5 ms·

We are talking radar here though. Radar in general is more sensitive to noise. It's not a digitally encoded signal like communications. It's looking at the refl

by anfilt 5y ago

We are talking radar here though. Radar in general is more sensitive to noise. It's not a digitally encoded signal like communications. It's looking at the reflections of radio waves and their strength among other things. This is an analog process the sensor is using. Before digitally encoded broadcasting was the norm the gaps between frequencies tended be wider to limit noise. For communication even with analog encodings like AM, FM ect... The noise floor is quite a bit higher before it becomes unusable for communications compared to radar. Radar needs to look at a wider band of spectrum as things will get shifted more from reflection and such. Further it needs to read the much weaker reflected signals. The signal is weaker because lots of energy will be lost by the time it's reflected. So it's quite possible some higher order harmonics operating on the edge of the 5G spectrum could bleed into the frequency area the radar is using above the expected noise floor. Keep in mind these sensors were designed and approved long before the FCC allocated this spectrum above this range to 5G. 200 Mhz may be massive for communications, but not that is not entirely true for radar.

darksaints 5y ago
If this were even remotely accurate, the s-band radars that are commonplace for weather radar and maritime radars would be rendered useless by high power telecom/TV BRS and EBS spectrum.
- mirashii 5y ago
  This is a pretty shallow dismissal that doesn't actually provide data to back up the claim. Why would the interference be substantial enough to make them useless? Which weather and maritime radar? There are many at many different bands. What is the meaningful impact of interference on those measurements, and how does it differ from the impact of interference on radio altimeters used in planes during take-off and landing? These are just a few of the reasonable questions that should be answered in such a strong dismissal. Instead, you called it inaccurate without providing any real analysis. RFI is a problem in these spaces [1]. It's been a problem for many years, and is an increasing one. [1] https://www.researchgate.net/publication/283760070_The_Threat_to_Weather_Radars_by_Wireless_Technology https://www.researchgate.net/publication/283760070_The_Threa...
  - darksaints 5y ago
    I already said s-band (which is 2-4ghz) specifically. BRS/EBS (2.5-2.7ghz) has no meaningful guard band to protect s-band radar, and the radars, at least when used on military vessels, are actually dramatically more sensitive than aviation radar altimeters. And not only has EBS/BRS been operating for decades without any problems, but a) the maritime radar emission/detection patterns are horizontal, just like telecom/television, which should produce more interference than the perpendicular patterns of radar altimeters, and b) EBS has operated for 3 decades using television broadcast power levels...1000s of watts as opposed to the 100s of watts that telecom uses. If interference was actually a problem, the DoD would have shut down those two bands back in the 80s. Franky speaking, everybody here is talking out of their asses, and just rehashing bullshit that they get told. I'll lowbrow dismissal that all day long because it doesn't deserve and inch more of effort.
    - amirhirsch 5y ago
      > Frankly … This whole discussion has a really low SNR
- anfilt 5y ago
  I did not say it was impossible to design a radar system to work with a higher noise floor. Also, if you have looked some of the concerns the FCC has in the S-Band range is radar. If you have looked at the frequency allocation rules and such. Quite often around certain frequencies you will see your not to cause interference with radar. However, a lot of radar in the S-Band pulse power tend be quite strong because of atmospheric attenuation, and also noise. Additionally, lot these weather radars are not trying to achieve same level of accuracy as an altimeter. If the weather formations ones monitoring is off by a little bit, but overall the data is close it's fine. Similar thing with a lot maritime radar, ships are not moving that fast so you also have a lot more time to average out the data. If an altimeter when landing is off by too much can cause a lot problems, especially when you factor in the speeds of an airplane it's you have less time to average out noise or doing other things to filter out spurious noise. Edit: Also one other thing about maritime radar there nowhere as much interference once you get outside ports or areas near land. Also it's not just Boeing here having issues, airbus also along with Boeing has brought this to the attention the FAA. The frequency bands in other countries for 5G tend to be farther away as well from the frequencies used by altimeters so it's not been issue in other countries. It only seems to be handful of air-frames that effected. Secondly, NOAA have been concerned about 5G, particularly near 24Ghz, again for possibly interfering with radar. All I was saying is radar is a lot more sensitive to the noise floor and if suddenly that noise floor is higher I am not surprised some equipment will have issues depending on it's use case. Especially, when things were designed long before these new spectrum allocations.
  - hilbert42 5y ago
    "Especially, when things were designed long before these new spectrum allocations." See my post about international radio regulations, the 5G service should not have been allocated if it was going to interfere with radar altimeters as that was (and still is) a 'going' service and it takes precedence under international radio treaties. Thus, an aircraft from a country that's not under US jurisdiction expects to be able to land in the US without its radar altimeter being jammed by some unexpected local service. (It stands to reason that the safety of an aircraft that travels across international borders should not be endangered because of inference to its navigation equipment.) This is the precise reason why international radio regulations exist! Under the treaty the 'interference' rules apply to ALL radiocommunication services. If such regulations (standards) are deemed 'old' then they have to be changed at an ITU WRC meeting in Geneva and the item must appear on the meeting's agenda some years before said meeting takes place to give all countries adequate time to respond (as I mentioned elsewhere, I was once involved in that process).
parsimo2010 5y ago
The FAA and FCC need to figure out what to do, but the above comment isn't the consideration they need to be worried about. > Radar in general is more sensitive to noise.... The noise floor is quite a bit higher before it becomes unusable for communications compared to radar This is not true. A radar has the huge advantage of knowing what signal was sent and only looks for returns that are similar- the FMCW radars in question often mix the return with the originating signal, and outside interference just doesn't cohere and won't affect the result unless the interference is very high power. Radars do not need to look at a wider band of spectrum, reflections do not change the frequency and the Doppler shift a radar altimeter (looking straight down) needs to accommodate is minuscule. A radar that can't handle noise is poorly designed, not that they are inherently susceptible to noise. Radar does require "high" transmit power to easily detect returns, but they can be robust to noise. A radar (including RA) can integrate over a relatively large window if needed and detect with an SNR <1 if it's designed that way. Whether the old 737 RA was properly designed and can handle the interference, or if they took shortcuts because saving a buck took precedence is a different discussion than the (incorrect) generalities about radar presented in the above comment. Whether the existing RAs can be retrofitted with better filters for cheaper than lobbying the FCC is also unclear, but we know what approach the airlines are attempting right now.
- wumpus 5y ago
  In support of this comment, as a radio astronomer I used to think I understood radar. Then I sat in a presentation by some smart people from MIT, and it turns out that I was really wrong, but the things actual radars do are built on top of things that I know from astronomy.
- anfilt 5y ago
  It is more sensitive noise. I did not say it was impossible though to design radar with a higher noise floor in mind. Like the main reason this is problem is not really Boeing's or Airbus's fault. It's only now the FAA went wait a minute when they could have been looking at this years ago since the FCC did announce this quite a bit ago now. The allocation here in the US is closer than to other countries so spurious emissions from 5G equipment and higher order harmonics could possibly be larger issue.
Dylan16807 5y ago
> 200 Mhz may be massive for communications, but not that is not entirely true for radar. The gap is the same size as the entire allocation, and the frequency difference is more than 4 percent. That's massive. > Radar needs to look at a wider band of spectrum as things will get shifted more from reflection and such. If you skew a radar signal by a thousand miles per hour, isn't that around 2 parts per million? Am I missing an effect here?
- anfilt 5y ago
  Radar for an altimeter since by it's very nature it needs to measure distance and not just speed is going to be frequency modulated. Unmodulated/continuous frequency radar can only measure speed. When you start processing the small amount of doppler frequency shift from reflections with a changing frequency things get tricky especially when one has have multi-path return (aka it's reflecting off multiple objects) This will lead two or more reflections with roughly the same frequency, but they have a phase shift. That's the kind of shift I am talking about here. The more noise you have the more difficult it is to separate these reflections when processing these signals. Edit: I will add the a fair amount of interference by multi path return can be mitigated by making the transmitter and receiver shielded from other directions and designed to be highly directional. This also is generally not to hard to do for an altimeter since the datatum one is measuring after all is just elevation over a small area, compared to something like radar for topography that wants to cover a large area.
  - Dylan16807 5y ago
    Okay, sure, you frequency modulate your output. But the frequencies you get back should be almost identical, right? That doesn't explain why something 200MHz away would interfere.
    - anfilt 5y ago
      Yes, they will be very close in frequency. If you have multi-path return you will have two more or more frequencies with different phases. The problem is not so much this alone. All transmitters will have spurious emissions of some level and also your going to have higher order harmonics of the root frequency that will be outside the allocated spectrum. The higher the transmit power of a transmitter the more higher order harmonics you will see above the noise floor. At least for low power transmitters you don't have to worry about more than 2nd order harmonic. Although if you have a powerful enough transmitter like a cell tower 3rd order harmonics can be a problem. Spurious noise is going to depend a lot the design of the transmitter and amplifiers and filters. So how much noise leaks into this bad used for aircraft altimeters is gonna depend alot on the quality of the equipment cellar providers use. Noise can effect radar more easily than normal communication transmissions.

by anfilt 5y ago

darksaints 5y ago

If this were even remotely accurate, the s-band radars that are commonplace for weather radar and maritime radars would be rendered useless by high power telecom/TV BRS and EBS spectrum.

mirashii 5y ago
This is a pretty shallow dismissal that doesn't actually provide data to back up the claim. Why would the interference be substantial enough to make them useless? Which weather and maritime radar? There are many at many different bands. What is the meaningful impact of interference on those measurements, and how does it differ from the impact of interference on radio altimeters used in planes during take-off and landing? These are just a few of the reasonable questions that should be answered in such a strong dismissal. Instead, you called it inaccurate without providing any real analysis. RFI is a problem in these spaces [1]. It's been a problem for many years, and is an increasing one. [1] https://www.researchgate.net/publication/283760070_The_Threat_to_Weather_Radars_by_Wireless_Technology https://www.researchgate.net/publication/283760070_The_Threa...
- darksaints 5y ago
  I already said s-band (which is 2-4ghz) specifically. BRS/EBS (2.5-2.7ghz) has no meaningful guard band to protect s-band radar, and the radars, at least when used on military vessels, are actually dramatically more sensitive than aviation radar altimeters. And not only has EBS/BRS been operating for decades without any problems, but a) the maritime radar emission/detection patterns are horizontal, just like telecom/television, which should produce more interference than the perpendicular patterns of radar altimeters, and b) EBS has operated for 3 decades using television broadcast power levels...1000s of watts as opposed to the 100s of watts that telecom uses. If interference was actually a problem, the DoD would have shut down those two bands back in the 80s. Franky speaking, everybody here is talking out of their asses, and just rehashing bullshit that they get told. I'll lowbrow dismissal that all day long because it doesn't deserve and inch more of effort.
  - amirhirsch 5y ago
    > Frankly … This whole discussion has a really low SNR
anfilt 5y ago
I did not say it was impossible to design a radar system to work with a higher noise floor. Also, if you have looked some of the concerns the FCC has in the S-Band range is radar. If you have looked at the frequency allocation rules and such. Quite often around certain frequencies you will see your not to cause interference with radar. However, a lot of radar in the S-Band pulse power tend be quite strong because of atmospheric attenuation, and also noise. Additionally, lot these weather radars are not trying to achieve same level of accuracy as an altimeter. If the weather formations ones monitoring is off by a little bit, but overall the data is close it's fine. Similar thing with a lot maritime radar, ships are not moving that fast so you also have a lot more time to average out the data. If an altimeter when landing is off by too much can cause a lot problems, especially when you factor in the speeds of an airplane it's you have less time to average out noise or doing other things to filter out spurious noise. Edit: Also one other thing about maritime radar there nowhere as much interference once you get outside ports or areas near land. Also it's not just Boeing here having issues, airbus also along with Boeing has brought this to the attention the FAA. The frequency bands in other countries for 5G tend to be farther away as well from the frequencies used by altimeters so it's not been issue in other countries. It only seems to be handful of air-frames that effected. Secondly, NOAA have been concerned about 5G, particularly near 24Ghz, again for possibly interfering with radar. All I was saying is radar is a lot more sensitive to the noise floor and if suddenly that noise floor is higher I am not surprised some equipment will have issues depending on it's use case. Especially, when things were designed long before these new spectrum allocations.
- hilbert42 5y ago
  "Especially, when things were designed long before these new spectrum allocations." See my post about international radio regulations, the 5G service should not have been allocated if it was going to interfere with radar altimeters as that was (and still is) a 'going' service and it takes precedence under international radio treaties. Thus, an aircraft from a country that's not under US jurisdiction expects to be able to land in the US without its radar altimeter being jammed by some unexpected local service. (It stands to reason that the safety of an aircraft that travels across international borders should not be endangered because of inference to its navigation equipment.) This is the precise reason why international radio regulations exist! Under the treaty the 'interference' rules apply to ALL radiocommunication services. If such regulations (standards) are deemed 'old' then they have to be changed at an ITU WRC meeting in Geneva and the item must appear on the meeting's agenda some years before said meeting takes place to give all countries adequate time to respond (as I mentioned elsewhere, I was once involved in that process).

parsimo2010 5y ago

The FAA and FCC need to figure out what to do, but the above comment isn't the consideration they need to be worried about. > Radar in general is more sensitive to noise.... The noise floor is quite a bit higher before it becomes unusable for communications compared to radar This is not true. A radar has the huge advantage of knowing what signal was sent and only looks for returns that are similar- the FMCW radars in question often mix the return with the originating signal, and outside interference just doesn't cohere and won't affect the result unless the interference is very high power. Radars do not need to look at a wider band of spectrum, reflections do not change the frequency and the Doppler shift a radar altimeter (looking straight down) needs to accommodate is minuscule. A radar that can't handle noise is poorly designed, not that they are inherently susceptible to noise. Radar does require "high" transmit power to easily detect returns, but they can be robust to noise. A radar (including RA) can integrate over a relatively large window if needed and detect with an SNR <1 if it's designed that way. Whether the old 737 RA was properly designed and can handle the interference, or if they took shortcuts because saving a buck took precedence is a different discussion than the (incorrect) generalities about radar presented in the above comment. Whether the existing RAs can be retrofitted with better filters for cheaper than lobbying the FCC is also unclear, but we know what approach the airlines are attempting right now.

wumpus 5y ago
In support of this comment, as a radio astronomer I used to think I understood radar. Then I sat in a presentation by some smart people from MIT, and it turns out that I was really wrong, but the things actual radars do are built on top of things that I know from astronomy.
anfilt 5y ago
It is more sensitive noise. I did not say it was impossible though to design radar with a higher noise floor in mind. Like the main reason this is problem is not really Boeing's or Airbus's fault. It's only now the FAA went wait a minute when they could have been looking at this years ago since the FCC did announce this quite a bit ago now. The allocation here in the US is closer than to other countries so spurious emissions from 5G equipment and higher order harmonics could possibly be larger issue.

Dylan16807 5y ago

> 200 Mhz may be massive for communications, but not that is not entirely true for radar. The gap is the same size as the entire allocation, and the frequency difference is more than 4 percent. That's massive. > Radar needs to look at a wider band of spectrum as things will get shifted more from reflection and such. If you skew a radar signal by a thousand miles per hour, isn't that around 2 parts per million? Am I missing an effect here?

anfilt 5y ago
Radar for an altimeter since by it's very nature it needs to measure distance and not just speed is going to be frequency modulated. Unmodulated/continuous frequency radar can only measure speed. When you start processing the small amount of doppler frequency shift from reflections with a changing frequency things get tricky especially when one has have multi-path return (aka it's reflecting off multiple objects) This will lead two or more reflections with roughly the same frequency, but they have a phase shift. That's the kind of shift I am talking about here. The more noise you have the more difficult it is to separate these reflections when processing these signals. Edit: I will add the a fair amount of interference by multi path return can be mitigated by making the transmitter and receiver shielded from other directions and designed to be highly directional. This also is generally not to hard to do for an altimeter since the datatum one is measuring after all is just elevation over a small area, compared to something like radar for topography that wants to cover a large area.
- Dylan16807 5y ago
  Okay, sure, you frequency modulate your output. But the frequencies you get back should be almost identical, right? That doesn't explain why something 200MHz away would interfere.
  - anfilt 5y ago
    Yes, they will be very close in frequency. If you have multi-path return you will have two more or more frequencies with different phases. The problem is not so much this alone. All transmitters will have spurious emissions of some level and also your going to have higher order harmonics of the root frequency that will be outside the allocated spectrum. The higher the transmit power of a transmitter the more higher order harmonics you will see above the noise floor. At least for low power transmitters you don't have to worry about more than 2nd order harmonic. Although if you have a powerful enough transmitter like a cell tower 3rd order harmonics can be a problem. Spurious noise is going to depend a lot the design of the transmitter and amplifiers and filters. So how much noise leaks into this bad used for aircraft altimeters is gonna depend alot on the quality of the equipment cellar providers use. Noise can effect radar more easily than normal communication transmissions.