3 ms·

As someone who doesn't know much about telecommunication tech, I have a hard time understanding how this happens. From what I've read, the frequency of 5G doesn

As someone who doesn't know much about telecommunication tech, I have a hard time understanding how this happens. From what I've read, the frequency of 5G doesn't interfere with the C-band, as the C-band is 4-8 GHz. This is between the ranges used by the mid and high band spec of 5G. So why do they auction off "parts of the C-band", and how is it connected to 5G? I'm sure I got something wrong, just curious what it is.

sidewndr46 6y ago
There is a letter here that does a better job than the linked article: https://transportation.house.gov/news/press-releases/chair-defazio-calls-on-fcc-to-postpone-tomorrows-scheduled-auction-of-a-portion-of-37-42-ghz-radio-frequency-spectrum-citing-new-research-that-amplifies-the-safety-concerns-of-the-aviation-community https://transportation.house.gov/news/press-releases/chair-d... Basically, it sounds like they agreed to auction 3.7 - 3.98 GHz & 4.0 GHz - 4.2 GHz. This leaves a gap in the region of 3.98 - 4.0 GHz for radar altimeters. The letter states "recent research confirms that even this spectrum gap is insufficient to ensure radio altimeters are protected from harmful RF interference by 5G networks". I don't know enough about a radar altimeter's operation to speculate as to why it could be an issue.
- smarx007 6y ago
  No, it's not the 3.98-4.0 GHz gap. > Given the neighboring aviation band (from 4.2 to 4.4 GHz) allocated exclusively for vital aeronautical radionavigation equipment They are afraid of (a) heavy use in the 3.7 - 3.98 GHz band (fundamental emissions) and (b) "accidental" broadcasts in 4.2 to 4.4 GHz range (spurious emissions). The study [1] on p. 12 in §6.1.1 basically assumes the 5G equipment will blast an interfering signal on altimeter band at full power. I think any frequency allocation request can be shut down this way. And I don't think 5G network will work properly if even a fraction of the assumptions in that report are true. If you look at https://en.wikipedia.org/wiki/5G_NR_frequency_bands https://en.wikipedia.org/wiki/5G_NR_frequency_bands, you will see that bandwidths are in 10s of MHz and the sub-carrier spacing [2] is as low as 15kHz. If 5G radios will be off by 15kHz (as opposed to 202000 kHz from 3.98 to 4.2 GHz as the aviation industry is implying), 5G network transmissions will not work properly. [1] https://www.rtca.org/wp-content/uploads/2020/10/SC-239-5G-Interference-Assessment-Report_274-20-PMC-2073_accepted_changes.pdf https://www.rtca.org/wp-content/uploads/2020/10/SC-239-5G-In... [2] https://info-nrlte.com/tag/subcarrier-spacings/ https://info-nrlte.com/tag/subcarrier-spacings/
  - NovemberWhiskey 6y ago
    >If 5G radios will be off by 15kHz (as opposed to 202000 kHz from 3.98 to 4.2 GHz as the aviation industry is implying), 5G network transmissions will not work properly. But that's not really how RFI tends to work - you can be perfectly aligned to your target frequency and generating the signals you expect there, but still be generating mixing or LO spurs, image response, third-order intermodulation products and so on which can be at totally different frequencies.
    - smarx007 6y ago
      Thank you for the correction. > mixing or LO spurs, image response, third-order intermodulation products Do I understand it correctly that it will only be a fraction of the output power and equation (6-1) in https://www.rtca.org/wp-content/uploads/2020/10/SC-239-5G-Interference-Assessment-Report_274-20-PMC-2073_accepted_changes.pdf https://www.rtca.org/wp-content/uploads/2020/10/SC-239-5G-In... must have a coefficient so that P_source would become P_source*k_spurious where k_spurious < 1 or even k_spurious << 1? Edit: by not working properly I did not mean that the signal on the target frequency will have severe interference but that just 15kHz apart there could be another target frequency that must be free of interference or otherwise multiple operators will not be able to use the 5G spectrum (12f, ie 180kHz seems to be an accepted guard band as per [1] and FCC seems to have mandated 23MHz guard band [2] which is 127x the norm) or even a single operator will not be able to use OFDM effectively. Am I still wrong in my understanding? Edit 2: with the interfence evaluation from [1], almost all evaluated guard bands result in -20dBm interference which would make proper k_spurious=0.01. My understanding remains that the aviation industry report used overzealous assumptions and overestimated the interference emission power by at least 100x. [1] E. Memisoglu, A. B. Kihero, E. Basar and H. Arslan, "Guard Band Reduction for 5G and Beyond Multiple Numerologies," in IEEE Communications Letters, vol. 24, no. 3, pp. 644-647, March 2020, doi: 10.1109/LCOMM.2019.2963311. [2] https://arstechnica.com/tech-policy/2020/05/millions-of-gps-devices-at-risk-from-fcc-approved-5g-network-military-says/ https://arstechnica.com/tech-policy/2020/05/millions-of-gps-...
      - NovemberWhiskey 6y ago
- mekkkkkk 6y ago
  Alright, thanks. But what is the use case for the 4 - 4.2 range? If it's outside of the spec, what would operators use it for?
  - drmpeg 6y ago
    Remember BUDs (Big Ugly Dishes) used for satellite reception? The band they operate on is 3.7 to 4.2 GHz. The FCC has repurposed 3.7 to 4.0 GHz (3.98 to 4 GHz is a guard band) for 5G. Satellite downlinks will continue to operate at 4.0 to 4.2 GHz.
karteum 6y ago
I am not familiar with this specific situation in the US but there are also similar topics in other parts of the world (with civil airborne radio-altimeters as a victim). It might seem odd to talk about interferences while 5G is in its dedicated band far away from radio-altimeters spectrum (> 4.2 GHz), however you have to understand there are 2 phenomenons to take into account : 1°/ transmitter unwanted emissions of 5G (i.e. how much out-of-band/spurious emissions "leak" outside of the 5G band i.e. above 3.8 GHz). This is most likely a negligible issue here even though there are still uncertainties whether the antenna pattern of AAS remains the same in-band and in spurious emissions 2°/ receiver blocking of the radio-altimeter (i.e. even if 5G emissions were perfect, the victim receiver can catch energy from other bands if it doesn't include a proper filter) : this is most likely the main issue here -> from my understanding, those airborne equipments have poor filters (the antenna limited passband probably helps a little bit which is why this blocking phenomenon did not happen previously with 4G in lower bands, but apart from that there is probably a low selectivity with regards to adjacent bands)
- mekkkkkk 6y ago
  Yeah, your second point didn't occur to me. It makes sense that filtering might be lacking if you have a relatively calm frequency neighborhood. Thank you!

sidewndr46 6y ago

There is a letter here that does a better job than the linked article: https://transportation.house.gov/news/press-releases/chair-defazio-calls-on-fcc-to-postpone-tomorrows-scheduled-auction-of-a-portion-of-37-42-ghz-radio-frequency-spectrum-citing-new-research-that-amplifies-the-safety-concerns-of-the-aviation-community https://transportation.house.gov/news/press-releases/chair-d... Basically, it sounds like they agreed to auction 3.7 - 3.98 GHz & 4.0 GHz - 4.2 GHz. This leaves a gap in the region of 3.98 - 4.0 GHz for radar altimeters. The letter states "recent research confirms that even this spectrum gap is insufficient to ensure radio altimeters are protected from harmful RF interference by 5G networks". I don't know enough about a radar altimeter's operation to speculate as to why it could be an issue.

smarx007 6y ago
No, it's not the 3.98-4.0 GHz gap. > Given the neighboring aviation band (from 4.2 to 4.4 GHz) allocated exclusively for vital aeronautical radionavigation equipment They are afraid of (a) heavy use in the 3.7 - 3.98 GHz band (fundamental emissions) and (b) "accidental" broadcasts in 4.2 to 4.4 GHz range (spurious emissions). The study [1] on p. 12 in §6.1.1 basically assumes the 5G equipment will blast an interfering signal on altimeter band at full power. I think any frequency allocation request can be shut down this way. And I don't think 5G network will work properly if even a fraction of the assumptions in that report are true. If you look at https://en.wikipedia.org/wiki/5G_NR_frequency_bands https://en.wikipedia.org/wiki/5G_NR_frequency_bands, you will see that bandwidths are in 10s of MHz and the sub-carrier spacing [2] is as low as 15kHz. If 5G radios will be off by 15kHz (as opposed to 202000 kHz from 3.98 to 4.2 GHz as the aviation industry is implying), 5G network transmissions will not work properly. [1] https://www.rtca.org/wp-content/uploads/2020/10/SC-239-5G-Interference-Assessment-Report_274-20-PMC-2073_accepted_changes.pdf https://www.rtca.org/wp-content/uploads/2020/10/SC-239-5G-In... [2] https://info-nrlte.com/tag/subcarrier-spacings/ https://info-nrlte.com/tag/subcarrier-spacings/
- NovemberWhiskey 6y ago
  >If 5G radios will be off by 15kHz (as opposed to 202000 kHz from 3.98 to 4.2 GHz as the aviation industry is implying), 5G network transmissions will not work properly. But that's not really how RFI tends to work - you can be perfectly aligned to your target frequency and generating the signals you expect there, but still be generating mixing or LO spurs, image response, third-order intermodulation products and so on which can be at totally different frequencies.
  - smarx007 6y ago
    Thank you for the correction. > mixing or LO spurs, image response, third-order intermodulation products Do I understand it correctly that it will only be a fraction of the output power and equation (6-1) in https://www.rtca.org/wp-content/uploads/2020/10/SC-239-5G-Interference-Assessment-Report_274-20-PMC-2073_accepted_changes.pdf https://www.rtca.org/wp-content/uploads/2020/10/SC-239-5G-In... must have a coefficient so that P_source would become P_source*k_spurious where k_spurious < 1 or even k_spurious << 1? Edit: by not working properly I did not mean that the signal on the target frequency will have severe interference but that just 15kHz apart there could be another target frequency that must be free of interference or otherwise multiple operators will not be able to use the 5G spectrum (12f, ie 180kHz seems to be an accepted guard band as per [1] and FCC seems to have mandated 23MHz guard band [2] which is 127x the norm) or even a single operator will not be able to use OFDM effectively. Am I still wrong in my understanding? Edit 2: with the interfence evaluation from [1], almost all evaluated guard bands result in -20dBm interference which would make proper k_spurious=0.01. My understanding remains that the aviation industry report used overzealous assumptions and overestimated the interference emission power by at least 100x. [1] E. Memisoglu, A. B. Kihero, E. Basar and H. Arslan, "Guard Band Reduction for 5G and Beyond Multiple Numerologies," in IEEE Communications Letters, vol. 24, no. 3, pp. 644-647, March 2020, doi: 10.1109/LCOMM.2019.2963311. [2] https://arstechnica.com/tech-policy/2020/05/millions-of-gps-devices-at-risk-from-fcc-approved-5g-network-military-says/ https://arstechnica.com/tech-policy/2020/05/millions-of-gps-...
    - NovemberWhiskey 6y ago
mekkkkkk 6y ago
Alright, thanks. But what is the use case for the 4 - 4.2 range? If it's outside of the spec, what would operators use it for?
- drmpeg 6y ago
  Remember BUDs (Big Ugly Dishes) used for satellite reception? The band they operate on is 3.7 to 4.2 GHz. The FCC has repurposed 3.7 to 4.0 GHz (3.98 to 4 GHz is a guard band) for 5G. Satellite downlinks will continue to operate at 4.0 to 4.2 GHz.

karteum 6y ago

I am not familiar with this specific situation in the US but there are also similar topics in other parts of the world (with civil airborne radio-altimeters as a victim). It might seem odd to talk about interferences while 5G is in its dedicated band far away from radio-altimeters spectrum (> 4.2 GHz), however you have to understand there are 2 phenomenons to take into account : 1°/ transmitter unwanted emissions of 5G (i.e. how much out-of-band/spurious emissions "leak" outside of the 5G band i.e. above 3.8 GHz). This is most likely a negligible issue here even though there are still uncertainties whether the antenna pattern of AAS remains the same in-band and in spurious emissions 2°/ receiver blocking of the radio-altimeter (i.e. even if 5G emissions were perfect, the victim receiver can catch energy from other bands if it doesn't include a proper filter) : this is most likely the main issue here -> from my understanding, those airborne equipments have poor filters (the antenna limited passband probably helps a little bit which is why this blocking phenomenon did not happen previously with 4G in lower bands, but apart from that there is probably a low selectivity with regards to adjacent bands)

mekkkkkk 6y ago
Yeah, your second point didn't occur to me. It makes sense that filtering might be lacking if you have a relatively calm frequency neighborhood. Thank you!