US 7260153

I'll search for current information on this patent, including USPTO records and any CAFC 2026 docket activity, before summarizing.

Let me check for any Federal Circuit (CAFC) 2026 activity and the current USPTO assignment record.

Let me confirm current legal status and check for any reexamination or appeal activity specific to this patent.

US Patent 7,260,153 B2 — Analyst Summary

Search scope note: I queried USPTO-derived records (Google Patents, PatentsView/Justia, PTAB/PTacts dockets) and NPE/litigation tracker databases. I found no Federal Circuit (CAFC) 2026 docket naming US 7,260,153 or its owner. The only post-grant/agency activity I could locate in 2026 concerns a different Fleet Connect patent (US 7,742,388), not the '153 patent. I flag this as a negative finding with residual uncertainty (see "Uncertainties" below).

1. Bibliographic Data (from the authoritative patent text)

Ownership chain (recorded assignments): MIMOpro Ltd → Kromisovich Data Ltd., L.L.C. (2008-10-24) → OL Security LLC (merger, 2016-01-22) → Intellectual Ventures Assets 160 LLC (2020-05-01) → Fleet Connect Solutions LLC (2020-06-05).

2. Abstract (verbatim)

"A Multi Input Multi Output (MIMO) Wireless Communication System method and apparatus are proposed whereby in a 2-way wireless communication system with scattering, random and imperfectly estimated propagation channels the ubiquitous and inherent MIMO cross-talk interference problem is solved so that robust and predictable Extended Communication Range and Extended Data Rate are achieved."

3. Plain-Language Overview of the Independent Claims

The claim set has 41 claims. The visible text runs through claim 28; claims 29–41 are truncated in the source I was given. On the visible record, the independent claims are 1, 21, and 28.

Claim 1 — Method of evaluating a MIMO channel

A four-step method:

1. Define a "channel matrix metric." For each data sub-stream, define a predefined function of the channel matrix's singular values that expresses a measure of the cross-talk SNR for that sub-stream.
2. Obtain an estimated channel matrix.
3. Perform an SVD on that estimate, yielding the estimated singular values plus the left-hand unitary weighting matrix and right-hand unitary weighting matrix and the diagonal singular-value matrix.
4. Compute a per-sub-stream cross-talk measure from the metric (step 1) and the estimated singular values (step 3).

In plain terms: take a noisy channel estimate, decompose it, and use the singular values to quantify how much the sub-streams will bleed into each other — a diagnostic/predictive step rather than a transmission step.

Claim 21 — Method of communicating data over a MIMO channel

A transmit-side method that builds on the metric of claim 1:

1. Obtain an estimated channel matrix.
2. Weight the data sub-streams by a unitary transmission weighting matrix equal to the conjugate transpose of the SVD right-hand unitary matrix (the classic V′ transmit-weighting).
3. Apply a pre-equalizer matrix to form a "pre-equalizer data stream." The pre-equalizer is derived by (a) defining the per-sub-stream cross-talk SNR metric, (b) computing per-sub-stream cross-talk measures from that metric and the estimated singular values, and (c) solving for a pre-equalizer that yields a "modified channel matrix" (the estimated channel pre-multiplied by the pre-equalizer) with improved cross-talk measures.
4. Transmit the pre-equalized stream.

In plain terms: pre-distort the transmitted signal with a matrix P (Hm = H·P) so the effective channel has better cross-talk behavior than the raw channel does, then send data through it.

Claim 28 — Transmission apparatus (device claim)

An apparatus version of the claim-1 concept, comprising:

• an SVD performer configured to perform the SVD of an estimated channel matrix (yielding the left-hand unitary, the diagonal singular-value matrix, and right-hand unitary matrices); and
• a measure calculator (associated with the SVD performer) configured to compute a per-sub-stream cross-talk measure from the estimated channel singular values and a channel matrix metric, the metric comprising a respective predefined f[unction]… (text truncated at this point).

In plain terms: a hardware/software unit that does the SVD and then applies the cross-talk SNR metric to produce per-sub-stream interference estimates.

Key dependent-claim themes (representative, not exhaustive):

• Cross-talk metric variants: expected value (claim 2), median (claim 3), or a value at a pre-determined probability threshold (claim 4) of the cross-talk SNR.
• Classifying channel "goodness" via iso-valued polynomial surfaces / discriminant functions dividing singular-value space (claims 5, 7, 8, 25) and the G = ∩Gi "Good" region intersection.
• Computing the pre-equalizer as products of the right-hand unitary matrix, inverse of the estimated singular-value matrix, and a desired singular-value matrix (claim 9).
• Optimizing a weighted sum of net total SNR gains (claims 10, 17, 26) and power-allocation constraints (claims 12–15, 22).
• Fallbacks: set pre-equalizer to identity / reduce sub-stream count if SNR loss is too high or net gain < 1 (claims 16, 17, 23, 24).
• A post-equalizer diagonal matrix computed from a transmitted training sequence and the reception weighting matrix, correcting magnitude/phase distortion per sub-stream (claims 18, 27).
• Time-domain (claim 19) vs. frequency-domain / OFDM (claim 20) processing.

4. Litigation / Enforcement Context (as found)

• Owner/assertion entity: Fleet Connect Solutions LLC, described in third-party databases as an NPE and an entity of Empire IP. Assertion campaigns have named Peloton, Rivian, USIC Locating Services, Aptiv, Penske Automotive, Heritage-Crystal Clean, Southern Tire Mart, Bose, Lenovo, Casio, Nikon, Lantronix, Ezurio, PACCAR/Peterbilt, Platform Science, Clarience, and others (Stanford NPE database; Unified Patents Patroll page).
• Unified Patents prior-art contest (2023–2024): A $2,000 Patroll contest sought prior art on at least claim 1 of US 7,260,153 B2; a winner was announced (contest page: patroll.unifiedpatents.com/contests/ux3ib2PAn5YAzeJRA).
• Most cases resolved by stipulation: Numerous district-court suits were dismissed with or without prejudice (e.g., Pittasoft — dismissal with prejudice, Jan 2026; Bose v. Fleet Connect; USIC Locating; Pegasus Transtech). None produced a merits ruling on the '153 patent's validity/infringement.
• Adjacent 2026 agency activity (not this patent): On 2026-03-26 the Central Reexamination Unit issued a notice of intent to issue a reexamination certificate cancelling claims of US 7,742,388 — a different Fleet Connect patent (Unified Patents, ex parte reexam 90/019,840). I found no equivalent reexam/IPR certificate for the '153 patent.

5. Uncertainties / Caveats

1. CAFC 2026 docket — negative finding. I did not locate any Federal Circuit appeal docketed in 2026 involving US 7,260,153 or Fleet Connect Solutions. This may reflect that no such appeal exists, or that my searches did not surface it. Verify directly via the CAFC docket (cafc.uscourts.gov) and PACER if a definitive answer is required.
2. Claims 29–41 are missing from the source text (it truncates mid-claim-28). Therefore additional independent claims (likely a receiver-side or post-equalizer apparatus/method claim) may exist beyond claims 1, 21, and 28. Do not treat the three identified independent claims as exhaustive.
3. Legal status. The listing states "Expired – Lifetime" with an adjusted expiration of 2025-04-22. Given the 2003-04-28 filing and 2002-09-09 priority, this reflects a 20-year term plus Patent Term Adjustment. Treat the status as an assumption of the database, not a legal conclusion.
4. Date discrepancy. The task states a current date of 2026-04-26, while the source-page metadata I retrieved is timestamped 2026-09-29. This affects only recency framing of the docket search, not the bibliographic facts.
5. Assignee field. Google Patents lists Fleet Connect Solutions LLC as current assignee; the underlying USPTO Assignment records I saw stop at the 2020-06-05 conveyance to Fleet Connect Solutions LLC. Confirm the live assignee at the USPTO Assignment Search if this matters.

Primary source: https://patents.google.com/patent/US7260153/en (full text used above).

I'll search litigation databases for cases specifically naming US 7,260,153.

Let me pull the authoritative per-patent case lists and additional cases naming the '153 patent.

Litigation Involving US Patent 7,260,153

Scope note: I searched Unified Patents' litigation portal (patent-filtered case list), the Stanford NPE Litigation Database per-patent page for 7260153, CourtListener/RECAP, Justia dockets, RPX Insight litigation documents, and the litigation links embedded in the Google Patents record for US 7,260,153. I filtered strictly to the '153 patent (application 10/423,447) and excluded cases asserting only sibling Fleet Connect patents (e.g., 6,549,583; 6,633,616; 7,058,040; 7,656,845; 7,742,388; 8,005,053) unless the '153 patent is also expressly named.

Date discrepancy flag (unchanged from prior section): the task states a current date of 2026-04-26, while the system date is 2026-09-29 and the source-page metadata is timestamped 2026-09-29. Where a case status could have changed between those dates, I mark it "as last reported."

1. Bottom line

US 7,260,153 has been asserted extensively — it is the MIMO/wireless-communications patent in Fleet Connect Solutions LLC's (Empire IP) portfolio. I identified dozens of district-court actions naming it. Every case I could find has been resolved by voluntary dismissal or stipulated dismissal (some with prejudice, some without); I found no merits ruling, no claim-construction order, no jury verdict, no damages award, and no PTAB/Court of Appeals decision on the '153 patent itself. I also found no Federal Circuit 2026 appeal involving the '153 patent (consistent with the prior section's negative finding).

2. Cases where US 7,260,153 is expressly identified by count or exhibit

These are the actions in which the record I retrieved explicitly ties the '153 patent (as an asserted patent or an "Exhibit — Evidence of Use"):

3. Additional actions linked to the '153 patent on the Google Patents litigation record

The USPTO/Google Patents record for US 7,260,153 embeds these case links (I did not independently verify the '153 count in each, but they are listed on the patent's own page, not just the family page). Grouped by court:

• E.D. Tex.: 2:22-cv-00160, 2:22-cv-00327, 2:23-cv-00210, 2:24-cv-00240, 2:24-cv-00465, 2:24-cv-00705, 2:24-cv-00718, 2:24-cv-00941, 2:24-cv-00961, 2:24-cv-00964, 2:24-cv-01029, 2:24-cv-01047, 2:24-cv-01049, 2:24-cv-01050, 2:24-cv-01054, 2:25-cv-00126, 2:25-cv-00351, 2:25-cv-00524, 2:25-cv-00616, 2:25-cv-00658
• W.D. Tex.: 6:20-cv-01096, 6:21-cv-01306, 6:23-cv-00484, 6:23-cv-00489, 6:23-cv-00623, 6:23-cv-00682, 7:25-cv-00264, 7:25-cv-00280, 7:25-cv-00342
• C.D. Cal.: 2:23-cv-09324, 2:24-cv-01273, 2:24-cv-05871, 2:24-cv-07486, 2:25-cv-04890, 5:24-cv-00376, 5:24-cv-01273, 8:25-cv-00053, 8:25-cv-00054
• N.D. Cal.: 3:22-cv-04791, 3:24-cv-09204, 4:22-cv-06083, 5:22-cv-04791, 5:22-cv-06083
• E.D.N.Y.: 1:24-cv-01258, 2:24-cv-01258, 2:25-cv-04453, 2:25-cv-05178
• S.D.N.Y.: 1:24-cv-08486
• N.D. Ga.: 1:24-cv-01935, 1:24-cv-05269, 1:25-cv-04786
• D. Mass.: 1:22-cv-11987, 1:25-cv-10240
• M.D. Fla.: 8:25-cv-01652
• D. Ariz.: 2:24-cv-02894
• E.D. Va.: 1:25-cv-01941, 2:25-cv-00712
• S.D. W. Va.: 3:25-cv-00497
• E.D. Wis.: 1:25-cv-00853
• N.D. Tex.: 3:25-cv-00233
• E.D. Tenn.: 1:25-cv-00004
• Worldwide family litigation: Darts-IP family ID 32397054 (https://patents.darts-ip.com/?family=32397054)

Named defendants appearing across these records include (non-exhaustive): Rivian, Peloton, USIC Locating Services, Aptiv, Penske Automotive, Heritage-Crystal Clean, Southern Tire Mart, Platform Science (as plaintiff in a declaratory-judgment action, S.D. Cal. 3:25-cv-00457), Enterprise Holdings/Enterprise Fleet Management (2:24-cv-00718), Powerfleet (5:24-cv-00376), C.R. England/Orbcomm, and GPS Insight (2:24-cv-02894).

4. Litigation posture / patterns

• Single-owner, single-patent-family enforcement. Every '153 case traces to Fleet Connect Solutions LLC (Empire IP). The '153 patent is asserted as part of a rolling portfolio of ~7–13 patents (notably 6,549,583; 6,633,616; 7,058,040; 7,206,837; 7,260,153; 7,593,751; 7,656,845; 7,742,388; 8,005,053; 6,961,586; 8,494,581; 6,647,270; 7,092,723; 7,741,968).
• Sector focus: fleet telematics / ELD / connected-vehicle and consumer electronics (dash cams, routers, laptops).
• Outcome pattern: overwhelmingly stipulated/voluntary dismissals (both with and without prejudice), consistent with negotiated licensing resolutions. Exceptions where a defendant answered and counterclaimed (e.g., CalAmp; Vehicle Tracking Solutions/IntelliShift) remained open at last report.
• No decision on the merits of the '153 patent in any court I could locate.
• No PTAB trial (IPR/PGR/CBM) on the '153 patent. The only agency-side '153 activity is Unified Patents' 2023 Patroll prior-art contest (US-7260153-B2, $2,000 prize, ended 2023-12-04, winner announced 2024-02-07) — a crowdsourced prior-art search, not an instituted proceeding. (The 2026 ex parte reexam cancellation involving US 7,742,388 is a different patent and does not concern the '153 patent.)

5. Negative findings and caveats

1. No Federal Circuit / CAFC 2026 appeal naming US 7,260,153 or Fleet Connect Solutions LLC was located (consistent with the prior section's finding). Verify against the CAFC docket and PACER if a definitive answer is needed.
2. Several "Closed" statuses are ambiguous in the aggregate trackers — a closure can mean voluntary dismissal, dismissal with prejudice, transfer, or consolidation. Where I could confirm the nature of dismissal from a PACER-derived quote, I noted it; otherwise I say "Closed (as last reported)."
3. Dates differ by source. Unified Patents, Stanford NPE DB, Justia, and the complaints occasionally show filing dates one day apart (e.g., 2025-08-06 vs. 2025-08-08 for the IntelliShift matter; 2025-11-02 vs. 2025-11-03 for the Xirgo E.D. Va. matters). I report both where material.
4. Case-number/venue hygiene: I noticed trackers transposing venues for the same case (e.g., 2:25-cv-00712 is variously indexed as E.D. Va. and N.D. Cal.; 2:24-cv-01258 appears under both E.D.N.Y. and E.D. Tex. in the raw data). These are indexing inconsistencies in the aggregators, not separate suits — treat venue as "as reported."
5. Not exhaustive. Given the volume (60+ docket links) and the step limit, the tables above are comprehensive for cases where the '153 patent is expressly named, but I cannot guarantee I captured every portfolio suit in which '153 appears only implicitly. A PACER "patent number" full-text search plus the Darts-IP family 32397054 export would close any residual gap.
6. No auto-correction performed. All patent numbers (including the sibling patents 7,742,388 / 7,260,153 and the Darts-IP family ID 32397054) are reported exactly as they appear in the sources.

Primary sources: https://portal.unifiedpatents.com/litigation/caselist?patents=726015 · https://npe.law.stanford.edu/patent/7260153 · https://patroll.unifiedpatents.com/contests/ux3ib2PAn5YAzeJRA · https://patents.google.com/patent/US7260153/en (litigation links) · CourtListener RECAP (E.D. Va. 1:25-cv-01941; E.D. Va. 2:25-cv-00712) · Dockets.Justia (E.D.N.Y. 2:25-cv-04453; E.D. Wis. 1:25-cv-00853) · RPX Insight litigation documents.

I'll verify the structured "no PTAB proceedings on file" finding against live sources before writing the disposition.

Let me run a few more targeted checks for any PTAB filing, ex parte reexam on the '153 patent, or Federal Circuit appeal.

PTAB proceedings on file — finding

I could not identify a single AIA trial proceeding (IPR, PGR, or CBM) naming US 7,260,153. This matches the canonical structured data block supplied in the prompt: "The USPTO ODP API returns no AIA trial proceedings for this patent as of the most recent ingest." My independent web checks surfaced no petition, institution decision, Final Written Decision, termination, or Federal Circuit appeal for this patent. I report that as a negative finding with the caveats below.

Proceedings overview

Total AIA trial proceedings on US 7,260,153: 0. Breakdown by status is therefore all zeros — 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denials. The patent has never been subjected to an IPR, PGR, or CBM. No claims have been canceled by the Board and no claims have been adjudicated patentable; the entire claim set (1–41) is legally untested at the PTAB. For a defendant, the defensive posture is neither "the patent is hardened by surviving IPRs" nor "the claims are dead." It is: an untouched, expired patent with a publicly available, crowdsourced prior-art package — but zero Board precedent and zero estoppel cutting either way.

Do not confuse this patent with US 8,032,153 (TracBeam). Search results for "7,260,153" readily surface IPR2015-01701 and IPR2015-01713, which are Apple's and T-Mobile's IPRs on TracBeam's US 8,032,153 — a completely different patent (same trailing digits). No proceeding there concerns Fleet Connect's '153. Per the operating rules, I flag rather than auto-correct, and I do not import those outcomes here.

The one "PTAB-adjacent" event: a Patroll prior-art contest (not a proceeding)

Because there is no trial to report, the only meaningful post-grant activity is worth documenting under the same rubric, clearly labeled as not an AIA proceeding.

Patroll contest — Unified Patents v. Fleet Connect Solutions LLC (US-7260153-B2)

• Type: Crowdsourced prior-art bounty (Unified Patents Patroll). Not an IPR/PGR/CBM; no filing fee, no Board, no petitioner, no estoppel.
• Filed / window: 2023-10-04 (start) to 2023-12-04 (end); winner announced 2024-02-07.
• Status: Finished — Winner Announced (verbatim from the contest page).
• Prize: $2,000. Contest targeted at least claim 1.
• Scope: prior art directed to the MIMO/2-way wireless cross-talk concept; entries via Unified's Pearl claim-charting tool; art previously cited during prosecution excluded.
• Outcome: A winning prior-art submission was identified and the winning references were published for public download.
• Appeal: N/A (not an appealable proceeding).
• Defensive value: High, as a starting kit. A ready-made, claim-charted invalidity package on at least claim 1 exists in the public domain, with no § 315(e) estoppel attached to anyone. It is a free head start for a defendant — but it is not a merits adjudication and carries no preclusive effect.

Sources: https://patroll.unifiedpatents.com/contests/ux3ib2PAn5YAzeJRA · https://www.unifiedpatents.com/insights/2024/2/7/empire-ip-entity-fleet-connect-solutions-communications-patent-prior-art-found

Adjacent USPTO/agency activity (verified as different patents — do not attribute to the '153 patent)

Source (reexam service record): https://ptacts.uspto.gov/ptacts/public-informations/petitions/1556913/download-documents?...

Strategic summary

Claim status of 7,260,153. No claim of the '153 patent is canceled, none is confirmed patentable, and none has been construed by the Board — claims 1–41 are all UNTESTED at the PTAB. The only validity-relevant agency outcome anywhere connected to the '153 patent is the un-used prior art found in the 2023–2024 Patroll contest. Contrast this with the sibling '388 patent, whose claims have been canceled through ex parte reexam — that is a different patent and gives zero cover against a '153 assertion. Compounding the picture, the '153 patent expired 2025-04-22 (per the prior section's bibliographic analysis), so the sole live exposure is past damages (subject to the 35 U.S.C. § 286 six-year lookback), not injunctive relief.

Estoppel landscape. Because no IPR or PGR was ever filed, § 315(e)(2) and § 325(e)(2) estoppel applies to no one. There is no petitioner and no privy who is barred from raising any § 102/§ 103/§ 112 ground in district court or before the Board. A current defendant faces no estoppel impediment whatsoever — every prior-art ground, including the Unified Patents contest art, remains fully available in any forum. (Ex parte reexamination, where it occurred at all, does not create § 315(e)(2)-style IPR estoppel.)

Pattern signals. No petitioner has filed even once against this patent, so there is no serial-petitioner pattern. Fleet Connect has not had to defend any PTAB appeal of a '153 FWD, because there has been no FWD. The patent is a heavy-assertion / zero-challenge outlier: dozens of district-court suits (see the litigation section) yet not a single AIA petition — a notable gap for a patent this widely asserted over ~4 years. Whether the explanation is a short cash-flow/long-tail strategy by Fleet Connect, quick settlements, or defendant inertia, it means the invalidity record is unusually thin relative to the assertion record. Unified Patents appears in the chain only as a Patroll bounty sponsor, not as a petitioner.

Recommended next steps

If you are a defendant currently facing a '153 demand or suit:

1. You have no PTAB estoppel against you. Nothing decided at the Board binds you. You may raise § 101, § 102, § 103, and § 112 freely — several Fleet Connect adversaries have already pleaded exactly those defenses/counterclaims against the '153 patent (e.g., § 101 abstract-idea, § 112 indefiniteness/written description, § 102/§ 103). Pull the winning Patroll references first; they are free and already claim-charted.
2. Calibrate the IPR calculus against expiration. The patent expired 2025-04-22, so IPRs are still permissible on an expired patent but there is no injunctive stake — weigh the cost of an IPR against a § 286 past-damages exposure. Note the § 315(b) one-year bar runs from service of a complaint alleging infringement of the '153 patent; if you have been served, confirm your window immediately.
3. If you file, note the statutory clock. Any instituted trial must reach an FWD within one year of institution (35 U.S.C. § 316(a)(11)); the institution decision is due ~6 months from the petition's filing date (with the § 316(a)(11)/§ 314(b) spillover and any extensions).

If you are evaluating exposure without a pending suit: the absence of any IPR is itself the signal — the patent has not been narrowed, so a straightforward "the claims are already dead in an IPR" defense does not exist. Build your invalidity case from scratch using the Unified Patents winning art plus independently sourced references.

Caveats / uncertainties

1. Negative findings. No AIA trial proceeding and no Federal Circuit appeal (2026 or otherwise) naming US 7,260,153 was located. This is a negative finding with residual uncertainty; confirm directly against PTAB E2E / PTActs (ptacts.uspto.gov) and the CAFC docket if a definitive record is required.
2. No invented identifiers. Per the constraints, I have not manufactured any proceeding number. The only "IPR" numbers appearing above (IPR2015-01701/01713) are flagged as belonging to a different patent (US 8,032,153).
3. Ex parte reexam 90/019,766 — the patent it targets is unconfirmed; do not assume it is the '153 patent.
4. Date discrepancy (carried forward). The task states a current date of 2026-04-26; the environment/system date is 2026-09-29. This affects only recency framing, not the absence-of-proceedings finding.
5. Claims 29–41 are truncated in the authoritative text; my claim references rely on the visible record and on third-party complaints (which cite claims 1, 2, 19, 28, and 39 as asserted). PTAB claim-number conventions would use the same numbers, but no FWD exists to quote.

Primary sources: structured "PTAB proceedings on file" block (USPTO ODP ingest) · https://ptroll.unifiedpatents.com/contests/ux3ib2PAn5YAzeJRA · https://www.unifiedpatents.com/insights/2024/2/7/empire-ip-entity-fleet-connect-solutions-communications-patent-prior-art-found · https://patents.google.com/patent/US7260153/en (litigation links).

(Correction note: the correct Patroll URL is https://patroll.unifiedpatents.com/contests/ux3ib2PAn5YAzeJRA; the "ptroll" variant above is a typo to disregard.)

I'll research the assignment record for US 7,260,153 — starting with the USPTO Assignment Center, then cross-referencing the ownership chain and correspondents.

Let me dig for the specific reel/frame numbers and correspondents on the recorded documents.

Assignment Record Reconstruction — US 7,260,153 B2

Retrieval caveat up front (do not skip). I confirmed that a five-link assignment chain exists for this patent (the Google Patents "Legal Events" record reproduces each conveyance with assignor, assignee, conveyance type, and event date). However, I was unable to retrieve the Assignment Center reel/frame numbers or the recorded "correspondent of record" for any link in this session. The USPTO Assignment API/index did not surface in my searches, and secondary aggregators (FreePatentsOnline, Justia, CourtListener) do not expose reel/frame or correspondent. Per your constraints, I have not fabricated reel/frame numbers or attorney names — every such field below is marked [NOT RETRIEVED] and must be filled by a direct pull from the Assignment Center. The event dates, assignor/assignee identities, and conveyance types are grounded in the sources cited.

Date discrepancy (carried forward from prior sections). The task states a current date of 2026-04-26; the system/source timestamp is 2026-09-29. This affects only recency framing, not the bibliographic facts.

Inventors

Pattern notes:

• Single-inventor patent. No co-inventor team, no evidence of a departed engineering group. This is a founder-driven filing, not a corporate R&D team filing.
• Late-recorded inventor assignment — flag. The inventor→MIMOpro assignment was recorded 2007-06-26, roughly four years after the 2003-04-28 filing and ~2 months before grant (2007-08-21). A four-year gap between filing and recordation of the base inventor assignment is atypical for a company that was already the named applicant at filing; it is consistent with a confirmatory/curative assignment executed late (a common artifact when a startup is being readied for sale). Execution date not retrievable — the Google event date is the recordation/reporting date.
• Departure-within-12-months signal: not determinable. Nothing in the record shows the inventor's employment end date or any transfer of the inventor's interest to a third party. I explicitly decline to infer a "founder left → fire-sale" narrative from the 2008 asset transfer alone.

Original assignee

Assessment: MIMOpro reads as an early-2000s wireless-IP startup that filed the application, took it to grant, and then sold the asset within about a year of issuance — the classic profile of a company whose value was its patent estate rather than a shipping product. I flag that this is an inference from the assignment timeline and the absence of product evidence, not a documented bankruptcy or wind-up.

Assignment timeline

Format note: reel/frame and correspondent are not available (see caveat). Dates below are the recorded/reported event dates from the Google Patents Legal Events record. Execution dates are [NOT RETRIEVED].

• Executed [NOT RETRIEVED] / recorded 2007-06-26 — Reel [NOT RETRIEVED]/[NOT RETRIEVED]

  • Conveyance: Assignment
  • Assignor: Nissani (Nissensohn), Daniel Nathan
  • Assignee: MIMOpro Ltd.
  • Correspondent: [NOT RETRIEVED] — the interrogatory most worth pulling manually; see §3 of the NPE signals below.
  • Context: Base inventor-to-company assignment, recorded ~4 years post-filing and ~2 months pre-grant; likely confirmatory/curative in nature.

• Executed [NOT RETRIEVED] / recorded 2008-10-24 — Reel [NOT RETRIEVED]/[NOT RETRIEVED]

  • Conveyance: Assignment
  • Assignor: MIMOpro Ltd.
  • Assignee: Kromisovich Data Ltd., L.L.C.
  • Correspondent: [NOT RETRIEVED]
  • Context: Transfer-out from the original operating developer to an LLC holding entity ≈14 months after issuance — transfer-to-holding-company / portfolio disposition (sale vs. securitization not determinable from the record).

• Executed [NOT RETRIEVED] / recorded 2016-01-22 — Reel [NOT RETRIEVED]/[NOT RETRIEVED]

  • Conveyance: Merger
  • Assignor: Kromisovich Data Ltd., L.L.C.
  • Assignee: OL Security Limited Liability Company
  • Correspondent: [NOT RETRIEVED]
  • Context: Internal reorganization by merger — no price/consideration event; the LLC is absorbed into an affiliate.

• Executed [NOT RETRIEVED] / recorded 2020-05-01 — Reel [NOT RETRIEVED]/[NOT RETRIEVED]

  • Conveyance: Assignment
  • Assignor: OL Security Limited Liability Company
  • Assignee: Intellectual Ventures Assets 160 LLC
  • Correspondent: [NOT RETRIEVED]
  • Context: Internal IV-side asset reorganization into a numbered IV asset-holding LLC, positioned for divestiture.

• Executed [NOT RETRIEVED] / recorded 2020-06-05 — Reel [NOT RETRIEVED]/[NOT RETRIEVED]

  • Conveyance: Assignment
  • Assignor: Intellectual Ventures Assets 160 LLC
  • Assignee: Fleet Connect Solutions LLC
  • Correspondent: [NOT RETRIEVED]
  • Context: Transfer-to-asserter. RPX reported the May 2020 assignment of nearly 50 US assets from Intellectual Ventures to Fleet Connect Solutions LLC, "a Texas entity recently formed by the principals of monetization firm Empire IP LLC." This is the operative link that created the litigation campaign.

Note on completeness: These five events are the complete set of post-issuance conveyances appearing on the patent's Legal Events record. I cannot rule out an unrecorded mesne assignment, a security interest, or a license that was never recorded — the Assignment Center pull is required to confirm the chain is closed.

Timeline diagram

timeline
    title Ownership of US 7260153
    2003 : Application filed by MIMOpro Ltd
    2007 : Granted
         : Inventor assigned rights to MIMOpro Ltd
    2008 : Assigned to Kromisovich Data Ltd LLC
    2016 : Merged into OL Security LLC
    2020 : Assigned to Intellectual Ventures Assets 160 LLC
         : Assigned to Fleet Connect Solutions LLC
         : First assertion suits filed
    2025 : Patent term expires

NPE / troll-pattern signals

Evidence standard applied. Naming alone is not a finding. I mark a signal present only where the record shows a concrete fact: a documented NPE listing, a documented asset-sale transaction, a dated transfer immediately preceding suit, or a chained multi-hop transfer. Where I lack the reel/frame or correspondent, I say so rather than upgrading the inference.

1. Shell-entity transfer — PRESENT

The patent moves from an operating developer (MIMOpro Ltd) to a chain of three successive holding/asset LLCs, ending at a licensing plaintiff: Kromisovich Data Ltd., L.L.C. (2008-10-24) → OL Security LLC (2016-01-22, merger) → Intellectual Ventures Assets 160 LLC (2020-05-01) → Fleet Connect Solutions LLC (2020-06-05).

• Concrete support: a documented asset-sale of ~50 IV assets to Fleet Connect Solutions LLC, a Texas entity formed by Empire IP principals (RPX, Empire IP's Fleet Connect Asserts Patents Of Disparate Origins, Jan. 2021). Fleet Connect is listed as a Non-Practicing Entity in the Stanford NPE Litigation Database with the '153 patent specifically named (npe.law.stanford.edu/patent/7260153).
• Weakness in the call: the intermediate entities (Kromisovich, OL Security, IV Assets 160) are named as transfer vehicles, but I did not retrieve their formation records, registered-agent addresses, or product-less status. The "shell" characterization for the intermediate links is inferential; the terminal link (Fleet Connect) is documented.

2. Known asserter in the chain — PRESENT

Two names in the chain match your enumerated NPE list:

• Intellectual Ventures Assets 160 LLC (recorded 2020-05-01) — a numbered Intellectual Ventures asset-holding entity. Intellectual Ventures is expressly on your list.
• Fleet Connect Solutions LLC (recorded 2020-06-05) — current assignee; identified by Unified Patents as "an NPE and entity of Empire IP" and by the Stanford NPE Litigation Database as an NPE patent asserter. Unified's Patroll entry states the patent "has been asserted against Peloton, Rivian, USIC Locating Services, Aptiv, Penske Automotive, Heritage-Crystal Clean, and Southern Tire Mart."

3. Repeat correspondent across the chain — UNCLEAR (data gap, not a negative finding)

I could not retrieve the recorded correspondent for any of the five conveyances; the Assignment Center fields were unavailable in this session. I therefore cannot state whether one attorney/firm recorded multiple links.

• Do not conflate two different roles. The litigation counsel repeatedly appearing for Fleet Connect — Rozier Hardt McDonough PLLC (Carey Matthew Rozier, James F. McDonough III, Jonathan L. Hardt) — is litigation counsel of record in the district-court campaign (e.g., E.D. Tex. 2:25-cv-00046, N.D. Cal. 3:24-cv-09204, E.D. Va. 2:25-cv-00837). That is not the same as the § 3.11 assignment-recordation correspondent on the reel/frame documents. If the Assignment Center shows Rozier Hardt McDonough (or Empire IP's transactional counsel) as correspondent on the 2020-06-05 reel/frame, that would be a genuine repeat-player finding spanning the recording and litigation roles — but it must be verified, not assumed.
• Action: pull the correspondent field for reels at 2007-06-26, 2008-10-24, 2016-01-22, 2020-05-01, and 2020-06-05, and compare against the IV/OL/Empire IP recording history on this site's other tracked patents.

4. Cascading transfers — PRESENT (partially)

The 2020 leg is a genuine cascade: 2020-05-01 (OL Security → IV Assets 160 LLC) followed by 2020-06-05 (IV Assets 160 LLC → Fleet Connect Solutions LLC) — two recorded hops in ~35 days, through a numbered IV asset LLC, terminating at a freshly formed Texas plaintiff. That is a textbook "stage then drop" sequence.

• Caveat: the earlier legs are not tightly cascaded — 2008-10-24 to 2016-01-22 is ~7 years and 2016 to 2020 is ~4 years. So the cascade signal supports the 2020-IV-to-Fleet-Connect leg specifically, not the whole 2008–2020 chain.

5. Pre-litigation transfer — PRESENT

Assignment to Fleet Connect Solutions LLC recorded 2020-06-05, and Fleet Connect's first assertion suits followed in the same campaign window, including cases naming the '153 patent that appear on the patent's own litigation record (W.D. Tex. 6:20-cv-01096, later 6:21-cv-01306; RPX reported the campaign targeting Melton Truck Lines, Dayton Freight, Efficient Fleets, Gentrifi, and Southern Glazer's). By RPX's account, the suits were filed in the months after the May/June 2020 assignment — i.e., within ~6 months of the transfer.

• Significance: the chain was arranged into a clean-standing, single-purpose plaintiff immediately before assertion — the pattern that matters for standing and for the § 315(b) one-year clock on any given defendant.

6. Bankruptcy fire-sale — NOT PRESENT (no evidence located)

No Chapter 7/11 filing by MIMOpro, Kromisovich, OL Security, or IV is in the record I retrieved. The 2008-10-24 MIMOpro→Kromisovich transfer has the shape of a portfolio disposition, but I have no bankruptcy docket, § 363 sale order, or assignment-recital evidence tying it to insolvency. Call: not present on the evidence retrieved; treat the 2008 transfer's motive as undetermined.

7. Privateering — UNCLEAR

The 2020 transfer originates from Intellectual Ventures, not from a product company asserting through an NPE against its own competitors. IV is a pure monetization/aggregation entity, so the classic privateering fact pattern (operating company funds an NPE against rivals) is not established. The MIMOpro→Kromisovich leg could theoretically be a privateering arrangement if MIMOpro retained an economic interest, but no SEC filing, retainer agreement, or revenue-share evidence exists to support that. Call: unclear / not present.

8. Defensive aggregator (anti-NPE) — NOT PRESENT

The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. It terminates at Fleet Connect Solutions LLC, an active plaintiff. (Unified Patents appears in proximity only as a Patroll bounty sponsor, not as a purchaser or licensee — that does not neutralize the patent.)

Verdict

NPE — high confidence.

At least two strong signals are independently documented: (i) a known asserter in the chain — the patent passed through Intellectual Ventures Assets 160 LLC (recorded 2020-05-01) to Fleet Connect Solutions LLC (recorded 2020-06-05), the latter listed as an NPE / Empire IP entity by Unified Patents and the Stanford NPE database; and (ii) a pre-litigation transfer — Fleet Connect's assertion campaign, including '153 suits such as W.D. Tex. 6:20-cv-01096, began within months of the 2020-06-05 assignment. These are reinforced by a cascading 2020 transfer (two hops in ~35 days through a numbered IV asset LLC) and by the operating developer MIMOpro Ltd divesting the asset just ~14 months after grant with no product evidence in the record.

Caveats that could change the grade: the terminal signal is strong, but I could not retrieve reel/frame numbers or correspondents for any link, so signal #3 (repeat correspondent) is unresolved rather than negative; and the 2008 MIMOpro→Kromisovich transfer's motive (sale vs. bankruptcy vs. privateering) is undetermined. If the Assignment Center shows a single recurring correspondent across the 2008/2016/2020 reels, this would harden to a maximal-confidence NPE finding; nothing I found would downgrade it below high confidence.

Verify at: https://assignment.uspto.gov/patent/index.html (search patent number 7260153; the same records surface at https://assignmentcenter.uspto.gov/). Pull the reel/frame and correspondent fields for the five events dated 2007-06-26, 2008-10-24, 2016-01-22, 2020-05-01, and 2020-06-05.

Primary sources: Google Patents Legal Events for US 7,260,153 (https://patents.google.com/patent/US7260153/en); RPX, Empire IP's Fleet Connect Asserts Patents Of Disparate Origins (Jan. 5, 2021); Stanford NPE Litigation Database (https://npe.law.stanford.edu/patent/7260153); Unified Patents Patroll contest page for US-7260153-B2; Unified Patents insight on Fleet Connect / Empire IP. Reel/frame and correspondent fields: not retrieved — no values fabricated.

I'll build out the prior-art citation record for US 7,260,153. Let me pull the patent's own citation lists and related sources.

I have the specification's own cited-reference list. Now let me pull the face-of-patent citation tables (the examiner-cited "References Cited" list) from Justia and the official record.

Let me pull Justia's dedicated page for the '153 patent, which lists both the "Patent Citations" (backward) and "Referenced By" (forward) tables, plus the examiner of record.

Prior Art — US 7,260,153 B2 (Application 10/423,447)

Scope / method note (read first). The prior-art record I was able to reconstruct comes from two distinct citation streams, and the distinction matters legally:

1. Applicant-cited references — the "CITED REFERENCES" block that appears on the face of US 7,260,153 (and its pre-grant publication US 2004/0047426 A1). I retrieved this list in full and verbatim.
2. Examiner-cited references — the "References Cited / U.S. Patent Documents / Foreign Patent Documents / Other Publications" table on the granted front page. I was not able to retrieve this table in this session. The Google Patents text supplied to me elides it, and my searches did not surface it. Every examiner-only citation is therefore marked [NOT RETRIEVED] rather than guessed. Do not treat the list below as the complete face-of-patent citation set until the front page or the file wrapper is pulled directly.

Identifier hygiene (per your rules): I searched the literal number 7260153. Two distractors surfaced and are flagged, not auto-corrected:

• JP 7260153 B2 — a Japanese patent titled "Semiconductor device and manufacturing method thereof" (application JP2019069320A), wholly unrelated. Same trailing digits, different office, different art.
• US 8,032,153 (TracBeam) — the patent behind IPR2015‑01701/01713; unrelated to this patent. (Already flagged in the prior PTAB section.)

Note also that Unified Patents' record for this patent lists a priority date of 2002-09-08, whereas the authoritative patent text and Google Patents state 2002-09-09. Reported as found; not reconciled.

Date discrepancy (carried forward from prior sections): task date 2026-04-26 vs. system/source timestamp 2026-09-29. Recency framing only.

1. Confirmed identification of the target patent

Framing: the analysis below is a pre‑AIA § 102 analysis, consistent with the effective filing date. For each reference I give the § 102 subsection that makes it available and identify the claim(s) it could reach — with an explicit verdict on whether it anticipates (discloses every element as arranged) or is only § 103 material.

2. Table A — Applicant-cited references on the patent face ("CITED REFERENCES")

Reproduced verbatim from the specification's own citation block (confirmed on the patent face and in the pre‑grant publication):

Foreign patent documents cited on the face: [NOT RETRIEVED] for the US face page. The family/PCT counterparts (PCT/IL2003/000993; WO 2004/049592 A1; AU 2003282366 A1) are the application's own family members and are not prior art to it.

Other publications cited by the examiner: [NOT RETRIEVED].

3. Table B — Patent-document candidates with pre-2003-04-28 effective dates (from the citation ecosystem; § 102(e) / § 102(b) potential)

These are not confirmed as face-of-patent citations; they surfaced from the patent's citation ecosystem (Unified Patents' "known references" graph, Google Patents "similar documents," and forward-citation tables). I include them because they are the most plausible § 102(e) candidates and should be checked against the file wrapper:

⚠️ Heavy caveat on Table B: I verified the patent numbers, assignees, titles, and priority dates as reported in the citation graph, but I did not read these documents, and I could not confirm which (if any) were cited by the examiner. Treat Table B as a checklist to verify, not as a settled citation list.

4. Per-reference § 102 assessment against the claims

Claim-element key (independent claims):

• Claim 1: (a) define channel-matrix metric = per-sub-stream function of singular values giving a measure of cross-talk SNR; (b) obtain estimated channel matrix; (c) SVD of it; (d) compute per-sub-stream crosstalk measure from (a)+(c). No transmission required.
• Claim 21: transmit method — V′ weighting + pre-equalizer P producing modified channel Hm = H·P with improved crosstalk measures.
• Claim 28: apparatus — SVD performer + measure calculator.

[1] US 6,097,771 (Foschini — BLAST)

Discloses: multi-element antenna arrays; layered space–time architecture; multi-element antennas at TX and RX; successive-cancellation / nulling detection of co-channel sub-streams; channel estimation at the receiver.
§ 102 verdict — ANTICIPATES NO CLAIM. The reference's detection paradigm is nulling/cancellation, not SVD. It does not disclose a singular-value-based cross-talk-SNR metric (claim 1(a)) or an SVD of the estimated channel (claim 1(c)). It is available as § 103 art against claims 1, 21, 28 (motivation: same field, same problem) and against interference-cancellation-adjacent dependents. Note the specification itself cites [1] as "less relevant to the present invention."

[2] US 6,317,466 (Foschini — MIMO at both ends)

Discloses: the closest of the three US patents — a space–time architecture with multi-element antennas at both the transmitter and receiver, receiver-side channel-matrix processing, and derivation of transmit/receive weighting from the estimated channel; it is the reference the specification cites for "no Channel Matrix information required at the transmitting side … known only at the Receiving side."
§ 102 verdict — ANTICIPATES NO CLAIM IN FULL, but is the best single-reference candidate for claim 21.

• Claim 21: '466 discloses the transmit-weighting step in substance; whether it discloses the weight as the "conjugate transpose of the right-hand unitary matrix of an SVD" is the crux. The specification's own Eq. 1 states that V-weighting is conventional — but that is an admission, i.e. § 103 material, not proof that '466 discloses SVD. I could not verify that '466 uses SVD terminology → no anticipation found; strong § 103.
• Claims 1(d), 27 (post-equalizer from a training sequence) are not disclosed.

[3] Telatar (1995) — Capacity of Multi-antenna Gaussian Channels

Discloses (high confidence): SVD of the MIMO channel; singular values as per-mode gains; per-mode SNR as a function of the singular values; water-filling power allocation; capacity.
§ 102 verdict — ANTICIPATES NO CLAIM IN FULL; closest single reference to claim 1's mathematical core.

• Claim 1: Telatar supplies (c) SVD and a singular-value function yielding a per-mode SNR. But the claim requires a measure of cross-talk SNR arising from imperfect channel estimation (claim 1(a)); Telatar's per-mode SNR is thermal-noise-limited and assumes a known channel. That gap defeats anticipation. Also (a) requires the function to be defined per data sub-stream as a metric — arguably read onto Telatar's per-eigenmode SNR, but the cross-talk character is missing.
• Claims 12, 13, 14, 22 (power allocation): Telatar's water-filling is the closest art; but these are dependents of claim 11 → claim 1, so no standalone anticipation.
• Claims 16, 17, 23, 24 (drop sub-streams when SNR too low / net gain <1): these recite the inherent consequence of water-filling, which Telatar teaches — the strongest per-claim § 102/§ 103 exposure in the set, but still dependent-claim exposure.

[4] US prov. 60/401,370 (Nissani) — NOT PRIOR ART (inventor's own disclosure; same inventive entity). No § 102 effect.

[5] Proakis, Digital Communications (1995)

Discloses: classical channel estimation (MLE/MMSE/LSE), equalization, training-sequence-based estimation — the standard toolbox the specification expressly invokes for claim 1(b) and the training-based steps of claims 18/27.
§ 102 verdict — ANTICIPATES NO CLAIM. Proakis is a scalar/SISO-to-general communications text; it does not disclose SVD-based MIMO sub-stream processing or a singular-value cross-talk metric.

[6] Van Trees, Detection, Estimation and Modulation Theory (1968)

Discloses: estimation-theoretic foundations; estimator covariance, error bounds.
§ 102 verdict — ANTICIPATES NO CLAIM. Available as § 103 support for claim 1(b) and claims 6, 18, 27.

[7] Stewart, Perturbation Theory for the SVD (1990)

Discloses: how the estimated singular subspaces (Un, Vn) rotate away from the true (U, V) as a function of the perturbation norm and the singular-value gaps — i.e., exactly the mathematical relationship between singular values and the loss of orthogonality that the '153 specification relies on (its Eq. 2 cross-talk analysis).
§ 102 verdict — ANTICIPATES NO CLAIM, but it is the single most important § 103 reference for claim 1.

• Stewart is a linear-algebra document: it says nothing about MIMO wireless, data sub-streams, or cross-talk SNR. So claim 1's (a) "measure of cross-talk SNR … for each data sub-stream" and (b) "estimated channel matrix" of a wireless system are not disclosed. Anticipation fails on the face of the reference.
• But because the applicant cited it, and because the applicant's own background admits "the achieved cross-talk SNR depends on the norm of the perturbation dH … as well as on certain relationships between the singular values of H" ([7] cited), Stewart converts claim 1's metric limitation into a predictable modeling step — strong § 103, weak § 102.

[8] Duda & Hart, Pattern Classification and Scene Analysis (1973)

Discloses: discriminant functions / decision surfaces partitioning a feature space (the specification's stated source for F₁(D)=0, F₂(D)=0 and the G = ∩Gi "Good" region).
§ 102 verdict — ANTICIPATES NO CLAIM. It is a pattern-recognition text; it does not disclose MIMO singular values. Even for dependent claims 5, 7, 8, 25 (iso-valued polynomial surfaces over singular-value space), those claims are dependent on claim 1, which Duda & Hart does not disclose. § 103 material only — and note the specification concedes the polynomial fits "can be found by … standard curve fitting techniques by anyone skilled in the art."

[9] Gallager, Information Theory And Reliable Communication (1968)

Discloses: the water-pouring algorithm the specification expressly attributes to [9] for the diagonal Power Allocation Matrix A.
§ 102 verdict — ANTICIPATES NO CLAIM STANDALONE. Directly relevant to claims 12, 13, 14, 22 (power allocation; allocation proportional to singular values), but those claims are dependents of claim 11 → claim 1, so Gallager cannot anticipate them alone. § 103 material; note claim 14 ("proportional to respective channel matrix singular values") is essentially water-filling restated.

[10] US 6,351,499 (Paulraj et al.)

Discloses: multiple antennas with adaptive control using channel state information — closed-loop feedback of channel information to the transmitter and adaptive transmit weighting to maximize a communication parameter; the specification cites it (with [1], [2]) as a known no-transmit-CSI / receive-CSI scheme.
§ 102 verdict — ANTICIPATES NO CLAIM, but is the best candidate for the transmit-side adjust claims.

• Its adaptive transmit processing is close to claim 21's "weighted data stream" and to claims 12–15's power allocation. But it does not disclose SVD of the estimated channel + a cross-talk-SNR singular-value metric + a pre-equalizer producing a modified channel Hm = H·P with improved crosstalk measures. No anticipation.
• Available under § 102(b) (issued 2002-02-26, before the 2002-04-28 cutoff) and as § 103 art.

[11] Nissani internal Technical Report (March 2003) — NOT PRIOR ART (inventor's own, non-public document). No § 102 effect.

5. Bottom line

No reference on the '153 patent's citation list anticipates any claim under 35 U.S.C. § 102.

The four elements of claim 1, taken together — a singular-value-based, per-sub-stream cross-talk-SNR metric applied to an SVD of an estimated channel — are not found in any single cited reference. That is consistent with the earlier Obviousness section's conclusion: the exposure on this patent is a § 103 exposure (applicant's own admissions + Stewart + Duda & Hart + a joint TX/RX linear-precoder reference), not a § 102 exposure. Because the patent expired 2025-04-22 and no IPR/PGR was ever filed, no § 315(e) estoppel attaches to anyone, so all of this art remains usable in district court or a fresh ex parte reexam.

6. What is missing and must be pulled (highest-value next steps)

1. The granted front page / file wrapper "References Cited" table (USPTO PatentCenter, App. 10/423,447). This is the authoritative examiner-citation list and the one thing I could not retrieve. Until it is in hand, the citation set above is incomplete by construction.
2. The "Known References" tab on the Unified Patents Patroll contest page (patroll.unifiedpatents.com/contests/ux3ib2PAn5YAzeJRA) — the page exposes such a tab; its contents did not render in my search. The contest rules state that "prior art that has been previously cited during prosecution will not be considered," which means the winning submission (announced 2024‑02‑07) is by definition art not on the face page — i.e. the most valuable lead, and it is publicly downloadable.
3. Verify the Table B § 102(e) candidates — especially US 6,944,917 and US 6,983,498 (both with August‑2002 priorities before the '153 priority of 2002‑09‑09) and US 6,785,341. Confirm actual filing dates and content before relying on any § 102(e) theory.
4. Confirm § 102(b) status of US 6,351,499 against the exact earliest priority date you intend to rely on (the 2002-09-08 vs. 2002-09-09 discrepancy noted above is immaterial to this cutoff of 2001-09-09, but flagging per the rules).

Primary sources: US 7,260,153 B2 full text and citation block — https://patents.google.com/patent/US7260153/en; pre-grant publication US 2004/0047426 A1 citation block — https://patentimages.storage.googleapis.com/23/1f/6a/e11c3a1773497c/US20040047426A1.pdf; patents‑review.com record (examiner of record Chieh M. Fan; adjusted expiration 2025-04-22) — https://www.patents-review.com/a/10423447-multi-input-multi-output-wireless-communication-method-rate.html; Unified Patents Patroll contest + § 102-known-references graph — https://patroll.unifiedpatents.com/contests/ux3ib2PAn5YAzeJRA.

Caveats (restated): (i) examiner-cited references [NOT RETRIEVED] — no values fabricated; (ii) filing dates for US 6,097,771 [NOT RETRIEVED]; (iii) issue dates for the three US patents are given as reported and should be confirmed on the face page; (iv) Table B documents were not read — numbers, titles, assignees, and priority dates are as reported in the citation graph; (v) all patent numbers reported literally, with the JP 7260153 B2 and US 8,032,153 distractors flagged rather than corrected.

I'll start by pulling the prior-art record for this patent, since the "Prior Art section" referenced in the task isn't included in the material passed to me. Let me ground this in searches.

Let me pull the specific "References Cited" on the patent's face and the Unified Patents winning-art announcement.

Let me try to access the patent's own citation lists directly (Justia and Google Patents "Patent Citations" tables).

Obviousness Analysis — US 7,260,153 B2 under 35 U.S.C. § 103

⚠️ Scope flag first (required by the task's own rules). The task instructs me to "use the results from the Prior Art section of this page." No Prior Art section was passed to me. The two previously-generated sections I was given are (a) the Patent summary and (b) PTAB challenges — neither contains a prior-art list. My own tool searches did not surface the face-of-patent "References Cited" list (the Google Patents text supplied to me elides it), and I could not retrieve the content of the Unified Patents Patroll winning submission. I therefore build this analysis from three explicitly-identified tiers of art (below) rather than from a supplied Prior Art section, and I flag every place where my confidence is less than full. This section should be redone if the actual Prior Art section is available.

1. Governing framework and putatively critical dates

Two consequences follow immediately. (i) Because there is no IPR and no estoppel, an invalidity theory can be built from any pre-2002 art, including the art the Unified Patroll contest surfaced — subject to § 282 and Rule 11 duties. (ii) Because the patent is expired, any new IPR would target an expired patent; the economic case for a fresh IPR is weak, so the § 103 battleground is district court / a fresh ex parte reexam.

2. The reference set I can actually work from (three tiers)

Tier 1 — Applicant-Admitted Prior Art ("AAPA") in the '153 specification itself

This is the strongest § 103 evidence, because it is the applicant's own characterization of the field, and it supplies claim limitations outright:

• SVD-based MIMO transmission with unitary weights — the background states: "the Transmitting side (say Left) applies the unitary complex matrix V as a transmission weight … and the Receiving side (say Right) applies the unitary complex matrix U′" and gives Eq. 1 (r = U′y = DAs) as known art. → reads on claim 21's V′ weighting and claim 27's U′ reception weighting.
• Water-filling / power allocation — expressly attributed to [9] Gallager as a known "water pouring algorithm," with diagonal matrix A. → claims 12–14, 22.
• Cross-talk caused by imperfect estimation — Eq. 2 and: "the achieved cross-talk SNR depends on the norm of the perturbation dH … as well as on certain relationships between the singular values of H." → this is the core of claim 1's metric limitation, admitted as known.
• Perturbation theory as the analytical tool — attributed to [7] Stewart.
• Scalar channel-estimation methods as well studied, attributed to [5] Proakis and [6] Van Trees.
• Discriminant functions / decision surfaces in a feature space — attributed to [8] Duda & Hart.
• Frequency-domain (OFDM) MIMO processing — attributed to the applicant's own [4] (prov. 60/401,370).

Tier 2 — US patents cited in the '153 specification

Tier 3 — Contemporaneous non-patent MIMO literature (pre-2002), cited from memory with hedging

I flag these as reconstructed from domain knowledge, not retrieved; exact pin cites must be verified before filing:

• Telatar (1995/1999) — capacity of multi-antenna Gaussian channels; SVD + water-filling; cited in the patent as [3]. (High confidence.)
• Foschini & Gans (1998) — capacity limits with multiple antennas. (High confidence.)
• Raleigh & Cioffi (1998) — spatio-temporal coding, eigenmode/SVD-based transmission. (High confidence.)
• Wolniansky et al. (1998) / Golden et al. (1999) — V-BLAST detection and laboratory results. (High confidence.)
• Yang & Roy (1994) — joint TX/RX optimization for MIMO transmission. (Moderate confidence.)
• Scaglione, Stoica, Barbarossa & Li (IEEE TSP, May 2002) — "Optimal designs for space-time linear precoders and decoders", joint linear TX/RX MMSE design. (Moderate–high confidence; before the 2002-09-09 priority.)
• Sampath & Paulraj (IEEE Comm. Letters, June 2002) — linear precoding with known channel statistics. (Moderate confidence.)
• Andersen (JSAC, Nov 2000) — array gain/capacity with arrays at both ends. (Moderate confidence.)
• Bolcskei & Paulraj (WCNC 2000) — space-frequency coded broadband OFDM. (Moderate confidence.)
• (Possibly) Onggosanusi, US 2003/0210750 A1 — "Multiple input, multiple output system and method," priority 2001-05-01 → potential pre-AIA § 102(e) art. Publication is 2003-11-13 (post-priority), so it is usable only as § 102(e)/§ 103 art as of its filing date. I could not verify its filing date or content — treat with caution.

3. Claim-construction predicates before the § 103 mapping

• Claim 1 is a pure "evaluating" method — it requires no transmission. It is a diagnostic/mathematical claim: define a per-substream cross-talk-SNR function of the singular values → estimate H → SVD → compute the per-substream crosstalk measure. This is the claim Fleet Connect asserted in Pittasoft.
• Claim 21 is the transmit-side method: V′ weighting + pre-equalizer P producing a "modified channel matrix" Hm = H·P with "improved crosstalk measures" + transmit.
• Claim 28 is the apparatus counterpart of claim 1 (an "SVD performer" + "measure calculator"). Truncated in the source; claims 29–41 remain unseen — likely a receiver/post-equalizer apparatus claim.
• Critical construction point: "pre-equalizer matrix" is not defined in the specification other than as a matrix P that transforms H into Hm with better metrics. The patent's own best-mode algebra is P_L = Vn Dn⁻¹ Dm (Eq. 10a). A broad construction of P covers essentially any transmitter-side linear pre-distortion that reshapes the effective channel — which is where the obviousness exposure is greatest.

4. Proposed § 103 grounds (ground-by-ground)

Ground A — Telatar [3] + Foschini '466 [2] + Gallager [9] (primary ground; targets claims 1, 21, 28 and dependents 12–14, 19, 22)

KSR rationale: all four references are in the same field (multi-antenna wireless), address the same problem (spectral efficiency/reliability with a channel known at both ends), and the improvement is in a predictable direction (better effective channel conditioning → higher per-substream SNR).

Ground B — Ground A + Stewart [7] (targets claims 3, 4, 5, 7, 8, 25)

Stewart's Perturbation Theory for the SVD — cited by the applicant — quantifies exactly how the estimated singular subspaces Un, Vn depart from U, V as a function of the singular-value gaps. A POSITA reading Stewart immediately has the missing link between estimation error and singular values, i.e., the "predefined function of channel matrix singular values" of claim 1. This collapses claims 1, 3, and 4 into a mathematical application of a cited reference, and supplies the "threshold" basis for the G/B partition of claims 5, 7, 8, 25 when combined with Duda & Hart (below).

Key point for the analyst: the applicant cited Stewart and Duda & Hart themselves — the two references that supply the metric definition and the classification machinery. That is a strong indication the examiner did not appreciate the combination, and equally strong § 103 material if the references were not made of record in an applied rejection.

Ground C — Ground B + Duda & Hart [8] (targets claims 5, 7, 8, 25)

Duda & Hart's Pattern Classification and Scene Analysis — cited by the applicant — teaches discriminant functions / decision surfaces that partition a feature space into regions. Substituting the singular-value vector D as the feature vector is the ordinary use of the reference; the "G = ∩Gi" intersection (claim 7/8) is simply the conjunction of decision regions. The specification concedes that the polynomial fits "can be found by … standard curve fitting techniques by anyone skilled in the art."

Ground D — Ground A + Yang & Roy (1994) and/or Scaglione et al. (2002) (targets claims 9, 10, 11, 15, 26)

These references teach joint transmitter/receiver linear design of precoder + decoder under algebraic constraints.

• Claim 9 (P = Vn · diag(Dn)⁻¹ · diag(Dm)) is the closed-form solution of Hn P = Un Dm Vm′ once you have decided to prescribe the desired singular values — a routine linear-algebra step.
• Claim 10 / 26 (maximize a weighted sum of net SNR gains) is the standard constrained-optimization formulation of a linear precoder — Gallager/Cover & Thomas optimization plus Scaglione.
• Claim 15 (thresholds + minimized SNR loss) is a constrained-optimization variation, an obvious design choice.

Ground E — Ground A + Proakis [5] / Van Trees [6] (targets claims 6, 18, 27)

• Claim 6 ("determine the necessity of a pre-equalizer from the crosstalk measures") — the ordinary if bad → fix branch; obvious.
• Claims 18 / 27 (transmit a training sequence, weight the received sequence by U′, compute a per-substream ratio, and form a diagonal matrix of those ratios) — this is textbook training-based channel estimation + per-substream magnitude/phase equalization, squarely taught by Proakis [5] / Van Trees [6] (both cited) and by the pilot-based equalization idiom of OFDM/V-BLAST systems. A diagonal correction matrix derived from a known training sequence is the most predictable possible solution.

Ground F — Ground A + Nissani [4] / Bolcskei & Paulraj (WCNC 2000) (targets claims 19, 20)

• Claim 19 (time domain) — Foschini '771/'466 are time-domain.
• Claim 20 (frequency domain / OFDM) — the specification cites the applicant's own [4] for exactly this, and Bolcskei & Paulraj (2000) teach space-frequency MIMO-OFDM.

Ground G — Telatar [3] / Gallager [9] mode-dropping (targets claims 16, 17, 23, 24)

"Set P = I if SNR loss too high" / "reduce the number of sub-streams if net gain < 1" are the textbook outcome of water-filling: modes whose SNR is below the water level are simply not allocated power, and a system that detects an unfavorable mode drops it. Telatar [3] teaches exactly this. These dependent claims are therefore the most vulnerable claims in the set — they recite an inherent consequence of a cited reference.

5. Motivation to combine — the KSR/PQAI rationales to assert

1. Same field, same problem, same solution family. All Tier-1/2/3 references are multi-antenna wireless; the problem (achieving capacity/reliability with a channel known at both ends) and the tool (SVD eigenspace transmission) are common.
2. The applicant's own admissions supply half the claimed subject matter. The background section states as known: SVD, V/U weighting, AAPA Eq. 1, water-filling, the cross-talk-from-imperfect-estimation problem, and the fact that cross-talk SNR depends on singular values. That last admission is essentially claim 1's metric limitation stated as prior art.
3. The applicant cited the two references that close the gaps — Stewart (perturbation theory ↔ singular values) and Duda & Hart (decision surfaces in a feature space). A POSITA would consult exactly the references the applicant did.
4. Predictable results. The patent's own FIG. 4 shows a ~20 dB improvement at 10⁻⁶ BER over a "naïve" MIMO system ([3]), leaving a ~5 dB gap to the ideal — a quantitative, in-the-predicted-direction improvement, not a new mode of operation. KSR: "[w]hen a work is available in one field … design incentives … may prompt variations … and if a person of ordinary skill can implement a predictable variation, § 103 likely bars its patentability."
5. Finite, identified set of solutions. Given the AAPA formulation r = Un′U · D · V′Vn · A s, the artisan faces a small, enumerable menu: (a) do nothing (naïve, [3]); (b) receiver-side interference cancellation (Foschini [1]/[2] / V-BLAST); or (c) pre-distort at the transmitter to condition the effective channel. Choosing (c) is not inventive.
6. No teaching away found. None of the candidate references (as far as I can verify) criticizes transmitter-side pre-equalization.

6. Anticipated rebuttal (secondary considerations) and how it fares

7. The strongest criticisms of this analysis (diligence required)

1. I do not have the face-of-patent "References Cited" list. If Foschini '771/'466, Paulraj '499, Telatar, Stewart, Duda & Hart, Proakis, Van Trees, or Gallager were made of record and addressed in a § 103 rejection during prosecution, the ground must be rebuilt on different art (or on a different combination). Verify via the file wrapper before writing an invalidity contention.
2. The Patroll rule is dispositive as to the winning art's provenance: "PRIOR ART THAT HAS BEEN PREVIOUSLY CITED DURING PROSECUTION WILL NOT BE CONSIDERED." → the Unified Patroll winning art is, by definition, NOT face-cited art. It is the single most promising lead and I could not retrieve its content. Harvest it.
3. The pre-equalizer-at-the-transmitter concept is the true novelty. My Grounds A/D assert it is obvious over joint TX/RX linear precoder design (Yang & Roy 1994; Scaglione et al. 2002). I have not verified that these references frame the design objective in terms of a cross-talk-SNR metric tied to singular values — that specific framing is where Fleet Connect will fight. A § 103 ground is strongest if the primary reference expressly says "adapt the transmit precoder to improve the effective channel's condition/number." Verify.
4. § 102(e) art unverified. US 2003/0210750 (Onggosanusi, priority 2001-05-01) and the Qualcomm applications surfaced (Ketchum, Sampath, Walton, Medvedev, 2003-published) may be § 102(e) art as of their pre-2002-09-09 filing dates. I could not confirm any filing date. These could be valuable anticipatory art if verified.
5. Claims 29–41 are unseen (source truncates mid-claim-28). An additional independent claim — plausibly receiver/post-equalizer apparatus — may exist and needs its own ground.
6. Date discrepancy (flagged in the prior section too): the task states 2026-04-26; the authoritative source page is timestamped 2026-09-29. Recency framing only.

8. Bottom line and recommended next steps

Bottom line. Claims 1, 21, and 28 are vulnerable but not cleanly anticipated: the strongest § 103 attack pairs (i) the applicant's own admitted prior art (SVD/MIMO unitary weighting; water-filling; cross-talk-from-imperfect-estimation; cross-talk-SNR-depends-on-singular-values) with (ii) Stewart [7] (perturbation ↔ singular values) and Duda & Hart [8] (decision surfaces) — both cited by the applicant — and (iii) a joint TX/RX linear precoder reference for the P matrix. The dependent claims are substantially weaker: 12–14/22 (water-filling) and 16/17/23/24 (mode-dropping) recite inherent consequences of cited references, and 18/27 (training-sequence diagonal post-equalizer) is textbook Proakis/Van Trees.

Do next:

1. Pull the file wrapper (USPTO PatentCenter, App. 10/423,447) → the true face-of-patent prior-art list and any § 103 rejections/allowance reasons. Nothing else matters until this is in hand.
2. Harvest the Unified Patroll winning submission (portal listing ux3ib2PAn5YAzeJRA; announcement 2024-02-07). Because no IPR was instituted on any of claims 1–41, no § 315(e)(2) estoppel attaches to anyone, so this art is fully usable in district court or a fresh ex parte reexam.
3. Verify the Tier-3 citations (Yang & Roy 1994; Scaglione et al. 2002; Sampath & Paulraj 2002; Andersen 2000; Bolcskei & Paulraj 2000) and check § 102(e) filing dates for Onggosanusi US 2003/0210750 and the Qualcomm 2003-published applications.
4. Build the ground around the claim actually asserted — claim 1 in Pittasoft-type complaints and claim 28 in Xirgo-type complaints — and remember claim 1 requires no transmission, which makes the Stewart + Duda & Hart + AAPA combination unusually clean.
5. Note the posture: patent expired 2025-04-22; no PTAB win or loss to inherit; dozens of Fleet Connect cases resolved by stipulated dismissal with no merits ruling. An obviousness challenge must be built from scratch.

Primary sources used: US7260153B2 full text (https://patents.google.com/patent/US7260153/en) · Unified Patents Patroll contest + 2024-02-07 announcement (https://patroll.unifiedpatents.com/contests/ux3ib2PAn5YAzeJRA; https://www.unifiedpatents.com/insights/2024/2/7/empire-ip-entity-fleet-connect-solutions-communications-patent-prior-art-found) · Google Patents "similar documents" tables showing US7260153B2 cross-citations.

Caveat (repeating the duty-to-disclose): every reference in Tiers 2–3 is characterized from the patent's own citation list and from domain knowledge; I did not retrieve or read copies of Yang & Roy, Scaglione et al., Sampath & Paulraj, Andersen, Bolcskei & Paulraj, or Onggosanusi here. Verify each before relying on it in a filed paper.