US 5727758

I'll run targeted searches for this patent number and check for any litigation activity.

Let me check a couple more targeted sources on litigation/assignment status.

US Patent 5,727,758 — Verification Summary

I searched for this exact number (not look-alikes). Findings below; where my search produced no authoritative result, I say so.

Bibliographic data (confirmed against Google Patents, Justia, USPTO.report, EveryPatent, and the patent's own front page)

Abstract (as issued)

A system and method for warning personnel in proximity to railroad tracks of an approaching train is disclosed. A transmitter disposed on a train transmits a warning signal. The personnel are each provided with a portable Radio Frequency (RF) receiver (pager) for receiving the warning signal. Upon receiving the warning signal, the portable RF receivers generate alarms to notify the personnel of the impending arrival of the train.

Independent claims — plain language

Claim 1 (system). The broadest claim. Two elements only: (a) a transmitter on the train sending a warning signal at a given power output, and (b) a portable RF receiver carried by a worker near the tracks that, on receiving the signal, generates an alarm. Notably, it is train-centric and requires no trackside infrastructure whatsoever.

Claim 13 (system). A narrower, more specific system claim adding three limitations beyond claim 1: the on-train RF transmitter transmits continuously during train operation; its power output varies as a function of train speed so the signal reaches the worker a predetermined amount of time before the train arrives; the receiver is portable and carried by the worker.

Claim 15 (method). A three-step method: (1) transmit a warning signal at a given power output from an approaching train; (2) receive it with a portable RF receiver carried by a worker near the tracks; (3) generate an alarm on receipt.

Dependent claims worth noting

• 2 / 16: continuous transmission during train operation.
• 3 / 14 / 17: speed sensor + on-train controller varying transmit power with speed, to yield constant warning lead time (specification gives 15 seconds as an example).
• 4 / 18: controller takes a portion of the transmitted signal as feedback to verify the transmitter works, and disables train operation if it does not.
• 5: wireless RF path.
• 6–9: a trackside "control base" transceiver that relays/repeats the warning to the pager; claim 7 covers coupling to the rails, claim 8 the wireless path, claim 9 a trackside alarm.
• 10–12 / 19–20: alarm type — visual, audio, or vibrative.

Observations and uncertainties

1. Title vs. claims mismatch. The title says "tactile and audible," but the claims cover visual, audio, and vibrative alarms; the "tactile/vibrative" feature (vibrative alarm 40) appears only in dependent claims 12 and 19. The "tactile" element is thus not mandatory in any independent claim.
2. Litigation / CAFC 2026 dockets — no hits found. My searches returned no district court action or Federal Circuit 2026 appeal involving US 5,727,758. The one 2026 PTAB item that surfaced (IPR2026-00134, Westinghouse Air Brake Technologies) concerns RE47,835 (Railware), and 5,727,758 appears there only inside an examiner's prior-art search-history string, not as a challenged or asserted patent. I cannot rule out an unindexed or very recent filing — treat "no litigation found" as a search result, not a certified clearance.
3. Enforceability note. With a 1997 priority date and an anticipated 2017-03-26 expiration, this patent is expired; any relevance today is as prior art rather than as an enforceable right.
4. CIP caveat. Because US 5,924,651 is a continuation-in-part, new matter added there carries a 1998 filing date rather than 1997. Do not assume the two share an identical effective filing date for every claim.
5. Claim 13 antecedent note. Claim 13 recites "an RF receiver" without expressly introducing its antecedent for the portable receiver; a strict reading is worth flagging in any validity analysis.

Sources: Google Patents (https://patents.google.com/patent/US5727758/en), Justia (https://patents.justia.com/patent/5727758), USPTO.report (https://uspto.report/patent/grant/5727758).

I'll search litigation databases for US Patent 5727758.

Let me search more specifically for litigation databases and the patent owner.

Let me try court records and litigation trackers directly.

Litigation Search Results for US Patent 5,727,758

Bottom line: I found no evidence of any litigation — district court, ITC, or PTAB — in which US Patent 5,727,758 was asserted or challenged as a patent-in-suit. I want to be explicit that this reflects the searches I was able to run within this session, not a certified negative from PACER/Docket Navigator/Unified Portal itself.

What the searches did show

Every "hit" was the patent appearing as a reference or citation, not as an asserted patent:

Note also the '758 patent's own continuity: US 5,924,651 (continuation-in-part, filed 1998-03-17, same inventors) and US 6,145,792 (Penza, 2000). If any Porcelain-family patent were asserted, it would more likely be '651 or '792 than the '758 — but I did not find assertions of those either in this session.

Why the absence is plausible

• Status: US 5,727,758 is expired (anticipated expiration 2017-03-26; "Expired – Lifetime"). Assertion value drops to zero after expiry except for back damages, and the patent is now ~29 years old.
• Assignment: Reassigned to Harsco Technologies LLC effective 2016-04-28 (REEL/FRAME 038438/0034), only ~11 months before expiration — leaving almost no window in which Harsco could have sued on it.
• No PTAB proceedings: No IPR/PGR against '758 surfaced.

Important caveats (do not treat this as a certified clearance)

1. Data completeness unknown. I could not query PACER, Docket Navigator, Unified Patents' litigation caselist, or CourtListener directly before hitting my tool-step limit. My results are from general web search and patent-aggregator pages, which do not comprehensively index pre-2010 district court dockets.
2. No case numbers, courts, or dates can be reported because none were found. I will not fabricate them. If a suit exists, it would most likely be an older (late-1990s/2000s) case involving the individual inventors Penza/Penza, and such cases are poorly indexed on the open web.
3. Search for similar numbers must be excluded, per your instruction, so anything turning up under '788, '752, '758-family cousins was disregarded.

Recommended verification steps

To get a definitive answer, check these directly (I could not reach them in-session):

• Unified Patents Portal: https://portal.unifiedpatents.com/litigation — search "5727758" exactly.
• CourtListener: https://www.courtlistener.com/?q=%225727758%22 (and the RECAP docket search).
• USPTO Patent Public Search / Assignment records for any recorded security interests or licenses that might hint at enforcement.
• PACER case search across districts for "Penza" and "Harsco Technologies" in the 1998–2017 window.

If you need, I can attempt the Unified Portal and CourtListener queries directly in a follow-up.

Conclusion: Based on available information, no known litigation involving US 5,727,758 is identified. The patent appears in the record only as a cited prior-art reference and in its own family documents.

I'll verify the structured-data default with targeted searches before writing anything.

No proceedings surfaced yet. Let me run additional targeted checks before concluding.

Let me run a final round of checks for appeals, litigation, and any sibling-patent PTAB activity.

PTAB Proceedings — US 5,727,758

Verification note up front: I ran multiple targeted searches (patent number alone; number + IPR/PGR/CBM/IPR20xx; number + Federal Circuit/CourtListener; number + assignee/litigation). I found no AIA trial proceeding in which US 5,727,758 was the challenged patent. This is consistent with the structured "PTAB proceedings on file" block, which reports no AIA trials from the USPTO Open Data Portal. The findings below reflect that absence — I have not invented proceeding numbers to fill the template.

---

Proceedings overview

Total AIA trials on US 5,727,758: 0. Breakdown by status: 0 active / 0 with claims invalidated / 0 with claims sustained / 0 settled / 0 institution denials — there is simply no PTAB record at all. The bottom-line defensive posture this gives a defendant is unusual and, on these facts, favorable: not because "the patent has survived and is hardened," but because every one of claims 1–20 is completely untested at the PTAB, and the patent's term expired on 2017-03-26, so the practical problem is not claim scope — it is whether any damages window remains open at all.

No proceeding to report

Because the canonical list is empty and my searches produced no contradicting hits, there are no per-proceeding entries (no petitioner, no FWD, no settlement, no CAFC appeal) to document for the '758 patent. For auditability, here is what the searches did surface, and why none of it is a proceeding against '758:

I could not confirm details of any district-court assertion of '758 either; my searches did not surface a complaint. I am flagging that as unverified, not as "never asserted."

Two data points worth reconciling

• Expiration. The structured data shows "Anticipated expiration 2017-03-26," legal status Expired – Lifetime. The sibling continuation-in-part, US 5,924,651, is listed with an expiration of 2017-03-25 (priority 1997-03-25) on the Unified Patents portal. I see no indication of patent term adjustment or extension in the record before me; if you need certainty, pull the face of the patent and the maintenance-fee/PTA record from Patent Public Search.
• Priority date. Google Patents states 1997-03-26; the Unified Patents portal for the '651 CIP states 1997-03-25. Minor, and it does not change the 2017 expiration reality either way.

---

Strategic summary

Claim status: 20 claims, zero tested. Claims 1–20 of US 5,727,758 have never been the subject of an AIA trial. None are CANCELED, none are SUSTAINED, all are UNTESTED. There is no IPR-narrowing to report, no claim-level FWD to quote, and critically no Federal Circuit mandate in this family. Compare that to the patent's own citation trail: '758 was substantial enough to be cited by examiners and by later railroad-safety applicants (Harsco, Bombardier, Metrom Rail, Miller Felpax, Stc, Westinghouse), yet no one ever took a shot at it at the Board. The most likely explanation is not that the claims are bulletproof — they are broad, pre-AIA, and were drafted before the modern anticipatory art (US 5,554,982 "Wireless train proximity alert system," the EBS optical-warning work gang references, and the Japanese train-approach alarms all pre-date it and appear in its own citation list) matured into litigation-grade invalidity charts. The more likely explanation is economics: the patent expired in March 2017, roughly nine and a half years before today, which removes most of the incentive to spend IPR fees on it.

Estoppel landscape: empty, and that cuts both ways. With no institution, no petitioner has triggered 35 U.S.C. § 315(e)(2) estoppel against this patent. There is therefore no estoppel-created safe harbor for a defendant and, conversely, no petitioner-side estoppel constraining you. Any § 102/§ 103 ground remains formally available in district court. The constraint you actually face is not estoppel but jurisdiction and remedy: under 35 U.S.C. § 286 a complaint filed today (2026-10-01) reaches back only to 2020-10-01, which is entirely after the 2017-03-26 expiration. Absent a term extension I have not seen, that leaves no recoverable infringement period for the '758 patent itself, which makes an invalidity fight largely academic — the sharper defense is non-infringement-in-time / no available remedy.

Pattern signals: no pattern, because there is no data. No serial petitioner, no second petition on the same patent, no patent-owner appeal practice, no aggregator challenge. The patent owner of record is Harsco Technologies LLC, assigned 2016-05-02 (effective 2016-04-28, ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR: PENZA, GEORGE GREGORY, Reel/Frame 038438/0034) — i.e., the inventor monetized the family to an operating company essentially at the end of the term, not to an NPE, and Harsco went on to build its own successor patents in this space (US 9,902,411 / US 10,752,272 / US 11,708,099, the "work block encroachment warning system"; US 10,179,596 / US 11,866,078, the "rail warning system and method"). If you are receiving demand letters citing '758 today, the live commercial risk is more plausibly that later Harsco family patents are being cited or that '758 is being used as prior art (as it is in the Westinghouse/Railware RE47,835 fight), not that '758 itself still supports a royalty demand.

---

Recommended next steps

1. Confirm the empty docket yourself, directly from the Office so your file has a clean record: PTAB E2E / PTAB Center and USPTO Patent Public Search for "5,727,758" in the trial proceedings field. Also confirm the maintenance-fee and expiration status (the structured data reports fee payments at years 4, 8, and 12 — the last on 2009-08-19 — consistent with a full-term, non-lapsed patent that simply ran its 20 years).
2. Run the § 286 / expiration analysis first. Build the timeline: filing date 1997-03-26 → expiration 2017-03-26 → six-year lookback from any hypothetical 2026 complaint = 2020-10-01. If, and only if, the patent carries PTA/PTE you can identify in Patent Public Search, redo the math with the adjusted date. Document that the paper trail shows none.
3. If you still need an invalidity position (e.g., because an expired claim is being asserted for past-era conduct or to clear a cloud over a later-generation product), note two practical points: (a) IPR of an expired patent is legally permissible, but under current Office practice institution decisions are centralized in the Director and age/settled-expectations discretionary-denial factors weigh heavily against old patents — see Unified Patents' 2025 in Review (record 607 denials in 2025) and the related congressional testimony on institution practice; and (b) if you are inside the § 315(b) one-year window from service, that clock is running, so docket it.
4. Screen the family, not just the patent. The continuation-in-part US 5,924,651 (also expired; expiration listed as 2017-03-25) and the later Harsco railroad-worker-safety patents are where any current assertion would have to live. For appeal-style verification on any future proceeding, CourtListener and the Federal Circuit docket are the places to check.

Bottom line for a defendant: there is no PTAB proceeding on US 5,727,758 — literally zero — so there is no FWD to link to, no canceled claim to attack a demand letter with, and no claim that has been sustained either. The dispositive leverage is not a PTAB verdict but the calendar: this patent expired 2017-03-26, and the six-year damages reach-back from today lands after expiration. Lead with that.

I'll verify the assignment record and cross-reference the parties before writing up.

Ownership Analysis — US 5,727,758 B1

Patent: "Tactile and audible warning system for railroad workers"
Appl. No.: 08/827,346 · Filed: 1997-03-26 · Issued: 1998-03-17 · Priority: 1997-03-26
Status: Expired – Lifetime (anticipated expiration 2017-03-26)

Data note: I could retrieve the recorded assignment's reel/frame, dates, assignor and assignee from the Google Patents legal-events mirror (which populates from the same USPTO assignment data as Assignment Center). I could not retrieve the correspondent of record for that recording, and I could not pull the raw Assignment Center record set directly. Where a field is not supported by a source I actually saw, I say so rather than filling it in.

---

Inventors

Pattern to flag: This is an individually owned case from day one. Google Patents lists the original assignee literally as "Individual" — there was no pre-issuance assignment to a company, so there is no "inventor departs the assignee" pattern to test. What is unusual is the reverse: the two inventors retained personal ownership for the entire 20-year term up to 2016, then assigned the patent to Harsco only ~11 months before it expired (2017-03-26).

Data gap (flagged, not resolved): The 2016-05-02 reassignment entry names only PENZA, GEORGE GREGORY as assignor, even though the face of the patent names two inventors. Either (a) a companion record for George Robert Penza exists that is not surfaced in the mirror I retrieved, (b) he had already conveyed his undivided interest under an unrecorded or separately recorded instrument, or (c) his interest was not conveyed. I did not verify a second recording, so I am not asserting any of these.

---

Original assignee

Entity named on the issued patent: the individual inventors themselves (Google Patents: "Original Assignee — Individual"). No operating company was the original assignee.

• Did they ship a product? There is no evidence that the Penzas commercialized under a corporate name at issuance. The commercial embodiment of the claimed subject matter that I can document is the Roadway Worker Protection (RWP) system sold by Harsco's Protran Technology unit (Harsco Rail division) — e.g., the RTD Denver fleet order announced 2017-01-09 and the WMATA/Sacramento orders announced 2017-09-19, both of which describe alerting track workers of an approaching train entering a work zone. That product post-dates the 2016 acquisition and belongs to the assignee, not the original individual owners.
• Current status of the chain terminus: Harsco Technologies LLC (assignee of record). It is an operating subsidiary, not a standalone IP holding vehicle: it appears as assignee on a large portfolio of rail-maintenance and work-block-encroachment patents, and its parent is Harsco Corporation, now Enviri Corporation (NYSE: NVRI), which reports Harsco Rail as a reportable segment. Per Enviri's Q1 2024 Form 10-Q, the previously announced sale of the Rail business was paused and Rail was reclassified from held-for-sale back to continuing operations. No bankruptcy is disclosed for Harsco/Enviri.

---

Assignment timeline

The Assignment Center record set for this patent is thin: exactly one recorded post-issuance conveyance. There is no original corporate assignment (the patent issued to individuals), no security agreement, no merger, and no change-of-name.

• 2016-04-28 (executed) / recorded 2016-05-02 — Reel 038438/0034
  • Conveyance: Assignment (ASSIGNMENT OF INTEREST)
  • Assignor: PENZA, GEORGE GREGORY
  • Assignee: HARSCO TECHNOLOGIES LLC (Minnesota)
  • Correspondent: Not retrievable from the sources I accessed. I am not able to name the attorney/firm of record for reel 038438/0034, so I cannot run the repeat-correspondent test on this chain. This is a limitation of my retrieval, not a finding of "no correspondent."
  • Context: Acquisition — the individual inventor's residual personal IP was picked up by the operating company that was concurrently assembling the Harsco Rail / Protran track-worker safety business. Note the timing: the patent expired 2017-03-26, roughly 11 months after this recording, so Harsco acquired a heading-toward-expiry asset.

Family cross-reference (not separate assignments to this patent, but same chain): the continuation-in-part US 5,924,651 A (filed 1998-03-17, issued 1999-07-20, same two inventors, claiming benefit of this application) and US 6,145,792 A (filed 1998-04-29, issued 2000-11-14, same inventors) round out the Penza family. Nothing in the records I retrieved shows those being routed through an NPE.

---

Timeline diagram

timeline
    title Ownership of US 5727758
    1997 : Filed by individual inventors Penza
    1998 : Patent issued March 17
    2016 : Assigned to Harsco Technologies LLC
    2017 : Patent term expires

---

NPE / troll-pattern signals

---

Verdict

Operating-company assertion — with the explicit caveat that the assertion prong is not satisfied on the record I could retrieve.

Justification (2–3 sentences): The single recorded conveyance is reel 038438/0034, executed 2016-04-28 and recorded 2016-05-02, moving the patent from individual assignor George Gregory Penza to Harsco Technologies LLC — an operating subsidiary whose parent (Harsco Corporation / Enviri, NYSE: NVRI) actually sells an embodiment of the claims through its Harsco Rail / Protran Technology Roadway Worker Protection product line. Every NPE-pattern signal above is either not present or non-assessable: there is no shell-entity transfer, no known asserter, no cascading LLC chain, no pre-litigation timing, and the patent expired 2017-03-26, roughly 11 months after the assignment, which removes any realistic assertion economics. The chain therefore reads as an ordinary operating-company acquisition of inventor-held IP, not an NPE monetization play; the only reason I do not call it "defensive/non-asserting" is that the defined option requires a defensive aggregator, which is absent here.

Verification links:

• USPTO Assignment Center: https://assignmentcenter.uspto.gov/
• USPTO Patent Assignment Search (patent-number query): https://assignment.uspto.gov/patent/index.html — search 5727758; expect a single hit at reel 038438/0034
• Google Patents legal events (source of the recording data cited): https://patents.google.com/patent/US5727758/en

Recommended next step to close the gaps: pull the full text of reel 038438/0034 in Assignment Center to capture the correspondent of record and the executed-vs-recorded stamp on the face of the instrument, and separately run PENZA, GEORGE ROBERT as assignor to determine whether a companion recording exists for the second inventor's interest.

I'll research each cited reference to build accurate prior-art profiles. Let me run targeted searches.

Let me research the remaining cited references.

Prior-Art Analysis — US 5,727,758 ("Tactile and audible warning system for railroad workers")

Building on the earlier verification summary (bibliographic data, claim map, and enforceability notes), this section addresses only the 9 references cited on the face of the patent. Per the Google Patents front page, all 9 carry the "* Cited by examiner" marker — i.e., they were the art the Examiner actually considered during prosecution, which is why they are the right starting set for a §102/§103 attack.

Scope/method note. I searched for each cited reference individually. I was able to retrieve substantive disclosure for the US references (Jepson, Shirkey, Michalek) and for DE 2405408 and DE 4219067 (title/abstract level). I was unable to retrieve full text for SU 374124 A2, JPH 04163273 A, JPH 061239 A, and WO 1994/021504 A1 — for those four I rely solely on the title, dates, and assignee as printed on the '758 front page, and I say so explicitly below rather than speculate about their contents. Also note a minor date inconsistency between my brief ("Current Date: April 26, 2026") and this session's timestamp (2026-10-01); it does not affect the analysis.

---

A. The 9 Examiner-cited references (in filing-date order)

All nine pre-date the '758 priority date of 1997‑03‑26, so all are §102(b)/§102(a) prior art of record. The three named in the patent's Background Art (US 5,415,369; JP 404126664; JP App. 0035594) are not in the formal citation list and are discussed only as background, not as anticipation.

---

B. Reference-by-reference analysis

1. US 2,762,913 — Jepson (1956) — strongest single-reference §102 candidate for claim 1

Disclosure. A railway train-proximity warning system in which the train carries a pulsed transmitter; a remotely located receiver detects the pulses and produces both aural and visual indications (speaker/tone, lamp, meter). A key feature is a self-test loop: a capacitive pick-off (66) takes a portion of the transmitter output and feeds it into the receiver input, periodically, to prove that the transmitter/receiver chain is functioning. A switch provides an indication when the transmitter is out of operation. (US2762913 PDF)

Mapping to '758 claims.

• Claim 1 — arguably fully met: on-train transmitter transmitting at a given output + remote receiver that, on receiving the signal, generates an alarm. The only stress point is whether Jepson's "observer" receiver is portable and carried by personnel in proximity to the tracks; the reference speaks of "observers" generally, which supports (but does not compel) that reading. Best §102 attack on claim 1 of any cited reference.
• Claim 4 — the pick-off/timer self-test that confirms "the transmitter is functioning properly" reads directly on §102 for the feedback/verification element (Jepson does not disable the train, so the "disabling" clause is not met).
• Claims 10 & 11 — aural and visual alarms disclose the "audio alarm" and "visual alarm" limitations.

Bottom line: a credible §102 reference against claim 1 and claims 10–11; a strong §103/§102(dependent) reference against claim 4.

---

2. US 5,554,982 — Shirkey et al. / Hughes Aircraft (1996) — strongest §102-type candidate for claim 1 architecture

Disclosure. A train-mounted transceiver (VPAS 22) transmits a train-proximity signal containing GPS-derived position and speed; a crossing-based transceiver (WAVS 36) computes train time-to-arrival and broadcasts a crossing position + warning-zone; a vehicle-based receiver (VPAS 54) produces an alarm (blinking light 64, beeper 66) to the operator when inside the warning zone. Critically for the '758 comparison, if the WAVS is disabled the train "broadcasts a general warning signal which is received by the RF receivers … of all vehicles within range." (US5554982 Google Patents; US5554982 PDF)

Mapping to '758 claims.

• Claim 1 — the fallback "general warning signal" broadcast from the train to RF receivers that then sound an alarm tracks claim 1's two elements closely; the receiver is vehicle-mounted rather than worker-carried, which is the distinguishing feature (a POSITA would argue portability, but Shirkey discloses no worker-worn pager). §102 anticipation of claim 1 is arguable but contestable; §103 against claim 1 is strong.
• Claims 10 & 11 — blinking light and beeper = visual and audio alarms.
• Claims 3 / 13 / 14 / 17 — Shirkey computes time-to-arrival within a range (20–30 s), i.e., a "predetermined amount of time ahead," but does so at the crossing using train GPS/speed, not by an on-train controller varying the transmitter's power output as a function of train speed. So it does not anticipate the variable-power-to-hold-constant-warning-time claims; it is at most §103 art for them.

Bottom line: the closest cited art to the core train-transmitter → receiver-alarm concept, but lacking the "portable worker pager" and "speed-controlled transmit power" features.

---

3. US 5,620,155 — Michalek (1997) — strong §103 art; arguable §102 on claim 1

Disclosure. A locomotive GPS receiver determines position and, at the appropriate point, the locomotive transmits a digitally-coded RF signal to activate upcoming crossing warning devices (lights/bells/gates). The crossing performs self-diagnostics and relays encoded status back to the locomotive for maintenance collection. Michalek also expressly discloses broadcasting the approach directly to "specially equipped motor vehicles … a receiver (8) … [that] would actuate dashboard visual or aural alert devices," with a mute/reset button. (US5620155 PDF; uspto.report 5620155)

Mapping to '758 claims.

• Claim 1 — the vehicle-receiver embodiment gives (a) an on-train transmitter and (b) a receiver producing a visual/aural alarm on receipt; again the "carried by personnel" limitation is the gap (Michalek's receiver is vehicle-installed). Arguable §102; strong §103.
• Claim 4 — Michigan relays subsystem health back to the locomotive (integrity monitoring), analogous to the '758 feedback/verification element, though Michalek's relays crossing health, not transmitter health, and does not disable the train. §103 art, not clean §102.
• Claims 6–9 (control base) — Michalek's wayside crossings receiving the train's RF signal are analogous to the '758 "control base … transceiver for receiving the warning signal … and then transmitting" concept; useful §103 art.

Bottom line: the best combined-art reference for the train-transmitter/wayside-relay aspects of the '758 system.

---

4. DE 2405408 A1 — Philips Patentverwaltung (1975) — relevant to the worker-warning problem, but missing the on-train transmitter

Disclosure. A portable warning system to protect track workers: automatic sensing stations at each end of the work section detect approaching trains and transmit to a portable central receiving station that produces an acoustic alarm. The central station has two wireless receivers, an analysis stage, and a relay circuit; each sensing station has a magnetic-field sensor and a weak transmitter. (DE-2405408-A1, Unified Patents)

Mapping to '758 claims.

• Claim 1 (§102) — fails: the train is detected by trackside sensing stations, not by a transmitter disposed on the train; and the alarm-bearing receiver is a central station, not a worker-carried portable RF receiver. No §102 anticipation of any independent claim.
• Claim 6 / Claim 9 (§103) — the "portable central receiving station … acoustic alarm" is close in spirit to the '758 "control base … generat[ing] an alarm to notify the personnel in proximity of the control base." Good §103 art against the control-base dependent claims, but only in combination with a train-borne transmitter reference.

Bottom line: the classic "warn track workers of approaching trains" prior art, but architecturally the wrong side of claim 1 (trackside detection, not train transmission).

---

5. DE 4219067 A1 — Wetzler (1993) — relevant to the rail-coupled alternative embodiment

Disclosure. Warns personnel and vehicles in the track area of railway lines against approaching, passing, and departing railway vehicles "using a source producing continuous body[-borne] sound signals in the track rails." (Title as printed on the '758 front page; no detailed text retrieved.)

Mapping to '758 claims.

• Claim 7 (§103) — this is the closest cited art to the '758 alternative in which the transmitter is coupled to the railroad tracks/loop and sends the warning signal along the rails. Wetzler's rail-borne body-sound signaling is conceptually analogous; useful §103 art against claim 7's "coupled to the railroad tracks" limitation.
• Claim 1 (§102) — fails: no portable RF receiver/pager carried by personnel, and the source is not clearly on the train.

Bottom line: niche but pointed art for the rail/loop-coupled claims; not a §102 reference.

---

6. WO 1994/021504 A1 — EBS GmbH für Eisenbahnsicherung (1994) — optical, not RF

Disclosure. "Optical warning installation for work gangs." I could not retrieve the specification within the tool budget; based on the title and classification it is an optical (light-based) trackside warning for track work gangs. As such it is directed to the same purpose as the '758 system (warning workers on/near the track) but by a different medium (optical rather than portable RF pager).

Mapping: relevant as §103 background / secondary art showing the well-known objective of warning track work gangs of approaching trains; not §102 of the RF-centric independent claims. Flagged as content-unverified.

---

7. JPH 04163273 A — Mitsubishi Cable (1992) — "Train approach alarming device" — content unverified

8. JPH 061239 A — Toshin Sangyo (1994) — "Train approach warning device" — content unverified

Both are examiner-cited Japanese train-approach alarm references from the early 1990s. I could not retrieve their full text; on title and date alone they are consistent with trackside/wayside "train approach warning" systems of the type the '758 patent distinguishes in its Background (the Japanese systems that "rely on track-mounted sensors in fixed positions"). As such they are best treated as §103 background art establishing that train-approach alarms were known, and not as §102 anticipation of the on-train-transmitter + portable-pager claims. Flagged: content-unverified — do not assert specific disclosures without retrieval.

9. SU 374124 A2 (1973) — content unverified, no title or assignee printed

A Soviet-era reference with no printed title, abstract, or assignee on the '758 front page. I could not retrieve content. Flagged: content-unverified. Cannot be mapped to claims on the present record.

---

C. Ranking — most relevant prior art for US 5,727,758

---

D. Bottom line and caveats

1. No single cited reference is a clean §102 anticipation of independent claims 1, 13, or 15. The recurring gap is the combination of (i) a transmitter on the train and (ii) a portable RF receiver carried by an individual worker that generates an alarm. Jepson and Shirkey come closest to (i)+(ii) but each is missing an express "worker-carried portable" disclosure; the Japanese/Philips references have the worker-warning purpose but use trackside detection.
2. The strongest §102 candidates for claim 1 are US 2,762,913 and US 5,554,982; the strongest §103 art is Shirkey and Michalek, alone or combined with Philips/Wetzler for the control-base and rail-coupling dependent claims.
3. Claim 13's distinguishing features (continuous transmission; power varied as a function of speed to hold a constant warning lead time; portable worker pager) are not disclosed by any cited reference; Shirkey's time-to-arrival is computed at the crossing using GPS, not by varying on-train transmit power. Claim 13 is the narrowest and best-defended independent claim.
4. Content-unverified references: SU 374124 A2, JPH 04163273 A, JPH 061239 A, and WO 1994/021504 A1 — I could not obtain their specifications; the above treatment of those four rests on titles/dates only and must not be represented as a disclosure analysis.
5. Status reminder (from the prior section): the '758 patent is expired (anticipated 2017‑03‑26), so its relevance today is as prior art, not as an enforceable right; and because related US 5,924,651 is a continuation-in-part, do not assume identical effective filing dates for every claim when citing these references.
6. The §102/§103 mappings above are my analyst's assessment, not an official USPTO determination.

Primary sources: Google Patents US5727758 (https://patents.google.com/patent/US5727758/en) — front-page citation list and dates; US 2,762,913 PDF (https://patentimages.storage.googleapis.com/6b/a8/96/62c763d391b90c/US2762913.pdf); US 5,554,982 (https://patents.google.com/patent/US5554982); US 5,620,155 PDF (https://patentimages.storage.googleapis.com/b9/39/05/94a59445b45a9d/US5620155.pdf); DE 2405408 A1 (https://portal.unifiedpatents.com/patents/patent/DE-2405408-A1).

I'll ground the analysis by pulling details on the cited prior-art references before building the combinations.

Let me ground the remaining cited references (trackside optical warning, rail-conducted signals, and the Japanese train-approach alarms).

Obviousness Analysis Under 35 U.S.C. § 103 — US 5,727,758

Analytical note up front. This is an independent §103 analysis built from the prior-art record on the patent's Google Patents page (the nine examiner-cited references) plus the two Japanese references the applicant discusses in its own Background, and the reference texts I was able to retrieve. I do not have the '758 file history in the provided materials (the "Prior Art" section of this page does not include prosecution papers), so nothing below should be read as a restatement of what the examiner actually did or did not reject on. All 20 claims issued. Treat the grounds below as hypothetical grounds a challenger or a later-art analyst would construct.

Date discrepancy flag: the task header states April 26, 2026, while the session environment states October 1, 2026. This does not affect the analysis (the patent expired 2017-03-26 either way), but I am flagging the inconsistency per my operating rules.

---

1. Framework

Effective filing date: 1997-03-26. Pre-AIA §103 applies. All nine examiner-cited references and both Japanese references discussed in the specification predate that date, so all are available as prior art; US 5,620,155 (Michalek) and US 5,554,982 (Shirkey) issued later but were filed earlier, making them available at minimum under pre-AIA §102(e).

Claimed subject matter (from the earlier summary, not repeated here): independent system claim 1 (on-train transmitter + worker-carried portable RF receiver + alarm), independent system claim 13 (continuous on-train RF transmission with speed-varying power yielding constant lead time), independent method claim 15, and dependent claims 2–12, 14, 16–20 covering continuous transmission, speed/power control, transmitter self-check with train disable, wireless vs. rail/loop paths, a trackside "control base" relay, and visual/audio/vibrative alarms.

Level of ordinary skill. A person having ordinary skill in 1997 would hold a bachelor's degree in electrical engineering (or equivalent) and roughly two to five years of experience in RF communications and/or railway signal systems, with working familiarity with: (a) commercial paging/beeper technology in the 900–980 MHz band; (b) RF propagation — specifically that received range is a monotonic function of transmit power and antenna gain; (c) railway grade-crossing and work-zone warning practice, including FRA/industry practice of warning workers at a defined time/distance before train arrival; and (d) the fact that workers wearing hearing protection or operating machinery cannot reliably rely on audible train horns or verbal lookouts.

All references are analogous art. Every cited reference is classified in or near B61L (railway traffic safety) and every one addresses detecting, signalling, or warning about an approaching rail vehicle. This is the same field of endeavor; there is no serious non-analogous-art defense for any of the nine.

---

2. Reference digest — what each reference supplies and what it lacks

(Numbers are reproduced literally as they appear on the page and in the specification; I have not "corrected" 404126664 or 0035594 even though they look like JP H04-126664 and JP 3-5594.)

---

3. Element availability for the independent claims

Claim 1 = (a) transmitter disposed on the train transmitting at a given power output; (b) portable RF receiver carried by personnel in proximity to the tracks; (c) alarm on receipt.

• (a): US 2,762,913; US 5,554,982; US 5,620,155 — each expressly mounts the transmitter on the train.
• (c): US 2,762,913 (lamp + speaker); US 5,554,982 (blinking light + beeper); US 5,620,155 (dashboard visual/aural); DE 2405408 (alarm instruments); WO 94/21504 (flashing lights, horns).
• (b): the gap. US 5,554,982 and US 5,620,155 put the receiver on a vehicle; DE 2405408 and WO 94/21504 protect workers but warn them from a central station. The only reference the record shows placing an RF alarm receiver on the worker is JP 404126664, as the applicant itself describes it.

Claim 13 adds continuous transmission + power varying with speed so the signal arrives a predetermined time ahead. Availability: continuous transmission (DE 4219067; US 2,762,913 front/rear transmitters radiating while moving); train broadcasts its speed (US 5,554,982 claim 2; US 5,620,155 GPS speed computation); warning-time-controlled-by-speed (US 5,554,982: speed-dependent range + "more than ten seconds in advance tend to be ignored" + claim 4's upper/lower time limits).

Claim 15 (method) is the same subject matter in method form; it adds nothing that changes the analysis.

---

4. Grounds of rejection

Ground 1 — Claim 1 (and 5, 10, 11, 15, 16, 20) obvious over US 5,620,155 (Michalek), alone or in view of DE 2405408

Michalek claim 1(a) is a "first transmitter located on a railroad train … transmitting first electromagnetic signals from said train," and claim 10 expressly adds "a third receiver located in said motor vehicles" that receives the train's digitally encoded signal and "control[s] said warning devices in said motor vehicles." The specification of Michalek states the same: the locomotive "signal[s] the approach of a locomotive directly to specially equipped motor vehicles," actuating "dashboard visual or aural alert devices."

The only unmet limitation is that the receiver is carried by personnel in proximity to railroad tracks rather than mounted in a motor vehicle. Replacing a vehicle-mounted receiver with a person-carried one is the classic simple substitution of one known element for another, yielding predictable results (KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398, 416, 417 (2007)). The art supplies the motivation directly: DE 2405408 exists precisely "for protecting workers engaged in work on the railway track, to eliminate the human observer and the associated shortcomings" — the same problem the '758 recites in its Background. And the EPO's own citation of Lengemann et al., "Automatische Rottenwarnsysteme — von der kollektiven zur individuellen Warnanlage" [Automatic gang warning systems — from the collective to the individual warning installation], Signal & Draht 79 (1987) 31–39 (cited in EPO T 1152/97 as D1 against the EBS application), shows that by the mid-1980s the field was already traveling from trackside collective alarms toward individual, person-specific alarms. That is an explicit, contemporaneous market/design incentive, not hindsight.

Expected patentee rebuttal: Michalek's alert is for motorists at grade crossings, a different regulatory and physical setting from a work zone. Assessment: weak. The problem Michalek and the '758 address — a moving train warning a party out of the crew's line of sight via RF — is identical, and the '758's own specification gives no structural difference between a dashboard receiver and a belt-carried pager beyond mounting.

Ground 2 — Claim 1 (and 5, 6, 8, 9, 10, 11, 12, 15, 19, 20) obvious over US 2,762,913 (Jepson) in view of DE 2405408 and/or WO 94/21504

Jepson supplies the on-train transmitter, the tuned receiver, the visual alarm (lamp 60), the aural alarm (speaker 64), and the alarm-at-a-distance concept. DE 2405408 supplies the entire rest of the concept: a portable warning system, wirelessly linked, whose stated object is to eliminate the human lookout and to alarm the workers. WO 94/21504 supplies the individual/person-directed warning displays (flashing-light units that "can be positioned along the work site according to the type of work site"), plus the express teaching that the alarm can be triggered when the train "reaches a predefined distance" from the gang.

Motivation: all three are in the same field and address the same problem; the only modification required is to mount the alarm on the worker rather than on a fixed display or a train — the very direction Lengemann/T 1152/97 documents as the field's trajectory. WO 94/21504 further explains why a fixed optical display is insufficient (the display must be perceptible to every person, and the system must remain reliable under fault conditions) — providing the design incentive to distribute the alarm to the individual.

Rebuttable point: Jepson's receiver is co-located with a transmitter and its "alarm" is really a proximity indicator on a train or a wayside indicator, not a worker pager. This complicates but does not defeat: DE 2405408 and WO 94/21504 supply the worker-directed alarm, so Jepson is doing no more work than the on-train-transmitter element.

Ground 3 — Claims 3, 13, 14, 17 obvious over any of Grounds 1–2 further in view of US 5,554,982

This is the strongest ground on the record, because US 5,554,982 addresses the exact purpose of the '758's speed-responsive power control.

• The problem is expressly stated in the art. US 5,554,982: "Alarms that occur more than ten seconds in advance tend to be ignored, while alarms less than six seconds in advance can fail to provide adequate response time."
• The solution concept is expressly stated in the art. Its alarm range "includes the vehicle's braking distance, a response distance for the driver and a distance for the controller to process the information and activate the alarm, which are a function of the vehicle's speed"; and claim 4 recites a response-time range with a lower limit to give adequate response time and an upper limit to induce the operator to react.
• The train already broadcasts its speed. US 5,554,982 claim 2: the train-mounted transmitter periodically transmits the train's position, "and estimate its speed and transmit the train's speed and position as said proximity signal."
• Range-by-power is admitted to be elementary. The '758 itself says: "As known to those skilled in the art, an RF signal having more power is able to reach a farther destination along an RF transmission path than an RF signal having less power." Varying transmit power to vary range is thus an admitted, conventional technique.

Combine: a POSITA seeking to keep warning lead time roughly constant (the art-recognized goal in US 5,554,982) and having an on-train transmitter already fed with train speed (Michalek's GPS-derived speed/direction, or US 5,554,982's own speed broadcast) would predictably implement the variation by any of a small number of known mechanisms, of which "drive the transmitter's power output from a speed signal" is the most straightforward. That is "applying a known technique to a known device ready for improvement to yield predictable results" (KSR, 550 U.S. at 417).

Note the claim-scope point: claims 3, 13, 14, and 17 require only "a predetermined amount of time," not fifteen seconds. The fifteen-second example appears only in the specification (FIG. 3 discussion). So the patentee cannot argue that the art's 6–10 second window is non-analogous; the claim is agnostic as to the number.

Expected patentee rebuttal (best available): US 5,554,982 varies the warning zone size and its deactivation, computing time-to-arrival at a fixed crossing, and its transmitter is a crossing-based WAVS unit in the principal embodiment — not a train whose transmit power is throttled by train speed. Assessment: partial merit. The reply is that (i) US 5,554,982's fig. 6 and claim 2 expressly place the transmitter on the train, (ii) its failure-backup mode has the train "broadcast[ing] a general warning signal … received by the RF receivers of all vehicles within range" — i.e., range-limited delivery from the train — and (iii) the '758's own specification concedes the power/range relationship as known. The patentee would need a convincing teaching-away or a strong secondary consideration, and the record shows neither.

Ground 4 — Claim 2 and 16 (continuous transmission) obvious over DE 4219067 (Wetzler)

Wetzler teaches a source producing "continuous body sound signals in the track rails" to warn personnel in the track area. Continuous transmission from the train-equipment source is also inherent in Jepson's front/rear transmitters, which radiate while the train moves and whose receivers stay on to "inform the observers of the presence of other trains." Claim 2/16's additional element (continuity during train operation) adds no patentable weight.

Ground 5 — Claims 6, 8, 9 (trackside "control base" that receives the train's warning signal and re-transmits it over a wireless RF path to the portable receiver; trackside alarm) obvious over DE 2405408 and/or WO 94/21504 and/or US 5,554,982

• DE 2405408 is a portable central station "in wireless contact" with two remote sensing stations, containing "two wireless receivers, an analysis stage and a relay circuit," producing an instrument alarm. That is a control base that receives a remote signal and re-alarms.
• US 5,554,982 is even closer: a crossing-based transceiver "receives the train's proximity signal and transmits" a signal onward to remote receivers, and in the failure mode the train broadcasts; the whole system exists to route a warning from the train to end-user receivers through a fixed relay.
• WO 94/21504 supplies the fault-mode alarms at the central unit and along the work site (claim 9's control-base alarm).

Claim 9 (control base generates its own alarm) is met by DE 2405408's acoustic alarm at the central station and by WO 94/21504's central-unit warning displays. Claim 8 (wireless path from train to control base to receiver) is met by the combination. The result is a two-hop RF relay, a conventional architecture.

Ground 6 — Claim 7 (train transmitter coupled to the railroad tracks; control base transceiver coupled to the tracks) obvious over WO 94/21504 in view of DE 4219067 and DE 2405408

WO 94/21504's background discussion describes two known track-coupled systems: (i) DE-AS 11 50 504, in which transmitter devices along the track are actuated by "inductive rail contacts" and the signal travels over a radio link; and (ii) a system in which a "transmission conductor [is] laid along the track" with a central unit "inductively coupled with this transmission conductor, so that the respective position of the train can be evaluated" and an optical + acoustic alarm is emitted "as soon as the train reaches a predefined distance." DE 4219067 independently teaches continuous signalling in the rails themselves.

So the record contains (a) signalling through/along the track and (b) a central unit coupled to that conductor. Claim 7's "transmitter coupled to the railroad tracks … control base transceiver coupled to the railroad tracks" is the union of those two teachings, with predictable results (the '758 concedes the same power/range principle applies "along a transmission line such as a railroad track").

Ground 7 — Claims 10, 11, 19, 20 (visual / audio / vibrative alarm) obvious

• Visual: US 2,762,913 (lamp 60 and meters); WO 94/21504 (flashing lights, expressly); US 5,554,982 (blinking light 64); US 5,620,155 (dashboard visual alert).
• Audio: US 2,762,913 (speaker 64); DE 2405408 (acoustic alarm); US 5,554,982 (beeper 66); US 5,620,155 (aural alert); WO 94/21504 (horn).
• Vibrative/tactile (claims 12, 19): this is the weakest ground in the set. None of the nine examiner citations as retrieved discloses a body-worn vibratory alarm. The strongest available support is the apparent EE-Signals GmbH / EBS-family disclosure (FPO US 5,907,294, claim 16), which describes a "sensitive alarm installation … a vibrator which is permanently insertable in a pocket of an article of work clothing" exerting "body vibrations of predetermined duration and magnitude," alongside body-worn flashing-light alarms — but I could not establish its priority/publication date, so I cannot confirm it is prior art to 1997-03-26, and I will not assert that it is. A secondary attack would be the KSR "predictable variation" route: the '758 itself specifies that the portable device is a commercial pager tuned to "the designated beeper or pager frequencies," and vibrate alert was a standard feature of commercial pagers by 1997, so adding a vibrate alarm is a predictable design choice — but that argument needs a documentary reference (e.g., a 1995–96 pager datasheet or the pager standards) to be reliable. Flag: this is the claim family most likely to survive on the current record.

Ground 8 — Claims 4 and 18 (transmitter feedback/self-check; disable train operation if the transmitter is faulty) partially obvious

The self-check half is squarely met: US 2,762,913 discloses taking "a small portion of the transmitter output … as a test signal, as by a capacity pick-off 66," routing it to the receiver input through relay 67 under timer 74, "for giving a periodic and automatic check on the operation of both the transmitter and receiver components," and a separate indicator giving "a warning that the transmitter is out of operation." Michalek's crossing self-diagnostics relayed back to the locomotive, and WO 94/21504's "monitoring the energy supply" and dedicated fault-operation warning displays, reinforce the general practice of fault-verifying the warning chain.

The disabling half is the problem. No cited reference teaches inhibiting train motion in response to a fault in the warning transmitter. The closest analogue is WO 94/21504's fault displays and Jepson's failure lamp, neither of which affects train operation. A challenger could argue from railway fail-safe first principles — the field's cardinal rule that a signal system "must revert to its safest condition" when an element fails (a principle stated in the railway-signal literature of the era, e.g., in the family of US 9,026,283) — that tying a train's operability to the health of its own warning transmitter is an obvious application of a known technique. That is a defensible argument, but it is the least well-supported ground because it requires importing a general design philosophy rather than a reference disclosure. I would rank claim 4/18 as the most likely surviving claim, along with 12/19.

---

5. Consolidated motivation-to-combine analysis

Critical point for the patentee: the '758's own specification criticizes the prior art for relying on "track-mounted sensors in fixed positions" and argues they would be impractical to install everywhere. That is a genuine distinguishing feature of independent claim 1 — claim 1 needs no trackside infrastructure at all. But the argument fails as a matter of law, because US 5,554,982 (fig. 6) and US 5,620,155 (claim 10) both disclose exactly the infrastructure-free architecture: an on-train transmitter broadcasting directly to mobile receivers. The applicant's own characterization of the art is therefore incomplete, and — see §7 — potentially itself a problem.

---

6. Claim-by-claim vulnerability summary

---

7. The largest single risk in the record — and a contradiction to flag

The '758's own Background contains what reads as an admission about JP 404126664:

"A track sensor receives an RF train passage signal and then subsequently transmits an RF alarm signal to an alarm receiver carried by each worker."

That describes: (i) an RF signal originating with the train; (ii) a receiver carried by each worker; (iii) an alarm signal delivered to that worker-carried receiver; and (iv) addressed/selective activation ("whether the alarm signal has been sent to its own station"). If the reference says what the applicant says it says, it comes very close to claims 1, 5, 6, 8, 10–12, and 15 — with a trackside relay in place of claim 1's direct train-to-worker path.

I flag a contradiction between two of my own prior sections and this one: the earlier Patent Summary and Litigation Summary sections do not mention JP 404126664 at all, even though it is cited and discussed in the specification's Background. That reference deserves its own entry in the prior-art inventory. Once it is added, the §103 picture shifts materially toward invalidity of claim 1 — and, more importantly, toward a §102 question on claim 1 that I cannot resolve because I do not have the reference text.

I also flag a tension within the '758's own Background: the applicant simultaneously (a) describes the Japanese system as transmitting "an RF alarm signal to an alarm receiver carried by each worker," and (b) asserts "[t]he alarm signal does not appear to be used for warning personnel of an incoming train." Those two statements are hard to reconcile. A challenger would use (a) as the admission and treat (b) as attorney characterization, not disclosure. I do not have the JP text, so I am not asserting that JP 404126664 anticipates or renders obvious any claim — I am flagging that the applicant's own description of it is the most probative prior-art statement in the four corners of this patent, and that it is absent from the prior sections of this analysis.

---

8. Anticipated patentee rebuttals, and my assessment

1. "The art is about motorists at crossings; the '758 is about track workers." Weak — DE 2405408 and WO 94/21504 are both track-worker systems, and US 5,554,982/US 5,620,155 supply the RF architecture. Combine the two purposes, and the result is a two-reference combination in the same field.
2. "The art requires fixed trackside infrastructure; claim 1 requires none." Weak as to claim 1 — US 5,554,982 fig. 6 and US 5,620,155 claim 10 are infrastructure-free, on-train transmissions to mobile receivers. This is the argument most likely to defeat claim 7 (which does require rail coupling), but not claim 1.
3. "The '758 is quieter and doesn't intrude on residential neighborhoods." Weak — the siren/annunciation problem is solved by DE 2405408's instrument alarm and by individual opto/acoustic alarms in WO 94/21504; and quietness is a result, not a claimed limitation.
4. "Varying transmit power (rather than warning-zone size) is a nonobvious mechanism." Moderate — the best patentee argument on claims 3/13/14/17, but undercut by the specification's own admission that the power/range relationship is "known to those skilled in the art," and by US 5,554,982's range-based backup broadcast.
5. "No reference teaches disabling the train." Good argument. I would expect claim 4/18 to survive a §103 challenge on the nine-reference record unless a fail-safe/positive-train-control reference is added.
6. "No reference teaches a vibratory alarm." Good argument on claims 12/19 pending a dated reference.

Secondary considerations are unlikely to help much: the patent's commercial history is unexamined in the materials I have, the assignment to Harsco occurred in April 2016 — roughly eleven months before expiration — leaving essentially no window for nexus evidence of commercial success, and I found no litigation in the earlier section from which licensing/acceptance evidence might be inferred. If a patentee asserted long-felt need or failure of others, the strongest counter is that DE 2405408 (1975) already identified and solved the "human observer" problem, and WO 94/21504 (1994) was already commercializing automatic worker-warning installations — meaning the field was not stuck.

---

9. Practical bottom line

• Claims 1, 2, 5, 10, 11, 15, 16, 19, 20 are highly vulnerable to a §103 attack built on Michalek (US 5,620,155) + DE 2405408, or on Jepson (US 2,762,913) + DE 2405408/WO 94/21504, with US 5,554,982 supplying the mobile-receiver and lead-time teachings.
• Claims 3, 13, 14, 17 are vulnerable but contestable, with US 5,554,982 as the linchpin because it articulates both the problem (early alarms are ignored; late alarms are useless) and the speed-dependent solution, and because the '758 concedes the power/range technique.
• Claims 4, 18 and 12, 19 are the most likely survivors on this record. Claim 4/18's train-disable-on-transmitter-fault limitation and claim 12/19's vibratory alarm appear not to be disclosed by any of the nine cited references, and I could not verify the only candidate reference for the vibrator.
• The single highest-value next step is to obtain and read JP 404126664 (and secondarily JP 0035594, JPH 04163273 A, JPH 061239 A, and SU 374124 A2, whose texts I could not retrieve). The applicant's own characterization of JP 404126664 is close enough to claim 1 that it belongs at the top of any validity or prior-art analysis of this patent.
• A validity finding does not neutralize the '758 as art. Even if every claim is obvious, the '758 remains a printed publication under §102(a)/(b) for its disclosed subject matter, and its disclosure remains citable against later filings that trace to it (the '758's "Cited By" list includes the Metrom Rail, Miller Felpax, Bombardier, STC, and BNSF worker-protection families). Obviousness of the claims and prior-art effect of the disclosure are independent questions — a point worth keeping straight if the '758 surfaces in a later validity dispute, as it did tangentially in the IPR2026-00134 exhibit record discussed in the earlier Litigation Summary.

---

Caveats

• I retrieved full or partial text for US 5,554,982, US 5,620,155, US 2,762,913, DE 2405408 A1, and WO 94/21504 A1 (the last as a machine translation from German, via Google Patents/EPO). I could not retrieve JPH 04163273 A, JPH 061239 A, SU 374124 A2, JP 404126664, or JP 0035594. Nothing above asserts specific disclosure content for those five.
• The EE-Signals GmbH vibrator disclosure at https://www.freepatentsonline.com/[5907294](/patent/5907294).html is undated in my retrievals; I have not treated it as prior art and it requires a priority-date check before use.
• Reference identifiers are reproduced exactly as they appear on the patent page and in the specification, without normalization.
• This is an analytical §103 assessment, not a validity opinion, a freedom-to-operate analysis, or a representation of the '758's actual prosecution record.

Sources used: https://patents.google.com/patent/US5727758/en · https://patents.google.com/patent/US5554982 · https://patents.google.com/patent/US2762913 · https://portal.unifiedpatents.com/patents/patent/US-[5554982](/patent/5554982)-A · https://portal.unifiedpatents.com/patents/patent/DE-2405408-A1 · https://patents.google.com/patent/WO1994021504A1 · https://patents.google.com/patent/EP0690803A1 · https://www.epo.org/en/boards-of-appeal/decisions/t971152du1 · https://www.everypatent.com/comp/pat5620155.html · https://uspto.report/patent/grant/[5620155](/patent/5620155) · https://patentimages.storage.googleapis.com/b9/39/05/94a59445b45a9d/US5620155.pdf · https://www.freepatentsonline.com/5907294.html