In 1912, Congress shoved amateurs onto the “useless” short waves above 1.5 MHz to get them out of the way of commercial and naval traffic — spectrum nobody important wanted. A century later, some of the best-funded firms on earth want it back. Not for voice, not for broadcasting, but to move a few bits of market data across an ocean a handful of microseconds faster than a fiber-optic cable can. This page is about that return: why a trading desk would put a 20-kilowatt transmitter on the HF bands, the experimental stations doing it right now, and the FCC petition — RM-11953 — that has the amateur radio community organized in opposition. It's a companion to the U.S. shortwave broadcasters, time stations, and coast stations pages — the latest chapter in the long story of who gets to use the high-frequency bands, and for what.
Noncommercial hobby compilation, gathered from public sources including FCC filings in proceeding RM-11953, the ARRL, Radio World, and the FCC Experimental Licensing System. Regulatory status changes — figures here reflect the public record as of mid-2026. Corrections welcome — chris@n6jet.com.
In scope: the use of the 2–25 MHz high-frequency bands by financial-trading firms for low-latency data — the physics that makes it worth doing, the experimental stations on the air today under FCC Part 5 licenses, and the Shortwave Modernization Coalition's pending petition (RM-11953) to make it permanent under Part 90, along with the amateur-radio opposition to it.
Not here: a tutorial on high-frequency trading itself, and the broadcast, time, and maritime uses of HF covered on the companion pages above. Nothing here is investment, legal, or regulatory advice.
Why a trading firm wants the ionosphere
The edge, in microseconds — light through glass vs. light through sky
The experimental fleet — who's on the air now
The amateur objection — adjacency, power, and the noise floor
Modern markets are an arms race measured in millionths of a second. When the same instrument trades on two exchanges — say Chicago and a venue in Europe — the firm whose price information arrives first can act on the gap before anyone else closes it. For two decades that race was run on fiber-optic cable and, on land, line-of-sight microwave relay, which is faster than fiber because the signal travels through air instead of glass. But microwave needs an unbroken chain of towers, and across an ocean there are no towers to build on. Over water, the fast options have been undersea fiber and satellite — and both, it turns out, can be beaten.
The beating is done by the oldest trick in long-distance radio: bouncing a signal off the ionosphere. A high-frequency transmitter aimed at the right angle puts a signal into the sky that refracts back to earth hundreds or thousands of miles away, no cable required. The trading firms aren't using HF the way a ham or a coast station did — no Morse, no voice, no conversation. They're sending narrow bursts of digital data, point to point, fixed station to fixed station, across exactly the over-water gaps where cable is slow and towers are impossible.
Two physical facts make HF attractive despite its tiny data capacity. First, speed. Light in an optical fiber crawls at roughly two-thirds of its vacuum speed, slowed by the glass; a radio wave through the thin upper atmosphere travels at very nearly the full speed of light. Same distance, the radio signal arrives sooner. Second, distance. A cable has to follow the seabed, routed around continents and trenches; an ionospheric hop follows the great-circle path — the actual shortest line over the curve of the earth. Shorter route and faster medium compound, and over a transatlantic or transpacific span the saving can run into milliseconds. In a business where a single millisecond is an eternity, that is the whole game.
The catch — and a knowledgeable ham will point this out immediately — is that HF buys independence, not reliability. It needs no cable, but it is utterly dependent on the ionosphere, which sags and shifts with the sun, the season, and the hour, and can be wiped out by a solar flare at exactly the wrong instant. The data rate is also minuscule by modern standards: a few kilobits per second, enough for a price and a timestamp, nowhere near enough for ordinary internet traffic. So HF doesn't replace the fiber and microwave networks — it rides alongside them as a faster path for the small, time-critical messages, accepting fragility in exchange for that latency edge.
None of this is hypothetical. Trading firms have been running HF data links for years under FCC Part 5 experimental licenses — the temporary, research-flavored authorizations that let a licensee test a system on spectrum it doesn't normally hold. The stations cluster, predictably, near the great exchanges: outside Chicago and across the Hudson from Wall Street, with a few far-flung sites positioned for specific over-water paths. Equipment details on the applications are routinely redacted, and the call signs are deliberately low-profile, so much of what's public has been pieced together from the license database and from hams who went looking for the signals.
The most-documented station, WH2XVO, is held by Virtu Financial (which took the license over from Services Development Company); it is authorized across roughly 6 to 24 MHz at up to 50 kW, with sites in Illinois, Alaska, and New Jersey. Public analysis of its grant shows segments such as 6.765–7.000 MHz and 10.150–11.175 MHz — which, as the next section explains, butt directly against amateur band edges.
| Experimental licensee | Call sign | What's publicly known |
|---|---|---|
| Virtu Financial | WH2XVO | The best-documented HF-trading license; ~6–24 MHz, up to 50 kW. Sites at Aurora & Chicago, IL; Homer, AK; Secaucus, NJ. Assumed from Services Development Co.; tied to New Line Networks, the Jump Trading / Virtu joint venture. Equipment redacted. |
| Toggle Communications | WI2XAJ | West Chicago site; experimenting with a similar system. Documented modes include FSK, AFSK, QPSK, and 8-PSK across the HF range. |
| Rockland Wireless | WK2XJK | Antenna mounted on the historic Armstrong Tower at Alpine, NJ — Edwin Armstrong's own 1930s FM test tower, now carrying trading data. |
| 10Band LLC | — | Cited in the coalition's own filings as an authorized experimental operator (referenced at 20 kW). |
| Skycast Services | — | Named as an SMC-affiliated experimental licensee; site/call details not public. |
| RCA Telecom | — | Named as an SMC-affiliated experimental licensee. |
| County Information Services | — | Named as an SMC-affiliated experimental licensee. |
| m-Wave Networks | — | Named as an SMC-affiliated experimental licensee. |
| Alpha Bravo Communications | — | Named as an SMC-affiliated experimental licensee. |
The coalition behind the permanent petition — see below — comprises the trading firms DRW Holdings, IMC Trading, Jump Trading Group, Virtu Financial, NLN Holdings, Optiver Services, and Tower Research Capital, which are affiliated “variously” with the experimental licensees above. The public record does not pin every firm to a specific call sign.
Experimental licenses are temporary and narrow. To put HF trading on a permanent footing, the firms organized as the Shortwave Modernization Coalition (SMC) and petitioned the FCC for a rule change. The Commission docketed it as RM-11953 and put it out for public comment.
The ask is specific, and the key to understanding the fight is that the SMC is not requesting brand-new spectrum. It wants to expand the use of the existing Part 90 “private fixed” HF bands — industrial/business allocations that already sit scattered between 2 and 25 MHz — with three changes that matter:
| Parameter | Today (Part 90 fixed) | SMC proposal |
|---|---|---|
| Eligible users | Industrial/business licensees | Expanded so trading firms qualify |
| Power | 1 kW PEP, typical | Up to 20 kW |
| Bandwidth | Voice-channel widths | Up to 50 kHz |
| Emission | Voice and data | Non-voice digital data only; no mobile |
The SMC's stated rationale leans almost entirely on efficiency and latency, not on resilience or backup value. Its filings describe “underutilized” HF spectrum and the benefit of moving “time-sensitive” financial data quickly, framing the change as letting markets price assets faster — to the benefit, it argues, of market participants generally. Notably absent is any “when all else fails” emergency-resilience argument; this is a pure speed-and-spectrum-efficiency case.
The petition says it “excludes amateur bands,” and technically it does. The objection — led by the ARRL and echoed in some 800-plus comments filed in the docket — is that exclusion isn't the same as separation. Many of the Part 90 fixed segments the SMC wants to use sit immediately adjacent to amateur allocations, sharing an edge with no guard band between them.
The WH2XVO segments noted earlier are the clean illustration: an authorization running up to 7.000 MHz ends exactly at the bottom edge of the 40-meter amateur band, and one beginning at 10.150 MHz starts exactly at the top edge of 30 meters. A transmitter operating legally just outside a ham band, but radiating tens of kilowatts, does not stop cleanly at the line — real signals have skirts, splatter, and intermodulation products that spill across an edge that exists only on paper.
And the power is larger than it looks. The headline figure is transmitter output, but a directional HF curtain array concentrates that power into a beam — so 20 kW into a high-gain antenna can mean an effective radiated power in the megawatt class along the beam's axis. Opponents argue that this, landing next to the band edge during the same propagation openings that hams rely on, is a recipe for interference that correlates precisely with the moments the bands are usable. The concern isn't theoretical: the FCC's own enforcement arm has already had to act.
In December 2025, the FCC's Office of Engineering and Technology sent letters to experimental HF licensees — in response to interference complaints from incumbent spectrum users — reminding them of an obligation they had apparently been neglecting: an experimental station must transmit its assigned call sign at the end of each transmission, at least once every 30 minutes, in clear voice or Morse, with all digital encoding and modulation disabled during the identification. It's a small administrative point with a large implication — the trading stations had been running effectively anonymously, and it took interference reports from the people next door on the dial to force them to announce who they were.
Step back and this is the same story the rest of these pages tell, viewed from the far end. The high-frequency bands were handed to amateurs in 1912 precisely because the experts of the day judged them worthless. The amateurs, and then the commercial operators, proved otherwise: the coast stations worked ships across whole oceans on a wire and a spark, the international broadcasters threw voices across borders, the militaries built global networks, and the same physics let a Cold War radar (the Soviet “Woodpecker”) stomp across the bands from the other side of the planet. Then satellites arrived and, one by one, made the old HF services obsolete — the AT&T High Seas radiotelephone stations went dark in 1999, the maritime Morse fell silent, and the bands emptied out.
What's happening now is the wheel coming back around. The very trait that made shortwave “worthless” in 1912 and indispensable ever after — that it needs nothing but the ionosphere, no cable, no infrastructure between the endpoints — is now a line item on a hedge fund's risk model. Satellite made maritime HF obsolete; HF now out-runs satellite for the one thing a trading firm cares about. Same spectrum, same physics, a century apart, found valuable again for the same fundamental reason. The only thing that's changed is who wants it, and what they're willing to pay.
As of mid-2026, RM-11953 remains a petition, not a rule. The FCC sought public comment, declined requests to extend the deadline, and the docket drew well over 800 filings — a heavy response for a technical petition, reflecting how thoroughly the amateur community mobilized. The record has stayed active into 2026, with the coalition filing further material and individual comments continuing to arrive, but the Commission has not issued a Notice of Proposed Rulemaking or adopted any rule. The experimental stations, meanwhile, keep operating under their Part 5 grants — now, at least, identifying themselves.
So the question is open. If the FCC moves the petition to a rulemaking, the fight shifts to the exact band edges, power limits, and out-of-band-emission masks — the technical fine print where interference is either contained or isn't. If it lets the petition sit, the experimental fleet continues in its current legal half-light. Either way, the broad-band-edge question that animates it — how much power you can put right next to a neighbor's allocation before “adjacent” becomes “inside” — is one the HF bands will be arguing about for a long time.
Companion pages: Shortwave from America: U.S. International Broadcast Stations · Time Stations: Standard-Frequency and Time-Signal Radio · Coast Stations: U.S. Ship-to-Shore Radio · AM Broadcast Band trivia.
Sources: FCC Electronic Comment Filing System, proceeding RM-11953 (petition, comments, and history report); the ARRL; Radio World; the FCC Experimental Licensing System; and independent technical analyses of the published experimental grants. Compiled by N6JET as a noncommercial hobby reference. Regulatory details change — verify current status against the FCC docket. Questions or corrections? Email chris@n6jet.com.