Item — HN Reader

20,000 watts!? The industry's petition is here: https://www.fcc.gov/ecfs/document/1042840187330/1

The petition requests "maximum transmitter power is 20,000 watts RMS" in the HF 2 – 25 MHz range. Wow. 20 kW is a LOT of power. Two amateurs in HF can talk to each other on opposite sides of the planet with 5 W and a tiny rig and a lot of skill. I think most amateurs would consider 100 W a lot of power for competent HF comms.

But American industry wants 20,000 W for coast-to-coast jibber jabber? Greedy. Lazy. And pollution.

The headline petitioners are:

  Eric Bellerive on behalf of DRW Holdings LLC

  Thomas Maxwell on behalf of IMC Trading
 
  John P. Madigan on behalf of NLN Holdings LLC
 
  Kevin Nielsen on behalf of County Information Services LLC,
  an affiliated company of Optiver Services B.V.
  
  Tom Proudley on behalf of Tower Research Capital LLC

7 comments

Why? 20 kW is very common for FM radio, which is also just streaming shitty ads to people. Warsaw had a 1 MW transmitter.

Power density scales with inverse of distance squared. At 1km, 20 kW is 0.001 W/m2. At 500 km it's a few nW/m2.

The issue here is the frequency used plus power and spurious out of band emissions. 'FM Radio' (really 150KHz wide FM modulated 88-108Mhz) has propagation generally limited to line of sight, perhaps a few hundred miles at most due to earth curvature and the signals go off through the atmosphere into space so do not interfere with anything beyond the immediate range. FM also has good interference immunity due to the capture effect of FM signals so spurious emissions nearby are typically ignored.

2-25Mhz reflects off the atmosphere and returns to earth and the multiple allocated bands in the range are typically less then 6KHz for broadcast AM and other bands often <3KHz SSB down to 100Hz CW can be completely trashed with spurious emissions. 60, 80m and 160m can be difficult to use already due to all sorts of noise from notionally low power devices in the mW range on nearby, or sometimes nowhere near, frequencies. The wide range is probably to enable use regardless of the current maximum usable frequency (MUF) to ensure atmospheric reflection. MUF varies, sometimes considerably, by time of day, season etc. However the reflections do not stop when you receive the signal, they continue to skip and be received in many additional locations that are not the intended destination.

Someone pumping 20KW across the range could cause an incredible amount of disruption, essentially worldwide. See Ringway Manchester's great series of videos on the Russian Woodpecker (it was speculated in the MW power range) https://www.youtube.com/watch?v=wB_jiigRq-c

20kW is a lot for HF because it's a "planet scale" frequency band. VHF signals fly off into space when they hit the curvature of the earth, but HF signals bounce off (technically are refracted by) the ionosphere, sending them back down to earth. Thus, VHF is only "line of sight"; if someone off the horizon abuses VHF, it's not a problem for your VHF communications on the same frequency.

HF signals can be reflected by the surface of the Earth after hitting the ionosphere and repeat this process until they've gone around the world (sometimes more than once!). This is not a rare quirk; amateur radio operators take advantage of this on a daily basis, with an absolute maximum power limit of 1500W. (But 1500W is uncommon; my radio does, at max, 10W. And I can communicate with Europe from NYC fairly regularly.) The result of all of this is you have to be careful about HF signals; you can ruin it for everyone in the world if you make a mistake. VHF, whatever.

VHF does occasionally propagate over long distances. It used to happen with analog TV and was quite the rare and amazing sight to behold (I've never seen it).

https://en.wikipedia.org/wiki/Tropospheric_propagation#Tropo...

https://en.wikipedia.org/wiki/Sporadic_E_propagation

TV moved from VHF to UHF during the digital transition, so this doesn't happen anymore.

Finally, there is a lot more VHF frequency available than HF. For example, one band that amateurs have is the 40 meter band, which is 7MHz to 7.3MHz. That is 300kHz of bandwidth for everyone in the world. Meanwhile, just 1 FM radio station is 270kHz!

> TV moved from VHF to UHF during the digital transition, so this doesn't happen anymore.

In the US, I don't think VHF ATSC is going anywhere, because there are only 23 UHF channels remaining (14-36) and VHF adds 12 more. VHF is less popular because you need a larger antenna to receive it.

You may know this already, but for others:

It's only higher of the two previously existing TV bands (channels 7-13, 174-216MHz) that is still allocated for terrestrial broadcasts in the US The old 54-88MHz (channels 2-6) are no longer used in the US for ATSC.

https://rabbitears.info/market.php shows a bunch of ATSC on channels 2-6.

FM is VHF, a completely different beast than HF.

Just wait for very high frequency trading firms to appear.

Let's give them EHF for future proof

> But American industry wants 20,000 W for coast-to-coast jibber jabber?

Chicago to New York, mostly. The derivatives markets are in Chicago and the cash markets are in New York. They've been testing at only 10Kb/s, and they're probably using forward error correction, which uses more bits. (Given the application, retransmission adds too much latency, so that's out.) 20KW would reduce the error rate.

I once read someone claiming that a paper study was done on the feasibility of drilling a small tunnel directly through the crust between Chicago and New York in a straight line to create an evacuated tunnel like a miniature hyperloop for light speed (in a vacuum) line of sight laser communication.

I think the proposed technique was laser or microwave vaporization tunneling at a small diameter.

Something tells me it was too expensive.

HFT is psychotic.

I mean, the market price in a major commodity like wheat or oil is the aggregate desires of 7.8 billion people. Move it a micropenny in either direction and it'll affect man-centuries of work. Market makers reduce the bid-ask spread of a commodity, which is good!

Narrowing the spread of the prices of a commodity doesn't have any immediate physical effect, but if you start wiping out billion dollar projects based on that, then that's the most office buildings, the ISS and LHC, the entire software industry...

I think HFT (i'm assuming high frequency trading) 'owns' all electricity

Can you imagine explaining something like that to an alien civilization

"You spent your limited resources and focused your technology on building this? Are you crazy?"

It is a lot, but I think they want to be able to blast through difficult sunspot cycle activity.

That's quite an electric bill, when you consider they're asking for up to 50 kHz bandwidth.

> That's quite an electric bill, when you consider they're asking for up to 50 kHz bandwidth.

Watts are watts, whether you spread them across 1 Hz or 1GHz.

I'm talking about the bandwidth of the individual signal. A 5 kHz wide signal uses a lot less electricity than a 50 kHz wide signal.

The same output power is the same output power.

In general, the opposite of what you say is true: the more occupied bandwidth, the less power you need for equivalent information transfer. So, e.g. when the FCC made everyone doing VHF/UHF voice comms narrowband from 25KHz to 12.5KHz, you needed basically double the power for equivalent analog voice.

Or for an extreme-- GPS uses 50MHz wide to send dozens of bits per second to allow much lower satellite output power and resilience to noise.

Are you saying that a transmitter with an output of 20,000 watts and a bandwidth of 50 kHz using digital modulation uses the same amount of electricity as a transmitter with 20,000 watts and a bandwidth of 1 kHz using digital modulation?

Yes. Watts are the unit of power. Either of them would probably use 22,000-35,000 watts of electrical input power to produce an output radio signal of 20,000 watts.

The former could be expected to transfer much more information, or alternatively to go further. (Or the former you could use less than 20,000 watts to get the same data transfer rate and range as the latter).

Thank you.

Throughput is proportional to bandwidth and is logarithmic with SNR (i.e. output power).

All being equal, you'd rather they use a narrow signal with a lot of power than a wider one. The greedy/lazy/polluting way would be to demand a wide allocation so that you can use low power.

Of course, a very loud signal right next to amateur allocations may mess with lower-quality receivers.

It's worth noting NLN Holdings is a joint venture of Jump and Virtu, also some big names within the HFT space.

20,000 watts is not a big deal. You want a strong Signal, all day between Chicago and New York? A power level with some design margin is what you need.

Provided the transmitter is 'clean' and modulated such that the signal stays inside the allocated bandwidth, it will not impact other users.

(except if you are < 1 Mile from the transmitter. the poor selectivity in your receiver may be overwhelmed by this power up close).

The question is how much bandwith and what kind is the modulation? Narrow band modes consume less spectrum, and they also transfer less data.

A 50-100 khz band could be carved out of HF and given to this services. 500khz would be way to much.

And you cannot take it from the Ham bands. They were there first.

The military already uses 15KW in HF as a fallback to satellite. I maintained those old systems for a while. They've been doing this since WWII. There is chatter every day confirming communications to Air Force 1 and 2. We used Chirpsounder to see which freqs would be best that day.

Interesting stuff, I bet! And I feel like it's much easier to justify high power carve-outs on public spectrum for the purpose of defending lives or country than for no purpose but to lower already low latency for opportunistic high frequency trading systems that exist primarily to make more money for very very few people. That's what this petition is all about as I read it.