The contested edges of this subject — directed-energy weapons, Havana Syndrome, "the Hum," "targeted individuals," and fast-moving data-center reports. For each: what's documented, what's claimed, and what's still unproven.
Most of this site is about the measurable, documented case: a neighbor's audio equipment producing low-frequency sound that travels through shared walls. This page is for everything adjacent to it that gets asked about — the claims that sit on the edge between established science and speculation.
We don't leave these on the floor, and we don't dismiss them as "crazy." Dismissing something real as a conspiracy damages credibility just as much as overclaiming does — directed-energy weapons are a live example: the U.S. military's Active Denial System "heat ray" is real, deployed technology. The useful question is never "real or fake"; it's what exactly is documented, what is only claimed, and how would you tell which one applies to your situation.
Each entry below is tagged with how well-supported it is. This page is a reference, not medical or legal advice. For the measurable core of the subject, start with the Research Library and Documented Cases.
Claimed — what people assert, in its strongest form.
Documented — what is actually established or on the record.
Unproven — the part that has not been demonstrated.
How to tell — how to work out whether it applies to you.
Claimed
A covert, concealable device is being aimed through walls at an individual — a microwave or "energy" weapon producing the residential symptom cluster (pressure, a sense of vibration, sleep loss).
Documented
State and military directed-energy systems are real. The U.S. Active Denial System is a vehicle-mounted 95 GHz millimeter-wave "heat ray" for crowd control; LRAD "sound cannons" are deployed acoustic devices (see the Edrei case); and the microwave auditory ("Frey") effect is a genuine laboratory phenomenon. These systems are large, power-hungry, and line-of-sight.
Unproven
That a concealable, civilian-deployable device is being used covertly, through walls, against private individuals to cause these symptoms. No such device has been demonstrated, and the size, power, and line-of-sight needs of the fielded systems argue against a hidden neighbor-scale version.
How to tell
Residential cases here have an identifiable acoustic source — usually a neighbor's audio equipment — that can be measured with a dB(C)/dB(Z) meter and tied to a specific unit. If there is a measurable acoustic signal with a source, you do not need an exotic weapon to explain it.
Claimed
Governments use 'sonic weapons' on crowds, and these devices cause lasting injury.
Documented
Long Range Acoustic Devices are real, fielded, and litigated: Pittsburgh paid $72,000 after a bystander alleged hearing loss at the 2009 G-20 (Piper); New York's painful alert tone was restricted and the city paid out (Edrei); Portland's mayor ordered police in June 2020 to use the LRAD for voice only, and the 2022 Don't Shoot Portland settlement limited its use for 14 months. In Belgrade, a sudden roaring sound at the 15 March 2025 vigil sent thousands fleeing; the European Court of Human Rights ordered Serbia to prevent any use of sound devices for crowd control pending the case (Đorović v. Serbia). Serbia admits owning LRAD units but denies using one. Israel fielded 'The Scream' at West Bank protests in 2005; Maldives police used an LRAD on 3 October 2025; Minnesota State Patrol used one on 26 January 2026 — video reviewed by Snopes shows voice announcements only.
Unproven
The Belgrade device (investigators judged a directional acoustic device 'highly likely' but could not prove LRAD use). Lasting injury from the Maldives, Israeli and Minnesota uses rests on self-reports with no medical confirmation or device settings on record. None of this involves residential buildings: these are line-of-sight, outdoor devices on vehicles or tripods.
How to tell
If the claim is 'the police used a sound cannon on a crowd', check the public record — it is often true and often litigated. If the claim is 'a sound cannon is being used on me at home', the device class does not fit: it is loud, audible, directional and outdoors.
Claimed
A foreign adversary is attacking U.S. personnel with a directed-energy or acoustic weapon.
Documented
The symptoms, reported by government personnel since 2016, are real. A January 2025 ODNI update found five agencies judge a foreign cause "very unlikely," while two judge a "roughly even chance" that a foreign actor had developed — and, in one view, used — a novel device, both with low confidence. NIH (2024) found no MRI-detectable brain injury. ODNI · NIH
Unproven
A specific weapon or perpetrator. The assessment is genuinely split and unresolved — not settled in either direction.
How to tell
This concerns government personnel and a possible novel device — a different context from a residential neighbor with consumer audio equipment. We keep the two separate so that residential complaints are not dismissed by association.
Claimed
Coordinated harassment by distributed groups of strangers using "voice-to-skull" (V2K) and civilian directed-energy weapons.
Documented
The psychiatric literature (Sheridan & James, 2015; Lustig et al., 2021) characterizes the gang-stalking / TI belief system as persecutory, with no validated mechanism for V2K against individuals. Separately, genuine, prosecuted acoustic harassment exists — the Seachrist case used ordinary speakers with an identifiable source. Courts have looked at the claim itself: in Ulibarri v. Winder (D. Utah, 2016) and Kelly v. Collins (E.D. Ky., 2022), federal judges dismissed infrasound / V2K / directed-energy complaints at the screening stage as factually frivolous or 'plainly implausible'. The belief can also appear in violent cases: after the 2013 Washington Navy Yard shooting the FBI reported the gunman had etched 'My ELF weapon!' on his shotgun and believed he was under an 'ultra low frequency attack', and a Pennsylvania man who shot at a neighbour's house in 2021 told police he believed the household was beaming radio waves at him (plea and sentence 2023). Those episodes show how the belief can escalate, not that any such weapon existed.
Unproven
Voice-to-skull, distributed civilian DEW networks, and the mind-control framing.
How to tell
Identifiable source + a measurable signal + correlation with a specific neighbor points to a credible residential case. Distributed strangers + no measurable source + symptoms that follow you everywhere is consistent with the TI pattern — and warrants a medical and audiological workup, not as a dismissal but to find the real cause. Two pleading rules from the dismissals: name a specific source and a measurement, not a technology; and describe what a meter recorded, not what a device must be doing.
Claimed
A mysterious global low hum, sometimes attributed to secret infrastructure or weapons.
Documented
Some hums are real and traceable: the Windsor Hum in Ontario was linked to U.S. Steel's Zug Island plant and stopped when it idled in 2020 (see Documented Cases). In Kokomo, Indiana a city-funded study (2003) measured 10 Hz and 36 Hz tones from two industrial plants and both were silenced by 2004 — yet most residents' symptoms persisted (Cowan, ICBEN 2008). In Taos, New Mexico a multi-agency 1993 study found no acoustic or electromagnetic signal at all in the 30–100 Hz range hearers described (Mullins & Poteet, JASA 1994). Styria's state hum project group (2021–24) logged 46 low-frequency complaints, investigated eight in depth and traced one to a neighbouring heat pump; a Dutch municipality (Lansingerland, 2025) measured well below legal limits and could not trace the frequencies to a single source. A July 2026 NTNU study links many 'Hum' reports to a form of low-frequency tinnitus — internal, not an outside sound (ScienceDaily). The volunteer World Hum Map holds about 4,500 self-reports (2018–24) with no measurements attached.
Unproven
A single global source, or a covert cause. Also unproven: that removing a measured source resolves every complaint — Kokomo is the cautionary case.
How to tell
Record with a C- or Z-weighted meter and check whether it stops when you leave the building, wear earplugs, or the suspected source is switched off. If nothing measures and the hum follows you everywhere, see an audiologist.
Claimed
A specific low frequency (the "brown note," around 7 Hz) can incapacitate a person or cause loss of bowel control, and infrasound can be used as a covert weapon through walls.
Documented
Militaries have researched acoustic weapons, and audible crowd-control devices (LRAD) are deployed — see the state-devices card below. High-level infrasound can cause annoyance, a pressure sensation, and, at extreme levels, measurable physical effects. The origin of the 'lethal 7 Hz' story is Vladimir Gavreau's 1960s CNRS lab: a historical review finds the reported injuries came from a 196 Hz whistle at 150–160 dB, and that his 7 Hz devices and 1966 patent show no evidence of effects on people or of ever being built and tested (Bowdler, 2018). Granted US patents (1971, 1982, 1999) describe acoustic-weapon concepts; a 2011 Raytheon application was abandoned in 2013. A patent certifies novelty, not a working device.
Unproven
The 'brown note': repeated testing has failed to show any frequency that reliably causes loss of bowel control. Covert infrasonic incapacitation of individuals through walls at residential levels is not established. The 'haunted lab' report (Tandy, 1998: a 19 Hz standing wave from an extractor fan) and the 2003 'Soundless Music' concert (about 22% of ~750 listeners reported more unusual feelings during pieces laced with ~17 Hz) are suggestive anecdotes, never peer-reviewed; later controlled work tied such sensations to suggestibility as much as to sound.
How to tell
Residential exposure is about annoyance, sleep disruption, and measurable low-frequency pressure — not incapacitation. If the claim involves being "knocked out" or physically disabled by a hidden tone, that is the myth, not the documented effect.
Claimed
Chronic low-frequency and infrasound exposure causes systemic tissue changes across organs ("vibroacoustic disease").
Documented
Proposed by Castelo Branco & Alves-Pereira (2004) from studies of aviation and other heavily exposed workers.
Unproven
The concept has not been independently replicated and is not accepted by mainstream acoustics or medicine.
How to tell
Treat VAD as a hypothesis, not a diagnosis. Document measurable exposure and well-established effects (sleep disruption, annoyance) rather than relying on VAD framing in a complaint or claim.
Claimed
Wireless, EMF, or 5G exposure causes the same symptom cluster as low-frequency noise.
Documented
EMF is real, and the symptoms reported under "electromagnetic hypersensitivity" are real to the people experiencing them.
Unproven
Double-blind provocation studies have not shown that people can detect EMF, or that EMF causes the symptoms. It is a distinct question from acoustic low-frequency noise.
How to tell
If symptoms track a measurable acoustic or vibration source and shared walls, that points to low-frequency sound — which you can measure and act on, a more productive path than EMF.
Claimed
Purpose-built devices sold online let a resident punish the flat above or below with knocking, vibration or bass that the user's own room barely hears — and nothing can be done about it.
Documented
They exist and courts treat them as harassment. A 'building shaker' (震楼器) is a motor or transducer clamped to the ceiling or a wall that hammers a rhythm, plays bass, or (in one Taiwan case) screams and animal cries into the structure. China's Supreme People's Court picked a Guangzhou shaker case as a 'typical case' in 2023 (Haizhu), courts have awarded emotional-distress damages (Leshan) and police have detained users (Shenyang); Taiwan convicted a ceiling-shaker user of coercion (Shilin); South Korea has fined ceiling-woofer builders (Seongnam). A March 2026 investigation by The Paper, reviewing 185 police penalty decisions and 131 judgments, counted at least 162 people warned or fined and 22 detained since 2018 (澎湃新闻). China's revised Public Security Administration Punishment Law (in force 1 January 2026, Art. 88) makes persistent social-life noise punishable by detention of up to 10 days (Xinhua). After the Haizhu court's judicial suggestion, major platforms blocked shaker keywords; a 2025 follow-up found one still did not, and sellers have re-labelled the devices as 'bone-conduction speakers' promising 45–80 dB delivered to the neighbour while the user's room stays quiet.
Unproven
That the devices are undetectable or legally untouchable. Every case above was traced by recordings, police attendance, a seized unit, or a structural-vibration survey. Outside East Asia the same effect is achieved with ordinary subwoofers face-down on the floor or speakers aimed at a party wall — see the UK, Spanish and US cases in the dataset.
How to tell
Structure-borne, rhythmic, clearly from one direction, and often timed to your own activity: that pattern points to a device, and it is prosecutable. Record with a dB(C)/dB(Z) meter and video, log times, and ask police to attend while it runs. We describe these devices so people recognise them; we do not link to listings.
Claimed
Specific continuous levels near AI and crypto data centers — e.g. facilities running up to ~96 dB (backup generators ~105 dB), and ~50–62 dB inside nearby homes — with a cluster of health effects.
Documented
Real litigation is underway and growing. Four 2026 class actions now sit in the dataset: Ostergaard v. Microsoft (Mount Pleasant, WI — the complaint pleads a 'low hum' of low-frequency sound 'not detected and reflected in dBA measurements'; settlement in principle reported Sept 2026), Montgomery v. DataOne (Vineland, NJ), and Haley / Carter v. X.AI (Southaven, MS and Memphis, TN — gas turbines, 'low-frequency rumbling, and tonal humming'), alongside Valenzuela (Dowagiac, MI) and the Granbury bitcoin-mine suit, which survived a motion to dismiss. Governments are acting: Prince William County, VA adopted dB(C) limits and Chandler, AZ requires baseline sound studies — see The dB(A) blind spot.
Unproven
Many specific decibel figures in circulation are single-source and not independently verified; the complaints are allegations, not findings; and health causation is not established. One online tracker wrongly reports Haley as dismissed — only SpaceX was dropped as a defendant.
How to tell
This is a live, fast-moving area. We keep the reports but flag the unverified figures; follow the court records as they develop.
Claimed
High-pitched 'anti-loitering' boxes and youth-dispersal devices are a form of sonic harassment aimed at the public.
Documented
The Mosquito emits 16–18.5 kHz at 75–95 dB, audible mainly to under-25s. It is legal in the UK, still runs in about 30 Philadelphia parks after a 2020 review, and the Council of Europe's Parliamentary Assembly urged a ban on rights and precautionary grounds while noting research found no hearing-loss risk (Recommendation 1930 (2010)); the Committee of Ministers declined to endorse a ban. In Pléneuf-Val-André, France, a mayor's ~17 kHz 'Beethoven' box aimed at teenagers was switched off after a court challenge (2008).
Unproven
Permanent hearing damage from these devices at their rated levels; and any connection to the low-frequency, structure-borne noise this site covers — they sit at the opposite end of the spectrum.
How to tell
If you can hear it and younger people hear it more, it is high-frequency and airborne, and a smartphone spectrum app will show a spike near 17 kHz. That is a civil-liberties and nuisance question, not a low-frequency one.
The core distinction between documented residential cases and adjacent claims.
Court-documented cases with outcomes and primary sources.
The peer-reviewed evidence base on infrasound and low-frequency noise.
How to measure a low-frequency source and build a record.
Know of a claim we should add, or evidence that moves one of these from "unproven" to "documented"? That's exactly the kind of thing this page should track.
Structured symptom logging, evidence export, and documentation tools for low-frequency noise cases.