4 ms·
How can this be a real problem with exponential signal loss over distance? Sure, two devices in close proximity may interfere with each other. But that is in no
by zaroth 10y ago
How can this be a real problem with exponential signal loss over distance? Sure, two devices in close proximity may interfere with each other. But that is in no way a "noise floor issue which will render certain frequencies unusable."
- krastanov 10y agoSignal strength does not decay exponentially, rather as distance^-3 for a dipole[1]. Most noise sources are probably well approximated by a dipole antenna[2]. Some fancy modern modulation scheme with extremely dense encoding of information might suffer stronger decay, but the article is about simple sources of noise, not about modern modulation schemes. [1]: https://en.wikipedia.org/wiki/Dipole#Vector_form https://en.wikipedia.org/wiki/Dipole#Vector_form [2]: https://en.wikipedia.org/wiki/Dipole_antenna#Hertzian_dipole https://en.wikipedia.org/wiki/Dipole_antenna#Hertzian_dipole
- Bromskloss 10y ago> rather as distance^-3 for a dipole Wait a minute. The radiated power in the far field goes like r^-2, right?
- kqr 10y agoThis is my intuition as well, because at a distance r around the source you have a sphere, and the area of the sphere will relate to the radius r as r^-2. That said, the equation linked in the comment does indeed contain an r^-3 factor. I don't know if it's cancelled out by something else – my physics game isn't strong enough.
- Bromskloss 10y agoThe expressions for the field include terms of different powers of r. At large distances (compared to the wavelength) from the transmitting antenna, only the r^-1 terms (which dies off with r more slowly than the others) remain with any appreciable strength. This is called the "far-field". The power per cross-sectional area is proportional to E·B (or, equivalently, to E^2 or to B^2), and therefore goes as r^-2.
- leni536 10y agoThe first link is for static dipole, no radiation. The second link shows that the for a dipole antenna there is a 1/r term in amplitude. It creates an 1/r^2 term in power so your intuition is right.
- krapht 10y agoMaybe you're not aware of this, but communication signals decay at the same rate as RF noise, so the overall effect is a push.
- zaroth 10y agoI think that's a tautology. I'm not sure what "overall effect is a push" means... Is "incidental noise" the dominant factor in the overall noise floor? Maybe you could call this another kind of "global warming". Any existing historical data readily available?
- tlrobinson 10y agoIs historical data necessary when you can just compare urban to suburban to rural areas? There's clearly a correlation to population (and thus noisy electronics) density. I've tried listening to the HF ham radio bands from my downtown SF apartment with very little success.
- Johnny555 10y agoThat's an awfully wide range of frequencies, from a couple Mhz to hundreds of GHz. I've been able to hit local repeaters on 2m and 70cm bands from my apartment in SF. I don't do much on HF though.
- tlrobinson 10y agoHF is ~3 to 30 MHz. I can hit VHF/UHF repeaters too, but thats an apples to oranges comparison.
- blackguardx 10y agoThe problem is if you are trying to receive a signal from far away and something in your house or your neighbors house is polluting the band. The interfering signal can be stronger than the desired signal.
- Gibbon1 10y agoCouple of things, there are just more and more devices that have high frequency electronics. So increasingly we're surrounded by a sea of radiators. A problem with all transmitters is they produce out of band signals. Mostly harmonics, but also lesser amounts of other frequencies. And another issue is that receivers only have limited ability to reject out of band signals. A strong out of band signal will reduce your receivers sensitivity. (Difficulty, out of band rejection comes at the expense of ultimate sensitivity, pick your poison) The problem then is that the amount of energy needed to raise the noise floor is tiny, minuscule. And we have all these devices that all add a tiny amount each, which then adds up to something. If you raise the noise floor a few db it matters a lot, because following reason. Consider this, if you double the distance from the transmitter, the signal strength drops by ~6db. So if interference reduces your receivers sensitivity from -120db to -114db it cuts the effective range in _half_[1]. [1] It's not quite half because in the real world you have air, trees, and buildings in the way which means your signal drops off faster than the inverse square. Instead the exponent is more like 2.5(outside) to 3.5(inside)
- VLM 10y agoThe recent problem isn't the decay rate of RF but the cost of noise attenuation per dollar. A legal switching power supply in a PC might knock out all AM radio reception on the same desktop. That's because the mfgr installed $5 worth of inductors and bypass capacitors in the power supply input section. So maybe 500 watts output, 4.999 watts of noise gets turned into heat by those interference filters, and 0.001 watts messed up the AM radio on the same desktop. (edited to add BTW if you think your cheap switching power supply is 99% efficient I have a bridge to sell you... the actual numbers are worse) The problem is those inductors and capacitors and ferrite cores are $5 of "lost" profit in China, and technically the power supply works with them replaced by pieces of copper wire. So the same 500 watts output, and 5 watts of noise broadcast into the neighborhood. Well, at least its not turned into heat inside the chassis, LOL. The problem is a 5 watt transmitter is going to knock out AM radio for some city blocks in every direction. In the 30s your average house had like one piece of electronics, probably an old console radio. But now houses are full of electronic junk from China so you're battling pretty much everyone who wants to get the lowest priced XYZ from walmart. Like some forms of pollution, RF interference is virtually impossible to clean up and the biggest violators have a significant payoff, higher profits at the factory and cheaper products on the shelf, or some continuum along that spectrum anyway. A good analogy would be trying to organic garden in an urban environment surrounded by pollution and pesticide spraying. You might not be dumping god knows what into your dirt, but your neighbors simply don't care and are constantly spraying or dumping who knows what, and you share the environment so even if per UDSA your dirt is legally organic because YOU didn't mess it up, you still know its a regular toxic brew in there because of your neighbors ...
- Gibbon1 10y agoI have no direct experience with power supplies radiating, except to note that switching frequencies are up significantly, from 20khz twenty years ago to 500khz to 2MHZ today. The AM band is something like 700khz to 1.4MHZ. Higher switching frequencies and the push for higher efficiency[1] also means very fast switch times, on the order of 20-100ns. High speed signals like that contain very strong harmonics which will leak out. [1] Old guy mentioned to me what you care about is not the efficiency, but the inefficiency. Because that's directly related to heat. Going tom 90% efficient to 95% efficient, who cares! Except you reduced the amount of heat you need to dissipate by half from 10% to 5%.