US 10352797

I'll run targeted searches for this patent number against USPTO and CAFC sources.

Let me run a few more targeted searches for USPTO records and any litigation/docket activity.

Search notes and caveats (read first)

• I searched for the literal string 10352797 and for "10,352,797" against USPTO/PTAB, Federal Circuit, and general patent sources. My searches surfaced a large amount of unrelated material, most notably a German application DE 10352797 A1 (a gas-pressure measuring device using an amperometric sensor — a completely different jurisdiction, subject matter, and family). That document is not the patent you asked about; I am flagging it so it is clear I did not conflate the two.
• No CAFC 2026 docket activity, no district court litigation, and no PTAB (IPR/PGR) proceeding involving US 10,352,797 was found in my searches. I could not confirm this as a negative with high confidence, because my tools did not return an authoritative Docket Navigator / PACER / USPTO Patent Center record directly. Treat "no litigation found" as a search limitation, not a verified absence.
• The bibliographic and claim details below are grounded in the full text of US10352797 as fetched from Google Patents (https://patents.google.com/patent/US10352797/en), which is authoritative here, plus corroborating hits on Justia Patents (https://patents.justia.com/patent/10352797) and the granted PDF (http://patentimages.storage.googleapis.com/8f/d7/4b/242a188deedb04/US10352797.pdf).

---

Bibliographic summary — US 10,352,797 B2

Abstract (as issued)

A tunable and resettable shock sensor using a parallel dipole line (PDL) trap system is provided. In one aspect, a shock sensor includes: a PDL trap having a pair of diametric magnets separated from one another by a gap g_M, and a diamagnetic rod levitating in between the diametric magnets; and contact pads below the PDL trap, wherein the contact pads are separated from one another by a space that is less than a length l of the diamagnetic rod. A shock monitoring system is also provided that includes a network of the shock sensors, as is a method for shock monitoring using the shock sensors.

---

Plain-language overview of the independent claims

Claim 1 — the sensor itself (apparatus)

A shock sensor with three cooperating elements:

1. A parallel dipole line (PDL) trap — a pair of transversely magnetized cylindrical ("diametric") magnets spaced apart by a variable gap g_M, with a diamagnetic rod (e.g., graphite) levitating in the field between them.
2. Electrically conductive contact pads below the trap, spaced apart by a gap smaller than the rod's length l, so that if a shock knocks the rod out of the trap, the rod lands across the pads and physically shorts them together — a binary, latchable "something happened" signal.
3. A variable-gap fixture to which each magnet is affixed, letting a user move the magnets closer together or farther apart. Changing g_M weakens or strengthens the trap's confinement potential, which tunes the shock-force threshold needed to dislodge the rod.

In short: a magnetic-levitation shock detector whose trip point is adjustable by mechanical gap tuning, and whose trip state is read out as a short between pads.

Claim 5 — the monitoring system

A network of the claim-1 shock sensors (each with the variable-gap PDL trap and the shorting contact pads) plus a blockchain ledger configured to record shock-trigger events reported by that sensor network. The point is tamper-resistance: because these sensors are resettable, someone could reset a tripped sensor to hide a damaging drop; an immutable ledger preserves the record.

Claim 10 — the method of shock monitoring

A four-step method:

1. Provide at least one sensor of the claimed type (PDL trap with variable-gap fixture + shorting contact pads).
2. Vary g_M using the variable-gap fixture to control the trap's confinement potential and thereby tune the shock threshold force above which the rod is knocked loose.
3. Place the tuned sensor with a product (in-package or in a shipping container).
4. Monitor the sensor's state (practically, by watching pad-to-pad resistance).

---

Dependent claims (brief)

• Claim 2 / 8: two pads below the center of the trap.
• Claim 3 / 9: a pair of pads below each end of the trap (horizontal-shock geometry).
• Claim 4 / 15: a second set of two pads above the center, enabling reset by flipping the sensor over.
• Claim 6 / 7: the sensor network is on a LAN, connected to the blockchain ledger via the Internet.
• Claim 11: determine whether a shock-trigger event occurred (rod falls and shorts the pads).
• Claim 12: monitor resistance between the pads.
• Claim 13: record the shock-trigger event in a blockchain ledger.
• Claim 14: reset the sensor and resume monitoring.
• Claim 16: resetting is accomplished by flipping the sensor.

---

Litigation / docket status

• No CAFC 2026 docket entry, district court case, ITC investigation, or PTAB proceeding referencing US 10,352,797 was located. IBM appears to hold and maintain this patent (4th-year maintenance fee paid January 2023) with no publicly indexed enforcement or validity challenge in the sources I could reach.
• The patent has 3 forward citations (US 11,835,333 B2 — IBM rotational oscillation sensor; US 12,135,398 B2 — IBM multiple dipole line trap system; US 12,099,997 B1 — Hoffberg tokenized fungible liabilities), none of which are litigation.

Explicit uncertainty flags

1. No authoritative USPTO Patent Center / Patent Trial and Appeal Board record was retrieved directly. Filing date, issue date, assignee, inventors, and claim text are taken from the Google Patents full text and corroborated by Justia; I did not independently verify the legal-status field ("Active") or the maintenance-fee history against the USPTO's own systems.
2. Absence of litigation is a search result, not a verified negative. A dedicated PACER/Docket Navigator or USPTO Patent Center query would be needed to state confidently that no 2026 CAFC matter exists.
3. Internal specification inconsistency worth noting for claim-construction work: the Abstract and claims 1/2 place the pads "below" the trap (centered), while the detailed description states the sensor "uses (electrically conductive) contact pads under opposite ends of the PDL trap," then describes FIG. 6B as two pads "beneath a center of the PDL trap." Claims 3/9 separately cover the "pair below each end" geometry. This mismatch between the descriptive text and the drawings/claims is characteristic of this patent and would be a natural focus in any future validity or infringement dispute.
4. The examiner identification (Tran M. Tran) comes from a third-party examiner-indexing site and is lower confidence than the bibliographic data above.

I'll search for litigation records involving US Patent 10,352,797.

The generic searches returned mostly irrelevant material. Let me target litigation-specific databases directly.

Let me verify through additional litigation-specific sources.

Result: No litigation found involving U.S. Patent No. 10,352,797

I searched for litigation records specific to U.S. Patent No. 10,352,797 ("Tunable shock sensor with parallel dipole line trap system," IBM, filed/priority 2017‑10‑10, granted 2019‑07‑16). I found no district court actions, no ITC investigations (Section 337), no CAFC appeals, and no PTAB post‑grant proceedings (IPR/PGR/CBM) naming this patent.

Requested case fields

What I checked and what the record shows

1. Google Patents litigation face for US10352797B2 (the authoritative text you provided) — the "Cited By" and "Families Citing this family" sections list later patents that cite the '797 patent, not lawsuits. Those are citations, not assertions.

   • Notably, the citing references include IBM's own later work (US 11,835,333 B2 "Rotational oscillation sensor with a multiple dipole line trap system"; US 12,135,398 B2 "Multiple dipole line trap system") and a third‑party citation by Shockwatch, Inc. (US 11,041,770 B2 "Impact indicator"). A citation by Shockwatch is not evidence that Shockwatch sued or was sued on the '797 patent.

2. Litigation-focused searches (Unified Patents portal/litigation pages, docket-oriented queries, CAFC/PACER-style queries) returned no docket for 10,352,797. Searches instead surfaced unrelated cases and unrelated patents — e.g., the Orange Electronics v. Autel TPMS matter (jury award of $6,616,397; patents at issue were U.S. 8,031,064-type claims, and later patents 10,397,374 / 11,849,977 appear in other disputes). None of these involve US 10,352,797 and should not be conflated with it.

3. International family — the same invention was filed via PCT/IB2018/057681 (WO2019073336A1), with national counterparts JP6998458B2, CN111194476B, DE112018004483T5, and GB2579531B. I found no opposition, nullity, or revocation litigation concerning these counterparts either.

Important limitations (so you can calibrate confidence)

• I could not directly query PACER, Docket Navigator, Lex Machina, or the Federal Circuit's docket in this session; my findings rest on publicly indexed web/aggregator results and the Google Patents record you supplied. Full-text docket databases occasionally include recently filed complaints that are not yet web-indexed.
• A negative finding is not proof of absence. If you need a certification-quality answer, run a docket-specific search on Docket Navigator / Lex Machina / CourtListener/RECAP against the exact string "10,352,797" and confirm no active or terminated case exists.
• Note also that the '797 patent is a shipping/impact-monitoring sensor patent held by IBM. IBM's practice for such patents has generally been defensive/licensing rather than asserting them in district court, which is consistent with (but does not by itself prove) the absence of litigation.

Bottom line

Based on the sources available to me, there is no known litigation involving U.S. Patent No. 10,352,797 as of this date. If you want, I can draft a formal "no litigation found" declaration, or attempt a deeper docket-only search if you can supply access to a litigation database.

Proceedings overview

Zero AIA trial proceedings are on file for US 10,352,797. The structured PTAB block in this prompt returns no IPR, PGR, or CBM for the patent, and my own web searches (PTAB-hosted petition documents, decisions, and aggregator indexes) surfaced no proceeding naming the '797 patent or its PDL-trap shock-sensor claims. Breakdown by status: 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denials.

Bottom line for a defendant: there is no PTAB-based shortcut here, and no estoppel either. All 16 claims are untested — issued 2019-07-16, still active, 4th-year maintenance fee paid 2023-01-17, anticipated expiration 2037-10-10. A demand letter citing any of claims 1–16 is not weakened by any prior Board ruling, but neither is the patent "hardened" by surviving an IPR. You get a clean slate: every § 102/§ 103/§ 112/§ 101 ground is still available to you, in the district court or in a first-filed IPR.

Because the source data is a negative, I want to be explicit about the epistemic status: ODP ingestion gaps and indexing limits mean "no proceedings" is the default and strongly indicated answer, not a mathematical certainty. Independent verification is a two-minute exercise — see Recommended next steps.

---

Proceedings

None to report. No petitioner has filed against US 10,352,797, so there is no proceeding number, panel, institution decision, Final Written Decision, settlement, or Federal Circuit appeal to describe. I will not invent one.

For completeness on the things that are on file and are frequently confused with PTAB activity:

• Post-grant "citations" that are not proceedings. The Google Patents record lists US 11,835,333 B2 ("Rotational oscillation sensor with a multiple dipole line trap system," IBM, 2023-12-05) and US 12,135,398 B2 ("Multiple dipole line trap system," IBM, 2024-11-05) as citing the '797 patent. These are later IBM PDL-trap family filings, not third-party challenges. Similarly, US 11,041,770 B2 (ShockWatch, Inc., 2021-06-22) cites the '797 disclosure — a competitor citing it, not attacking it.
• Foreign counterparts, no PTAB. The '797 family includes GB2579531B (granted 2020-11-11), JP6998458B2 (granted 2022-02-04), CN111194476B (granted 2023-07-25), DE112018004483T5 (pending), and WO2019073336A1 (status: ceased). A validity ruling on a counterpart is not a PTAB proceeding, but if you find one it is useful comparative material — I found none.
• Prosecution history. The examiner's cited art includes US 5,168,183 (Whitehead), US 5,315,247 (Kaiser), US 5,623,248 (Min), US 6,472,864 (Emo), US 7,810,373 (Seagate, MEMS shock sensors), EP 846952 A1 (Oki Electric), and US 9,316,550 (STMicroelectronics, bistable shock sensor) — plus the non-patent literature Gunawan 2015 (Appl. Phys. Lett. 106, 062407), Gunawan 2017 (J. Appl. Phys. 121, 133902), McDonald 1999, and Crosby et al. 2016 (blockchain). That is your starting § 102/§ 103 art set, and notably it is all art the patent owner itself put before the Office — which cuts both ways: no § 325(d) "already considered" excuse if you cite it, since the examiner did consider it.

---

Strategic summary

Claim status. Every claim of US 10,352,797 is UNTESTED. Nothing is CANCELED, nothing is SUSTAINED by any tribunal. The claim set is: independent claim 1 (PDL trap with variable gap g_M + electrically conductive contact pads spaced less than rod length l, rod shorting the pads on shock, variable-gap fixture tuning the shock threshold); dependent claims 2–4 (pad placement below center, pairs below each end, and pads above center for flip-reset); independent claim 5 (network of such sensors + blockchain ledger) with dependents 6–9; and independent claim 10 (monitoring method: vary g_M to tune threshold, place with product, monitor) with dependents 11–16. Note that claims 5 and 10 were only added/narrowed to recite the variable-gap-fixture limitation during prosecution — the CN counterpart (CN111194476A/B) shows a broader claim set (its claim 1 has no gap-fixture element, and "gap g_M is variable" appears only as a dependent claim). That asymmetry is a real prosecution-history and family-prosecution-disclaimer lever worth preserving.

Estoppel landscape. Because no IPR/PGR was ever instituted, § 315(e)(2) estoppel is empty. As a defendant today you may raise any prior-art ground under §§ 102/103, any § 112 written-description/enablement/definiteness theory, any § 101 challenge (the blockchain-ledger claims 5–9 and the method claims 10–16 have meaningful Alice exposure), and any inequitable-conduct or derivation theory, in district court, without worrying that an earlier petitioner's reasonably-could-have-raised grounds were foreclosed. The corollary: if you do file an IPR and lose, you will hand the patent owner the estoppel weapon — the first petitioner bears the full cost of that asymmetry.

What is foreclosed is not a matter of estoppel but of statute:

• § 315(b) time bar — if you or a real party in interest/privy were served with an infringement complaint more than one year before your petition, you are barred from filing an IPR (joinder under § 315(c) is the only escape).
• PGR — unavailable; the 9-month § 321(c) window from issuance (2019-07-16) closed in 2020, and PGR requires a § 112/§ 282(b)(3)(C) or first-to-file § 102(a)(2) basis anyway.
• CBM — unavailable on two independent grounds: the '797 patent does not claim a financial product or service (it is a magnetic shock sensor plus a distributed ledger), and the CBM transitional program sunset for new petitions on 2020-09-16.

Pattern signals. No petitioner has filed even once, so there is no repeat-petitioner pattern, no multi-petition serial attack, and no defensive aggregator (Unified Patents, RPX IPR fund, etc.) anywhere in the chain. The patent owner is IBM, the original and current assignee, with inventors Oki Gunawan and Norma E. Sosa — not a litigating NPE. The whole PDL-trap portfolio (US 8,895,355; 9,093,377; 9,236,293; 9,263,669; 9,978,493; 10,022,350; 10,234,286; 10,352,797; 10,460,863; 10,564,175; 10,564,303; 10,584,966; 10,593,456; 11,835,333; 12,135,398) is internal IBM R&D that has been licensed rather than enforced — Gunawan's own seminar materials describe the PDL Hall system as licensed to a semiconductor instrument company. There is no Federal Circuit appeal to track, and the patent owner has shown no appetite for PTAB appeals on this family. The practical inference is that this patent is most likely to reach you as a shelf asset in an IBM divestiture or portfolio sale, at which point the buyer's NPE would be free to assert claims that no one has ever stress-tested.

---

Recommended next steps

If you are a defendant and want a decisive invalidity posture:

1. Confirm the negative before you rely on it. Run the patent number through the PTAB End-to-End (E2E) system and the "Proceedings" / "Transactions" tabs in USPTO Patent Center, and cross-check CourtListener's docket search for any Federal Circuit appeal touching the '797 patent. If E2E also returns nothing, the "no PTAB activity" conclusion is safe to plead around.
2. File first if you are inside the § 315(b) window. With zero estoppel and zero Board precedent on these claims, a well-funded IPR on claim 1 is a genuinely open question. The strongest candidate art is the art already in the file — Gunawan 2015 and Gunawan 2017 for the PDL trap and camelback potential, US 7,810,373 (Seagate, MEMS shock sensors) and US 9,316,550 (STMicroelectronics, bistable shock sensor) for threshold-triggered shock detection, and US 5,623,248 (Min, miniature electromagnetic impact sensor) and EP 846952 A1 (Oki Electric, shock/acceleration sensor) for contact-closure shock sensing — plus the IBM PDL portfolio itself (US 9,978,493, "Parallel dipole line trap with variable gap and tunable trap potential") as § 102(a)(1) art against the variable-gap-fixture limitation of claim 1, since 9,978,493 shares the same priority chain and issued 2018-05-22.
3. Press the § 101 angle on claims 5–9 and 10–16. The blockchain-ledger limitation in claim 5 ("a blockchain ledger configured to record any shock-trigger events") is a generic, conventional data-recording element bolted onto a sensor network, and the method claims 10–16 recite monitoring and recording steps that map cleanly onto Alice step two. This is the highest-value, lowest-cost invalidity theory and it does not depend on finding new art.
4. Preserve the family-prosecution record. The narrower US claim 1 (with the variable-gap-fixture element) versus the broader CN111194476 claim 1 (gap variation only in a dependent claim) is ammunition for a § 112 or claim-differentiation argument. Pull the CN and JP file wrappers via the Global Dossier links on the Google Patents page before the patent owner can characterize the difference.
5. Do not build a defense on PTAB history, because there is none. If opposing counsel asserts that the patent has been "validated" by the Board, that is false — no panel has ever looked at these claims.

If you are the patent owner reading this: the absence of any IPR on a 2019 IBM sensor patent is a monetization-ready posture, but the untested claim 1 is also the softest target in the portfolio — expect the first serious defendant to file on the Gunawan publications and the IBM variable-gap patent, and consider whether claim 1's variable-gap-fixture limitation is narrow enough to survive a § 103 combination of Gunawan 2015 with a conventional mechanical adjustment mechanism.

I'll research the assignment record for US 10,352,797, starting with the USPTO Assignment Center and cross-referencing legal events already present in the authoritative full text.

Let me verify whether any post-issuance assignments exist and check for NPE patterns in the chain.

Assignment & Ownership Analysis — US 10,352,797 B2

Scope note / source limitation. I was able to confirm the recording via the legal-events block in the authoritative Google Patents full text you supplied, which reproduces the USPTO assignment metadata for the single recorded conveyance. I was not able to open the live Assignment Center record (assignmentcenter.uspto.gov / assignment.uspto.gov) for reel/frame 043828/0470 in this session, so the correspondent-of-record field could not be captured. I flag every place that matters below rather than infer it. Verify at: https://assignmentcenter.uspto.gov/ (search by patent number 10352797) or https://assignment.uspto.gov/patent/index.html.

---

Inventors

Unusual-pattern check — no red flags. Both inventors are career IBM research personnel and named as assignors on the original assignment (see timeline). There is no evidence that either inventor left IBM within 12 months of the 2017-10-10 filing, and no inventor-side assignment or departure is recorded. Gunawan in particular remained at IBM Watson past 2020 (awards and publications). This is the opposite of the "inventors depart → portfolio fire-sale" pattern.

---

Original assignee

• Entity on the issued patent: International Business Machines Corporation, New Orchard Road, Armonk, NY 10504 (US).
• Primary line of business: Diversified enterprise technology — hardware, software, cloud, and a large research/licensing arm (IBM Research; IBM's patent portfolio is a revenue line via licensing).
• Does IBM ship a product embodying these claims? No. The claims cover a magnetic parallel dipole line (PDL) trap shock sensor (claims 1, 5, 10). This is a research-instrumentation invention from IBM Watson (Gunawan's PDL-trap program); there is no commercial IBM product identified that practices a PDL-trap shock sensor. The PDL-trap family is research/defensive IP, not a product line.
• Current status: Operating; solvent. 4th-year maintenance fee paid 2023-01-17; legal status listed Active; anticipated expiration 2037-10-10. No bankruptcy, no acquisition, no dissolution.

---

Assignment timeline

Exactly one (1) assignment is recorded against this patent. The Google Patents legal-events block shows a single "AS – Assignment" entry and no subsequent conveyance of any type (no change of name, no security interest, no merger, no license, no release).

• 2017-10-06 → 2017-10-09 (inventor execution dates; Gunawan signing and Sosa signing) / recorded 2017-10-10 — Reel 043828 / Frame 0470
  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)"
  • Assignor: Oki Gunawan; Norma E. Sosa (joint inventors)
  • Assignee: International Business Machines Corporation, New York (Armonk, NY)
  • Correspondent: Not captured — the correspondent / submitting-attorney field was not retrievable from the sources reached in this session. Do not assume an in-house IBM correspondent without checking reel 043828/0470 directly. (If retrieved, this single entry would be the baseline against which any later chain is compared; a lone appearance is not a recurrence signal.)
  • Context: Original inventor→employer assignment. Standard employment/obligation-to-assign transfer, recorded on the same day the application was filed. Not a fire-sale, reorg, securitization, or transfer-to-asserter.

No post-issuance assignment exists on the record I could reach. No conveyance recorded after the 2019-07-16 grant. The only other legal events are administrative (notice of allowance 2019-03-05; issue-fee receipts June 2019; patent grant 2019-06-26; maintenance fee 2023-01-17) — none of which is an ownership transfer.

Because there is no second link in the chain, the chain effectively terminates at the original assignee, IBM.

---

Timeline diagram

timeline
    title Ownership of US 10352797
    2017 : Inventors execute assignment
         : Recorded reel 043828 frame 0470
         : Filed by IBM 10 Oct 2017
    2019 : Patent granted 16 Jul
    2023 : 4th year maintenance fee paid
    2037 : Anticipated expiration 10 Oct

---

NPE / troll-pattern signals

Assessed against the single recorded link (reel 043828/0470, recorded 2017-10-10) and the absence of any later link.

1. Shell-entity transfer — not present. The only assignee is International Business Machines Corporation, an operating S&P 500 company at a real corporate HQ (Armonk, NY). No "IP / Holdings / Licensing / Ventures" suffix, no registered-agent address, no single-purpose LLC appears anywhere in the chain (reel 043828/0470).

2. Known asserter in the chain — not present. IBM is not on any of the named NPE lists (Acacia, Marathon, IV, IPNav, Wi-LAN, Conversant/Mosaid, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities). No such entity appears as assignor or assignee on reel 043828/0470 or thereafter.

3. Repeat correspondent across the chain — unclear / not assessable. With only one recorded link and the correspondent field unavailable, recurrence cannot be tested. There is no chain in which a correspondent could recur. (A single IBM in-house or outside-counsel appearance on the original assignment would not, standing alone, be a signal.)

4. Cascading transfers — not present. Zero consecutive assignments, let alone multiple LLC-to-LLC hops within 24 months. Reel 043828/0470 is the first and only entry.

5. Pre-litigation transfer — not present. No infringement suit naming this patent was identified (see prior litigation section), and there is no assignment within 6 months of any such suit. Nothing to arrange for venue or standing.

6. Bankruptcy fire-sale — not present. IBM has not filed Chapter 7/11; the patent was not sold in proceedings. Maintenance fee paid 2023-01-17 confirms continued upkeep (the opposite of an abandoned/divested asset).

7. Privateering — not present. No transfer of this patent to any NPE asserting on IBM's behalf. None of IBM's later PDL-trap continuations (US 11,835,333; US 12,135,398 — both IBM-owned) reflect an outbound transfer of the '797 family. The third-party forward citations (e.g., Shockwatch, Inc. US 11,041,770 "Impact indicator"; Hoffberg US 12,099,997) are citations, not assignments — they do not place any right in this patent.

8. Defensive aggregator — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at IBM itself.

---

Verdict

Insufficient data — specifically, only the original assignment record exists, and no post-issuance transfer was found.

Justification: the sole recorded conveyance is the inventors' original assignment to International Business Machines Corporation, executed 2017-10-06 to 2017-10-09 and recorded 2017-10-10 at reel 043828 / frame 0470, with no subsequent assignment, security interest, change of name, or release on the record. This is a clean single-link, operating-company-held chain — none of the eight NPE signals is present, and the patent remains under active maintenance by IBM (fee paid 2023-01-17; status Active; expiration 2037-10-10). The "insufficient data" verdict reflects the absence of any transfer history to analyze, not any indication of NPE activity; on the evidence reached, the patent is defensively held research IP at IBM, not an assertion vehicle.

Verification: https://assignmentcenter.uspto.gov/ → search patent number 10352797; confirm reel/frame 043828/0470 is the only assignment and capture the correspondent field that this session could not retrieve.

I'll research the prior-art references cited on the face of US 10,352,797. Let me run targeted searches for the most technically relevant families (PDL-trap references and shock/impact sensors).

Let me characterize the key shock-sensor references and the electromagnetic shock sensor art.

Prior-Art Analysis — US 10,352,797 B2 ("Tunable shock sensor with parallel dipole line trap system")

0. Patent identification (literal, no auto-correction)

I analyzed the reference record for US 10,352,797 B2 specifically — application 15/729,403, filed/priority 2017-10-10, issued 2019-07-16, inventors Oki Gunawan; Norma E. Sosa, assignee International Business Machines Corporation. The citation lists, dates, and titles below come from the authoritative full text you supplied (https://patents.google.com/patent/US10352797/en), which is the ground truth here.

Do not conflate with similar numbers. A prior search in this matter surfaced DE 10352797 A1 (a German gas-pressure/amperometric-sensor document) and unrelated US numbers such as 10,397,374 / 11,849,977 / 6,616,397 that appear in other disputes. None of those are US 10,352,797, and none are treated as its prior art. The only numbers used here are the literal patent citations printed on US 10,352,797.

---

1. §102 framework applied

Under 35 U.S.C. §102, a single reference anticipates a claim only if it discloses every element of that claim as arranged as claimed. The three independent claims each require the combination of:

• (A) a PDL trap — a pair of diametric magnets separated by a variable gap g_M with a diamagnetic rod levitating between them;
• (B) electrically conductive contact pads below the trap, spaced less than the rod length l, such that the dislodged rod shorts the pads (claims 1, 5, 10);
• (C) a variable-gap fixture affixed to each magnet, movable to vary g_M and thereby tune the shock-threshold force (claims 1, 5, 10);
• (D) (system claims 5–7) a blockchain ledger recording shock-trigger events;
• (E) (method claims 10–16) place-with-product, monitor, determine-trigger, record-in-ledger, and reset-by-flipping steps.

The crucial observation: the cited art splits cleanly into two non-overlapping groups. The magnetic-trap references disclose (A) and (C) but never (B)/(D)/(E). The shock-sensor references disclose threshold-based shock detection with an electrical (often contact-closure) readout — i.e., functional analogs of (B) — but never a PDL/diametric-magnet levitation trap (A). No single cited reference discloses the full combination, so no cited reference anticipates any claim under §102. The most relevant references are §103 combination candidates. I flag below where a reference is the strongest primary reference for a given claim.

Date check (critical for §102(a)(2)): the effective filing date is 2017-10-10. References with effective filing before that date — including US 9,978,493 (filed 2016-04-18), US 10,082,408 (2016-04-18), US 2017/0299410 (2016-04-18), US 10,031,058 (2015-08-14), and US 2019/0017965 (priority 2017-07-12) — are available as §102(a)(2) art even though several published after 2017-10-10.

---

2. Tier 1 — Most relevant art: the IBM PDL-trap family (discloses elements A and C)

Non-patent citations (Tier 1):

• Gunawan et al., "A parallel dipole line system," Appl. Phys. Lett. 106, 062407 (Feb. 2015) — discloses the PDL trap and camelback potential; supports element (A). Not anticipating (no pads, no gap-tuning for shock).
• Gunawan et al., "The one-dimensional camelback potential in the parallel dipole line trap: Stability conditions and finite size effect," J. Appl. Phys. 121, 133902 (Apr. 2017) — stability/camelback analysis; supports (A)/(C). Not anticipating.
• K. T. McDonald, "Long Rod with Uniform Magnetization Transverse to its Axis," Princeton Univ., Nov. 1999 — the 2D magnetic-field model (Eq. 1 of '797). Background physics; not anticipating.

---

3. Tier 2 — Shock/impact sensor art (discloses analogs of element B, never A)

---

4. Tier 3 — Magnetic levitation (no PDL, no shock detection)

---

5. Tier 4 — Network / blockchain and other oblique citations

---

6. Bottom-line §102 conclusions

Independent claims:

• Claim 1 (sensor): Not anticipated by any cited reference. US 9,978,493 discloses the PDL trap + variable gap + tunable confinement (elements A, C) but lacks the contact pads/shorting (B) and the shock-threshold framing. The shock-sensor references (US 6,142,007; US 9,316,550; US 6,480,081; EP 0 846 952) disclose contact/threshold detection (B analog) but lack the diamagnetic-magnet PDL trap (A). A §103 combination of US 9,978,493 (primary) + a contact-closure shock reference (e.g., US 6,142,007 or US 9,316,550) would be the natural validity attack.
• Claim 5 (system with blockchain): Not anticipated. Requires the claim-1 sensor network plus a blockchain ledger. Crosby et al. supplies only the ledger; the PDL sensor combination is missing from every cited reference. §103 candidate = US 9,978,493 + a shock-contact reference + Crosby et al.
• Claim 10 (method): Not anticipated. The tune-g_M step is disclosed by US 9,978,493, but the "place with product / monitor / shorting pads" steps are not in any single reference.

Dependent claims (no independent anticipation either):

• Claims 2, 8 (two pads below center) and 3, 9 (pair below each end): the specific contact-pad geometries are not disclosed by any cited reference; the nearest functional analogs are the reed/contact sensors (US 6,142,007, US 6,480,081) and the resistive-link sensor US 9,316,550 — §103 candidates only.
• Claim 4 / 15 (second pad set above center for flip-reset): no cited reference discloses symmetric above/below pads; §103 candidate.
• Claims 6, 7 (LAN + Internet to ledger): US 7,250,855 (sensor network) + Crosby et al.; §103 only.
• Claims 11–14, 16 (determine trigger, monitor resistance, record in ledger, reset, flip-to-reset): US 9,316,550 (latching threshold with resistance readout) is the most relevant for 11–12; Crosby et al. for 13; none disclose flip-to-reset. §103 only.

Most relevant prior art overall (ranked):

1. US 9,978,493 B2 (Gunawan/IBM) — primary reference for the PDL trap, variable gap, and g_M-tuning of confinement/fall.
2. US 9,316,550 B2 / Frangi et al. 2015 — closest art for a latching, threshold, two-contact-resistance shock sensor (and the reset/tamper concern).
3. US 6,142,007 A and US 6,480,081 B1 — contact-closure shock sensors (analog of the pad-shorting readout).
4. Gunawan et al. 2015 / 2017 and the Cao et al. PDL patents (US 8,895,355 family) — foundational PDL physics/trap.
5. Crosby et al. 2016 — blockchain ledger element (claims 5, 13) only.

---

7. Confidence and caveats

• Source authority: All citations, dates, and titles are taken from the authoritative Google Patents full text of US 10,352,797 you supplied; I did not retrieve each cited reference's own full text individually this session, so descriptions of the less relevant references (Tiers 3–4) rest largely on their titles and CPC context — treat those as moderate confidence. The Tier 1 and Tier 2 characterizations are corroborated by full-text searches (US 9,978,493, US 9,236,293, US 6,142,007, EP 0 846 952, US 9,316,550).
• "Potentially anticipates" is used strictly: §102 requires all elements in one reference. On this record no cited reference anticipates any claim. I have reported the closest §102 candidates and flagged where the analysis is really §103.
• Reference list reproduced literally. Note the patent's own citation tables are internally inconsistent (a "Citations (33)" list and a "Patent Citations (41)" list, plus several IBM PDL applications identified by their pre-grant publication numbers rather than grant numbers, e.g., US 2017/0299410 ↔ US 10,082,408, and US 2019/0017965 ↔ the later force-gauge grant). I did not auto-correct any of these.
• The internal specification inconsistency flagged in the earlier section stands: the claims/Abstract place pads "below the PDL trap"/"below a center," while the description also says pads are placed "under opposite ends." Claims 3/9 separately recite the "pair below each end" geometry. This mismatch is a claim-construction/validity focal point independent of the prior-art analysis above.
• I could not, in this session, pull the examiner's actual §102/§103 rejections or the IDS subsections; if you need the as-applied art (what the examiner actually relied on), that requires the file wrapper from USPTO Patent Center, which my tools did not reach.

I'll build on the prior-art section, but first I need to verify a few dates that are decisive for § 103 (which references are even available as prior art).

Obviousness Analysis — U.S. Patent No. 10,352,797 B2 under 35 U.S.C. § 103

---

0. Scope, and how this differs from the earlier prior-art section

This builds directly on the prior-art section already generated (which mapped the cited references into Tiers 1–4 and concluded no cited reference anticipates). I do not repeat that mapping. Instead I do three new things:

1. Run the availability screen first. § 103 analysis is worthless until you know which references are legally available as prior art against an effective filing date of October 10, 2017 (the '797 has no earlier priority; filing date = priority date = 2017-10-10). This screen changes the answer materially.
2. Build the actual combination grounds — reference-by-reference, element-by-element, with the KSR/MPEP rationale and the expectation of success for each.
3. Grade each ground by how likely it is to carry, and identify the single weakest link.

Contradiction flags against the earlier sections (as instructed)

Three corrections, all verified against the fetched sources this session:

---

1. Legal framework and the level of ordinary skill

Statute. The '797 has an effective filing date after March 16, 2013, so AIA § 102/§ 103 govern, including the § 102(b)(1) grace-period and § 102(b)(2)(C) common-ownership exceptions. There is no pre-AIA § 103(c) analysis.

Standard. Graham v. John Deere, 383 U.S. 1 (1966); KSR Int'l v. Teleflex, 550 U.S. 398 (2007). The question is whether the differences between the claim and the prior art, the scope/content of the prior art, and the level of ordinary skill make the subject matter as a whole obvious — with no requirement of an explicit teaching, suggestion or motivation in the references. MPEP §§ 2141–2144 supply the operative rationales:

• (A) known elements combined by known methods → predictable result;
• (B) simple substitution of one known element for another;
• (C) use of a known technique to improve a similar device in the same way;
• (D) applying a known technique to a known device ready for improvement;
• (F) "obvious to try" — varying a result-effective variable is routine (In re Aller, 220 F.2d 454 (CCPA 1955); MPEP § 2144.04(V));
• (G) design need / market pressure (KSR).

PHOSITA. I would define the notional person as holding an M.S. (or B.S. + 3–5 years) in applied physics, mechanical engineering, or electromechanical instrumentation, with working familiarity with (i) diamagnetic levitation/magnetic trap design, (ii) analog instrumentation and threshold/contact-closure sensing, and (iii) low-cost packaging of shipping indicators. This is a reasonably capable, mechanically sophisticated person — importantly, the '797's own specification is written to that person (it uses the full Gunawan magnetic-field model rather than explaining it).

Critical posture for § 103. Where the parties will actually fight is a single clause: "the diametric magnets are each affixed to a variable gap fixture configured to move the diametric magnets closer together or farther apart … to thereby tune a shock threshold force." Everything else in claim 1 — a PDL trap with a levitating diamagnetic rod, and a pair of pads spaced closer than the rod is long — is either squarely in the art or is mechanical arithmetic. If the variable-gap clause falls, the patent falls.

---

2. Availability screen — what a § 103 ground can actually use

2.1 The deadline that controls everything

Two IBM publications of the same subject matter family landed October 19, 2017:

• US 2017/0301445 A1 — "Parallel Dipole Line Trap with Variable Gap and Tunable Trap Potential" (application 15/131,566, filed 2016-04-18; the application expressly incorporated by reference into the '797 at col. ~7).
• US 2017/0299410 A1 — "Voltage-Tunable 1D Electro-Magnet Potential and Probe System with Parallel Dipole Line Trap."

The '797 was filed October 10, 2017. That is a nine-day margin. IBM filed the shock-sensor application just ahead of its own family publications moving into the public domain. Whether deliberate or lucky, the consequence is decisive: those publications are not § 102(a)(1) art, and they are § 102(a)(2) art only if the § 102(b)(2)(C) exception does not apply.

2.2 The § 102(b)(2)(C) exception almost certainly applies

§ 102(b)(2)(C): subject matter disclosed in a US patent/application publication that is otherwise § 102(a)(2) art is not prior art if, "not later than the effective filing date of the claimed invention," the disclosed subject matter and the claimed invention "were owned by the same person or subject to an obligation of assignment to the same person."

• US 2017/0301445 names Oki Gunawan alone → it names "another inventor" relative to the '797 (Gunawan + Sosa), so it would be § 102(a)(2) art but for the exception.
• US 2017/0301445 was filed by International Business Machines Corporation (the '566's continuation, US 10,460,863, lists Applicant/Assignee IBM; Gunawan is an IBM employee under an obligation to assign).
• The '797 was filed 2017-10-10 by IBM, with inventor assignments executed 2017-10-06 to 2017-10-09 and recorded 2017-10-10 at reel 043828/0430 (frame 0470).

Common ownership as of 2017-10-10 → § 102(b)(2)(C) applies → US 2017/0301445 (and US 9,978,493, and US 10,460,863, and US 2017/0299410, and US 2017/0031716) are removed from the prior art. Confidence: high on the legal rule; medium-high on the factual predicate (I have not opened the '566 assignment record; the '863 front page and the '797's own assignment of record make common ownership very likely but not yet proven from the reel itself).

This is the single most important fact in the whole § 103 analysis, and it is the opposite of what the earlier prior-art section assumed. The best art on the variable-gap limitation is IBM's own, published nine days too late, and disqualified even if timely.

2.3 The § 102(b)(1)(A) grace-period wrinkle cuts the other way

Every PDL publication authored by Gunawan (a joint inventor of the '797) published within one year of 2017-10-10 is potentially excepted under § 102(b)(1)(A) if the disclosure was "made by the inventor or joint inventor":

Caveat, stated honestly: the exception's application to a multi-author disclosure where only one author is a joint inventor of the challenged claim is contestable (the analysis turns on whether the subject matter was obtained from the inventor; MPEP § 2153.01(a)). Gunawan and Virgus are both IBM Watson researchers who jointly performed the work, so a patent owner could plausibly argue the virgus-attributable subject matter was not "obtained from" Gunawan and is therefore not excepted — which would keep US 2017/0045433 in the art. Conversely, Gunawan could invoke the exception. Treat this as genuinely undecided and verify with the file wrapper. The practical effect of resolution is modest: neither the viscometer publication nor Gunawan 2017 discloses the variable-gap fixture, and the stoppers and tilt-platform features they do disclose are not claimed in the '797.

2.4 Net available-art ledger

---

3. Claim 1, decomposed

Two claim-construction notes that matter for § 103. (i) "…the diamagnetic rod will connect the pads… forming a short" is capability language; it is met by structure that inherently performs the function (MPEP § 2111.04). (ii) "…configured to … tune a shock threshold force" is an intended-result recitation; the structural predicate is only the magnets-plus-variable-fixture.

---

4. GROUND A — Claim 1: the primary § 103 attack

Combination: Gunawan 2015 (or US 8,895,355 / 9,093,377 / 9,236,293, Cao et al.) + US 6,142,007 (Nippon Aleph) or US 6,480,081 (Oki Electric), and/or US 9,316,550 (STMicroelectronics) — with the variable-gap fixture supplied by the '797's own admissions and ordinary mechanical skill.

4.1 Element mapping

4.2 Why a POSITA would combine — the motivation chain

This is where the ground is strongest, because the motivations come from four independent sources, three of which are admissions in the patent itself.

1. AAPA — magnetically displaced shock sensors were already a known category. The '797's Background lists conventional shock sensors including "magnetic balls which can be dislodged from a holder." That is a magnetically restrained mass dislodged by shock. Claim 1 differs only in what restrains the mass (diamagnetic levitation in a PDL trap rather than mechanical/magnetic capture). Under KSR rationale (B) (substitution of one known element for another to obtain a predictable result), this is the heart of the case.
2. AAPA — the need for a tunable threshold is admitted. The Background states: "Different items are, however, sensitive to different levels of shock. Since the sensitivity of conventional shock sensors is not tunable, a different sensor must be used to change the sensitivity." This is textbook KSR (G) design-need/market-pressure motivation, and it is the applicant's own words.
3. Same field of endeavor / reasonably pertinent art. The IBM PDL body of work is sensor-transduction work (viscometer, pressure gauge, inclinometer, force gauge), and the shock-sensor art is impact-indication work. Under the KSR/MPEP § 2141.01(a) test, both are "reasonably pertinent to the problem" of measuring a physical quantity with a levitated mass.
4. Result-effective variable (§ 2144.04(V)). Gunawan 2015's own model expresses the total trap field as the superposition of two dipole-line fields at positions ±a: B_T(x,y,z) = B_DM(x−a,y,z) + B_DM(x+a,y,z), and the '797 reproduces that equation. The half-separation a is therefore a parameter the art itself identifies as controlling the field at the rod — and the field at the rod controls the restoring force. A POSITA wishing to change the barrier would vary a. That is the definition of a result-effective variable, and optimizing it is routine.

4.3 Expectation of success — no "unpredictable arts" defense

The dissuasions available to the patent owner are thin, because the physics was fully quantified before 2017-10-10. Gunawan 2015 and Gunawan 2017 give closed-form field expressions, the equilibrium height y₀ (Eq. 4 of the '797), and the stability/camelback barriers across rod lengths and radii. The '797's own Eq. 5 gives the critical gap g_C at which the rod drops — precisely the fall condition the claim turns on. A POSITA therefore had both (i) a quantitative prediction that increasing the magnet gap weakens the barrier until the rod falls, and (ii) a routine mechanical means for varying the gap. Reasonable expectation of success is high, and predictability cuts against nonobviousness.

4.4 Weakness of Ground A

The only genuinely missing link is a printed publication that states "vary the gap between the PDL magnets." That link must be carried by: (i) the '797's express AAPA reference to the variable-gap fixture, (ii) the § 2144.04(V) optimization rationale, and (iii) expert testimony in district court. In an IPR, this is materially harder — the Board's petitions must rest on patents/printed publications, and AAPA-style reasoning is argued only as evidence of the POSITA's knowledge. Expect the patent owner to argue the art "teaches magnets in contact" (all Cao patents; Gunawan 2015's reference setup uses center-to-center separation 2a = magnets touching) and that nothing suggests they be separated.

---

5. Ground B — Claim 1 with a levitated-mass inertial-sensor reference

If US 8,109,142 B2 ("Inertial sensing system with a curved base and a diamagnetic mass"), US 7,252,001 B2 (active magnetic levitation for inertial sensing), and/or US 7,143,648 B2 (magnetofluidic accelerometer with capacitive sensing of the inertial body) verify out, the combination becomes considerably stronger: they supply the missing teaching that a levitated mass can serve as an inertial/proof mass, removing the need to rely on the "magnetic balls dislodged from a holder" admission alone. These are medium-confidence candidates — they surfaced via similarity association, not full-text review — and must be verified before being relied on.

Grade: would move claim 1 from contestable to more likely than not obvious, if the references verify and if the variable-gap issue is carried.

---

6. Dependent claims 2–4 (and 8, 9, 15, 16)

---

7. Claims 5–9 — the network + blockchain system claim

Combination: Ground A sensor network + Crosby et al. 2016 (blockchain immutability) ± US 7,250,855 (networked sensors) for claims 6–7.

• Element (D) — "a blockchain ledger configured to record any shock-trigger events": supplied entirely by Crosby 2016. The '797 cites Crosby for exactly this purpose; the reference is solid § 102(a)(1) art (June 2016 > 1 year before filing).
• Motivation: the '797 itself articulates it — "Since the present sensors can be reset, there is a concern of device tampering whereby someone might impermissibly reset the sensor in order to cover up the fact that the sensor had already been tripped." The patent owner thereby concedes the problem. A POSITA in 2016–2017 supply-chain monitoring, aware of both (a) resettable indicators and (b) distributed-ledger immutability as taught by Crosby, would combine them for the predictable result of a tamper-evident log. KSR (A) + (C).
• Claims 6/7 (LAN; Internet link): US 7,250,855 (sensor network) plus a routine networking choice — strong.
• Weakness: the motivation is partly hindsight-shaped — Crosby 2016 is about finance, not shipping. A patent owner will argue no reference suggests applying distributed ledgers to package shock events, and that the linkage is only visible with the '797 in hand. Also note the substantial § 101 exposure of this limitation (a generic data-recording function appended to an apparatus), which is a separate and cheaper attack than § 103 — but it is not the same analysis and should not be conflated.

Grade for claim 5: Moderate.

---

8. Claims 10–16 — the monitoring method

---

9. Consolidated motivation ledger (the § 103 argument in one place)

---

10. The patent owner's rebuttals, and how much they bite

1. "The variable-gap PDL art is disqualified / is our own co-pending work." This argument is correct and it is the patent owner's best one. It removes the only clean documentary teaching of the gap fixture. It does not remove the AAPA admission that such fixtures were known, nor the optimization rationale — but it shifts the burden onto the challenger's expert.
2. "Every PDL reference teaches or shows the magnets in contact." True (Gunawan 2015's reference setup; all Cao patents). This is the closest thing to a teaching away argument, though not a strong one, because contact is simply the zero-gap case and Gunawan's own equation is written in terms of a.
3. "MEMS is the mainstream shock-sensor platform; nothing points to magnetic levitation." Weakened substantially by the '797's own Background admission of magnetic-ball-dislodgement sensors and by Gunawan's own statement that the PDL trap is being investigated "for various sensing applications."
4. "The examiner considered the entire IBM PDL family — including US 9,978,493, US 10,031,058, US 2017/0299410 and US 2019/0017965, all cited on the '797's face — and still allowed." This is a real point. It is non-binding, but it (i) supports the patent owner rhetorically, and (ii) triggers the Board's § 325(d) discretionary-denial machinery for any IPR built on the same art. A petitioner should bring art the examiner did not have.
5. Secondary considerations. There is no objective evidence on the record: no litigation, no commercial embodiment, no licensing of this specific patent, no unexpected-results data. That means the patent owner has no objective indicia to rebut a prima facie case — but equally, the challenger has no admissions of commercial success to exploit. The strongest available "unexpected result" the owner could assert — high sensitivity and tunability from a levitated graphite rod — is asserted in the specification without comparative data and would need testing to support.

---

11. Bottom line by claim

Single most important takeaway: the '797's § 103 vulnerability is not about whether someone thought to put contact pads under a magnetic trap — that is nearly foregone. It is about whether the variable-gap fixture and the "tune the shock threshold" function were obvious on 2017-10-10. IBM's own best art on that point is disqualified by § 102(b)(2)(C), and the remaining case must be built on the applicant's own admissions (the Background section's magnetic-ball sensors and non-tunable-sensitivity lament, and the express incorporation-by-reference of the variable-gap fixture), read with the § 2144.04(V) result-effective-variable rationale. That is a workable district-court case and a difficult IPR case, and any § 325(d) exposure from the PDL art already of record must be managed by bringing new art.

---

12. Verification checklist (do these before relying on any of the above)

1. Assignment records. Pull the '566 (now US 9,978,493) and '620 (now US 10,082,408) assignment reels from USPTO Assignment Center to prove common ownership with the '797 as of 2017-10-10 — this is the linchpin of the § 102(b)(2)(C) conclusion.
2. Search for any pre-2017-10-10 public disclosure of the variable-gap PDL trap — a PCT (check PATENTSCOPE/Espacenet for a WO filed April 2017 claiming priority to 15/131,566; the US counterpart published 2017-10-19, so a PCT would have published ~the same day or later, but confirm), conference abstracts (APS March Meeting 2016/2017), ARPA-E reports (the Gunawan 2017 paper acknowledges ARPA-E Award No. ARPA-DE-AR0000540), IBM Research blog/seminar materials, or a Gunawan presentation. Any such disclosure before October 10, 2016 would be un-excepted § 102(a)(1) art and would largely decide the case.
3. Verify the candidate levitated-mass inertial-sensor art (US 8,109,142; US 7,252,001; US 7,143,648; US 7,597,002) in full text, and confirm pre-2017-10-10 publication.
4. Pull the file wrapper for the '797 from USPTO Patent Center to confirm (a) which art the examiner applied, (b) whether the variable-gap-fixture clause was added to overcome art (the earlier section's family-prosecution point, based on the broader CN111194476 claim 1), and (c) the IDS contents.
5. Pull the GB, JP and CN counterparts' search reports and examination citations — foreign examiners searched the tunable-threshold concept without the IBM-family disqualification constraints that apply in the US, and may have located third-party art on adjustable-threshold shock indicators.
6. Resolve the § 102(b)(1)(A) question for US 2017/0045433 and Gunawan 2017 (whether the grace-period exception applies to a Gunawan+Virgus disclosure). This affects the marginal grounds only, but it affects them in both directions.
7. Note the date discrepancy in the task framing (the working session is dated both April 26, 2026 and September 29, 2026); nothing above turns on which is correct, except that the '797 remains unexpired (anticipated expiration 2037-10-10) and unlitigated as of the earlier sections' searches.