US 10795107

I'll verify this against live sources before summarizing.

Now let me check for any 2026 Federal Circuit activity and verify the post-IPR claim status.

Let me verify the adverse judgment details and whether the 2026 Federal Circuit appeal involves this patent.

US 10,795,107 B2 — Verification Summary

Important caveat on dates: You state today is April 26, 2026. My searches returned a Federal Circuit order dated June 29, 2026, which post-dates your stated current date. I am reporting what the dockets show rather than silently reconciling this.

Bibliographic data (from the patent text + Google Patents/PTAB records)

Data discrepancy flagged: Unified Patents' portal record lists priority date 2017-10-02, application date 2018-09-26, grant date 2020-10-05, and expiration 2038-09-26 — i.e., one day earlier across the board than the Google Patents/USPTO values above. This looks like a timezone/normalization artifact, but per your instruction I am not auto-correcting it; I'm reporting both.

Abstract (as issued)

A modular fiber optic cassette system, method and cassette is disclosed. The system comprises a case defining a tray receiving space and an opening to a tray receiving space, at least one tray slideably received within the case for movement between a stored position wherein the tray is inside the case and an accessible position wherein a front end of the tray is in front of the opening and outside of the case, a plurality of fiber optic cassettes arrangeable side by side along the tray front end, a first of the cassettes comprising a first cassette width of one of one, two, three, four or six standard width units and a second of the cassettes comprising a second cassette width different from the first cassette width and one of one, two, three, four or six standard width units, and a fastener assembly for removeably securing each of the cassettes to the tray.

Independent claims — plain language

Claim 1 (system). A housing (top, bottom, two side walls) that encloses a slidable tray which pulls out through a front opening from a fully-stored position to an accessible position with the tray's front end projecting outside the housing. A set of fiber optic cassettes sits side by side along the front of the tray; each cassette has a row of plug receptacles (optic plugs) on its front face. The set includes at least two cassettes of different widths, and each width must be one of 1, 2, 3, 4, or 6 "standard width units." A fastener releasably holds each cassette to the tray. Added during prosecution: when mounted, each cassette must physically touch at least one neighboring cassette (the "snug" side-by-side limitation).

Claim 8 (method). A method of organizing a tray in a fiber optic interconnect system: (1) define a "standard width unit" such that the tray's front-edge width is evenly divisible by that unit; (2) pick cassettes for the tray from a set of cassettes of several different widths, each width being an even multiple of that unit; (3) releasably secure the picked cassettes along the tray's front edge — again with the requirement that each secured cassette touches at least one other cassette.

Claim 14 (cassette per se). A modular cassette for releasable mounting on a ferrous-metal tray that has spaced slots. The cassette has: an elongate rectangular fiber-receiving housing; a row of optic plug receptacles along the front; a multi-fiber connector receptacle at the rear; optical fibers internally connecting each front receptacle to the rear multi-fiber connector; a magnet on the underside of the housing; and a pair of stabilizing bosses on the underside, one on each side of the magnet, each boss seating in a respective slot in the tray when the cassette is placed down.

Dependent claims (brief): 2 — tray sized for ≥12 width units; 3 — rear multi-fiber connector receptacle with interconnecting fibers; 4 — four front receptacles per standard width unit; 5 — rack-mounting brackets on the case side walls; 6 — ferrous tray + magnet fastener; 7 — a one-width-unit-wide cable management bracket with a cable loop; 9 — the set includes widths 1, 2, 3, 4 and 6; 10 — all selected cassettes same width; 11 — two different widths; 12 — front edge divides into max 12 width units; 13 — a third, different width; 15 — cassette width is 1/2/3/4/6 width units.

⚠️ Critical legal-status development (post-issue)

This patent is no longer fully intact. Verified live:

• IPR2024-01218 — Petitioner CommScope Technologies LLC, CommScope, Inc., and CommScope, Inc. of North Carolina filed July 26, 2024, challenging claims 1–5 and 8–13. Board instituted Feb. 27, 2025; proceeding status is now "Terminated – Adverse Judgment" (termination date Sept. 8, 2025) — ipverse.greyb.com IPR2024-01218.
• Statutory disclaimer: Belden disclaimed claims 1–5 and 8–13 under 35 U.S.C. § 253(a) / 37 C.F.R. § 1.321(a) (Google Patents legal events show the disclaimer docketed Aug. 26, 2025, effective July 7, 2025; Belden's updated mandatory notices reference disclaimers filed July 17, 2025). Belden then moved to terminate by adverse judgment (Aug. 7, 2025) because "no claims remain in this trial."
• IPR certificate (K1) issued Dec. 12, 2025 (Trial No. IPR2024-01218).
• Practical consequence: only claims 6, 7, 14 and 15 were not disclaimed. Because claims 6 and 7 depend from disclaimed claim 1, they are effectively inoperative; the only claims with real remaining scope are claim 14 (cassette with magnet + stabilizing bosses) and claim 15 (the width-unit dependent claim). That last sentence is my analytical inference, not an official USPTO statement.

Related litigation / family

• D. Del. Belden Canada ULC v. CommScope, Inc. et al., No. 1:23-cv-00810-RGA — the '107 patent was one of four asserted patents; the case was stayed pending four related IPRs (IPR2024-01218, -01227, -01228, -01382). A second suit, No. 1:25-cv-00870-RGA, asserts newer family members.
• Family continuations: US 11,435,542; 11,656,422; 11,740,423; 12,313,896; 12,332,491; and pending US 2025/0306324 A1.
• CAFC 2026 dockets: The only 2026 Federal Circuit matter I found involving these parties is 26-1255 / 26-1285, Belden Canada ULC v. CommScope, Inc., dismissed by joint stipulation under Fed. R. App. P. 42(b) on June 29, 2026 (CAFC order PDF). Uncertainty: that appeal originates from D. Del. 1:22-cv-00782-RGA, not the '107 patent's 1:23-cv-00810-RGA case, and the order does not name the asserted patents. I therefore cannot confirm it concerns US 10,795,107. I found no Federal Circuit appeal taken from IPR2024-01218.

Claim-construction note (context for the disclaimers)

In the Delaware action, "standard width unit" was disputed. CommScope argued it means "an actual width standardized in the industry (e.g., SC Duplex footprint)," otherwise indefinite; Belden proposed "a unit of width that accommodates a single receptacle module, such as a LC Quad (SC Duplex) technology footprint comprising four connectors." (Joint Claim Construction Chart, C.A. No. 23-810-RGA.) "Cassettes arrangeable side by side along said tray front end" was stipulated to mean "on or beside the front end of the tray."

Uncertainties I am explicitly flagging

1. The IPR adverse judgment and disclaimers mean the '107 patent's claim set described in the printed patent text is no longer the operative claim set; the Google Patents page you supplied still shows status "Active" and lists all 15 claims.
2. I could not confirm whether the June 29, 2026 CAFC dismissal (26-1255) involves this patent.
3. The one-day date discrepancies between Unified Patents and Google Patents/USPTO are unexplained; I did not resolve them.

I'll search for litigation involving US patent 10795107 across patent litigation databases.

Let me dig deeper into the Delaware case and search for additional litigation.

I have strong initial results. Let me verify additional details about the Delaware case outcome, potential appeals, and any other litigation.

Based on my searches of patent-litigation databases (Unified Patents, PACER docket summaries, PTAB/IPR records, and CAFC dockets), here is what I found for US 10,795,107 B2 ("Modular fiber optic cassette, system and method," Belden Canada ULC). I've limited results strictly to the '107 patent and flagged where aggregator sources appear to conflate dockets.

1. District Court Litigation

Belden Canada ULC v. CommScope, Inc.; CommScope, Inc. of North Carolina; CommScope Technologies LLC

Notable development: On July 17, 2025, Belden filed statutory disclaimers under 37 C.F.R. § 1.321(a) disclaiming claims 1–5 and 8–13 of the '107 patent (this is reflected in the USPTO Patent Trial and Appeal Board record and in the Google Patents legal-events data, and it corresponds to the "Disclaimer filed" entry dated Aug. 26, 2025 in the patent's file). The complaint had asserted at least independent Claim 1, so the disclaimer removed the primary asserted claims of the '107 patent.

2. PTAB (Inter Partes Review) Proceedings Involving the '107 Patent

CommScope Technologies LLC v. Belden Canada ULC — IPR2024-01218

This proceeding is what triggered the district-court stay. Per the Petitioner's Updated Mandatory Notice, four parallel IPRs were filed on the asserted patents:

• IPR2024-01218 — US 10,795,107
• IPR2024-01227
• IPR2024-01228
• IPR2024-01382 — US 11,740,423

(For completeness, the related-patent IPRs suggest the '542 and '422 patents were challenged in the -01227/-01228 proceedings, though I could not fully confirm the exact patent-to-IPR mapping within the source limits.)

3. Additional Related (Non-'107) Matter — Noted to Avoid Confusion

A separate Belden v. CommScope action exists that does not involve the '107 patent:

• Belden Canada ULC v. CommScope, Inc. et al., No. 1:22-cv-00782-RGA (D. Del.) — asserted US 6,409,547 B1 ("high-performance structured cabling connector"), not the '107 patent. This case proceeded to trial-stage motions in October 2025 and produced a Federal Circuit appeal (Nos. 2026-1255, 2026-1285), which the CAFC dismissed under Fed. R. App. P. 42(b) (order dated June 29, 2026).

Several third-party litigation blogs (e.g., PatSnap) appear to conflate dockets 1:23-cv-00810 and 1:22-cv-00782 and even attribute the '547 patent to the Propel™ case. Treat those conflated summaries with caution — the '107 patent was only asserted in 1:23-cv-00810.

4. Other Sources / Caveats

• The Google Patents page for US 10,795,107 lists exactly these litigation flags: the Delaware District Court case 1:23-cv-00810, the PTAB case IPR2024-01218, and a Darts-IP "first worldwide family litigation" link. This corroborates the two matters above.
• Canadian family members (e.g., CA 3,221,590; CA 3,019,081; CA 3,221,599; CA 3,221,718) are listed as active, but I found no separate litigation on the Canadian counterparts.
• Uncertainty flags: (i) the exact complaint filing date appears as both July 27 and July 28, 2023 depending on the docket entry vs. PACER header; (ii) the precise claims/patents asserted in each of the parallel IPRs beyond -01218 is not fully confirmed by the sources retrieved.

Summary

Known litigation/contested proceedings specifically involving US 10,795,107:

1. Belden Canada ULC v. CommScope, Inc. et al., D. Del. 1:23-cv-00810-RGA — filed ~July 27, 2023; stayed/administratively closed pending IPR. Asserts '107 plus three family members.
2. CommScope Technologies LLC v. Belden Canada ULC, PTAB IPR2024-01218 — filed July 26, 2024; instituted Feb. 27, 2025; terminated by adverse judgment (Sept. 8, 2025); IPR certificate issued Dec. 12, 2025, canceling claims 1–5, 8–13 of the '107 patent.

No other litigation asserting the '107 patent was found.

I'll reconcile the structured "no proceedings" block against the litigation/legal-event data embedded in the patent text, then search for the public record on IPR2024-01218.

Key discrepancy already: the structured block says "no proceedings," but the patent text's own litigation/legal-events data shows IPR2024-01218. Let me pin down the details and check the sibling petitions.

Now let me nail down the challenged claims, grounds, panel, and any appeal.

⚠️ Data-integrity flag before anything else

The structured "PTAB proceedings on file" block in this prompt states the ODP API returns no AIA trial proceedings for US 10,795,107. That is wrong. The authoritative patent text supplied in this prompt — the Google Patents legal-events table and the "Family has litigation" block — itself records:

• 2024-09-24 | IPR | TRIAL NO: IPR2024-01218; Opponent: COMMSCOPE TECHNOLOGIES LLC, COMMSCOPE, INC. OF NORTH CAROLINA, COMMSCOPE, INC. (effective date 2024-07-26)
• 2025-08-26 | DC | DISCLAIM THE FOLLOWING COMPLETE CLAIMS 1-5, 8-13 OF SAID (effective 2025-07-07)
• 2025-12-16 | IPRC | INTER PARTES REVIEW CERTIFICATE; TRIAL NO. IPR2024-01218 … CERTIFICATE ISSUED DEC. 12, 2025
• "Family has litigation — PTAB case IPR2024-01218 filed (Adverse Judgment)" → https://portal.unifiedpatents.com/ptab/case/IPR2024-01218

Public sources confirm the same. Do not rely on the ODP "no proceedings" default for this patent. Treat the ODP ingest as stale/incomplete here — this is exactly the failure mode the prompt told me to flag.

Proceedings overview

One AIA trial has been on file against US 10,795,107 — IPR2024-01218 (CommScope v. Belden Canada) — and it ended not on the merits but by adverse judgment following Patent Owner's own statutory disclaimer of claims 1-5 and 8-13, so the defensively relevant fact is that the bulk of the patent is now treated as if it never existed, leaving only narrow, untested, largely magnet/bracket-dependent claims. Breakdown: 0 active, 0 settled, 0 institution-denied, 0 claims invalidated by a Final Written Decision, and 1 terminated by adverse judgment entered against the patent owner after it disclaimed all challenged claims. Bottom line for a defendant: if a demand letter or complaint cites claims 1-5 or 8-13 of the '107 patent, it has no case — those claims are canceled by disclaimer — and any remaining exposure is confined to claims 6, 7, 14 and 15.

IPR2024-01218 — CommScope Technologies LLC et al. v. Belden Canada ULC et al.

• Type: Inter Partes Review (35 U.S.C. §§ 311–319)
• Filed: 2024-07-26 (petition filing date; PTAB docket reflects filing date accorded 2024-08-30). Consolidated with three companion petitions on sibling family patents: IPR2024-01227 ('542), IPR2024-01228 ('422), IPR2024-01382 ('423).
• Status: Terminated-Adverse Judgment (Board paper "Adverse judgment: post-institution," 2025-09-08; termination date 2025-09-08). Plain English: the patent owner killed the trial by disclaiming the challenged claims, and the Board entered judgment against it rather than issuing a merits decision. An Inter Partes Review Certificate (kind code K1) issued 2025-12-12. Source: https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2024-01218
• Judge panel: Not confirmed in the sources I retrieved. A "Panel Change Order" was entered 2025-04-18, so the panel was reconstituted mid-trial. I will not guess names; pull the panel directly from PTAB E2E (https://ptacts.uspto.gov/ptacts/). For reference, the sibling case IPR2024-01382 shows APJs Christopher M. Kaiser, Josiah C. Cocks, Kimberly McGraw and Sharon Fenick — but I am not attributing that panel to -01218.
• Petition grounds: The petition's exhibit list includes US 10,101,551 (Fukui), US 9,690,064 (Sauter), US 10,215,944 and EIA rack standards, and it cross-references the co-pending Delaware claim-construction record. I could not confirm, within my search budget, the ground-by-ground § 102/§ 103 mapping or the exact challenged-claim list from the institution decision PDF. What is confirmed is that the set of claims Belden subsequently disclaimed — claims 1-5 and 8-13 — matches the set that would have been under challenge, and that Belden's dismissal-by-disclaimer of all claims then in the trial is what triggered § 42.73(b). Do not treat the "Fukui/Sauter" reference list as established grounds without pulling Paper 2 and the institution decision.
• Institution decision: Instituted 2025-02-27 ("Decision Granting Institution of Inter Partes Review," Board paper). The sources I retrieved confirm institution occurred; I could not retrieve the panel's stated reasoning text. No institution denial, and no partial-institution finding, was reported.
• Final Written Decision: None issued. This is important and I will not paper over it — because the trial terminated before the § 318(a) deadline, there is no merits verdict on any claim of the '107 patent. No claim was held unpatentable; no claim was held patentable; no independent claim was canceled by the Board. The claims were removed by the patent owner's own disclaimer, not by a PTAB ruling.
• Settlement / termination: Not a settlement. Timeline from the PTAB docket: Non-Contingent Motion to Amend (PO) 2025-05-21 → statutory disclaimer of claims 1-5, 8-13 recorded 2025-07-07 → Board order authorizing withdrawal of the Motion to Amend 2025-08-04 → PO Motion to Withdraw MTA and PO Motion to Terminate (with disclaimer exhibits) 2025-08-07 → adverse judgment: post-institution 2025-09-08. The disclaimer was filed under 35 U.S.C. § 253(a) / 37 C.F.R. § 1.321(a); under 37 C.F.R. § 42.73(b), disclaiming a claim such that the party has no remaining claim in the trial is construed as a request for adverse judgment, which is what the Board entered.
• Appeal: No Federal Circuit appeal identified in any source I reviewed. Belden voluntarily disclaimed, so there was no FWD to appeal; an adverse judgment entered on a disclaimer is appealable (the Federal Circuit so held for pre-institution adverse judgments in Arthrex, Inc. v. Smith & Nephew, Inc.), but nothing indicates one was taken here. Verify on the CAFC docket and CourtListener before relying on this.
• Defensive value: Claims 1-5 and 8-13 are dead. A statutory disclaimer "cancels the claims from the patent and the patent is viewed as though the disclaimed claims had never existed" (Guinn v. Kopf, 96 F.3d 1419, 1422 (Fed. Cir. 1996)). Any infringement theory built on those claims is non-viable, and the patent owner is barred by 37 C.F.R. § 42.73(d)(3)(i) from obtaining in any patent a claim not patentably distinct from them. The residual risk is confined to the four never-challenged claims.

Strategic summary

Claim-level status of the '107 patent after IPR2024-01218

• CANCELED (disclaimed 2025-07-07, adverse judgment 2025-09-08): claims 1, 2, 3, 4, 5, 8, 9, 10, 11, 12, 13. These are treated as if they never existed.
• SURVIVING / UNTESTED — never challenged in any AIA trial: claims 6, 7, 14, 15.
  • Claim 15 depends from claim 14 and adds only the "standard width unit" width limitation.
  • Claim 14 is the self-contained, carve-out claim: a cassette for a ferrous-metal tray with spaced slots, requiring a magnet on the underside and a pair of stabilizing bosses on opposite sides of the magnet engaging those slots. This is the claim most likely to be asserted, and it is narrow.
  • Claim 6 ("tray manufactured from a ferrous metal and said fastener comprises a magnet") and claim 7 (cable management bracket) depend from claim 1 — which is now treated as never having existed. Dependent claims are not disclaimed merely because their parent is, but their scope incorporates a non-existent claim's limitations, so their enforceability and scope are genuinely contestable. Do not assume the whole patent is dead, but do not treat claims 6 and 7 as clean residual risk either. Flag this as an open issue for claim-construction analysis rather than a settled point.
• Not asserted and not tested: everything else. Note that the original Delaware action asserted 32 claims across four patents (per the parties' Joint Claim Construction Chart, C.A. No. 23-810-RGA, D.I. filed 2024-06-21). Belden has since disclaimed essentially that entire asserted perimeter — claims 1-5, 8-13 of the '107 patent; claims 1-6, 10-14, 16, 18, 20-28, 32-36, 38, 40-42, 44-56, 60-62 and 64 of the '422 patent; and claims 1-30 (all) of the '423 patent, all on 2025-07-17 — a portfolio-wide defensive retreat forced by CommScope's four petitions.

Estoppel landscape

• § 315(e)(2) estoppel against CommScope is, at minimum, questionable — and moot. Statutory IPR estoppel in litigation attaches after the petitioner "receives a final written decision under section 318(a)." No § 318(a) FWD ever issued here; the trial died on an adverse judgment after disclaimer. So CommScope would likely argue no statutory estoppel attached, and it would have a decent textual argument. Practically it does not matter: the challenged claims no longer exist, so there is nothing left to invalidate.
• The estoppel that does bite runs against the patent owner, not the defendant. The Board's adverse judgment triggers 37 C.F.R. § 42.73(d)(3)(i): Belden may not take action inconsistent with the judgment, including obtaining in any patent a claim not patentably distinct from a canceled claim. A continuation or reissue claim that resurrects the disclaimed standard-width-unit subject matter is vulnerable — the Board has been hostile to exactly that maneuver (see the October 2025 Amgen v. BMS "whac-a-mole" adverse-judgment decisions).
• For a new defendant (not CommScope or its privies), all prior-art grounds remain open — no estoppel runs to them, and no FWD means no claim was adjudicated unpatentable, so there is no judgment to leverage. But any residual claims (6, 7, 14, 15) are untested, meaning neither side has a merits ruling to point to.

Pattern signals

• Same petitioner, coordinated campaign, not follow-on petitioning. CommScope Technologies LLC (with CommScope, Inc. of North Carolina and CommScope, Inc. as RPIs; Belden Canada ULC and Belden Inc. as PO RPIs) filed one IPR per asserted patent — IPR2024-01218 ('107), IPR2024-01227 ('542), IPR2024-01228 ('422), IPR2024-01382 ('423) — rather than stacking multiple petitions on one patent. That avoids General Plastic follow-on risk. All four petitions were instituted the same day (2025-02-27) and all four terminated the same day (2025-09-08) by adverse judgment.
• This is a competitor campaign, not a defensive aggregator. Unified Patents appears only as a data source in the patent's litigation block (creative-commons PTAB/litigation data); it is not a petitioner. No defensive-aggregator IPR exists on this patent.
• The patent owner did not fight to a merits decision. Belden filed a non-contingent Motion to Amend, then withdrew it and disclaimed instead. Belden has not (per available sources) pursued a PTAB appeal.
• Parallel litigation posture: Belden Canada ULC v. CommScope, Inc. et al., C.A. No. 1:23-cv-00810-RGA (D. Del.) was stayed on 2024-09-04 (D.I. 63) pending the four IPRs and then "ADMINISTRATIVELY CLOSED," with the court directing the parties to notify it when the PTAB action resolves so the case may be reopened. That stay condition has now been satisfied — the PTAB actions terminated in September 2025 and the IRP certificate issued in December 2025. Whether Belden seeks reopening (with claims 1-13 disclaimed, its case is materially gutted) is the live question. Sources: https://portal.unifiedpatents.com/litigation/Delaware%20District%20Court/case/1%3A23-cv-00810

Recommended next steps

1. If you are a defendant and a demand letter or complaint cites the '107 patent, demand the claim numbers. If it cites claims 1-5 or 8-13, the answer is dispositive: those claims were disclaimed effective 2025-07-07 and are treated as though they never existed (Guinn v. Kopf), and an adverse judgment was entered against Belden on them on 2025-09-08. Cite the Board's adverse-judgment paper and the IRP Certificate of 2025-12-12. Pull both from PTAB E2E: https://ptacts.uspto.gov/ptacts/ (docket "CommScope Technologies LLC et al. v. Belden Canada ULC et al., IPR2024-01218"). Asserting disclaimed claims after that notice is a fee-shifting/§ 285 exposure.
2. If the assertion is on claims 6, 7, 14 or 15, note these are untested — no FWD, no claim construction from any tribunal, and the Delaware case was administratively closed before claim construction was decided. The parties' Joint Claim Construction Chart shows the fight would center on the "standard width unit" term (Belden: "a unit of width that accommodates a single receptacle module, such as a LC Quad (SC Duplex) technology footprint comprising four connectors"; CommScope: indefinite) and on "a width of the front edge is substantially evenly divisible by said standard width unit" (claim 8 — now disclaimed). Indefiniteness under § 112(b) on the surviving claims' "standard width unit"-derived limitations is a live defense.
3. If you are considering your own IPR on claims 6/7/14/15, calibrate expectations hard. It is now late 2026. The patent issued 2020-10-06 with a 2017-10-03 priority date, so it is well past the six-year mark and would face the "settled expectations" discretionary-denial factor, alongside the restored Fintiv framework. Institution rates collapsed under Acting Director Stewart and Director Squires (reported ~32% "net" under Stewart and near-zero in early Squires-era summary denials, with denials issued without reasoning and effectively unreviewable on rehearing). Budget for a discretionary-denial brief and a real possibility of denial without a merits ruling. Sotera stipulations are no longer dispositive.
4. Watch the docket for claim resurrection. The highest-value monitoring target is whether Belden files a reissue or reexamination, or prosecutes the pending family members (US 18/328,431 → US 12,313,896; US 18/232,729 → US 12,332,491; US 19/239,502 → US 2025/0306324) toward claims that are not patentably distinct from the disclaimed '107 claims. That would implicate 37 C.F.R. § 42.73(d)(3)(i) estoppel and the Board's adverse-judgment precedent on disclaim-and-refile tactics. Also confirm no CAFC appeal was docketed by 2025-11-10 (60 days from the 2025-09-08 judgment) on CourtListener or the Federal Circuit docket.

Confidence and limits. Confirmed: proceeding number, parties, filing date, institution date, termination-by-adverse-judgment date, the disclaimed claim set, the IRP certificate date, and the related Delaware stay. Not confirmed within my search budget: the exact challenged-claim list and ground-by-ground art mapping from the institution decision, and the panel composition for -01218 specifically (closest analog I found was the sibling -01382 panel). Retrieve Papers 2, 7/8 (institution) and the 2025-09-08 adverse-judgment paper from PTAB E2E before relying on those details, and do not quote grounds or a panel I have not read.

I'll start by verifying the litigation, the IPR, and the recording details that aren't fully exposed in the patent text I was given.

Let me check the assignment records themselves and Belden's corporate status.

Let me make one targeted attempt at the recorded correspondent field before writing.

Ownership & Assignment Reconstruction — US 10,795,107 B2

Patent: US 10,795,107 B2 — "Modular fiber optic cassette, system and method"
App. No.: 16/143,831 · Filed: 2018-09-27 · Granted: 2020-10-06 · Priority: US 62/567,339 (2017-10-03)
Source fetched: https://patents.google.com/patent/US10795107/en (2026-09-26)

Inventors

Pilon is the only named inventor on the '107 patent and on the entire US family. Employer-at-filing is inferable from the application data sheet (applicant of record is BELDEN CANADA ULC, Saint-Laurent) and from the nunc pro tunc assignment to Belden Canada Inc. effective 2018-10-01 — i.e., an employer-employee invention assignment, executed the day of filing. The prior-art search result also shows Pilon's inventor declaration listing him personally at the Lachine address, consistent with a rank-and-file engineer rather than a founder-assignor.

Pattern notes:

• No co-inventor cluster, so the classic "all inventors departed within 12 months" fire-sale precursor is not applicable / not observable.
• No evidence in any retrieved record of Pilon assigning to, or being employed by, any entity other than Belden. No inventor-side follow-on entity exists in this chain.

Original assignee

Entity named on the issued patent: BELDEN CANADA ULC (Saint-Laurent, Quebec, Canada). The application was filed by BELDEN CANADA INC. and the patent issued to its successor BELDEN CANADA ULC following the 2020-03-20 conversion/continuance recorded at reel 053639/0483.

Corporate parent: Belden Inc. (NYSE: BDC), a Delaware corporation, St. Louis, MO — confirmed by Belden's Rule 7.1 corporate disclosure filed in Belden Canada ULC v. CommScope, D. Del. 1:23-cv-00810 (("identifying Corporate Parent Belden Inc. for Belden Canada ULC")). The PTAB mandatory notice in IPR2024-01228 likewise names "Belden Canada ULC and its parent company, Belden Inc." as real parties-in-interest.

Primary line of business: Belden Inc. designs and manufactures network infrastructure — copper and fiber optic cable, connectors, interconnect panels, racks/enclosures and cable management ("Enterprise product lines include copper cable and connectivity solutions, fiber cable and connectivity solutions, interconnect panels, racks and enclosures…" — Belden Form 10-K). Belden Canada ULC is the Canadian operating/subsidiary arm that holds and commercializes the fiber-cassette portfolio.

Did they ship a product embodying the claims? Yes — high confidence. The asserted patents cover the same high-density modular cassette/panel architecture Belden sells into data-center and enterprise cabling; the parallel litigation accuses CommScope's competing "Propel™" panels and modules, i.e. Belden is enforcing in the exact product market it sells into. (PatSnap/Ex Parte complaint analysis of 1:23-cv-00810.)

Current status: Operating. Belden Inc. is a publicly traded going concern; Belden Canada ULC is an active, fee-paying patent owner (4th-year maintenance fee paid 2024-03-25). No insolvency, receivership or assignment-for-benefit-of-creditors event appears anywhere in the record.

Assignment timeline

Three assignment records exist for app. 16/143,831, plus one superseded/erroneous record. All are internal to the Belden corporate family — there is no third-party transfer of any kind.

• 2018-10-01 (executed) / recorded 2018-10-22 — Reel 047254/0973

  • Conveyance: Assignment (recorded as Nunc pro tunc assignment)
  • Assignor: Pilon, Vincent (individual inventor)
  • Assignee: BELDEN CANADA INC. (Canada)
  • Correspondent: Not retrievable. The Assignment Center correspondent field for this reel/frame was not exposed in any source I could reach; the public prosecution record shows the filing correspondence address as Lavery, De Billy, LLP, 1 Place Ville Marie, Suite 4000, Montreal, QC H3B 4M4, Canada (Customer No. 25545), docket G13569-00485-HM — but I cannot state as fact that Lavery, De Billy was the recording correspondent without seeing the reel/frame image. Flagging as an open verification item rather than asserting it.
  • Context: Ordinary inventor-to-employer invention assignment executed the day of filing, back-dated nunc pro tunc — boilerplate, not a transfer of interest to a third party.

• ⚠️ 2020-03-20 (executed) / recorded ~2020-08 — Reel 053614/0001 — DEFECTIVE, SUPERSEDED

  • Conveyance: Assignment (incorrect record)
  • Assignor: BELDEN CANADA INC. → Assignee: BELDEN CANADA ULC
  • Note: this record misidentified the application number as 16/143,381 (the '107 app is 16/143,831). It was affirmatively corrected by the next entry. Listed here because it appears in the chain of record and is the reason for the corrective filing.

• 2020-03-20 (executed) / recorded 2020-08-31 — Reel 053639/0483

  • Conveyance: Correction ("Corrective assignment to correct the application number 16/143,381 previously recorded on reel 053614 frame 0001")
  • Assignor: BELDEN CANADA INC.
  • Assignee: BELDEN CANADA ULC
  • Correspondent: Not retrievable from the sources available to me.
  • Context: Change of name / internal reorganization only — fixes the mis-recorded application number; no change in beneficial ownership.

• 2020-03-20 (executed) / recorded 2020-12-04 — Reel 054592/0263

  • Conveyance: Assignment ("Assignment of Assignors' Interest")
  • Assignor: BELDEN CANADA INC.
  • Assignee: BELDEN CANADA ULC
  • Correspondent: Not retrievable from the sources available to me.
  • Context: Internal corporate reorganization — the Inc. → ULC conversion of the same Belden Canada legal entity (same address, same ultimate parent Belden Inc., same effective date 2020-03-20 as the corrective record). Not an arm's-length transfer.

Related but NOT an assignment (noted for completeness): a statutory disclaimer of complete claims 1–5 and 8–13 was filed by the patent owner on 2025-08-07 in the IPR, recorded at USPTO ~2025-08-26, effective 2025-07-07. This is a substantive narrowing, not a change of ownership — only claims 6, 7, 14 and 15 now survive, and asserted independents 1 and 8 are gone.

Correspondent caveat: the US Patent Assignment Search record's correspondent field (the attorney/agent who filed each recording) was not exposed in Google Patents' legal-events table, and I could not retrieve the reel/frame images with my remaining steps. I am therefore not able to name a recording correspondent for any of the three links. I will not substitute the prosecution firm (Lavery, De Billy) for a recording correspondent I did not see. Verifiable at the Assignment Center search by patent number (https://assignmentcenter.uspto.gov/ · https://assignment.uspto.gov/patent/index.html).

Timeline diagram

timeline
    title Ownership of US 10795107
    2017 : Provisional filed by Vincent Pilon
    2018 : Non provisional filed
         : Assigned to Belden Canada Inc
    2020 : Patent issues to Belden Canada ULC
         : Internal reorg from Inc to ULC
    2023 : Belden Canada ULC sues CommScope in Delaware
    2024 : CommScope files IPR against the patent
    2025 : Claims 1 to 5 and 8 to 13 disclaimed

NPE / troll-pattern signals

Additional non-NPE-specific observation: the IPR2024-01218 proceeding (CommScope Technologies LLC et al. v. Belden Canada ULC et al., filed 2024-07-26, instituted 2025-02-27) terminated by adverse judgment on 2025-09-08, immediately following the patent owner's statutory disclaimer of claims 1–5 and 8–13 (filed 2025-08-07). This weakens the enforcement value of the '107 patent but is entirely orthogonal to ownership: the surviving claims 6, 7, 14 and 15 remain held by Belden Canada ULC.

Verdict

Operating-company assertion

Belden Canada ULC — a product-shipping subsidiary of NYSE-listed Belden Inc. — is both the sole post-issuance owner and the plaintiff, having asserted US 10,795,107 against direct competitor CommScope in Belden Canada ULC v. CommScope, Inc. et al., No. 1:23-cv-00810 (D. Del., filed 2023-07-27). Every recorded assignment (reels 047254/0973, 053639/0483, 054592/0263) is internal to the Belden family: one inventor-to-employer nunc pro tunc assignment and two Belden Canada Inc. → Belden Canada ULC reorg/correction records effective the same day, with no third-party, shell, aggregator or known-asserter entity ever appearing. Zero of the eight NPE signals are present, and the corporate-parent disclosure in the Delaware docket independently confirms the chain terminates at an operating manufacturer rather than a licensing vehicle.

Caveat on scope: the assignment-center correspondent field is the one element I could not verify to the reel/frame level; the three ownership determinations above rest on the recorded reel/frame numbers, execution dates and assignee names, which are solid.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (mirror: https://assignment.uspto.gov/patent/index.html) — search application 16/143,831 or patent 10,795,107.

I'll verify the patent record and then investigate the substantive prior-art references cited on its face.

Now let me verify the remaining cited references and the IPR's asserted art.

Prior-Art Analysis — US 10,795,107 B2

1. Record verified (and scope control)

I searched for patent number 10,795,107 specifically. The record returned is the correct one and I am not conflating it with any similar number:

Sources: Google Patents US10795107B2; Unified Patents portal record. Unified's portal repeats the one-day-earlier dates (priority 2017-10-02, app. 2018-09-26) already flagged in the earlier section; I have not reconciled them.

Framework applied. Effective filing date 2017-10-03 → AIA §102 applies. I therefore test each reference under §102(a)(1) (patented/published before 2017-10-03) and §102(a)(2) (U.S. patent or published application effectively filed before 2017-10-03, naming another inventor), with the §102(b)(2)(C) common-ownership exception and §102(d) provisional-support requirement noted where they bite. Anticipation requires every limitation in one reference.

⚠️ Cross-reference, not repeated here: the earlier section establishes that claims 1–5 and 8–13 were statutorily disclaimed (effective 2025-07-07) following IPR2024-01218, where CommScope's grounds 1–3 used Fukui and grounds 4–6 used Sauter (US 9,690,064) as primary references. That reframes everything below: a §102 showing against claims 1–5/8–13 is now largely academic, and the operative anticipation question is claim 14 and claim 15.

2. The face citations — full analysis

Google's "Citations (19)" table for the '107 patent contains a date anomaly that materially affects the §102 answer, which I flag up front rather than silently correcting: six of the nineteen entries have filing/priority or grant dates that post-date the '107's 2017-10-03 priority date (including one filed in March 2019). Such documents cannot be §102(a)(2) art against the '107 claims. The most plausible inference is that Google is aggregating family-level citations (the '107, '542, '422, '423 and later continuations share one citation pool), so my table separates "plausible prior art to the '107" from "cannot be prior art to the '107 on these dates." I state this as an inference, not a verified fact.

2.A Group A — pre-2017 art (available as §102(a)(1) references)

2.B Group B — §102(a)(2) art (effectively filed before 2017-10-03, published after)

2.C Group C — listed as citations but not prior art to the '107 patent

3. Claim-by-claim anticipation conclusions

4. Most relevant prior art — ranked

Tier 1 — real validity risk (the art that actually mattered):

1. US 9,690,064 B2 — Leviton Manufacturing Co., Ltd. (Sauter et al.), "Multi-gang cassette system," filed 2015-11-10, granted 2017-06-27. This is in the '107's face citations, and it is the single most dangerous reference — which is exactly why CommScope chose it as "Sauter" (Ex. 1006) for IPR grounds 4–6. It expressly teaches "cassettes of various sizes… mounted simultaneously… on a given tray," single-gang and multi-gang cassettes being integer multiples of one gang width, on the same tray without modification, with a spring-loaded latching protrusion engaging an aperture in the tray and a front release latch. That maps uncomfortably closely onto claim 1's "first cassette width… and a second cassette width different" + "fastener for removeably securing each of said cassettes." This reference should be treated as the closest art and given a full §102/§103 mapping. (Text; CommScope Ex. 1006 copy)
2. US 2019/0004268 A1 — AFL Telecommunications (Vogel) — §102(a)(2) art; slidable tray + modular cassettes with front LC/rear MPO + latches in a rackmount enclosure.
3. US 10,215,944 B2 — Panduit (Sedor et al.) — §102(a)(2) art directed at the very problem the '107 claims to solve (multiple module sizes on one tray).
4. Corning US 9,075,216 B2 — modules/cassettes in a slideable shelf within a housing.
5. Corning US 2016/0062050 A1 — front duplex + rear MPO module architecture (claims 3–4).

Tier 2 — element-level material: US 7,509,015 (Ortronics latch/cassette), US 8,913,866 (Corning movable adapter panel), US 2016/0033732 (Corning module holder), US 9,008,485 (Corning housing), US 2007/0131628 (rack-mount shelf / claim 5).

Tier 3 — background only: US 4,792,203, US 2009/0290842, US 7,331,722, US 2002/0131730.

Also worth pulling (from the family "Fiber Cites Families" list, not face citations): US 9,690,065 (Panduit, high-density fiber enclosure), US 9,885,845 (CommScope, module & assembly for fiber optic interconnections), US 8,712,206 (Corning, high-density fiber optic modules — the "7.0 mm per connection" family litigated in Panduit v. ITC), and NL 2008155 C2 ("Optical patch cassette"). These are family-level citations and are the natural place to look next for §102 art against claims 1/8.

5. Two consequences that follow from the disclaimer posture

1. For claims 1–5 and 8–13, this §102 exercise is now moot. Belden disclaimed them effective 2025-07-07 (docketed on Google Patents 2025-08-26) after the Board instituted IPR2024-01218 on 2025-02-27; the IPR certificate issued 2025-12-12. A disclaimed claim cannot be infringed and needs no prior-art defense. The anticipation analysis you asked for is most useful as a record of why the claims were vulnerable — and Leviton '064 / AFL '268 / Panduit '944 are that record.
2. For claims 14 and 15, the cited art is essentially irrelevant. None of the nineteen references discloses the magnet-plus-boss-pair securing structure. My analytical inference, not an official statement: claims 6 and 7 depend from disclaimed claim 1 and are therefore effectively inoperative, leaving claim 14 (and 15) as the only claims with live scope — and they survived the disclaimer precisely because the magnet/boss combination was outside the art of record.

6. Uncertainties I am explicitly flagging

1. Verification depth. I retrieved and confirmed the full text of US 10,215,944 (Panduit), US 2019/0004268 (AFL), US 9,690,064 (Leviton/Sauter) and US 9,075,216 (Corning). My tool budget was exhausted before I could open the full text of the remaining fifteen. Their descriptions above rest on the titles, assignees, dates and abstracts shown in the authoritative patent text plus my own knowledge, and should be re-verified against the full specifications before being relied on in a validity opinion.
2. The citation-list date anomaly (§2.C) is unexplained. I have not confirmed whether Google is aggregating family-level citations; I flagged it rather than "correcting" any identifier.
3. US 10,281,672 B1 may claim benefit of a provisional I could not verify; if so, it could qualify as §102(a)(2) art.
4. §102(d) caveat for the two Group B references. Their §102(a)(2) status depends on the relied-on disclosure being supported by their 2016 (Panduit) / mid-2017 (AFL) priority filings.
5. I could not retrieve the full institution decision in IPR2024-01218, so I cannot state definitively which of the six grounds the Board instituted on — only that institution issued 2025-02-27 and that the proceeding terminated by adverse judgment.

Primary sources: US 10,795,107 B2 · IPR2024-01218 status · Belden Canada ULC v. CommScope, No. 1:23-cv-00810-RGA (joint claim construction chart, Ex. 1011)

I'll ground this obviousness analysis in the prior-art references cited on the patent page and in the IPR record. Let me verify key references first.

The IPR petition grounds are directly on point. Let me pull the petition's claim charts and identify the references.

Obviousness Analysis — US 10,795,107 B2 under 35 U.S.C. § 103

Builds on the two prior sections (bibliographic/claim summary; litigation summary). I do not repeat their content.

0. Two contradictions to flag before proceeding

1. Date conflict. This task states the current date is April 26, 2026, but my system date is Sept. 28, 2026, and the prior sections already logged a CAFC order dated June 29, 2026 and an IPR certificate dated Dec. 12, 2025. The events relied on below (IPR2024-01218 institution, adverse judgment, disclaimer of claims 1–5 and 8–13, K1 certificate) are all dated before April 26, 2026, so this analysis is unaffected — but the inconsistency is unresolved.
2. The "operative claim set" problem carried forward. Per the prior sections, claims 1–5 and 8–13 are statutorily disclaimed and were cancelled by the K1 certificate. Because claims 6 and 7 depend from claim 1, the only claims with self-standing scope are 14 and 15. An obviousness opinion on the printed claims 1–13 is still valuable, however, because five co-pending family members (US 11,435,542; 11,656,422; 11,740,423; 12,313,896; 12,332,491) carry substantially the same subject matter. I analyze all claims, flagging where the posture differs.

1. Level of ordinary skill (POSITA)

As of the Oct. 3, 2017 priority date, a POSITA would hold a bachelor's degree in mechanical or electrical engineering (or equivalent practical experience) plus roughly two to four years designing fiber-optic enclosures, patch panels, and cassette/module trays; alternatively, a senior field/design technician with commensurate hands-on experience. This is a mechanical packaging art — the references are simple structural documents, so the POSITA definition is unlikely to be outcome-determinative. (Cf. the comparable definition offered in the parallel CommScope litigation: bachelor's degree in physics or engineering plus industry cable-installation experience.)

2. Claim-construction anchor that drives the whole §103 question

Per the prior section, "standard width unit" was disputed (CommScope: an industry-standardized actual width, e.g. an SC-Duplex footprint, else indefinite; Belden: a unit of width accommodating one receptacle module, e.g. an LC-Quad/SC-Duplex footprint of four connectors). This is the pivotal issue for §103: if the term is construed broadly (any repeating pitch), the prior art's "gang" cassettes read on it directly. If it is construed narrowly to a specific standardized four-connector footprint measured outer-edge to outer-edge, the prior art must be shown to produce cassettes that are exact integer multiples of that footprint — and Petitioners in the sibling IPR struggled precisely there (see §6.3).

3. Prior-art inventory (drawn from the prior-art section of the patent page)

A. Cited in the patent's own IDS ("Citations (19)") and thus of-record:

B. Prior art located from the "Family Cites Families (22)" and "Cited By (16)" sections and the IPR record:

C. IPR2024-01218 petition grounds (CommScope v. Belden) — retrieved from the PTAB petition document at
https://ptacts.uspto.gov/ptacts/public-informations/petitions/1556417/download-documents (Petition, IPR2024-01218):

The petition's own statement of the allowance rationale is the single most probative item in the file:

"The examiner allowed the claims because the closest art he was aware of had rails between the cassettes, and he concluded it would not be obvious to remove the rails to have the modules touch. 'Fukui' is a new reference. It does not have any rails between the cassettes and discloses the exact limitations the examiner believed were missing."

The Board instituted on Feb. 27, 2025 — i.e., it found a reasonable likelihood that at least one challenged claim is unpatentable.

Verification limits I must state. The petition identifies Fukui as US 10,101,551 and asserts it is §102(a)(2) art (i.e., prior art only via an earlier effective filing date, since its Oct. 16, 2018 issue date post-dates the '107's Oct. 3, 2017 priority date). I could not independently verify Fukui's effective filing date, its full disclosure, or the identity/exhibit numbers of Laarveld and Sauter within my search budget. The Sauter reference is described in the sibling POPR (IPR2024-01228) as disclosing "single-gang, double-gang, triple-gang and other multi-gang" cassettes mounted with rails/rail guides/latch, including "BASE-8" (2 W/4 W) and "BASE-12" (3 W/6 W) cassettes. Those descriptions are Petitioner- and Patent-Owner-characterizations, not my own reading of the documents.

4. Claim 1 — element-by-element

4.1 Motivation to combine (KSR rationales)

1. Same field of endeavor and same problem, expressly acknowledged by the '107 itself. The '107's Background states the drawback that cassettes for different fiber counts "are generally incompatible with one another and additionally require trays with different spacing, securing mechanisms and the like." Sedor '944 states the identical problem ("trays are currently designed to only support one size of fiber optic module") and offers the same class of solution. KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398, 417 (2007) (references addressing the same problem in the same field).
2. Predictable result. Selecting module widths as integer multiples of a common pitch is the standard modularity technique in this art (19″ EIA rack units; gang cassettes; 12-fiber trunks fanned to 6× duplex). Using a known technique (ganging) to improve a similar device in the same way is a classic KSR ground.
3. Finite number of predictable solutions. Once the tray is 12 units wide and the unit must divide the front edge evenly (claim 8; claims 2, 12), the workable widths collapse to {1, 2, 3, 4, 6, 12}. Enumerating a subset is the "finite number of identified, predictable solutions" scenario.
4. Design need / market pressure. Data-center migration among 8-, 12-, and 24-fiber architectures (US 2016/0062050 is literally directed to "migration between duplex and parallel multi-fiber solutions") created pressure to populate one tray with mixed-density cassettes — the '107's stated goal of "optimizing the use of tray space."
5. Eliminating inter-cassette rails is a known density technique. Where the reference set includes both rail-separated cassettes and rail-less abutting cassettes, selecting the higher-density arrangement to maximize ports per rack unit is the predictable use of a known design option, not an inventive step. The Board's institution decision is consistent with that view.

Confidence: High that claims 1–5, 8–13 would have been held obvious on this record — the patent owner's own disclaimer, following the Board's institution and the adverse judgment, functionally removes the question for those claims. Moderate-to-high as a prospective §103 holding for the family continuations, subject to §6 below.

5. Dependent claims 2–7, 9–13

6. Where the §103 case is vulnerable (and what claim 14 turns on)

6.1 The "standard width unit" construction cuts both ways

If the term is construed narrowly (outer-edge-to-outer-edge multiple of an LC-Quad/SC-Duplex footprint), the prior art must produce exact multiples. The Patent Owner's POPR in IPR2024-01228 hammered exactly this point — that Sauter's cassettes contain gaps between receptacle modules, so the double-gang cassette is not exactly twice the single-gang cassette, and the "width as measured between outer side edges" limitations are unmet. That argument was aimed at the continuation claims (which recite "outer side edge … directly adjacent"), not at '107 claim 1, which says only "touches." Flagging the contrast: the '107's "touches" language is broader and harder to distinguish than the family's "directly adjacent outer side edges" language.

6.2 Inconsistent modification across grounds

The POPR repeatedly attacked the petition for "selectively proposing modifications for some claim elements in isolation, while ignoring those same modifications elsewhere." Specifically: the petition needed a rail/latch on the right-hand side of the double-gang cassette to meet some elements, then proposed removing the rail/latch to meet the width elements. A court applying KSR's "predictable results" requirement — and the In re Ratti / bodily-incorporation line of cases — could find that this combination requires altering a reference's principle of operation rather than combining teachings. This is the strongest non-obviousness argument available, and it is the same intuition the examiner used in allowing the claims.

6.3 Teaching away

Sedor '944's entire inventive point is removable rails repositioned to accommodate different module sizes. The '107's claim requires cassettes to touch, which requires removing those rails. Belden can argue the primary reference teaches away from the claimed arrangement, since eliminating the rails defeats the reference's stated purpose. A Petitioner can rebut this (the rails are a design choice, and the '107 itself says the goal is "optimizing the use of tray space"), but the argument is real and the examiner already credited it once.

6.4 Claim 14 (and 15) — the magnet + stabilizing bosses

Claim 14 was never challenged in IPR2024-01218 (challenged claims were 1–5 and 8–13). It is therefore the claim with the best survival prospect, and the only one where I have no verified invalidating combination.

Elements and available art:

• Elongate rectangular fiber-receiving housing; front receptacles; rear multi-fiber connector with interconnecting fibers → US 2019/0004268; Telect '562.
• Magnet on the underside; tray is ferrous metal → I could not verify a specific prior-art reference in my search budget (my magnet-specific searches were cut off by the step limit). The general-knowledge theory is: replacing a known latch/tab-and-slot fastener (Sedor '944; US 2019/0004268's release slide; US 7,509,015's secure cassette assembly) with a known magnetic fastener on a steel tray, to obtain tool-less removal, is the substitution of one known fastener for another with a predictable result. Note also that the '107 specification itself presents the magnet as an alternative to the tabs/latch of the first embodiment — which, while not prior art, undercuts any argument that the magnet was an unobvious departure.
• Pair of stabilizing bosses on opposite sides of the magnet, each engaging a respective spaced slot in the tray → Telect '562 discloses a cassette with a detent engageable with a tray wall detent and a groove engageable with a tray protrusion; Sedor '944 discloses modules engaging cutouts; the '107's own slot 70 / boss 72 arrangement (FIG. 6B) is a routine anti-rotation/indexing feature. Two elements performing their known functions (fastening + anti-rotation) is a classic predictable aggregation.

Recommendation: claim 14 should be evaluated against a targeted prior-art search for (a) magnetically retained fiber-optic modules/cassettes, and (b) steel patch-panel trays with magnetic module retention. Until then, I would not opine that claim 14 is obvious, and I would rate claim 15 (which merely imports the 1/2/3/4/6 width limitation) as obvious only to the same extent as §4–§5 above.

6.5 Secondary considerations

Belden could assert commercial success, industry adoption, or long-felt need. Two practical realities weigh against it: (i) the prior sections record that Belden disclaimed the very claims it was asserting and moved to terminate by adverse judgment, which is at minimum strong practical evidence that the merits were not defensible; and (ii) the '107's stated problem (mixed-fiber-count incompatibility) was expressly recognized in the prior art (Sedor '944; US 2016/0062050), which tends to defeat long-felt-need. A statutory disclaimer is not an admission of invalidity, and I do not treat it as one — but an adverse judgment in IPR2024-01218 is a binding adjudication of unpatentability as to claims 1–5 and 8–13.

7. Bottom line

1. Claims 1–5 and 8–13 should be treated as unpatentable, and are in any event cancelled (disclaimer + adverse judgment + K1 certificate, Dec. 12, 2025).
2. The best §103 combination in the record is: Sedor (US 10,215,944 B2) as the primary reference for a modular tray accepting differently-sized cassettes at a common pitch, in view of US 2019/0004268 (AFL) or Telect '562 for slidable trays + modular cassettes with front LC/MPO receptacles and latchable mounting, and in view of Fukui (US 10,101,551) for the rail-less, cassette-touches-cassette arrangement that the examiner identified as the only missing limitation — with WO 2011/143537 supplying claim 7's cable-management bracket and US 2007/0131628 / Sedor '944 supplying claim 5's rack brackets.
3. The single genuinely contestable limitation is "touches" (the no-rails requirement), and the attack on it depends on whether KSR-style predictable-result reasoning survives the teaching-away and inconsistent-modification attacks. The Board's Feb. 27, 2025 institution decision indicates it likely does.
4. Claim 14 is the open question. It was outside the IPR and I found no verified magnet-cassette reference. It is the claim on which the family's remaining commercial value most likely rests, and the one warranting a dedicated prior-art search.
5. Do not overlook the §102(a)(2) trap: Sedor '944, Fukui, and the Telect '562 all issued/published after Oct. 3, 2017, so their status depends entirely on earlier effective filing dates. Any opinion relying on them must document those dates.