US 10965512

I'll search for authoritative information on this specific patent number, then synthesize.

Let me check for Federal Circuit activity and the IPR outcomes specifically.

Let me do one targeted check for any Federal Circuit (CAFC) 2026 activity on this patent.

US Patent 10,965,512 — Analyst Summary

Note on the ID. I interpreted "10965512" literally as US 10,965,512 B2. Search results are consistent with this identification; no similar-numbered patent was substituted.

1. Bibliographic Data

Prosecution chain (all claiming the 2004-01-29 priority): 60/540,032 → PCT/US2005/001939 → 10/583,530 (US 8,009,660) → 13/212,116 (US 8,432,891) → 13/874,278 (US 8,934,473) → 14/595,132 (US 9,065,614) → 14/746,676 (US 9,749,168) → 15/688,441 (US 10,326,631) → 16/440,754 (abandoned) → 17/012,813 (this patent). A sibling continuation, 17/201,703, issued as US 11,388,034 B2.

⚠️ Uncertainty on assignee history: The Google Patents page lists Neo Wireless LLC as original assignee, but the earlier family members (e.g., US 8,009,660) name Neocific, Inc. The Unified Patents Patroll page describes Neo Wireless as "a Softbank entity." I could not verify the chain-of-title from a primary USPTO assignment record in these searches; treat the intermediate ownership history as unconfirmed.

2. Abstract (verbatim)

"A multi-carrier cellular wireless network (400) employs base stations (404) that transmit two different groups of pilot subcarriers: (1) cell-specific pilot subcarriers, which are used by a receiver to extract information unique to each individual cell (402), and (2) common pilots subcarriers, which are designed to possess a set of characteristics common to all the base stations (404) of the system. The design criteria and transmission formats of the cell-specific and common pilot subcarriers are specified to enable a receiver to perform different system functions. The methods and processes can be extended to other systems, such as those with multiple antennas in an individual sector and those where some subcarriers bear common network/system information."

3. Plain-Language Overview of the Independent Claims

The specification's core idea is splitting pilot subcarriers into two functional classes: cell-specific pilots (used to distinguish and estimate a particular cell's channel; interference management needed) and common pilots (identical across synchronized base stations, so energy from multiple cells combines coherently for frequency/timing estimation). Notably, the issued claims of this continuation do not use the phrase "common pilots" — they instead recite "second pilots of a second type," which is the literal claim language.

Claim 1 — Base station apparatus. An OFDMA-compatible base station (subcarriers in frequency, time slots in time) with:

• multiple antennas and a transmitter coupled to them;
• the transmitter inserts first pilots of a first type onto a first set of subcarriers, where those first pilots are cell-specific pilots;
• the transmitter inserts data and second pilots of a second type onto a second set of subcarriers;
• at least some subcarriers of either set are beam-formed;
• the antennas transmit both subcarrier sets in at least one time slot;
• the second pilot type is different from the first type, and the first pilots do not interfere with the second pilots.

Claim 8 — Base-station method. The method counterpart of claim 1 (insert first/cell-specific pilots; insert data + second-type pilots; beam-forming on at least some subcarriers; transmit both sets in at least one time slot via multiple antennas; types differ; no interference).

Claim 15 — Mobile station apparatus. An OFDMA-compatible mobile station with at least one antenna and a receiver configured to:

• receive first pilots of a first type (cell-specific pilots) on a first set of subcarriers;
• receive second pilots of a second type plus data on a second set of subcarriers, both sets received in at least one time slot;
• at least some subcarriers of either set are beam-formed;
• recover the data using channel estimates from at least the second pilots;
• recover cell-specific information using the cell-specific pilots;
• second type differs from first type; first pilots do not interfere with second pilots.

Claim 23 — Mobile-station method. The method counterpart of claim 15.

Dependent-claim themes (symmetric across all four independent claims):

• Complete non-overlap of the two subcarrier sets (claims 2, 9, 16, 24).
• Specific pilot spacings: second set includes n and n+18; first set includes m and m+20 (claims 3, 10, 17, 25) — i.e., 18- and 20-subcarrier periodicity, reminiscent of LTE-style reference-signal patterns.
• Each cell-specific pilot unique to a respective cell (4, 11, 18, 26).
• First set not frequency-aligned with another cell's cell-specific pilot subcarriers (5, 12, 21, 29).
• The second subcarrier set is beam-formed (6, 13, 20, 28).
• Cell-specific pilots convey cell-specific information (7, 14); extraction of that information (22, 30).
• First and second sets transmitted/received at the same time (19, 27).

4. Litigation, PTAB, and CAFC Status (searched)

• PTAB — IPR2022-01539 (Volkswagen Group of America v. Neo Wireless, filed 2022-09-15, patent 10,965,512; later joined by Ford, GM, Nissan, Tesla, Honda; Mercedes via IPR2023-00079 and IPR2023-00961). Institution granted 2023-05-02. Terminated 2024-02-07 after the parties' settlement/joint stipulation resolving all disputes as to the '512 patent. (Docket Alarm termination decision)
• District court — Part of MDL 3034, In re Neo Wireless, LLC Patent Litigation, No. 2:22-md-03034-TGB (E.D. Mich., Judge Terrence G. Berg). The '512 patent was asserted against numerous automakers. The Volkswagen case (2:22-cv-11404) was dismissed with prejudice on 2024-01-11 under FRCP 41(a)(2) with each side bearing its own costs. The Unified Patents Patroll contest page records the patent as "owned by Neo Wireless, LLC, a Softbank entity" with "several pending litigations and IPRs." (Patroll)
• CAFC 2026 dockets — I did not find any authoritative Federal Circuit docket entry, opinion, or appeal notice in 2026 for US 10,965,512. The Federal Circuit activity I surfaced in searches concerns other Neo Wireless patents (e.g., the '450 patent) and unrelated cases (Neonode/Apple). Per your strict rule, I will not infer a 2026 CAFC appeal exists. Two caveats: (a) my search tools returned one query that hit a step limit, and (b) the IPR relating to this patent was terminated by settlement, which typically removes it from further PTAB→CAFC appeal paths. Treat "no CAFC 2026 activity found" as a search result, not a guarantee.

5. Explicit Uncertainties

1. Assignee chain: Google Patents lists Neo Wireless LLC as original assignee; the family's earlier members name Neocific, Inc. I could not confirm the assignment record from a primary source.
2. CAFC 2026: No authoritative record found; I cannot affirmatively state the absence of a docket entry.
3. Term: Google Patents shows anticipated expiration 2025-01-20 (20 years from the 2005-01-20 PCT filing), consistent with the "Expired – Lifetime" status. This is Google's assumption, not a legal conclusion.
4. Claim terminology vs. title: The title and abstract emphasize "common pilot subcarriers," but the issued independent claims recite only "second pilots of a second type" — the claims do not literally require the second pilots to be common across cells. This gap (title/abstract vs. claim scope) is itself a notable prosecution-drafting observation and was a claim-construction battleground in the IPRs (e.g., "cell-specific pilots," "beam-formed," "at least one of the time slots").

I'll search for litigation involving US patent 10965512.

Let me gather more specific case details and outcomes.

Let me search for the broader campaign and current status of the MDL.

Litigation Involving U.S. Patent No. 10,965,512

Patent identified (interpreted literally)

The number you gave, 10965512, corresponds in the records to U.S. Patent No. 10,965,512 B2 (rendered "10,965,512" in court filings). It is not any of the co-asserted Neo Wireless patents (8,467,366 / 10,447,450 / 10,771,302 / 10,833,908 / 10,075,941), which I have kept separate below.

• Title: "Method and apparatus using cell-specific and common pilot subcarriers in multi-carrier, multi cell wireless communication networks"
• App. No.: 17/012,813; Issued: March 30, 2021; Priority: January 29, 2004
• Assignee / patent holder: Neo Wireless LLC (patents acquired from Neocific IP in Nov. 2019; Neo Wireless is described as a Fortress Investment Group–affiliated entity)
• Source: https://patents.google.com/patent/[US10965512](/patent/US10965512)/en

The '512 patent was asserted by Neo Wireless LLC as plaintiff against automakers/connected-car OEMs, in a campaign consolidated into an MDL.

District court litigation

Consolidated proceeding: In re Neo Wireless, LLC, Patent Litigation, MDL No. 3034, Case No. 2:22-md-03034-TGB, U.S. District Court for the Eastern District of Michigan (Detroit), before Judge Terrence G. Berg. All member cases below were transferred into this MDL (June–August 2022).
(Source: https://www.docketalarm.com/cases/Michigan_Eastern_District_Court/2--22-md-03034/NEO_Wireless_LLC_Patent_Litigation/docs/197.pdf)

Sources: Google Patents litigation links (https://patents.google.com/patent/US10965512/en); Docket Alarm MDL docket (doc. 197) listing all member cases; Law360 (https://www.law360.com/cases/62b487855cdd0d02dec0f43a/); USPTO AO-120 forms at ptacts.uspto.gov and docketalarm.com; IPWatchdog docket roundups (Apr. 7, 2022 and Jul. 8, 2022).

Outcome detail

• Volkswagen (2:22-cv-11404): The parties jointly moved to dismiss and the court dismissed all claims/counterclaims with prejudice on Jan. 11, 2024, each side bearing its own costs. This followed a Dec. 2023 joint request to stay pending settlement.
• Mercedes-Benz (2:22-cv-11769): Resolved by agreement; the court granted a joint Rule 41(a)(2) motion and dismissed with prejudice (each party bears its own costs). Law360 reported Mercedes "was dropped from the case in June" 2023.
• Remaining defendants (Ford, Honda, Nissan, Toyota, GM, Tesla, FCA): I found no public final judgment on the '512 patent in these cases. They remained MDL member cases as of the most recent docket records I could retrieve. (Note: the related '512 IPR was terminated by joint motion of Ford, GM, Nissan, Tesla, Honda, and Neo Wireless on Feb. 7, 2024 following settlement — a strong signal of resolution, but I cannot confirm the corresponding district-court dispositions from my sources.)

Related PTAB (IPR) proceedings on the '512 patent

These are administrative validity challenges, not district-court litigation, but they are the "case" history tied to '512:

Sources: Google Patents "litigation" panel for US10965512; Docket Alarm (https://www.docketalarm.com/cases/PTAB/IPR2022-01539/...02-07-2024-Board/Termination_Decision__Post_DI_Settlement-51...); Unified Patents PTAB portal links.

Summary of known litigation

Yes — litigation is known. Every identified suit was brought by Neo Wireless LLC (plaintiff) against an automaker, and all cases were consolidated into MDL 3034 in the Eastern District of Michigan (Judge Terrence G. Berg). Nine district-court actions were identified in which '512 was or was likely asserted:

• Volkswagen (2:22-cv-11404) — settled/dismissed with prejudice (Jan. 11, 2024)
• Mercedes-Benz USA (2:22-cv-11769) — settled/dismissed with prejudice (2023)
• Ford (2:22-cv-11402), Honda (2:22-cv-11403), Nissan (2:22-cv-11405), Toyota (2:22-cv-11406), GM (2:22-cv-11407), Tesla (2:22-cv-11408), FCA/Stellantis (2:22-cv-11770) — no confirmed final judgment; '512 IPRs terminated by settlement.

Caveats and confidence

1. No CAFC appeal found on the '512 patent itself. The Federal Circuit activity I located (appeals filed Jan. 2025) concerns Neo Wireless's '450 patent, not the '512 patent.
2. Tesla ('512) is uncertain. The USPTO AO-120 for Tesla's original E.D. Tex. case (2:22-cv-00095) lists five asserted patents without the '512 patent. I could not confirm whether '512 was later added to the Tesla MDL case, so treat item #9 as unverified.
3. The Stanford NPE Litigation Database (https://npe.law.stanford.edu/patent/10965512) lists only two cases for this patent — GM (2:22-cv-00094) and Mercedes-Benz (2:22-cv-11769) — which appears incomplete relative to the AO-120 forms and IPWatchdog docket listings. I have prioritized the court filings.
4. Ongoing status as of April 2026 for the non-settled automaker cases could not be confirmed from the sources retrieved; the last records I have are from late 2023 through 2025. I am not asserting those cases are closed.
5. Some of the above (e.g., PatSnap case narratives) are secondary sources; the primary support is the PACER/docket and USPTO AO-120 records cited.

If you need confirmation of the current (2026) disposition of the Ford/Honda/Nissan/Toyota/GM/Tesla/FCA MDL member cases and any CAFC appeals, that will require a direct PACER docket pull for each of the case numbers above.

I'll reconcile the structured "no PTAB activity" block against the litigation data on the patent page, then verify each proceeding with targeted searches.

I have the four proceedings. Let me verify no FWD issued, check for any CAFC appeal, and confirm the joinder panel details.

⚠️ Flagged contradiction before the analysis

The structured "PTAB proceedings on file" block in this prompt states: "The USPTO ODP API returns no AIA trial proceedings for this patent as of the most recent ingest… the default should be 'no PTAB activity on file.'"

That is wrong for this patent. Live search evidence — Board decisions, PTACTS artifacts, PTAB docket pages, and the district court record — establishes four AIA trial proceedings on US 10,965,512. I am overriding the structured block per the operating rule that live results govern, and I am flagging the discrepancy rather than silently burying it. (This also means the ODP ingest is likely missing the entire VW/Mercedes/Ford/GM-Nissan-Tesla-Honda automotive campaign on this family.)

Second flag: the previously generated "Patent summary" section is materially incomplete on the PTAB question. It listed only IPR2022-01539, and it noted the four "Settlement" flags on the Google Patents page as litigation metadata without enumerating them. The correct count is four proceedings across seven petitioner groups, all terminated post-institution by settlement. Its statement that no Final Written Decision exists is correct and is the single most important fact below.

Proceedings overview

Four AIA trials — all Inter Partes Reviews — all instituted, all terminated by settlement before any Final Written Decision. Breakdown: 0 active, 0 claims invalidated, 0 claims sustained, 4 settled, 0 institution denied. Bottom line for a defendant: the '512 patent has never been adjudicated on the merits by the PTAB. There is no FWD to quote, no canceled claim to point at, and no § 315(e) estoppel attached to anyone. This is not "hardened" and it is not "dead" — it is untested, and the only reason it is untested is that it expired on 2025-01-20, which killed the defendants' incentive to keep litigating. A demand letter citing the '512 patent today is a past-damages-only play capped by the 35 U.S.C. § 286 six-year lookback.

IPR2022-01539 — Volkswagen Group of America, Inc. (later joined by Mercedes-Benz USA, Ford, GM, Nissan, Tesla, Honda) v. Neo Wireless LLC

• Type: Inter Partes Review
• Filed: 2022-09-15
• Status: Terminated-Settled (verbatim docket status). This was the anchor proceeding; the other three were joinders into it.
• Judge panel: Hyun J. Jung (writing), Jo-Anne M. Kokoski, Matthew S. Meyers. (Note: Patexia's summary lists Boudreau in place of Kokoski for this case; the termination decision itself names Jung/Kokoski/Meyers, so I use that.)
• Petition grounds: Challenged claims 1–30 (all claims), pre-AIA § 103(a) obviousness, in two grounds. Ground 1 turned on "cell-specific pilots" and "beam-formed"; Ground 2 turned on "cell-specific pilots" and "in at least one of the time slots." Art relied on included Ketchum (EX1006, filed 2003-06-30, pub. 2004-09-16), Kim (EX1004, filed 2002-11-29, pub. 2004-06-10), Li (US 2002/0163879), Ma, Hara, and Van Nee. Supported by the declaration of Dr. Paul Min (EX1003).
• Institution decision: Instituted on all claims and all grounds on 2023-05-02 (Paper 7). The Board resolved claim construction against the Patent Owner on "at least one of the time slots," holding that "the plain language of the claims does not require transmitting the subcarriers in the same time slot" and rejecting Neo's proposed narrowing to "at least one of the same time slots." The Panel also rejected Neo's arguments that the art failed to show "cell-specific pilots" and "beam-formed" at the institution stage. Institution on all grounds followed SAS Institute v. Iancu.
• Final Written Decision: None. No FWD ever issued on any claim. Statutory FWD deadline would have been ~2024-05-02; the case terminated short of it.
• Settlement / termination: Staged. VWGoA terminated individually on 2024-01-09 (Paper 49, joint motion granted). The remaining petitioners — Ford, GM, Nissan, Tesla, Honda — then jointly moved to terminate, and the Board terminated the entire proceeding on 2024-02-07 (Paper 51, Termination Due to Settlement After Institution of Trial, 35 U.S.C. § 317; 37 C.F.R. § 42.74). The Board found the joint stipulation "resolves all pending disputes" as to the '512 patent. The settlement agreement (EX1055 / EX2100) is confidential and sealed "Filing Party and Board" only.
• Appeal: None. No FWD = no appealable judgment. No CAFC docket was found. Termination by settlement removed any PTAB→CAFC path.
• Defensive value: Low as a merits precedent, high as an estoppel-free zone. Because no FWD issued, § 315(e)(2) estoppel does not bar these petitioners from re-raising these grounds in a later civil action — but they contractually gave that up: the parties stipulated they would "not initiate or participate in any other invalidity challenges" to the '512 patent. For a new defendant, this proceeding is essentially a free look at a full, well-funded validity attack that was never judged — the petition is a usable prior-art roadmap, and nothing in it has been rejected on the merits.

Links: Termination Decision, Paper 51 · Patexia case summary

IPR2023-00961 — General Motors LLC, Nissan North America, Inc., Tesla, Inc., and American Honda Motor Co., Inc. v. Neo Wireless LLC

• Type: Inter Partes Review (petition + motion for joinder into IPR2022-01539)
• Filed: 2023-06-02
• Status: Terminated-Settled (terminated 2024-02-07, Paper 10)
• Judge panel: Hyun J. Jung, Stephen E. Belisle, Jo-Anne M. Kokoski (per Patexia; the joinder/institution ruling was entered as Paper 29 in IPR2022-01539)
• Petition grounds: Claims 1–30, pre-AIA § 103(a), substantively identical grounds to the Volkswagen petition. Petitioner expressly told the Board its petition tracked the anchor case.
• Institution decision: Instituted and joined on 2023-09-21. The Board granted joinder, ordered that "the grounds on which trial in IPR2022-01539 were instituted are unchanged, and no other grounds are added," and bound petitioners to the existing 1539 scheduling order.
• Final Written Decision: None. Terminated ~4.5 months into trial, well before the FWD deadline.
• Settlement / termination: Terminated 2024-02-07 as part of the global '512 settlement. The district court Joint Stipulation (public, Dkt. 219 in 2:22-md-03034) recites that these defendants "stipulate and agree to promptly contact the PTAB… and to request, within seven (7) days… that the Patent Trial and Appeal Board terminate their respective IPR proceedings on the '512 Patent."
• Appeal: None.
• Defensive value: Same as 1539 — no estoppel, no merits ruling. But this is the proceeding that produced the published covenant not to sue: Neo Wireless covenanted that neither it nor its subsidiaries/successors-in-interest "will sue any of the Defendants or their current affiliates for infringement of the '512 Patent, or any of the following foreign counterparts: EP1712089 B1, KR100808462 B1, CN1879426 B, or CN101854188 B." If your client is a current affiliate of GM, Nissan, Tesla, or Honda, that covenant is worth reading before you respond to anything.

Links: IPR2023-00961 case details · Patexia docket summary · District court Joint Stipulation (Dkt. 219)

IPR2023-00764 — Ford Motor Company v. Neo Wireless LLC

• Type: Inter Partes Review (petition + conditional motion for joinder into IPR2022-01539)
• Filed: 2023-03-28
• Status: Terminated-Settled (terminated with the rest of the family on 2024-02-07)
• Judge panel: Not confirmed in the sources I retrieved (the joinder decision was entered as Paper 15 in IPR2022-01539). I will not guess the panel.
• Petition grounds: Claims 1–30, same grounds/art as Volkswagen. The petition is expressly a "verbatim copy of Volkswagen's Petition" apart from mandatory notices and the Fintiv discussion; Ford's expert, Dr. Todor Cooklev (EX1035), filed a declaration adopting Dr. Min's opinions paragraph-for-paragraph. Ford took an "understudy role" and said it would rely on its own expert only if VW and Mercedes exited the proceeding first.
• Institution decision: Instituted and joined. The Board granted Ford's motion, joined it as a petitioner in IPR2022-01539, changed the caption, and directed all further filings into 1539. (The public decision text I retrieved did not carry an entry date; treat the exact date as unconfirmed.)
• Final Written Decision: None.
• Settlement / termination: 2024-02-07 by joint motion under 35 U.S.C. § 317.
• Appeal: None.
• Defensive value: A pure copycat joinder — Ford added no independent art or argument. Its only value to a defendant today is as a cautionary note on joinder strategy: the Board granted it, then the whole thing died at settlement, meaning Ford paid for an IPR that produced nothing except (by its own stipulation) a contractual bar on further challenges. This is also the fact pattern that drove the Board's General Plastic/Valve "significant relationship" fight against Neo in the sibling IPR2023-00763 ('450 patent), where the Board initially denied institution — a decision the Director later vacated as an improper extension of the significant-relationship doctrine.

Link: Ford joinder/institution decision (PTACTS)

IPR2023-00079 — Mercedes-Benz USA, LLC v. Neo Wireless LLC

• Type: Inter Partes Review (petition + motion for joinder into IPR2022-01539)
• Filed: 2022-10-19
• Status: Terminated-Settled (terminated 2023-08-15, Paper 13)
• Judge panel: Hyun J. Jung, Jo-Anne M. Kokoski, Matthew S. Meyers, Charles J. Boudreau
• Petition grounds: Claims 1–30, pre-AIA § 103(a), two grounds. Neo's Preliminary Response (2023-02-15) framed the fight as: Ground 1 fails on "cell-specific pilots" and "beam-formed"; Ground 2 fails on "cell-specific pilots" (Ketchum alone, and Ketchum + Li) and on "in at least one of the time slots." Mercedes' expert was Mr. McNair.
• Institution decision: Instituted and joined on 2023-05-05 (Paper 11 — Decision Granting Institution and Granting Motion for Joinder). The panel incorporated the Volkswagen institution analysis wholesale and instituted on all claims and all grounds, rejecting Neo's argument that institution was improper. Neo itself conceded in its papers that the petition "is of the sort that has been called a 'copycat' petition."
• Final Written Decision: None.
• Settlement / termination: 2023-08-15 — this was the first '512 settlement. The Board granted termination as to Mercedes-Benz only and ordered the settlement agreement (Ex. 2100) treated as business confidential, available only to federal agencies or on a showing of good cause. IPR2022-01539 remained pending as to Volkswagen and Ford.
• Appeal: None.
• Defensive value: Mercedes bought out early and got the cleanest exit. Because the proceeding was a joinder, the institution decision here is textually the Volkswagen analysis, so cite it and 1539 together, not separately. Again: no estoppel, because no FWD.

Link: Termination Decision, Paper 13 · Mercedes Preliminary Response

Strategic summary

Claim-level status of US 10,965,512. Independent claims 1, 8, 15, 23 and dependent claims 2–7, 9–14, 16–22, 24–30 are all UNTESTED. Nothing is canceled; nothing is sustained. All 30 claims were challenged in four separate IPRs, all four were instituted on all claims and all grounds, and none of the four reached a Final Written Decision. Stated plainly: the Board found a reasonable likelihood that every claim is unpatentable, four times over, and then never decided the question. The two claim-construction rulings that do exist — (i) "at least one of the time slots" does not require the same time slot, and (ii) the Board's preliminary rejection of Neo's attempt to read "cell-specific pilots" and "beam-formed" narrowly enough to escape the art — both cut against the patent owner. Those are institution-stage constructions, not final ones, and they carry no preclusive weight, but they are the best available signal of how a panel reads these terms.

Estoppel landscape. No FWD means no § 315(e)(1) or § 315(e)(2) estoppel attached to any petitioner, on any ground, in any forum. The prior-art grounds from the VW petition (Ketchum, Kim, Li, Ma, Hara, Van Nee, in the two Ground-1/Ground-2 combinations) remain fully available to any party that did not settle — and to those who did, subject only to their contractual covenant (the settling defendants agreed not to "initiate or participate in any other invalidity challenges" to the '512 patent). For a defendant currently facing assertion: § 315(b) is your real constraint — one year from service of a complaint alleging infringement of the '512 patent — not any estoppel from these proceedings. There is also no § 325(e) PGR estoppel, as no PGR was filed.

Pattern signals. (1) This was a coordinated automotive-OEM attack, not a lone petitioner: VW filed first, Mercedes and Ford filed copycat petitions and joined, then GM/Nissan/Tesla/Honda filed a fifth copycat and joined — the Board itself noted Ford's petition was a "verbatim copy" and that Mercedes' was "nearly identical." (2) No defensive aggregator is visible in the '512 chain — the named petitioners are all operating companies and their litigation counsel (Sterne Kessler/Bush Seyferth for VW and Ford; Latham & Watkins for Ford's joinder; Caldwell Cassady Curry for GM et al.; Lowenstein & Weatherwax for Neo Wireless). Unified Patents' Patroll listed the patent but the IPRs here were filed by OEMs, not by Unified. (3) The patent owner litigated aggressively and won the fights that mattered elsewhere: the Board denied institution on Neo's sibling '450, '366, and '771,302 patents, and Neo defeated the '450 IPRs on the merits (Nov. 2024 and Sept. 2025 FWDs) — both now on appeal at the CAFC as of Jan. 2025. Those are different patents; treat them as background on Neo's litigation posture, not as '512 precedent. (4) No CAFC appeal exists for the '512 patent — there is no appealable PTAB judgment to appeal.

One open thread worth pulling. The MDL defendant list also included Toyota and FCA US, and the public '512 dismissal stipulation is signed by GM, Ford, Nissan, Tesla, and Honda (plus earlier VW and Mercedes settlements). I did not confirm whether Toyota and FCA settled the '512 assertions, were dismissed, or remain in the case. Do not assume the covenant not to sue extends to them.

Recommended next steps

1. Lead with expiration. The '512 patent expired on 2025-01-20 (Google Patents: "Expired – Lifetime"; the district court stipulation itself recites "the '512 Patent is anticipated to expire on January 20, 2025"). Any exposure is past damages only, limited to the 35 U.S.C. § 286 six-year window (roughly since 2020-09-25) and ending 2025-01-20. Neo Wireless will have to prove pre-expiration infringement; there is no injunction and no ongoing royalty.

2. Do not represent that any claim is invalidated. There is no FWD on this patent. If someone tells you "the PTAB killed the '512 claims," that is false — I searched the PTAB record, the docket aggregators, and the district court record. Cite the four termination orders instead: IPR2022-01539 Paper 51 (2024-02-07) and IPR2023-00079 Paper 13 (2023-08-15). Both say on their face: "This Order does not constitute a final written decision pursuant to 35 U.S.C. § 318(a)."

3. No pending AIA trials exist to track. All four are terminated; there is no institution deadline, no oral hearing, and no FWD due date to calendar. Any new trial would have to be a fresh, properly-filed petition, subject to § 315(b) and Fintiv/General Plastic discretion — and the Board has shown willingness to deny on § 314(a)/Fintiv when the patent is at or near expiry. Expired patents can be IPR'd, but the practical value is limited to foreclosing past damages.

4. Check the covenant before you draft a response. Pull Dkt. 219 and its footnote on "current affiliates." If your client is an affiliate of GM, Ford, Nissan, Tesla, or Honda, Neo Wireless has already contracted away the '512 patent against you — including the foreign counterparts EP1712089 B1, KR100808462 B1, CN1879426 B, and CN101854188 B.

5. Use the VW petition as a free validity roadmap. The full Volkswagen petition and Dr. Paul Min's declaration are public and were never adjudicated. They are the best available starting point for a § 282 invalidity case in litigation, where the same art can be run without PTAB estoppel and without the Board's claim-construction constraints.

6. Uncertainties I am not papering over. (a) The exact entry date of the Ford/I-764 institution-and-joinder decision is not confirmed in the sources retrieved. (b) I found no CAFC appeal for the '512 patent, and given that no FWD exists, none should be expected — but treat that as a search result, not an affirmative guarantee. (c) Toyota's and FCA's '512 disposition is unconfirmed. (d) The settlement agreements are sealed; do not assume the financial terms are public — they are not.

I'll research the assignment record for this patent using authoritative sources.

Let me dig deeper into the specific chain of title and the Neo Wireless / Fortress structure.

Assignment Record Analysis — US 10,965,512 B2

Method note / provenance caveat (read first). The Assignment Center's per-patent record for the '512 patent itself was not directly retrievable in this session (one search hit the step limit; the returned hits were prosecution-file and court-document mirrors rather than the Assignment Center UI). The reel/frame data below is reconstructed from the USPTO-recorded chain-of-title as recited in (a) 37 CFR 3.73(b)/(c) "Statement of Assignee" filings in the family's prosecution files (quoted in PTAB exhibits) and (b) district-court infringement complaints that recite the recorded chain. Where the '512 patent's own record differs, I flag it. Verify at the USPTO Assignment Center: https://assignmentcenter.uspto.gov/ (search "10965512") or https://assignment.uspto.gov/patent/index.html.

Inventors

Residences are taken verbatim from the recorded inventor-to-assignee assignment (quoted in a USPTO filing: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1519096](/patent/1519096)/download-documents — the same four inventors and the same Bellevue/Kirkland addresses appear across the Neocific/Waltical portfolio).

Unusual-pattern check — inventors departing within 12 months: Not present. The founder-inventors did not walk away. Neo Wireless's own Rule 26 initial disclosures (E.D. Mich. MDL 2:22-md-03034) identify Dr. Xiaodong (Alex) Li as "co-founder of Neocific, Inc. and former governor of Waltical Solutions, Inc." and Dr. Titus Lo as a "founding employee of Neocific, Inc. and former agent," both reachable only through Neo Wireless's litigation counsel (Caldwell Cassady & Curry). Li is further identified as Co-CEO of Neo Wireless. The inventorship therefore migrated with the asset rather than decoupling from it — the opposite of the fire-sale precursor pattern. (Source: https://www.docketalarm.com/cases/Michigan_Eastern_District_Court/2--22-md-03034/NEO_Wireless_LLC_Patent_Litigation/docs/98/3.pdf)

Original assignee

Two distinct answers, because the "issued-patent assignee" and the "original-assignee-of-the-invention" are different entities:

• Entity named on the issued '512 patent: Neo Wireless LLC (Google Patents lists it as both original and current assignee). Neo Wireless LLC is a non-practicing licensing entity — it ships no products, and its own outbound licensing letter describes the portfolio as acquired and licensed for revenue ("Neo's portfolio has been licensed by multiple handset manufacturers, for approximately $65 M USD"). Its recorded address of use abroad is 123 West Wayne Avenue, Floor 1, Wayne, PA 19087 (from a Belgian patent-register entry). Neo Wireless is publicly characterized as a Fortress Investment Group / SoftBank-affiliated assertion vehicle: Unified Patents describes the patent as "owned by Neo Wireless, LLC, a Softbank entity" (https://patroll.unifiedpatents.com/contests/R7igaMTfN8bgXQbaq), and IPWatchdog tags the plaintiff as "Neo Wireless LLC [Softbank Group, Fortress IP]." RPX states the portfolio "was acquired by Neo USA (under its prior name CFIP NCF LLC) in November 2019 from Neocific IP" (https://litigation.rpxcorp.com/news/81426-...). Status: operating as a licensing/assertion entity; no products; not dissolved.

• Original assignee of the invention (family-level): Waltical Solutions, Inc. (formerly Walbell Technologies, Inc.), a Washington corporation at Suite D159, 1750 112th Ave. NE, Bellevue, WA 98004, which took the inventors' rights and then conveyed them to Neocific, Inc. Neocific was the operating startup — a Bellevue, WA OFDMA/wireless R&D company founded by Drs. Li and Lo with Wi-Max development experience. Status: no longer an operating product company; it sold the portfolio in Nov. 2019. I did not find a primary-source bankruptcy or dissolution record, so I characterize Neocific as "ceased operating / portfolio divested," not "bankrupt."

Assignment timeline

All four transactions below are the recorded chain common to the Neo Wireless patents asserted alongside the '512 patent; the JPML MDL transfer order expressly finds that the asserted patents "share common inventors and a common assignment history from Neocific, Inc." (In re Neo Wireless, MDL 3034, 2022-06-14).

• 2005-04-08 (executed) / recorded in Washington-law assignment — Reel 051258 / 0162

  • Conveyance: Assignment
  • Assignor: Xiaodong Li, Titus Lo, Haiming Huang, Kemin Li
  • Assignee: Waltical Solutions, Inc. (f/k/a Walbell Technologies, Inc.)
  • Correspondent: not captured in the source document; the assignment was executed contemporaneously with the application and authorizes insertion of the application number.
  • Context: initial inventor-to-company assignment of the founding patent application.

• 2005-12-14 (executed) — Reel 051258 / 0850

  • Conveyance: Assignment
  • Assignor: Waltical Solutions, Inc.
  • Assignee: Neocific, Inc.
  • Correspondent: not captured in the source document.
  • Context: internal reorganization — moving the portfolio from the Waltical holding corporation into the operating company Neocific.
  • Note: a parallel record, Reel 051256 / 0932, reflects a direct INVENTORS → NEOCIFIC, INC. assignment used for a different family member (the '450/14-720,554 line).

• 2019-11-22 (executed) — Reel 051116 / 0396

  • Conveyance: Assignment
  • Assignor: Neocific, Inc.
  • Assignee: CFIP NCF LLC
  • Correspondent: not captured — ⚠️ the Assignment Center "correspondent of record" field for this recording was not retrievable in this session; do not treat any name below as the assignment correspondent.
  • Context: transfer to a licensing/financial entity — the divestiture of the whole Neocific portfolio to the Fortress-backed acquirer. This is the pivotal event.

• 2020-01-23 (executed) — Reel 051971 / 0226

  • Conveyance: Assignment (court complaints describe it as an assignment; RPX characterizes CFIP NCF LLC as Neo Wireless's "prior name", which could instead indicate a Change of Name)
  • Assignor: CFIP NCF LLC
  • Assignee: Neo Wireless LLC
  • Correspondent: not captured (see warning above).
  • Context: transfer-to-asserter / intra-group renaming — completion of the acquisition into the named plaintiff vehicle.

• No separate post-issuance assignment recorded for US 10,965,512 itself. The '512 application (17/012,813) was filed 2020-09-04, i.e. after the Jan. 2020 completion of the portfolio acquisition. Google Patents lists Neo Wireless LLC as original assignee, consistent with the '512 patent having been filed by Neo Wireless as a continuation, inheriting the family chain rather than receiving its own recorded transfer. If the Assignment Center shows only the family-level records above (or none for '512 itself), that is expected and is itself the finding.

Recurring attorney/firm tell (prosecution-side, not assignment-correspondent): The 37 CFR 3.73(b)/(c) ownership showings and the large-entity status declarations for Neo Wireless in these files were signed by Christina R. Walsh (Reg. No. 61553) and John C. Donch Jr. of Volpe and Koenig, P.C., Philadelphia, PA (Customer No. 3624), with Stephen Bishop (Reg. No. 28320) also appearing on ADS filings. ⚠️ This is a recurrence finding (same firm on multiple family prosecution/ownership filings), but it is not evidence of the assignment-record correspondent; I could not confirm the recorded correspondent field for any reel above.

Timeline diagram

timeline
    title Ownership of US 10965512
    2004 : Provisional filed by four inventors
    2005 : Inventors assign to Waltical Solutions
         : Waltical assigns to Neocific Inc
    2019 : Neocific divests portfolio to CFIP NCF LLC
    2020 : CFIP NCF LLC transfers to Neo Wireless LLC
         : Continuation filed by Neo Wireless
    2021 : Patent issued to Neo Wireless
    2022 : Asserted against major automakers

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT. The operating startup (Neocific) transferred the portfolio to a licensing-only LLC sequence with no products: Neocific → CFIP NCF LLC (Reel 051116/0396, executed 2019-11-22) → Neo Wireless LLC (Reel 051971/0226, 2020-01-23). Neo Wireless ships no products (self-described portfolio licensor), and the recorded corporate address of use is a Wayne, PA office suite. The "CFIP" prefix and the Fortress/SoftBank parentage are corroborated by Unified Patents and RPX. Caveat: the LLCs are not anonymous shell names — they are publicly tied to a named parent — but the "operating → licensing-only" movement is concretely evidenced.

2. Known asserter in the chain — PRESENT. Current assignee Neo Wireless LLC is a recognized assertion entity: it appears in Stanford's NPE Litigation Database as a "1 — Acquired patents" asserter (https://npe.law.stanford.edu/patent/10965512), is tracked by Unified Patents' Patroll, and is identified by RPX and IPWatchdog as a Fortress/SoftBank vehicle. It is not on the legacy lists (Acacia, Marathon, IV, Wi-LAN, etc.), but it is squarely a high-frequency plaintiff.

3. Repeat correspondent across the chain — UNVERIFIED / partial. The recorded assignment-correspondent field could not be retrieved for Reels 051258/0162, 051258/0850, 051116/0396, or 051971/0226, so I cannot certify a repeat assignment correspondent. On the prosecution/ownership side, Volpe and Koenig, P.C. (Christina R. Walsh, Reg. 61553) recurs across the family's ownership filings — a genuine recurrence, but not the same field the task asks for. Marked unclear pending Assignment Center retrieval.

4. Cascading transfers — PRESENT. Two consecutive transfers within ~62 days: Neocific → CFIP NCF LLC (2019-11-22) → Neo Wireless LLC (2020-01-23). Even accounting for the possible change-of-name interpretation of the second hop (RPX's "prior name" phrasing), the Nov. 2019 step and the Jan. 2020 step together form a rapid two-hop chain into the assertion vehicle.

5. Pre-litigation transfer — NOT PRESENT (strict 6-month test). The transfers completed Jan. 23, 2020. The '512 patent did not issue until 2021-03-30 and was first asserted in the 2022 waves (automaker suits filed March and June–August 2022). The transfer predates the first '512 suit by roughly two years, so it does not meet the "within 6 months before suit" test. Indirectly, the acquisition was arranged to enable the licensing campaign (the first handset suits followed in early 2021), but the timing is outside the strict window. (Note: the 2020 acquisition did precede the Nov. 2020/Jan. 2021 handset suits — e.g., Neo Wireless v. LG, W.D. Tex. 6:21-cv-00025; Neo Wireless v. Apple, 6:21-cv-00026 — but those campaigns did not assert the not-yet-issued '512 patent.)

6. Bankruptcy fire-sale — NOT PRESENT. No Chapter 7/11 record was found for Neocific or Waltical; the portfolio moved by negotiated sale to a financial acquirer, not through a bankruptcy estate. (Contrast the Kodak/Nortel/Polaroid model.)

7. Privateering — UNCLEAR / weak. The recurring fact pattern (a Fortress-managed fund acquiring a small defunct-ish operating company's standard-essential portfolio and asserting against automakers/handset makers) resembles privateering, but classic privateering requires the originating operating company to remain in business and benefit from the NPE's assertions. Neocific appears to have ceased operating, and the assertion campaign serves the acquirer, not an operating competitor. I do not score this as a confirmed privateering signal.

8. Defensive aggregator — NOT PRESENT (inverse). The chain terminates at Neo Wireless LLC, an active plaintiff; it does not end at RPX, AST, LOT, Unified, or OIN. Note the inverse context: Unified Patents is running a prior-art contest against this patent (Patroll R7igaMTfN8bgXQbaq), i.e., a defensive aggregator is on the opposing side — the mirror image of a defensive termination.

Verdict

NPE — high confidence.

Two or more strong signals are concretely evidenced. (1) Shell/operating-to-licensing transfer: recorded Reel 051116/0396 (Neocific → CFIP NCF LLC, executed 2019-11-22) and Reel 051971/0226 (CFIP NCF LLC → Neo Wireless LLC, executed 2020-01-23) moved the patent from the Bellevue, WA operating startup to a no-products licensing entity. (2) Known asserter: current assignee Neo Wireless LLC is documented by Unified Patents as "a Softbank entity," by RPX/IPWatchdog as Fortress-backed, and by Stanford's NPE database as an "Acquired patents" plaintiff, and it has run an extensive multi-district campaign (MDL 3034) against handset makers and automakers. The only counter-indication — the absence of a strict pre-litigation-transfer (within 6 months) and the fact that the founder-inventors stayed with the asset — does not outweigh the transfer-to-licensing-vehicle and known-asserter signals.

Verify at: https://assignmentcenter.uspto.gov/ (search patent 10,965,512).

I'll verify the patent against USPTO/authoritative sources and dig into the specific cited prior art. Let me run parallel searches.

Strong results — the IPR petitions reveal the genuinely operative prior art. Let me pin down details on the key references.

Prior-Art Analysis — US 10,965,512 B2

Scope note (applies to everything below). US 10,965,512 B2 is a continuation claiming priority to 2004‑01‑29 (provisional 60/540,032 → PCT/US2005/001939). The claims are therefore governed by pre‑AIA 35 U.S.C. §§ 102/103, and a reference is only § 102(a)/(b) art if it published (or issued) before 2004‑01‑29; a U.S. patent/publication can additionally be § 102(e) art if its effective U.S. filing date predates 2004‑01‑29, even though it published later. I flag date problems explicitly where they arise, because a large fraction of the references printed on the '512 face are not available as art against the 2004 priority date.

Two distinct citation sets must be kept separate:

1. The '512 face citations ("Citations (69)" / "Patent Citations (87)" on Google Patents) — art the examiner/third parties listed. Most of these are background or distinguishing art; several post‑date the priority date.
2. The operative art actually used to attack the claims — the references in the PTAB petitions (IPR2022‑01539 VW; IPR2023‑00079 Mercedes; IPR2023‑00764 Ford/GM/Nissan; IPR2023‑00961 GM). These are the genuinely "most relevant" references and are the answer to the question of what threatens the claims. (IPR2022‑01539 exhibit list; Unified Patents Patroll)

Part A — The Operative Prior Art (from the IPRs): most relevant references

These are the references the automaker petitions relied on. Note the legal posture: the petitions used § 103 obviousness combinations, not § 102 anticipation. No single reference was asserted to anticipate independent claims 1/8/15/23; two references were combined per ground. So the "§ 102 anticipation" mapping below is necessarily conditional — I state where a reference is a genuine single‑reference (anticipatory) threat vs. only a § 103 component.

Key procedural fact (building on the earlier section): the Board instituted IPR2022‑01539 on 2023‑05‑02 and the proceeding was terminated 2024‑02‑07 after settlement. Because the automaker IPRs settled, none of these references ever produced a Final Written Decision of unpatentability for the '512 patent. The Patent Owner's response attacked exactly the two weak links: (a) Kim discloses cell‑specific pilot patterns, not cell‑specific pilots, and (b) Tong's adaptive beam‑forming requires advance channel knowledge and cannot be grafted onto Kim's initial‑sync/cell‑search pilots. (POR) This is directly relevant to the § 102 question: the "cell‑specific pilots" limitation is the crux, and a pure‑anticipation theory on Kim is contested.

Part B — The '512 Face Citations: the most relevant of the patent's own listed references

The face list is dominated by pre‑2004 background art plus a substantial number of post‑2004 items that cannot be § 102 art against 2004‑01‑29. I give the most claim‑relevant references full treatment and group the rest.

B1. Most relevant (same problem space; pre‑2004)

B2. Other pre‑2004 background references on the face (lower relevance)

US 5,905,436 (Telia, 1999‑06‑01); US 6,181,717 (DLR, 2001‑02‑13); US 6,175,550 (Lucent, 2001‑01‑16); US 6,430,364 (Ericsson, 2002‑08‑13); US 2002/0141483 (Doetsch, 2002‑10‑03, channel impulse estimation); WO 2003/058881 (Philips, 2003‑07‑17); US 2003/0179776 (Matsu shita, 2003‑09‑25, multicarrier transmitter/receiver); CN 1407745 A / CN 1445949 A / CN 1452326 A (2003); KR 20030060892 A (France Télécom, 2003‑07‑16, pilot‑only channel estimation). These are general OFDM/multi‑carrier/pilot background and do not map to the distinctive two‑pilot‑type or beam‑forming limitations.

B3. Face references that publish after 2004‑01‑29 — not available as § 102(a)/(b) art

Examples visible in the face list: US 7,039,001 (Qualcomm, 2006 — but priority 2002‑10‑29, so § 102(e)‑eligible); US 7,142,239 (Nortel, 2006‑12‑12); US 7,274,652 (Conexant, 2007); US 7,414,897 (Samsung, 2008); US 7,418,042 (Atheros, 2008); US 7,555,268 (Broadcom, 2009); US 7,639,660 (ETRI, 2009); US 7,647,647 (NTT DoCoMo, 2010); US 7,738,437 (Nortel, 2010); US 7,907,592 (LG, 2011); US 8,102,832 (Qualcomm, 2012); US 2005/0111397 (2005‑05‑26, priority 2002‑12‑06 → § 102(e)‑eligible); US 2006/0245409 (2006, priority 1999); US 2004/0085946 A1 (2004‑05‑06, priority 2001‑12‑07 → § 102(e)‑eligible); EP 1 650 891 (2006).

Each must be screened: those with pre‑2004 U.S. priority survive as § 102(e) art; those without do not qualify at all.

Part C — Anticipation (§ 102) bottom line

• No single face citation or petition reference anticipates the independent claims (1, 8, 15, 23). Every challenge was run as an obviousness combination (Ground 1: Kim + Tong; Ground 2: Ketchum + Li), and none of the references individually contains: (i) cell‑specific first pilots, (ii) a second, different pilot type collocated with data, (iii) beam‑forming on at least some subcarriers, all in one OFDMA/multi‑antenna time‑slotted signal.
• Best single‑reference (§ 102) candidates, each incomplete:
  • Kim (WO 2004/049618) — strongest on the two‑pilot‑type concept (Q₁ common + Q₂ cell‑specific, same‑slot), but lacks beam‑forming; also contested on "cell‑specific pilots" vs. "patterns."
  • Baum (US 5,867,478) — strongest on synchronized common pilots on aligned subcarriers, but has no cell‑specific pilot class and no beam‑forming.
  • Ma (US 2003/0072255 / 7,248,559) — strongest on multiple antennas + dedicated (BTS‑specific) pilot + common sync channel, but no beam‑forming limitation and a different (preamble‑centric) structure.
• Dependent‑claim anticipation is more realistic where a single reference squarely discloses the added feature — e.g., claims 2/9/16/24 (non‑overlap) and 4/11/18/26 / 5/12/21/29 (cell‑specific, non‑aligned) read well on Kim; the n/n+18 and m/m+20 spacing claims (3/10/17/25) were argued obvious over Kim + Tufvesson, not anticipated.

Explicit uncertainties / flags

1. Kim applicant identity: one search result attributes WO 2004/049618 to ETRI (Elec. & Telecom. Research Inst.); the petition caption says "Kim et al." I could not confirm the applicant from a primary WO cover page in these searches — treat the assignee line as provisional.
2. Ketchum (US 2004/0179627) date risk: if its U.S. filing date is March 2004, it is after the 2004‑01‑29 priority date and may not qualify as § 102(e) art at all. I could not verify the exact filing date; this is a material § 102 vulnerability I did not resolve.
3. Tong (US 7,120,395): I confirmed the content relied on (beam‑forming via multi‑antenna constructive combining) from the petition text, but did not independently verify its exact issue/priority dates; the face of the '512 does not list it, so it is a petition‑only reference.
4. Date verification incomplete for the ~50 lower‑relevance face items; my "post‑2004" grouping is based on the publication dates shown in the supplied Google Patents text, which I did not independently re‑verify for every entry.
5. No contradiction found with the previously generated section: it correctly noted the claims recite "second pilots of a second type" rather than "common pilots," and that the IPR turned on "cell‑specific pilots," "beam‑formed," and "at least one of the time slots" — all confirmed by the petition/POR text above. One minor discrepancy to note: the task header says "Current Date: April 26, 2026," while my environment date is 2026‑09‑25; I used the supplied priority/date logic, not the calendar, for the § 102 analysis.

I'll ground my obviousness analysis in the actual prior-art references listed on the patent page. Let me verify the teachings of the most relevant references.

Let me examine the actual IPR petition grounds against this patent family, which represent real-world obviousness contentions, plus check the Nortel scattered-pilot reference.

Obviousness Analysis — US 10,965,512 B2 under 35 U.S.C. § 103

Building on the earlier claim summary (independent claims 1, 8, 15, 23; dependent claims 2–7, 9–14, 16–22, 24–30). This section adds the § 103 combination analysis and does not repeat the claim/limitation mapping already provided.

1. Framework

Governing law. The application was filed 2020-09-04 but is a continuation claiming priority to 2004-01-29 (provisional 60/540,032) through PCT/US2005/001939. Because every claim has an effective filing date before 2013-03-16, pre-AIA § 102/§ 103 applies (consistent with how the parallel Neo Wireless IPRs were pleaded — the '450-family petitions expressly invoked "pre-AIA § 103"). Prior art therefore must qualify under pre-AIA §§ 102(a)/(b)/(e)/(f)/(g), and the KSR flexible TSM standard governs the combination question. KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007).

Level of ordinary skill. For a Jan-2004 priority date, a POSITA would have a B.S. in EE (or equivalent) plus ~2–4 years' experience in multicarrier/OFDM cellular system design, or an M.S. with less experience — familiar with OFDM/OFDMA pilots, channel estimation, cell search, MIMO, and beamforming. (This is the profile Dr. Min was retained to apply in the IPRs.)

Claim-construction pressure points. The claim-construction disputes actually litigated on this patent frame the analysis: "cell-specific pilots," "beam-formed," "at least one of the time slots," and whether the first/second subcarrier sets must be co-slot. See Patent Owner's Response, IPR2022-01539 (Aug. 10, 2023), TOC (arguing Petitioner failed on "cell-specific pilots," "beam-formed," and "in at least one of the time slots") (Docket Alarm POR).

2. The Prior-Art Landscape (from this page's Prior Art section, plus the real IPR record)

The Google Patents "Citations (69)" and "Family Cites" lists are unusually rich, and several listed references were in fact the exact references Petitioners used. From the IPR2022-01539 exhibit list (Joint Motion, Ex. 1001–1021):

The page's prior-art section therefore already contains, in its own citation list, references that (a) teach synchronized multi-cell OFDM with cell-distinguishable pilots, (b) teach dual pilot classes (dedicated/cell-specific and common/shared), (c) teach MIMO beamforming with pilots, and (d) expressly motivate selecting pilot spacing based on channel variation. That is essentially the entire claim stack.

3. Combination A — Ground 1 analog: Kim + Tong (+ Ma '559)

(the primary ground Petitioners advanced; institution granted 2023-05-02)

What each reference supplies

• Kim (WO 2004/049618) — an OFDMA/cellular reference teaching cell-specific pilot/scramble patterns distinguishable per cell — mapped to "first pilots… are cell-specific pilots," inserted on a first set of subcarriers, with data and a second pilot type on a second set.
• Tong (US 7,120,395) — beamforming. The Board credited Petitioner's showing that "Tong indicates beam-forming provides benefits without having to receive feedback on channel conditions," and that on the preliminary record the evidence for "wherein at least some subcarriers… are beam-formed" was sufficient for institution. (Law360/Sterne Kessler).
• Ma '559 — antenna-specific, periodically spaced scattered pilots (supports the pilot-set and spacing limitations).

Why a POSITA would combine. Both are in the same field (multi-cell OFDM/OFDMA), address the same problem the '512 specification itself identifies — that in a multi-cell environment "signals originated from the base stations at different cells interfere with each other" — and the combination is a mere substitution of one known multi-antenna technique (beamforming) into a known multicarrier pilot framework, with the predictable result of higher SINR/reduced inter-cell pilot interference. KSR at 416–17 (predictable variation of known elements; "design incentives" and "market forces" supply motivation). There is a reasonable expectation of success because beamforming with dedicated pilots was itself conventional.

Claim mapping. Claims 1/8: OFDMA BS with plural antennas + transmitter inserting cell-specific pilots + data/second pilots + beamforming + common time-slot transmission + different pilot types + no interference. Claims 15/23: mirror image at the mobile station (recover data from at least the second pilots; recover cell-specific info from cell-specific pilots). Dependent claims 2/9/16/24 (disjoint sets), 4/11/18/26 (cell-unique pilots), 6/13/20/28 (second set beam-formed) are met by, or obvious in view of, the same combination.

4. Combination B — Ground 2 analog: Ketchum + Li

(the secondary ground; institution granted)

• Ketchum (US 2004/0179627) and Li (US 2002/0163879) — Qualcomm MIMO/OFDMA references supplying beamformed transmission and pilot-based channel estimation.
• Motivation. Adding the other reference's pilot structure (or cell-specific scrambling) to a beamformed OFDMA downlink is the kind of combination of known elements a POSITA would pursue to serve both synchronization and cell identification functions — exactly the two functions the '512 specification assigns to its two pilot groups.

Why this ground matters to § 103. Ground 2 shows the claimed subject matter is reachable from multiple independent, unrelated starting points — a hallmark of non-novelty-grade advance under KSR. Notably, PO's Response conceded the art teaches "cell-specific pilot patterns" but argued the claims require something more (see §6 below) — an admission that the structural elements are present.

5. Additional / backup combination available straight from the page's own Citations

Even setting aside the IPR ground set, the page's citations independently support a § 103 rejection:

Baum '478 + Ma '255 + Qualcomm '706 (or Qualcomm MIMO WLAN '301) + Fujitsu EP 1 650 891

• Baum '478 supplies the multi-cell, base-station-synchronized-to-a-common-source framework and the aligned/common pilot concept ("pilot codes … transmitted on a same plurality of predetermined subcarriers for each OFDM transmitter"), plus the express teaching that pilot spacing is chosen "according to the expected rate of channel variation" — directly motivating the n/n+18 and m/m+20 spacings of claims 3/10/17/25 as a design choice.
• Ma '255 supplies both pilot classes in an OFDM multi-antenna context: a dedicated (BTS-specific, i.e., cell-specific) pilot channel and a common synchronization channel that is "common across multiple base stations." This maps almost one-to-one onto "first pilots… are cell-specific pilots" and "second pilots of a second type" that "do not interfere" with the first.
• Qualcomm '706 supplies the "at least some subcarriers… are beam-formed" element (full-CSI eigen-beamforming with pilots for channel estimation).
• Fujitsu EP 1 650 891 supplies the sector/cell-specific pilot rationale ("it is necessary to use pilots that differ for every sector") and multiple pilot types multiplexed in one OFDM system.

Motivation to combine: all four are in OFDM/MIMO cellular; the '512's own stated design goals (frequency synchronization via common pilots; cell identification via cell-specific pilots) are exactly the functions Baum '478 and Ma '255 attribute to their pilots; adding beamforming is an obvious capacity/SINR-improvement measure.

6. Where obviousness is contestable (honest assessment)

The Record shows the PO mounted specific rebuttals, and a rigorous analysis should credit them:

1. "cell-specific pilots" vs. cell-specific pilot patterns. PO argued the invention improved on prior-art "cell-specific pilot patterns," so the reference's pattern disclosure does not meet the claim term as construed. If "cell-specific pilots" is construed to require pilots carrying cell-unique information (not merely scrambled patterns), some references may fall short on the literal limitation. Mitigated by claim 4/11/18/26 and Ma '255's BTS-specific PN pilot channel.

2. "beam-formed." PO argued beamforming requires advance knowledge of the user's channel, and that initial-synchronization/cell-search pilots cannot implement adaptive beamforming. This is the single most contested limitation. Mitigated by Tong's teaching that beamforming is beneficial without feedback, and by the claim's broad "at least some … or" wording (only some of either set needs to be beamformed).

3. "in at least one of the time slots." PO argued the prior art does not co-locate the first and second subcarrier sets in a common slot as claimed. This is a real literal-infringement-style gap and should be checked against each reference's frame structure.

4. Specific spacings n/n+18 and m/m+20 (claims 3/10/17/25). PO specifically argued Ground 1 does not render these obvious — consistent with the fact that no cited reference appears to disclose these exact periods. The strongest § 103 path here is design-choice/optimization: Baum '478 expressly frames pilot spacing as selected per channel variation, and there is no evidence the specific values are critical or produce unexpected results. Absent criticality, a numerical range/period chosen by routine optimization is obvious. KSR; In re Applied Materials.

5. "first pilots do not interfere with the second pilots." Vague; likely read as "the two pilot types are designed to be distinguishable/non-colliding." Claim 2's "all… different than all…" reinforces non-overlap, which almost every cited reference satisfies.

6. Pre-AIA § 102(b) status of some references. Ma '255, Baum '478, Qualcomm '706, Fujitsu are all published >1 year before 2004-01-29 and qualify under § 102(b). Ketchum (US 2004/0179627, pub. 2004-09-16) and Kim (WO 2004/049618, 2004) publish after the priority date; Petitioners must rely on § 102(e)/§ 102(a)-with-priority-dating (their earlier provisional/filing dates) or on the '512's inability to swear behind certain dates. This is a genuine vulnerability of Ground 1/2 that should be verified against each reference's earliest effective date.

7. Secondary considerations

No strong objective indicia are apparent on the face of the record. The patent was commercially asserted (MDL 3034 against ~15 automakers), which could be argued as commercial success, but the nexus to the claimed pilot dual-structure is unproven, and the technology was widely adopted as standardized 3GPP LTE functionality — i.e., practice may reflect the standard, not the patent's incremental advance. Note the related '450 patent was upheld by the PTAB in 2024–2025, but that turns on different claims/terms (e.g., "starting time-frequency coordinate," "mobile device") and is not probative of the '512's pilot claims.

8. Bottom line

• Claims 1, 8, 15, 23 are, on the weight of the page's own citation set and the actual IPR record, likely obvious over at least Kim + Tong (+ Ma '559) (Ground 1) and Ketchum + Li (Ground 2), and independently over Baum '478 + Ma '255 + Qualcomm '706 + Fujitsu EP 1 650 891. The combination rationale is strong: same field, same stated problem, predictable substitution of beamforming into a known synchronized multi-cell pilot framework.
• Claims 2, 4–7, 9, 11–14, 16, 18–22, 24, 26–30 (disjoint sets, cell-unique pilots, cell-info extraction) are very likely obvious as express disclosures or mere design choices.
• Claims 3, 10, 17, 25 (n+18 / m+20 spacing) are the hardest to invalidate; they are vulnerable only as an obvious optimization of pilot spacing (well supported by Baum '478's teaching), not by direct anticipation.
• Caveat on the merits record: the IPR2022-01539 challenge to this patent settled and terminated on 2024-02-07 before any final written decision, so the PTAB never adjudicated the merits of these grounds for the '512. The institution decision (finding a reasonable likelihood on, inter alia, the "beam-formed" limitation in view of Tong) is the strongest available indication of the merits, but it is not a holding of unpatentability.

9. Explicit uncertainties

1. Reference-internal teachings I could not fully verify: the precise disclosures of Kim (WO 2004/049618), Ketchum (US 2004/0179627), Li (US 2002/0163879), and Tong (US 7,120,395) are characterized here from the IPR's ground structure and secondary reporting, not from full-text review; specific limitation-by-limitation mappings would need the petition and exhibits.
2. Exact § 102 basis for the post-priority references (Kim, Ketchum) is unverified; the analysis flags it rather than assuming it.
3. No final merits ruling exists for the '512 (settlement). Anything stated as "likely obvious" is an analytical conclusion, not a PTAB or court determination.
4. The earlier-generated summary's finding of no 2026 CAFC activity on the '512 stands; the CAFC appeals reported on Lexology involve the '450 patent, not this one.