual links are to the buffering/retransmission elements of claims 2/6 and the interface elements of claims 18–20, but there is no disclosure of recognizing a selected event and pattern and injecting a transmission error. § 102 not met.

6. US 5,524,218 — Unisys, "Dedicated point to point fiber optic interface"

• Citation: US 5,524,218 A; filed 1993‑12‑23; granted 1996‑06‑04; Unisys Corporation.
• Description: A dedicated point‑to‑point fiber‑optic link interface (optics, serialization/deserialization, and likely 8B/10B‑type coding). Full text not retrieved; description from title/assignee — lower‑confidence.
• Anticipation assessment: None on the available record. Could overlap the 8B/10B and serializer/deserializer subject matter of claims 19 and 20 (FOSA encoder/decoder, serializer/deserializer), but the claims as a whole require the injection architecture, which this reference does not disclose. Potential § 103 combination material for claims 19/20; not § 102 anticipation.

7. US 5,396,357 — Goossen et al., "Fault tolerant optical cross‑channel data link"

• Citation: US 5,396,357 A; filed 1994‑01‑25; granted 1995‑03‑07; Honeywell Inc.; inventors Goossen, Nelson, Woods. (Expressly cited in the '073 Background as "Goosen et al.")
• Description: A wavelength‑division‑multiplexed, bidirectional cross‑channel data link for triple‑modular‑redundant (fly‑by‑wire) computers, using QAM‑over‑FM double modulation. It includes diagnostics: noise‑density (BIT) measurement, transmitter‑power sweeps to find a soft‑error threshold, and loopback/echo of received data for diagnostic purposes. It "guarantees" identical data across the link despite any single‑point failure.
• Anticipation assessment: None. It performs measurement and loopback diagnostics, but it does not deliberately inject a transmission error into the optic stream upon recognizing a protocol event/pattern. It is the opposite philosophy (avoid/recover from errors vs. systematically induce them for fault‑tolerance testing). § 103 background at most.

8. US 5,706,278 — Raytheon, "Deterministic network protocol"

• Citation: US 5,706,278 A; filed 1995‑07‑20; granted 1998‑01‑06; Raytheon Company; inventors Robillard and Morrison.
• Description: A deterministic, fault‑tolerant control‑by‑light network protocol: nodes interconnected by fiber (often in a ring), a timing beacon for synchronization, and preassigned/contention time slots for critical vs. non‑critical messages; uses state machines in the transceivers and voter switching.
• Anticipation assessment: None. A media‑access/synchronization protocol; no error injection, no event‑then‑pattern recognition, no FOSA‑based injector. § 103 background only.

9. US 5,774,242 — Northern Telecom, "Eye mask for measurement of distortion in optical transmission systems"

• Citation: US 5,774,242 A; filed 1996‑05‑28; granted 1998‑06‑30; Northern Telecom Limited.
• Description: An optical test/measurement technique — an "eye mask" used to measure distortion in an optical transmission system (signal‑quality assessment at a receiver/analyzer).
• Anticipation assessment: None. It outwardly measures signal quality; it does not inject an error. It is the closest cited reference to the general theme of "optical testing," but it does not teach any claim element beyond the generic context of testing an optical transmission system. No § 102.

Non‑patent literature cited by the '073 patent

• Widmer, A. X., and Franaszek, P. A., "A DC‑Balanced, Partition‑Block, 8B/10B Transmission Code," IBM Journal of Research and Development, Vol. 27, No. 5, Sept. 1983, pp. 440–450. Cited in the '073 Background/specification. This is § 102(b) prior art (printed publication, well over one year before filing). It is the source of the 8B/10B DC‑balanced transmission code relied on in claims 19 and 20 (the "(8,10) code" encoder/decoder). It discloses the coding, not the error‑injection method, so it does not anticipate claims 1/12/17; it is the primary § 103 reference for the encoding limitations of claims 19/20, which were the well‑known standard of the day.

Bottom line

• Most relevant "prior art" in the § 102 sense: the two Glista references — US 5,229,875 (fault‑tolerant fiber optic coupler/repeater) and US 5,228,105 (programmable electro‑optic packaging/interconnect) — are the references closest to the fiber‑network fault‑handling setting of the '073 patent, and US 5,229,875 is the one the patent itself singles out in its Background. The Goossen US 5,396,357 optical cross‑channel data link is the closest in optical‑link diagnostics.
• No single cited reference anticipates any claim of US 6,233,073 under 35 U.S.C. § 102. Every one of them either (a) detects and reports or avoids/reconfigures around faults (Glista, Goossen, Teradata, Raytheon), (b) measures optical signal quality (Northern Telecom '242), or (c) is directed to unrelated interface/frame/protocol management (Unisys '667/'218, IBM '963). None discloses the deliberate injection of a transmission error into the optic stream in response to recognizing a selected pattern after a selected event — the distinguishing feature of independent claims 1, 12, and 17 and their dependents.
• Best § 103 combination targets, if one were to be built: the 8B/10B art (Widmer/Franaszek NPL plus the Unisys interfaces '667/'218) for the codec/SERDES claims (19/20), and the Glista/Goossen optical fault‑handling art for the general network environment — but combining them still would not supply the claimed injection step, which is the point of novelty. This is consistent with the patent having issued over these nine references.

Caveats / lower‑confidence items (explicitly flagged):

• I retrieved substantive text for US 5,228,105, 5,229,875, 5,396,357, and 5,706,278. For US 5,321,813 (Teradata), 5,555,963 (IBM), 5,602,667 (Unisys), 5,524,218 (Unisys), and 5,774,242 (Northern Telecom) I did not retrieve the full specifications in this session; their descriptions rest on titles, assignees/dates from the controlling full text, and my general knowledge, and are marked lower‑confidence. A definitive § 102 element‑by‑element comparison for those five would require pulling each specification.
• Not prior art — excluded from § 102 analysis: the "Cited By" documents, e.g., US 6,560,720 (IBM, "Error injection apparatus and method") and US 7,669,095 (IBM, "Methods and apparatus for error injection"). These post‑date the '073 patent and are therefore not § 102/§ 103 prior art against it; they are instead useful as evidence of the later development the '073 patent is said to have enabled.
• All bibliographic and citation data above are taken from the authoritative full text you supplied (source: patents.google.com/patent/US6233073/en), treated as controlling over conflicting web snippets. No patent number or identifier was auto‑corrected.

I'll research the key prior-art references before building the §103 analysis.

Obviousness Analysis — US 6,233,073 B1 under 35 U.S.C. § 103

Scope note / caveats up front

• This analysis draws on the Prior Art section of the US 6,233,073 ("'073") page — i.e., the nine (9) references in the Patent Citations list, the background references discussed in the specification, and the Widmer/Franaszek article cited in the description. I supplemented with content retrieved for the key references.
• The "Cited By" list is not prior art to '073: every entry there (e.g., US 6,560,720, US 6,631,481, US 7,669,095, FR 2,870,958) has a priority date after '073's 1998‑07‑30 filing date. Those references can illustrate how the field later developed, but they cannot be used for § 102/§ 103 against '073. I flag this because a naive read of the Google Patents page conflates "Patent Citations" and "Cited By."
• Two references I use to supply the error‑injection teaching (US 4,759,019, Bentley; EP 376,034 / EP 388,526) are not on the face of '073 but predate the 1998 filing. I identify them explicitly as off‑the‑face prior art rather than pretending they were cited.
• Minor date discrepancy: the search task header says "Current Date: April 26, 2026," while the document metadata says the page was fetched 2026‑10‑01. Nothing in this analysis turns on that difference.
• I could not retrieve the full text of every cited reference (e.g., Teradata US 5,321,813; Unisys US 5,524,218 / US 5,602,667; Raytheon US 5,706,278; IBM US 5,559,963). Characterizations of those are based on their titles/abstracts in the citation list and the '073 specification's own description of them, and should be verified against full text before being relied on in a legal opinion.

1. Legal standard and the person of ordinary skill

§ 103 standard (post‑KSR). A claim is obvious if the differences between the claimed subject matter and the prior art are such that the subject matter as a whole would have been obvious to a person having ordinary skill in the art ("POSITA") at the time of the invention. The analysis rests on the Graham factors: (1) scope and content of the prior art; (2) differences between the prior art and the claims; (3) the level of ordinary skill; and (4) objective evidence (secondary considerations). Under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), a combination is obvious where the elements were known, the combination achieves no more than predictable results, and there existed a design need or market pressure to solve the problem — "if a technique has been used to improve one device, and a person of ordinary skill in the art would recognize that it would improve similar devices in the same way, using the technique is obvious."

POSITA here. A person with a bachelor's degree in electrical/computer engineering and ~2–5 years of experience in high‑speed serial data links and storage/network interfaces (Fibre Channel, ESCON, SSA, ATM, FDDI), familiar with 8B/10B coding (Widmer/Franaszek), FOSAs, SERDES, CRC/ordered‑set protocol structures, and the ordinary toolbox of fault‑injection testing. This is essentially the "similar serial transport layer" audience the '073 specification itself describes.

2. Scope and content of the prior art (the page's Prior Art section)

Off‑the‑face error‑injection art (all pre‑1998), used below to supply the injection teaching:

• US 4,759,019 (Bentley), "Programmable Fault Injection Tool" (PFIT), issued 1988‑07‑19 — event‑compare hardware, user‑programmable to compare a stored event (a digital word or a sequence of words) with an event in the system under test; includes a mask memory to ignore inconsequential bits; error‑injection hardware controlled by the event‑compare hardware, injecting a fault for a user‑specified time duration and only after a user‑specified time delay. Its stated purpose is to test the diagnostic capabilities of a system.
• EP 376,034 A3 (IBM), "Event‑controlled error injection system" (pub. 1990) — "programmable masks at ascending levels … support the programmed injection of errors in response to machine events and machine states and in synchronism with machine operation," with variable temporal and duration characteristics.
• EP 388,526 B1 (IBM), "Programmable channel error injection" — error injection under microcode control, immediate or delayed, "layered" so an event triggers a later injection, with recovery timing measurement.

3. Element mapping — the independent claims

Claim 1 (method; error injected into the optic stream)

Claim 12 (method; error injected into the electrical stream, then converted)

Every element of claim 1, plus the explicit inject‑then‑convert order. This is the natural architecture of an inline optical regenerator/repeater that converts optical→electrical, processes, then re‑encodes to optical — precisely the Glista '875 (electrical‑select embodiment, FIG. 11) / Unisys '667 retransmit model with the injection stage inserted. Claim 12 is, if anything, more plainly obvious than claim 1.

Claim 17 (apparatus)

4. Combinations that render the claims obvious, and the motivation to combine

Combination A (primary) — Fiber‑optic interface art + event‑controlled fault injection + 8B/10B

Glista '875 (or Unisys '218/'667) + Bentley '019 (or IBM EP 376,034) + Widmer/Franaszek.

Motivation (KSR rationales):

1. Same problem, known solution. Glista, Goosen '357, and the '073 specification itself all frame the goal as building fault‑tolerant optical links that recover from faults. Once such links exist, verifying that the fault‑detection/recovery actually works is the recognized, self‑evident next step, and fault‑injection testing was the established way to do it (Bentley: to test "the diagnostic capabilities" of a system; EP 388,526: to "measure the time it takes a channel to recover from an injected error"). The '073 Summary states the same object verbatim — "verify that a fault tolerant fiber optic network design has successfully implemented error detection schemes." A POSITA would have been motivated to port the known injection test into the optical link.
2. Predictable result / use of a known technique in an analogous structure. Convert optical→electrical, run the stream through a comparator + mask + injection stage, then re‑encode/retransmit — every stage existed and behaved predictably. Per KSR, applying a known fault‑injection technique to the analogous fiber‑optic data link yields nothing more than expected results.
3. Bentley '019 supplies nearly the entire trigger architecture the '073 claims as novel: user‑programmable event compare (selected event/sequence), mask memory (masked pattern comparison, claim 9), injection controlled by the event compare, and programmable delay and duration (the '073 "trigger delay counter" and stop counter). IBM's own EP 376,034 goes further, using event‑controlled programmable masks to trigger injection — the two‑stage event→mask→inject flow of claims 1/12/17.
4. Reasonable expectation of success. The optical‑domain pieces were standard and interoperable: FOSAs (Glista/Unisys), the 8B/10B code (Widmer/Franaszek, already used by "most fiber optic protocols" per '073), and clock recovery (O'Sullivan '242). No undue experimentation to combine.

Combination B — Optical test/measurement art as the explicit combining suggestion

O'Sullivan '242 + Glista '875/'105 + Widmer/Franaszek.

O'Sullivan '242 is itself optical transmission test equipment that (i) launches a signal into a path, (ii) receives the degraded signal downstream, and (iii) converts the optical signal to electrical to analyze it against a reference mask. It therefore supplies (a) the testing motivation in the optical domain and (b) the optic→electrical conversion/inspection step. Combining it with Glista's inline coupler/repeater — which already sits mid‑fiber, compares signals to presets, and retransmits — gives a POSITA a clear path to placing an error‑injecting stage in the optical link at a selected point. The difference between O'Sullivan's measure/characterize and '073's inject/observe the target's response is the kind of substitution of one known testing modality for another that KSR treats as obvious, particularly where the reference mask / preset‑value comparison logic of Glista and O'Sullivan provides the "recognize a pattern" step.

Combination C — Protocol/fault‑tolerance framing

Teradata '813 + Raytheon '278 + Goosen '357 + Bentley '019/EP 376,034.
These establish that (i) fault‑tolerant interconnect protocols were known and (ii) the field was already concerned with guaranteeing correct behavior "irrespective of any single point failure" (Goosen). A POSITA validating such a design would look to the existing fault‑injection toolset (Bentley/IBM) and apply it at the link level — supplying the "selected event related to the optic stream defined protocol" (protocol semantics) and the injection mechanism.

5. Dependent claims

None of the dependent claims adds an element that is not either (a) disclosed by Bentley '019/EP 376,034, (b) conventional in 8B/10B serial‑link design, or (c) disclosed by the cited fiber‑optic interface references.

6. Counter‑arguments a challenger/owner would raise

1. None of the nine on‑the‑face cited references teaches deliberate error injection. This is the strongest non‑obviousness argument: the cited art is about detecting/flagging/tolerating faults (Glista alarm; Goosen redundancy) or measuring link quality (O'Sullivan), not about intentionally corrupting a live stream to exercise a target's recovery. An owner would argue the examiner never found an injection teaching in the references of record, and that the "select event → select pattern → inject" chain is the point of novelty.
   • Rebuttal: the injection teaching is supplied by off‑the‑face prior art that plainly predates 1998 (Bentley '019; IBM EP 376,034; IBM EP 388,526), and the motivation is explicit in the art itself ("determining their effectiveness," "measure … recovery time"). Under KSR this is enough; a reference need not have been before the examiner.
2. Different technical field (digital fault injection vs. inline optical link). An owner might argue Bentley injects faults into digital logic by shorting voltage nodes and is not analogous to an inline optical stream.
   • Rebuttal: the problem (testing fault tolerance via controlled error injection triggered by a recognized event with a mask) is the same, and O'Sullivan '242 confirms the optical field already used optic→electrical conversion and mask comparison for testing. KSR expressly countenances combining elements from analogous fields where the technique's improvement would be recognized.
3. Secondary considerations. If the record contained evidence of unexpected results, commercial success of a licensed product, copying, or industry praise, that could rebut. I found no such evidence in the sources searched, and the patent is expired (anticipated expiration 2018‑07‑30), so an obviousness challenge would in any event be academic for enforcement purposes. Neither the patent summary nor the litigation summary located any assertion of the patent, which is consistent with there being no developed secondary‑considerations record.

7. Conclusion

On the record available, the independent claims (1, 12, 17) of US 6,233,073 are vulnerable to a § 103 obviousness challenge. The cited fiber‑optic art supplies the link/FOSA/retransmit environment (Glista '875, Unisys '218/'667), the protocol/frame and preset‑value semantics (Raytheon '278, Teradata '813, IBM '559,963, Goosen '357), the optical‑test/convert‑to‑electrical motivation (O'Sullivan '242), and the 8B/10B coding (Widmer/Franaszek). The event‑triggered, mask‑based, delayed error‑injection architecture comes from the pre‑1998 fault‑injection art (Bentley '019; IBM EP 376,034 / EP 388,526). The motivation to combine is strong and, notably, is stated in the '073 specification's own object: to verify that a fault‑tolerant fiber‑optic design actually detects and recovers from errors — exactly the purpose for which the fault‑injection references existed. The result of placing a known injection stage inside a known inline optical repeater would have been predictable, with a reasonable expectation of success and no demonstrated secondary considerations.

Confidence: moderate‑to‑high on the legal framework and the mapping of claim elements to the cited references; the analysis is only as strong as the full‑text content of the individual references. I recommend verifying the full text of US 5,524,218, US 5,602,667, US 5,321,813, US 5,706,278, and US 5,559,963, and confirming the publication data for Bentley US 4,759,019, EP 376,034, and EP 388,526, before relying on this for any formal opinion.