US 11402458

I'll search for information on this patent number, including any litigation or CAFC dockets.

Let me check for any litigation or post-grant proceedings specific to this patent.

Summary — U.S. Patent No. 11,402,458 (interpreted literally as "11402458")

Verification note: I did not find any CAFC 2026 docket, district-court litigation, or PTAB/IPR proceeding naming U.S. Patent No. 11,402,458. Searches for the exact number returned only patent-document aggregator hits (Google Patents, Patexia) and unrelated matters. I therefore cannot confirm any 2026 Federal Circuit activity for this patent, and I do not have authoritative litigation information — treat the "no litigation found" statement as a negative search result, not a certainty.

Bibliographic data (grounded in the patent text provided)

Abstract (verbatim gist)

Mechanisms for using compressive sampling to detect direction of arrival (DoA) of a signal of interest (SoI): in each of a plurality of receiver paths, an antenna receives the SoI and produces a received signal; a modulator receives a modulator input signal (MIS) based on the received signal, modulates the MIS at multiple points in time based on different pseudo-random numbers, and produces modulated output signals; a summer sums across paths, for each point in time, to produce sum signals; and a hardware processor performs a compressed sensing recovery algorithm to recover the DoA of the SoI.

Plain-language overview of the independent claims

Claim 1 (circuit, apparatus):

• Multiple receiver paths, each with (a) its own unique antenna that receives the signal of interest and produces a received signal, and (b) a modulator that takes a modulator input signal derived from that received signal, modulates it at multiple points in time using different pseudo-random numbers (one per time point), and produces a set of modulated output signals for that path.
• A summer that, for each point in time, adds up one modulated output signal from each path to produce a sum signal per time point.
• A hardware processor that receives the sum signals and runs a compressed sensing recovery algorithm to recover the DoA of the signal of interest in physical space with respect to the antennas.
• Key structural limits: "a unique one of a plurality of antennas" per path (one antenna per path), and the DoA is recovered in physical space relative to the array.

Claim 11 (method):

• Mirrors claim 1 as a method: in each of multiple receiver paths, receive the SoI and produce a received signal using a unique antenna; use a modulator to receive a modulator input signal based on the received signal, modulate it at multiple points in time based on different pseudo-random numbers, and produce multiple modulated output signals; use a summer to sum, for each time point, one modulated output signal per path to produce sum signals; and perform a compressed sensing recovery algorithm to recover the DoA in physical space with respect to the antennas.

Notable dependent claims (framing, not separate inventions):

• Cl. 2 / 12: a mixer downconverts the received signal to produce the modulator input signal.
• Cl. 3 / 13: the modulated outputs are current signals, and the summer is a connection that combines currents (wired summing).
• Cl. 4 / 14: an ADC samples the sum signals.
• Cl. 5 / 15: the processor is a digital signal processor.
• Cl. 6 / 16: the recovery algorithm is orthogonal matching pursuit (OMP).
• Cl. 7 / 17: the modulator is a vector modulator.
• Cl. 8 / 18: the modulator is a phase shifter; cl. 9 / 19: it also modulates at other times based on a desired phase shift; cl. 10 / 20: processor also performs conventional beam forming (CBF) — i.e., the dual-mode (CS-DoA + CBF) "direct space-to-information converter" (DSIC) concept described in the specification.

Technical context from the specification

The patent describes a DSIC architecture unifying conventional beam forming (delay/phase-shift-and-sum) with compressive-sampling DoA finding, usable in two modes (CBF mode and CS-DoA mode) with shared hardware, using Rademacher-based PN sequences for the phase modulation, and needing as little as one bit of resolution in the vector modulators for a full field-of-view DoA scan.

Uncertainty / caveats

• The full text above is the authoritative source I used; item-by-item claim scope should be read from the actual issued claims.
• I found no evidence of CAFC 2026 docket entries, IPRs, or litigation for 11,402,458. If you need a definitive litigation check, that should be run against PAIR/PatentCenter, the PTAB P-TACTS system, and PACER/CAFC docket databases directly, since my searches did not surface authoritative records.

I'll search for litigation involving US patent 11402458.

Let me search more targeted litigation databases.

Let me run a few more targeted searches to confirm whether any litigation exists.

Litigation Search: US Patent 11,402,458

Result: No litigation identified.

Based on my searches of patent litigation resources (including the Unified Patents litigation portal, Patexia, Google Patents, PACER-adjacent sources, and CAFC opinion databases), I was unable to identify any litigation involving US Patent 11,402,458. I want to be explicit about the limits of this conclusion rather than imply a definitive negative.

Patent Reference Data (as confirmed by search)

Per the Google Patents record and Patexia, the patent is:

Case List

No cases to report. There are no plaintiff(s), defendant(s), jurisdiction, case number, filing date, or outcome to provide, because no litigation asserting or challenging this patent (e.g., a district court infringement action, declaratory judgment action, ITC action, or PTAB post-grant proceeding) appeared in my search results.

Important Caveats

1. Searches are not exhaustive. I did not obtain a confirmed "zero results" from a dedicated litigation database (e.g., a direct Docket Navigator, Lex Machina, or Unified Patents caselist query keyed exactly to "11402458"). Absence of results in web searches does not guarantee no litigation exists — it may simply be absent from indexed/returned content, or too recent to be indexed.
2. Strict ID matching. Some search hits contained the string "11402458" in unrelated contexts (e.g., a UK Companies House gazette list of registration numbers, and a separate "IPR2016-01464" Unified Patents case re US 7,454,002). These are not references to the patent at issue and should not be treated as such.
3. Citation of the patent elsewhere is not litigation. The patent appears in the "citations/family" tables of other Google Patents pages (e.g., as a citation on US 9,077,443 and US 2014/0269849). Citation relationships are prior-art/family links, not litigation.
4. Related corporate context (unverified for this patent). The same inventors/assignee family (Columbia University, Kinget et al.) is associated with other receiver/interferer patents (e.g., US 8,594,603; US 2017/0026066) that have drawn post-grant activity, but I found no evidence tying any such activity specifically to US 11,402,458.

Bottom line

On the basis of the searches conducted, I find no known litigation involving US Patent 11,402,458. Anyone relying on this should independently confirm against a paid litigation database (Docket Navigator / Lex Machina) and the USPTO Patent Center / PTAB docket before treating the absence as definitive.

I'll start by verifying the canonical (empty) proceedings list against live sources.

Let me run a few more targeted checks to make sure nothing is hiding (including any terminated/settled IPRs and related-patent activity).

PTAB Proceedings Report — U.S. Patent No. 11,402,458 B2

"Circuits and methods for using compressive sampling to detect direction of arrival of a signal of interest"
Patent Owner: The Trustees of Columbia University in the City of New York
Inventors: Matthew Bajor; John Wright; Tanbir Haque; Peter R. Kinget
Appl. No. 16/420,108 · Filed 2019-05-22 · Priority 2018-05-22 · Granted 2022-08-02 · Adjusted expiration 2040-07-04

Proceedings overview

Total AIA trial proceedings on file: 0 — zero active, zero claims invalidated, zero claims sustained (there is nothing to sustain), zero settled, zero institution denials. The USPTO Open Data Portal returns no AIA trial proceedings for this patent, and independent web searches for IPR/PGR/CBM petitions naming U.S. 11,402,458 (and its application number 16/420,108 and publication 2019/0361088) surfaced nothing — no petition, no institution decision, no FWD, no termination, no Federal Circuit appeal.

Bottom-line defensive posture: this is not a hardened, IPR-tested patent — it is an untested patent. No claim of the '458 has been canceled, narrowed by amendment, or adjudicated unpatentable by the Board. Critically, that also means no petitioner has accrued any § 315(e) estoppel, so a defendant today retains the full universe of prior-art grounds and all 20 claims remain live and challengeable. The absence of PTAB activity is itself a signal (see Pattern signals below): well-asserted patents in this space tend to attract IPRs once a real dispute arises. There is no public evidence of assertion activity against an implementer, which likely explains the clean PTAB docket.

For each proceeding

None. There are no proceedings to enumerate. I will not manufacture proceeding numbers, panels, or dispositions. Per the operating constraint, the canonical source ("PTAB proceedings on file") is empty, and web search did not contradict it.

Verification trail

Source of record: https://patents.google.com/patent/[US11402458](/patent/US11402458)/en · PTAB case portal (P-TACTS, successor to PTAB E2E): https://ptacts.uspto.gov/ptacts/

⚠️ Flag — things I found but that are NOT this patent. Columbia is a frequent and heavily-litigated patent owner, so search results are thick with Columbia PTAB matters. None involve the '458, and I want to be explicit so nothing gets misattributed:

• Illumina v. Columbia — IPR2012-00006, IPR2012-00007, IPR2013-00011, IPR2013-00128, IPR2013-00266, IPR2013-00517, and the 2018 wave IPR2018-00291/00318/00322/00385/00797, plus the follow-on IPR2020-01177 — all directed to Columbia DNA-sequencing patents (e.g., U.S. 9,718,852; 9,719,139; 9,708,358; 9,725,480; 9,868,985). Affirmed in Trustees of Columbia Univ. v. Illumina, Inc., 620 F. App'x 916 (Fed. Cir. 2015).
• Columbia v. Symantec/Norton (Gen Digital) — IPRs against Columbia's cybersecurity patents U.S. 8,601,322 and 8,074,115 left only claims 2, 11, 27 ('322) and claim 2 ('115) for trial; the resulting $185,112,727 verdict was vacated by the Federal Circuit on § 101 grounds (Nos. 2024-1243, 2024-1244, Mar. 11, 2026).
• Prior art cited during prosecution of the '458: U.S. 2017/0026066 A1 (Kinget, "Circuits and methods for detecting interferers") and U.S. 8,594,603 (Columbia, "Systems and methods for cancelling interferers in a receiver") appear in the citation list. These are prosecution references, not PTAB proceedings — but they are the most natural starting art for any future IPR and share overlapping inventors/specification lineage.

If a client's invalidity team has found a petition that I could not locate (e.g., a very recent filing not yet indexed in ODP), treat it as a live-updating gap, not a contradiction of this report.

Strategic summary

Claim-by-claim status of 11,402,458. All 20 claims are UNTESTED and INTACT. Claim 1 (independent, circuit) and claim 11 (independent, method) are alive, together with dependents 2–10 and 12–20. No claim is CANCELED. No claim has been SURVIVED (nothing was attacked). The claims as granted recite: a plurality of receiver paths, each with a unique antenna producing a received signal and a modulator that modulates at multiple points in time based on different pseudo-random numbers, producing multiple modulated output signals per path; a summer that sums across all paths for each point in time to produce multiple sum signals; and a hardware processor performing a compressed sensing recovery algorithm to recover the DoA "in physical space with respect to the plurality of antennas." The unique-antenna and physical-space limitations were apparently the examiner's line of distinction over the art — note that the pre-grant publication abstract (2019/0361088) describes the broader mechanism, and the granted claim 1 added "a unique one of a plurality of antennas" and the "physical space" language. Those limitations are the likely battleground for any future § 103 challenge and, equally, the likely center of any infringement dispute involving a multi-antenna RFIC with per-element PN phase modulation plus a shared summer (a "direct space-to-information converter" architecture).

Estoppel landscape. There is nothing to work around. Because no AIA petition was ever filed against the '458, no petitioner (or privy) has triggered 35 U.S.C. § 315(e)(2), and no IPR final written decision exists to shut down grounds. A defendant's prior-art options are therefore unrestricted: § 102 anticipation, § 103 obviousness, § 112 enablement/written-description/indefiniteness, and § 101, in any forum, on any reference. The only real limitation on a defendant is the § 315(b) one-year clock running from service of a complaint alleging infringement, plus estoppel arising from that defendant's own past IPRs on this patent (of which there are none). Conversely, the patent owner has not had any claim narrowed by adverse Board construction, so there is no FWD holding to exploit.

Pattern signals. (i) Same petitioner, multiple IPRs on this patent? No — no petitioner at all. (ii) Has Columbia pursued PTAB appeals aggressively? Columbia litigates hard and extensively (see the Illumina sequencing campaign and the Symantec/Gen Digital cybersecurity case, both of which cycled through PTAB and the Federal Circuit), but that aggressiveness is on other patents and runs mostly in the direction of defending its portfolio. There is no Columbia appeal involving the '458. (iii) Defensive aggregator in the chain? No. There is no evidence of Unified Patents, RPX, or any defensive aggregator challenging the '458, and unlike Columbia's sequencing and cybersecurity patents, this one does not appear to have drawn any ex parte reexam or covered-business-method activity either. Notably, the CBM lane is closed: CBM review sunset on 2020-09-16 and the '458 (a radio-direction-finding circuit) is not a "covered business method" financial patent in any event. PGR is also time-barred: the nine-month § 321(c) window closed 2023-05-02. IPR is the only AIA vehicle now available, and it remains fully open for any defendant that is not yet served, or for a served defendant within its one-year § 315(b) window.

Recommended next steps

If you are a defendant being asserted against on the '458:

1. Do not expect an IPR-based easy win — and do not expect the patent owner to have a roadmap from prior PTAB findings. There is no FWD to quote and no canceled claim to point to. Any demand letter citing claims 1–20 is citing claims that have never been adjudicated unpatentable; there is no "claim 1 is dead" argument to make. Conversely, there is also no adverse Board claim construction binding the patent owner, so you are litigating on the full asserted scope.

2. Move fast on the § 315(b) clock. If you have been served with a complaint alleging infringement of the '458, you have one year from service to file an IPR or be forever barred. Because no prior IPR exists, a well-funded defendant's IPR is arguably the single most valuable lever here — it is the only AIA path left (PGR time-barred since 2023-05-02; CBM unavailable).

3. Start from the prosecution record and the overlapping Columbia/Kinget family. The most fruitful prior art will likely be the interferer-detection lineage cited against the '458 — U.S. 2017/0026066 A1 (Kinget) and U.S. 8,594,603 — plus the compressive-sensing/DoA literature the patent itself acknowledges (OMP, Rademacher-based sensing matrices, sub-Nyquist array processing). Attack the two limitations that appear to have carried the allowance: "a unique one of a plurality of antennas" and recovery of DoA "in physical space with respect to the plurality of antennas." Also mine § 112: claim 1's modulator/summer/hardware-processor architecture and the claim 9/19 "desired phase shift" mode-switching language are worth a written-description and indefiniteness review.

4. Preserve § 101 as a parallel theory. The Federal Circuit's March 2026 decision vacating the $185M Columbia v. Gen Digital verdict shows that Columbia's patents can fail Alice step one at the Federal Circuit. The '458 claims a signal-processing algorithm on conventional receiver hardware; a § 101 challenge is non-frivolous, though the "physical space with respect to the plurality of antennas" and hardware-centric claim language cuts against it far more than the claims at issue in Gen Digital.

5. Watch for a late-filed petition. A very recent IPR (or an ex parte reexam) may not yet appear in the ODP ingest. Re-check PTAB P-TACTS (https://ptacts.uspto.gov/ptacts/) and CourtListener (https://www.courtlistener.com/) before finalizing any invalidity or estoppel position, and calendar a re-verification before the § 315(b) deadline.

If you are the patent owner: the clean PTAB docket is a genuine asset — all 20 claims survive, no prosecution disclaimer was forced, and no estoppel constrains you. But recognize that the same facts mean the patent is untested against a serious invalidity challenge; validity of claims 1 and 11 has never been adversarially vetted, and the unique-antenna / physical-space limitations that distinguish the granted claims from the application as published are exactly the kind of narrowing that invites a § 112 or § 103 attack the moment you assert.

If you simply need the answer: as of 2026-09-29, there is no PTAB activity on US 11,402,458. No IPR, no PGR, no CBM, no FWD, no appeal — the patent is unlitigated at the Board.

Caveat on completeness: I could not obtain a direct, primary-source P-TACTS case-list retrieval for the patent number in this session; the "no proceedings" conclusion rests on (a) the empty canonical ODP block and (b) the absence of any contradicting hit across multiple web searches, including searches keyed to the patent number, the application number, the publication number, and the assignee/inventors. If a filing exists that is newer than the ingestion window, it would not appear here, and I have flagged that gap rather than asserting it away.

I'll verify the assignment record before constructing the report.

Ownership & Assignment Analysis — US 11,402,458 B2

Patent: US 11,402,458 B2 — "Circuits and methods for using compressive sampling to detect direction of arrival of a signal of interest"
Application: US 16/420,108, filed 2019-05-22 · Provisional priority US 62/674,930 (2018-05-22) · Granted 2022-08-02
Primary source for record data: Google Patents legal events for US11402458B2 · USPTO Assignment Center · legacy assignment search

Retrieval caveat up front. The Assignment Center search UI is not directly reachable from this analysis environment. The assignment data below is drawn from the Google Patents legal-events mirror of the USPTO assignment record. That mirror confirms one post-filing assignment event but does not expose a reel/frame number or the correspondent of record. I am not able to verify or supply those fields, and I will not invent them. Any reel/frame cited below as "not retrieved" should be confirmed by directly querying the Assignment Center by patent number.

Inventors

Pattern notes:

• All four inventors share a single institutional employer (Columbia University / Columbia Integrated Systems Lab). This is the classic single-employer academic roster, not a multi-company joint-invention roster.
• No "inventor exodus" pattern. Nothing in the record indicates inventors departed the original assignee within 12 months of filing. Kinget remained at Columbia (chair 2017–2020) and is still there; Wright remains Columbia faculty. Bajor was a graduating doctoral student, which is normal and not indicative of a portfolio fire-sale.
• Commercialization lineage (context, not an assignment finding): Kinget has co-founded startups that license Columbia technology (Seamless Devices, 2014; Concurrent Wireless, 2016; WiLO Networks, 2021) and lists "Wright, John / Bajor, Matthew / Haque, Tanbir" among Columbia technology inventors. None of these entities appears as an assignee on this patent. Columbia's licensing arm is Columbia Technology Ventures (this invention is catalogued as CU18364, "Direct space-to-information converter for efficient interference sensing").

Original assignee

The Trustees of Columbia University in the City of New York, New York, NY.

• Entity type: University (non-profit, chartered 1754 as King's College; "The Trustees of Columbia University in the City of New York" is the legal corporate name; "Columbia University" is the operating name). Operating and solvent — no bankruptcy, no dissolution, no acquisition.
• Product embodying the claims: None in commerce by Columbia itself. The technology was reduced to silicon: Columbia's CISL designed, fabricated (65 nm CMOS) and tested a "DSIC" RF ASIC operating 1–3 GHz with 8 antenna elements, consuming 158 mW (~19.8 mW/element) and switching between CBF and CS-DoA modes in 1 µs. That is a research demonstrator/prototype, not a shipped commercial product. Columbia's business model is non-exclusive/exclusive technology licensing, not product sales.
• Note on the two assignee name forms: The published application (US 2019/0361088 A1) lists the applicant as "Columbia University in the City of New York," while the granted patent and the recorded assignment name "The Trustees of Columbia University in the City of New York." This is a naming-formalization, not a change of ownership.

Assignment timeline

Only one assignment event is recorded against this patent, plus the underlying filing/grant events.

• 2022-06-14 (executed and recorded; reel/frame not retrieved — see caveat)

  • Conveyance: Assignment — "Assignment of Assignors' Interest" (inventor-to-employer / obligation assignment)
  • Assignor: John Wright, Tanbir Haque, Matthew Bajor, Peter R. Kinget (all four named inventors)
  • Assignee: The Trustees of Columbia University in the City of New York
  • Correspondent: Not exposed by the source I could access. No flag: I cannot state that any correspondent recurs here because I could not retrieve the correspondent field for this recording. (For calibration only: a different Columbia patent, app. 16/293,403, was recorded at reel 048685/0319 with correspondent WILMERHALE, 60 State Street, Boston MA, attn. Michael E. Connors — that is a different technology and I have no evidence it applies to this patent.)
  • Context: Internal/obligation assignment confirming the university's title — the inventors assign to their employer. Not an acquisition, fire-sale, securitization, or transfer-to-asserter.

• 2019-05-22 — Non-provisional application 16/420,108 filed (not an assignment event; recorded for chain context).

• 2022-08-02 — Patent granted to The Trustees of Columbia University in the City of New York (not an assignment event).

• 2040-07-04 — Adjusted expiration (terminal disclaimer / PTA-adjusted term); not an assignment event.

There are no post-issuance assignments. No transfer to an IP-holding LLC, no security agreement, no merger, no license recorded, no correction. As of this analysis, Columbia still owns the patent outright.

Timeline diagram

timeline
    title Ownership of US 11402458
    2018 : Provisional filed by Columbia inventors
    2019 : Non-provisional filed by Columbia
    2022 : Inventors assign to The Trustees of Columbia University
         : Patent granted
    2040 : Adjusted expiration

NPE / troll-pattern signals

Additional corroborating data points (not signal-scored):

• US 11,402,458 appears as item 43 in a PTAB filing exhibit list (Petition, PTABS accession 1557168) alongside its sibling US 11,374,599 ("Circuits for identifying interferers using compressed-sampling") — i.e. it sits in a the same Columbia compressive-sampling family that has drawn third-party PTAB attention. Being challenged does not equate to being asserted; it is context only.
• Columbia is a genuinely litigious university patent owner in other technical areas (e.g. Trustees of Columbia Univ. v. Symantec/Gen Digital, Fed. Cir. 24-1243; the long-running Norton/Symantec inventorship dispute; Mercedes-Benz-related PTAB matters). This establishes that Columbia does enforce patents directly and through outside counsel — but a university enforcing its own patents is not the classic NPE model, because Columbia is the original owner, is non-operating, and licenses rather than sells product. Classify carefully.

Verdict

Insufficient data — only the original assignment is on record.

The chain is one link long: all four named inventors assigned to The Trustees of Columbia University in the City of New York on 2022-06-14 (recorded; reel/frame not retrievable from the sources I could access). There is no assignment to any shell LLC, no known asserter in the chain, no cascading transfer, no bankruptcy sale, and no defensive-aggregator termination — the file simply reflects an ordinary academic inventor-to-university title confirmation. With a single recorded link and no evidence of any post-issuance transfer, the NPE-signal analysis is underdetermined; the correct call per the rubric is Insufficient data, not a negative finding of NPE behaviour. If a licensing or assertion activity against this patent later surfaces, the relevant question is university licensing/enforcement, which should be evaluated on its own terms rather than scored against the NPE-troll framework.

Verify directly: USPTO Assignment Center — search "11402458" (or the legacy interface at https://assignment.uspto.gov/patent/index.html). Querying by patent number is required to obtain the reel/frame and correspondent fields that this analysis could not retrieve.

Method note: names, dates, and entity identities above are taken from the authoritative patent text supplied for this analysis and from the Google Patents legal-events mirror. Reel/frame numbers and correspondent names are deliberately left unstated where they could not be sourced, per the no-fabrication constraint.

I'll research this patent and its cited prior art. Let me run several searches.

Let me gather details on a few more cited references and check for any direction-of-arrival-specific prior art.

Prior-Art Analysis — US 11,402,458 B2

1. Confirmation of the subject patent (identifier interpreted literally)

Identifier note: I did not auto-correct or normalize the number. Everything below concerns the literal number 11402458 / US 11,402,458 B2 only.

Claim architecture (for §102 mapping). Independent claim 1 (circuit) requires: (a) a plurality of receiver paths, each with a unique antenna and a modulator that (i) receives a modulator input signal based on the received signal, (ii) modulates it at multiple points in time based on different ones of a plurality of pseudo‑random numbers, and (iii) produces a plurality of modulated outputs; (b) a summer that, for each point in time, sums across the receiver paths to produce a plurality of sum signals; and (c) a hardware processor that performs a compressed‑sensing recovery algorithm to recover the DoA in physical space with respect to the antennas. Independent claim 11 is the method counterpart. Claims 2–10 depend from 1; claims 12–20 depend from 11 (mixer downconversion; current‑summing summer; ADC sampling; DSP; OMP recovery; vector modulator; phase shifter; additional phase‑shift modulation; conventional beam forming).

2. Important framing / caveat

The examiner‑cited art for this patent is not a DoA/antenna‑array body of art. It is almost entirely (i) the same Columbia/Kinget interferer‑detection and compressive‑sampling line of work, (ii) general compressive‑sensing references, and (iii) component‑level receiver art (LNAs, mixers, LO generators, filters, ADCs, wake‑up receivers, AGC). The claimed novelty — pseudo‑random spatial modulation across a plurality of antenna paths whose outputs are summed per time point and CS‑recovered into a direction of arrival in physical space — is not taught by any single cited reference I can identify. Consequently, the realistic §102 exposure is on the dependent claims (component choices) and possibly on claims 1/11 only if one reads a frequency‑domain interferer detector as inherently disclosing the array/DoA elements (it does not, because it uses a single antenna path).

I am providing a complete citation inventory plus a focused validity analysis of the most relevant references. For the low‑relevance component references, descriptions are drawn from titles/assignees/dates in the patent's own citation list; I flag items where I could not independently retrieve the full text (my retrieval of US 8,457,579, US 8,836,557 and EP 2,662,706 was cut short by tool limits, so those three are lower confidence).

3. Most relevant prior art (Tier 1) — with claim mapping

3.1 US 2017/0026066 A1 — Circuits and methods for detecting interferers (the closest reference)

• Full citation / URL: Kinget, P. R. et al.; US 2017/0026066 A1; pub. 2017‑01‑26; priority 2014‑09‑12; also granted as US 9,762,273 B2 (2017‑09‑12) and in the family WO 2016/040958 A1. Assignee: The Trustees of Columbia University in the City of New York. https://patents.google.com/patent/US20170026066A1/en ; related US 11,374,599 B2 (Kinget).
• Description: A receiver that downconverts an RF signal to complex baseband, multiplies the I and Q signals by m unique pseudorandom‑noise (PN) sequences (e.g.,Gold/m‑sequences), low‑pass filters, digitizes with ADCs, pairwise combines, and uses a support‑recovery / CS block to identify interferers' frequency locations. It is a compressed‑sampling, PN‑modulated, mixer‑downconversion architecture with PRN generation and a sparse‑recovery processor.
• Potential §102 relevance:
  • Claims 2/12 (mixer downconverting received signal to the modulator input) — directly disclosed.
  • Claims 4/14 (ADC sampling) and 6/16 (greedy/CS recovery) — disclosed (OMP/greedy support recovery).
  • Claims 1/11 — partial only. It discloses a "modulator… modulate the modulator input signal… based on… pseudo‑random numbers" and CS recovery, but it is a single‑antenna, frequency‑domain detector; it does not disclose "a plurality of receiver paths, each [with] a unique one of a plurality of antennas," a summer "sum[ming] across each of the plurality of receiver paths" per time point, or recovery of "direction of arrival… in physical space." Because §102 requires every element in one reference, this reference anticipates the dependent‑claim elements but not claims 1/11 as a whole; it is the primary §103 combination reference against 1/11 (same inventors/assignee, same CS/PN-modulation core).
• Note: Same inventive entity lineage (Kinget) as the subject patent — expect this to be the central prosecution/validity reference.

3.2 US 8,750,438 B2 / US 2012/0314822 A1 — Interference resistant compressive sampling

• Full citation / URL: NewLANS, Inc.; US 2012/0314822 A1 pub. 2012‑12‑13; granted US 8,750,438 B2 (2014‑06‑10). https://patents.google.com/patent/US20120314822
• Description: Random‑Modulation‑Pre‑Integration (RMPI) compressive sampler: the input is multiplied by a PN/chipping sequence (mixer 502), integrated (504), and sample‑and‑held (506); the measurement operator is built from antipodal/random chipping sequences; reconstruction via L1 (Lasso) or orthogonal matching pursuit. Interference cancellation is pushed into the RF front end.
• Potential §102 relevance:
  • Claims 1/11 (modulator that modulates a signal at successive time intervals based on a pseudo‑random chipping sequence; CS recovery) — strong element‑level disclosure, but it is a single‑channel, single‑antenna sampler with no per‑antenna array and no DoA; so again dependent‑claim and §103 exposure, not full anticipation of 1/11.
  • Claims 6/16 (OMP) — expressly disclosed.
  • Claims 3/13 (integration/combination; "connection that combines" signals) — arguably disclosed at the integrator.

3.3 US 9,413,420 B1 — Adaptive interference removal for compressive signal detection and reconstruction in real time

• Full citation / URL: Kong, C. (X.) et al.; HRL Laboratories, LLC; filed 2011‑04‑20, granted 2016‑08‑09. https://patents.google.com/patent/[US9413420B1](/patent/US9413420B1)/en
• Description: Compressively senses a wideband input by mixing with a random sequence and sampling at a sub‑Nyquist rate, then projecting the samples onto a set of sinusoids to locate a dominant signal's center frequency; iteratively removes detected signals and re‑detects. Explicitly discusses PN/random‑sequence modulation front ends (Tropp; Mishali & Eldar).
• Potential §102 relevance: Claims 1/11 (random‑sequence modulation + CS detection/reconstruction) at element level; again frequency‑domain and single‑channel, so §103, and possibly §102 only against dependent claims dealing with CS processing (claims 6/16).

3.4 US 8,594,603 B2 — Systems and methods for cancelling interferers in a receiver

• Full citation / URL: Kinget, P. R. & Balankutty, A.; The Trustees of Columbia University; filed 2010‑11‑08, granted 2013‑11‑26. https://patents.google.com/patent/[US8594603B2](/patent/US8594603B2)/en
• Description: Receiver with a main path and an alternate path; the alternate path mixes with a phase‑shifted LO, and the two paths are combined to cancel an interferer (gain/phase matching calibration disclosed).
• Potential §102 relevance: Claims 9/19 (modulating based on a desired phase shift) and 10/20 (combining/CBF‑like processing) at element level — §102 candidate against the dependent claims; the phase‑shifted‑LO interferer‑cancellation architecture is the same applicant's prior art and supports a §103 attack on claims 9/10, 19/20.

3.5 US 2004/0266356 A1 — Multiple antenna apparatus and method to provide interference detection and cancellation

• Full citation / URL: Javor, R. D.; pub. 2004‑12‑30 (filed 2003‑06‑27).
• Description: Multiple‑antenna apparatus for detecting and cancelling interference.
• Potential §102 relevance: This is one of the few cited references that involves multiple antennas; worth a full‑text claim‑chart review against claims 1/11 (plurality of antennas + interference/DoA handling). On the title alone I cannot assert anticipation; flag for detailed review.

3.6 General compressive‑sensing / sparse‑sampling references (element‑level §102/§103 candidates)

• US 8,457,579 B2 — Technion R&D Foundation; "Efficient sampling and reconstruction of sparse multi‑band signals"; filed 2009‑02‑18, granted 2013‑06‑04. (lower confidence — could not retrieve full text) CS sampling/reconstruction architecture → element‑level relevance to claims 1/11, 6/16.
• US 8,836,557 B2 — Technion R&D Foundation; "Sub‑Nyquist sampling of short pulses"; filed 2010‑10‑13, granted 2014‑09‑16. (lower confidence) Sub‑Nyquist/CS sampling → claims 1/11, 4/14.
• WO 2013/152022 A1 — InterDigital Patent Holdings; "Method and system for wideband spectrum scanning employing compressed sensing"; filed 2012‑04‑03, pub. 2013‑10‑10. CS‑based spectral scanning → element‑level relevance to claims 1/11 (CS scanning) but frequency‑domain.
• US 2011/0221518 A1 (Allegro) switched‑capacitor notch filter; US 8,166,084 B2 (Intersil) calibration of adjustable filters; US 9,065,504 B2 (Broadcom) transmitter front end notch filter → filter/notch component art, generally only relevant to any filter‑reciting claims (none are express in 1–20), so mostly background.
• US 5,001,441 A — Allied‑Signal; "Operational transconductance amplifier programmable filter"; filed 1989‑10‑30, granted 1991‑03‑19. A variable‑transconductance programmable filter → potential §102 candidate against claim 7/17 (modulator is a vector modulator built from variable transconductors), if combined with the independent‑claim structure.
• US 5,629,956 A (Omnipoint, 1997) noncoherent serial correlation of a CPM signal; US 5,640,416 A (Comsat, 1997) digital downconverter/despreader for DSSS → PN‑correlation/despreading art; element‑level relevance to "pseudo‑random numbers" modulation (claims 1/11) and mixer/downconversion (claims 2/12).

4. Full citation inventory (all 57 examiner citations in the patent file)

Dates are the patent's listed publication dates (grant/publication); where a US pre‑grant publication is cited I use its publication date. "§102 target" indicates claim(s) for which the reference is at most an element‑level or dependent‑claim §102 candidate; no single reference anticipates independent claims 1 or 11.

A. Compressive sensing / PN‑modulation / interference — core

B. Same‑applicant / receiver & interferer‑cancellation art

C. Component‑level / receiver front‑end art (mostly background; at most dependent‑claim relevance)

(The patent's own citation list contains 56 citations in one rendering and 57 in another; US 8,571,504 B2 appears only in the "Patent Citations (57)" rendering, and US 2004/0266356 A1 and US 2008/009? entries shift accordingly. Lists that differ by one document are common between the "Citations" and "Patent Citations" renderings on Google Patents; this is an artifact of the two renderings, not an identifier error.)

5. Bottom line

1. No cited reference anticipates independent claim 1 or independent claim 11 in full. Each lacks at least the "plurality of receiver paths, each with a unique antenna" + "sum across receiver paths per time point" + "recover the direction of arrival in physical space" combination. The cited CS references (NewLANS US 8,750,438 / US 2012/0314822; HRL US 9,413,420; Technion US 8,457,579 and US 8,836,557; InterDigital WO 2013/152022) are frequency‑domain, single‑channel CS systems.

2. The strongest and most dangerous reference is US 2017/0026066 A1 (Kinget/Columbia), with its family US 9,762,273 B2 and US 11,374,599 B2. It shares the CS + PN‑modulation + mixer/ADC + sparse‑recovery architecture and (because it shares the Kinget inventive lineage) is the natural §103 combination basis for claims 1/11 and a direct §102 reference for the dependent claims 2/12, 4/14, and 6/16.

3. Dependent claims carry the realistic §102 exposure: claims 2/12 (downconverting mixer — Kinget US 2017/0026066, Comsat US 5,640,416); 3/13 (current summing/combination — RMPI integrator art); 4/14 (ADC — Kinget, and most component references); 5/15 (DSP — ubiquitous); 6/16 (OMP/greedy CS recovery — NewLANS, HRL, Technion); 7/17 (vector modulator — Allied‑Signal OTA US 5,001,441); 8/18–10/20 (phase shifter / desired phase shift / CBF — Kinget US 8,594,603 and general beam‑forming art).

4. Two items deserve a full‑text claim chart before I would assert §102: (a) US 2004/0266356 A1 (Javor) because it is the only cited reference framed around multiple antennas for interference detection/cancellation; and (b) the Technion US 8,457,579 / US 8,836,557 and EP 2,662,706 (SELEX ES) references, on which my retrieval was incomplete, so my descriptions are from the patent's citation metadata and general knowledge and are lower confidence.

Confidence statement: The identification of the patent, its claims, and its complete citation list is high confidence (taken from the authoritative full text). The §102 assignments above are analytical candidates, not examiner conclusions — a true §102/§103 determination requires full element‑by‑element claim charts against each reference's complete specification, which I have not performed for the lower‑tier component references. Three references (US 8,457,579; US 8,836,557; EP 2,662,706) could not be fully retrieved during this session; treat their descriptions as provisional.

I'll research the key prior-art references cited on this patent's page to ground the obviousness analysis.

Obviousness Analysis — U.S. Patent No. 11,402,458 B2

"Circuits and methods for using compressive sampling to detect direction of arrival of a signal of interest"
Assignee: The Trustees of Columbia University in the City of New York · Inventors: Bajor, Wright, Haque, Kinget · Priority: 2018‑05‑22 (Prov. 62/674,930) · Filed: 2019‑05‑22 · Granted: 2022‑08‑02

Framework: Because the effective filing date is after 2013‑03‑16, the AIA version of 35 U.S.C. §§ 102/103 governs. All references analyzed below have priority/filing dates well before 2018‑05‑22 and are therefore available as § 102(a)(1)/(a)(2) prior art.

Scope note / caveat: The "Prior Art" (Citations) section of the Google Patents page for US11402458 lists ~56–57 U.S./foreign references, but the page content I retrieved is truncated mid‑list (the last fully readable entries are around US20130136154A1). I therefore base the analysis on the fully readable cited references and flag where I supplement with a closely related document found by search. I have not independently verified every reference's complete disclosure beyond the text retrieved; statements about a reference's disclosure are grounded in the quotes/URLs cited below.

1. The claims at issue (literal text)

Claim 1 (independent, circuit):

• a plurality of receiver paths, each comprising:
  • a unique one of a plurality of antennas receiving the signal of interest and producing a received signal; and
  • a modulator that (i) receives a modulator input signal based on the received signal, (ii) modulates at multiple points in time based on different ones of a plurality of pseudo‑random numbers, and (iii) produces a plurality of modulated output signals for the path;
• a summer that, for each point in time, sums across each receiver path one of the modulated output signals corresponding to that point in time → a plurality of sum signals; and
• a hardware processor receiving the sum signals and performing a compressed‑sensing recovery algorithm to recover the direction of arrival … in physical space with respect to the plurality of antennas.

Claim 11 is the method counterpart. Claims 2–10 / 12–20 add: downconverting mixer; current‑summing summer ("connection"); ADC; DSP; OMP; vector modulator; phase shifter; re‑use of the modulator for a desired phase shift; and a conventional beam‑forming (CBF) process.

The only claim element that arguably distinguishes the claims from the cited CS art is the spatial/angular framing: the CS measurements are taken across antenna paths and the recovered sparse vector is a DoA in physical space (the "angular spectrum"), rather than the frequency spectrum. Everything else — PN/modulator per branch, per‑time‑point summing, ADC, DSP, OMP — is squarely in the cited CS receiver art.

2. Hypothetical person having ordinary skill in the art (PHOSITA)

A PHOSITA here is an RF/mixed‑signal engineer (M.S. + ~3–5 years) with working knowledge of (a) array/antenna signal processing and DoA estimation (delay‑and‑sum and phase‑shift beamforming, array factor), and (b) compressive sensing / analog‑to‑information converters (RMPI/random‑demodulator, modulated wideband converter, PN modulation, sub‑Nyquist sampling, OMP/L1 recovery). This is a narrow, well‑populated intersection; the patent's own background concedes that CBF DoA scanning is conventional ("slow and require[s] significant power").

3. Element mapping to the cited references

4. Primary § 103 combination (strongest ground)

Ground A: Kinget '066 + Javor '356 + NewLans '822 (optionally + HRL '420)

Why each reference is analogous art: All three are RF receiver / analog‑to‑information converters addressing the same problem space the patent addresses — detecting and locating unwanted signal energy efficiently (fast, low‑power) using compressive/PN‑based sampling. They are from the same field of endeavor and reasonably pertinent to the problem.

What the combination teaches:

• Kinget '066 (Columbia; common assignee, overlapping inventors) discloses the core architecture: downconvert an RF signal with an LO; in each of multiple branches multiply by a distinct PN sequence; combine/convert with ADCs; and use a hardware processor performing support recovery on the CS measurements (see https://patents.google.com/patent/US20170026066A1/en). It expressly frames the mechanism as a "compressed sampling (CS) time‑segmented quadrature analog‑to‑information converter."
• Javor '356 supplies the multi‑antenna, per‑path front‑end architecture: "a multiple antenna and multiple receiver apparatus … to improve a radio's resilience to multi‑path fading and interfering signals," each antenna feeding a dedicated LNA→mixer→filter→ADC path into a processor (https://patents.google.com/patent/US20040266356A1/en).
• NewLans '822/US8750438 supplies (i) the recognition that a random antipodal/PN sequence is "universally incoherent with any time‑frequency basis" — a teaching a PHOSITA would read as generic to any sparsifying dictionary, including the angular dictionary — and (ii) an explicit recitation that "greedy algorithms such as orthogonal matching pursuit will be employed" for recovery (https://patents.google.com/patent/US20120314822A1/en).

Motivation to combine (KSR rationales):

1. Same problem, known technique: The patent's own background states DoA detection by CBF "is slow and requires significant power," and that swept CBF scan energy scales as N² (Eq. 9). CS was the known technique for reducing the number of measurements — a PHOSITA would apply it to the angular scan to cut N scans to m ≪ N scans.
2. Design incentive / predictable result: NewLans teaches that random modulation is a universal incoherent measurement; applying it across antenna branches (instead of frequency bins) to form "composite antenna patterns" is a predictable application of a known technique to a known field. The patent itself states the composite‑pattern concept (Eq. 12) as the natural result.
3. Same assignee/inventor overlap: Kinget '066 is Columbia work by the same research group; a PHOSITA (and certainly these inventors) would look to it and to Javor's multi‑antenna front end.
4. Reasonable expectation of success: PN‑modulated CS measurement + OMP reconstruction were demonstrated (HRL '420; Kinget '066; NewLans '822); substituting an antenna array for the single antenna to make the sparse domain spatial requires only routine engineering, since the array's angular response and array factor are well‑characterized (see the patent's Eqs. 11–12).

Result: Every element of claim 1 is disclosed or rendered obvious; the sole "new" element — labeling the recovered sparse vector a DoA in physical space — is the predictable consequence of applying the known random‑modulation CS measurement across the known multi‑antenna array.

Ground B: HRL '420 + Javor '356 + NewLans '822 (+ Technion '579/'557)

• US9413420B1 (HRL) claims and describes a CS front end that "mix[es] the input signal x(t) … with a random sequence" and samples "at a sub‑nyquist rate," then reconstructs in real time, including an interference‑removal loop (https://patents.google.com/patent/US9413420B1/en). This is squarely the modulator + summer + CS‑recovery core of claim 1 (and supports claims 6 and 10's interferer‑suppression motivation).
• Javor '356 again supplies the multi‑antenna receiver‑path architecture.
• US8457579B2 / US8836557B2 (Technion, Mishali/Eldar) — the modulated wideband converter / sub‑Nyquist sampling front ends — reinforce that PN modulation + combining + sub‑Nyquist sampling is a standard CS acquisition "front end."
• Motivation: same as Ground A (speed/power of DoA scanning; universality of random modulation).

5. Ground C — the closest spatial‑CS art (supplement found by search)

US9660693B1 (HRL, "Spatio‑temporal signal monitoring," Ni & Rao; filed 2014‑07‑10, granted 2017‑05‑23) is highly material even though I could not confirm it appears in the truncated citation list: it discloses an antenna array with beamforming → random mixing → CS sub‑Nyquist sampling → identification of a "signal of interest: frequency band, angle of arrival," with a hardware processor generating estimates of the angle‑of‑arrival (see https://patents.google.com/patent/US9660693B1/en and the USPTO copy at patentimages.storage.googleapis.com).

• Combination C: US9660693B1 + Kinget '066 + Javor '356. HRL '693 already ties CS measurements to an antenna array and DoA/angle‑of‑arrival output; Kinget '066 supplies the per‑branch unique‑PN‑sequence modulator architecture and support recovery; Javor supplies the per‑antenna receiver paths. This combination addresses the only remaining element (spatial DoA recovery) with an express teaching, making claim 1 at minimum obvious and arguably anticipated in its broader aspects.
• Note: I flag that I did not verify whether US9660693B1 is among the examiner‑cited references on the truncated list; it is nonetheless § 102(b) prior art (granted 2017).

6. Dependent claims

None of claims 2–10 recites a structure or step absent from the combination of RF front‑end art (Javor, Kinget), CS recovery art (NewLans, HRL), and conventional beamforming (admitted prior art). The dependent claims are therefore obvious over Grounds A/B/C applied to the independent claim.

7. Anticipated patentee rebuttals and responses

1. "Prior‑art CS is frequency‑domain, not spatial." — Weak. NewLans '822 expressly teaches the random antipodal sequence is universally incoherent with any basis; the angular/array‑factor dictionary is just another sparsifying basis. HRL '9660693 (and the array literature) expressly performs CS for angle‑of‑arrival. The patent's own specification (Eqs. 10–12) frames the angular problem in the same standard CS formalism as the frequency case, confirming it is a routine extension.
2. "Motivation was lacking; it was only the inventors' insight to unify CBF and CS in one DSIC." — The independent claim does not require a unified/dual‑mode architecture; the DSIC unification and single‑bit vector modulators appear in the specification/figures, not in claim 1. Claim 1 reads on the generic CS‑DoA pipeline and is therefore obvious even if the specific DSIC implementation is clever. (If the patentee seeks to rely on the DSIC/dual‑mode features for non‑obviousness, they are not claimed in a way that distinguishes over the cited art.)
3. Common‑ownership disqualification of Kinget '066. — As commonly owned § 102(a)(2) art, US20170026066A1 might be disqualified under AIA § 102(b)(2)(C). However, because it was published 2017‑01‑26, before the 2018‑05‑22 priority date, it is independently § 102(a)(1) "printed publication" art, to which the (b)(2)(C) exception does not apply. So it remains available for § 103. (Grounds B and C do not depend on Kinget '066 at all.)

8. Secondary considerations (for completeness)

I have no evidence of record here of objective indicia (unexpected results, industry praise, licensing, copying). Columbia has published DSIC work (e.g., the ISSCC/papers around the chip described in FIG. 5), so the patentee may point to commercial success or praise; any such argument would need a nexus to the claimed subject matter. On the record available, no nexus is established, and the DSIC‑specific features that might support a nexus are not captured by claim 1.

9. Bottom line

• Claim 1 / 11 are obvious under 35 U.S.C. § 103 over:
  • Kinget US20170026066A1 + Javor US20040266356A1 + NewLans US20120314822A1 (Ground A), or
  • HRL US9413420B1 + Javor US20040266356A1 + NewLans US20120314822A1 (Ground B), or
  • HRL US9660693B1 + Kinget US20170026066A1 + Javor US20040266356A1 (Ground C), each with the Technion references (US8457579B2 / US8836557B2) as cumulative evidence of the PN‑modulated, sub‑Nyquist CS front end.
• Claims 2–10 / 12–20 are obvious as routine design choices over the above, with OMP (claim 6/16) expressly taught by NewLans and the CBF features (claims 8–10/18–20) admitted prior art.
• Key residual uncertainty: whether the examiner‑cited list (partly truncated in my source) already includes the HRL spatio‑temporal CS reference (US9660693B1), and the precise disclosure scope of references I could only review at the snippet/abstract level. If the spatial‑CS reference is not of record, Ground C supplies a reason to add it.

Primary sources used (Google Patents / cited pages):

• US11402458B2 — https://patents.google.com/patent/US11402458/en
• US20170026066A1 (Kinget, Columbia) — https://patents.google.com/patent/US20170026066A1/en
• US9413420B1 (HRL/Kong) — https://patents.google.com/patent/US9413420B1/en
• US20120314822A1 / US8750438 (NewLans) — https://patents.google.com/patent/US20120314822A1/en
• US20040266356A1 (Javor) — https://patents.google.com/patent/US20040266356A1/en
• US9660693B1 (HRL, spatio‑temporal CS / AoA) — https://patents.google.com/patent/US9660693B1/en

If you'd like, I can produce a downloadable claim chart mapping each element of claim 1 (and each dependent claim) to specific paragraphs/figures of each reference, or run a focused search to confirm whether US9660693B1 and the Technion references appear in the examiner‑cited "Prior Art" list.