4 ms·

The future are super dense networks Lifi, 60GHz, then 5Ghz, then 2.4-1.8GHz because that will give the most bandwidth for the buck. You always want higher frequ

by acd 11y ago

The future are super dense networks Lifi, 60GHz, then 5Ghz, then 2.4-1.8GHz because that will give the most bandwidth for the buck. You always want higher frequency first because it can carry more and then fall back to lower frequencies if you are out of coverage of the high freq. Outside developed areas there will be internet by solar drones and lower orbit satellite internet.

dijit 11y ago
LiFi is such a sketchy idea to me, due to how IrDA has been in the past. Maybe somehow I'm the last generation who remembers really shitty remote controls for televisions. :\
ucaetano 11y ago
The bandwidth doesn't depend on the frequency you're occupying, but on the amount of spectrum available: you "usually" get in the order of 1 bps for every Hz of spectrum available for mobile: a 20Mz chunk of spectrum will give you ~20Mbps, no matter if it is 700MHz or 5 GHz. (Please notice that YMMV, LTE gets several times that, illustrative, bla bla bla, etc.). Higher frequencies have awful penetration and range, that's why today you define who wins in the mobile game by the amount of 700MHz and 800MHz spectrum they own. In other words, lower frequency spectrum is (within certain limits) always better.
- Armisael16 11y ago
  That may be true at present, but higher frequency systems do ideally allow for higher data rates. EDIT: No, I'm wrong.
  - gsteinb88 11y ago
    That's simply false. What is true is that per-band allocations tend to be larger at higher passband frequencies, but the width of the baseband in absolute terms is what determines information capacity. See: The Shannon limit
  - ucaetano 11y ago
    Not really, the bandwidth doesn't depend on the frequency, but on the how much spectrum you have. The frequency impacts in other ways (noise, penetration, etc.) but not in the amount of raw data you can put through a channel.
- rayiner 11y ago
  Cellular network capacity is a function of channel width, modulation, and spatial reuse. Because there are many more low-frequency channels of a given width than high frequency ones, it's much easier to justify very wide channels at higher frequencies. The better propagation of lower frequencies is also a double edged sword. It's much easier to build a very dense cellular network if cell sites far apart can't hear each other (causing co-channel interference). Also, because higher frequencies propagate more directionally, it's also easier to take advantage of spatial reuse techniques.
  - LeifCarrotson 11y ago
    > there are many more low-frequency channels of a given width than high frequency ones, it's much easier to justify very wide channels at higher frequencies. Right conclusion, wrong (mis-stated?) premise. There are more high-frequency channels of a given width than low-frequency ones. 700-800 MHz is a whopping 12.5% change, and it gets you 100 MHz to work with, but 5.725-5.875 is only a 2.5% change and gets you 150 MHz to work with.
    - rayiner 11y ago
      Yeah, meant that the other way around.
  - ucaetano 11y ago
    > because higher frequencies propagate more directionally Again, right conclusion but wrong premise. The don't propagate more directionally, but as they have limited penetration and high attenuation, they are more useful for point-to-point with line-of-sight than for broadcast.
- auganov 11y ago
  Anyone got a resource that explains the relationship between spectrum/frequency/modulation-tech/etc and bandwidth in an understandable way for someone not-so-physics/signal-processing-literate? My [uninformed] intuition tells me that you can have theoretically infinite bandwidth over any spectrum or modulation scheme, but the cost/size/energy-requirement quickly gets prohibitively high because of the "sensitivity" such a device would have to have?
  - samcheng 11y ago
    That's why they call it "bandwidth" - the width of the spectrum band is what determines the information throughput. Wikipedia might work as a primer. See Nyquist & Shannon: https://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem#Sampling_of_non-baseband_signals https://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampli... For example, in order to sample the FM radio signals in the frequency range of 100-102 MHz, it is not necessary to sample at 204 MHz (twice the upper frequency), but rather it is sufficient to sample at 4 MHz (twice the width of the frequency interval).
    - typon 11y ago
      When I first learned this it blew my mind at it's implications and then obviousness. Shannon was a once-in-a-century genius...in a century of many once-in-a-century geniuses.
    - ucaetano 11y ago
      Also take a look at spectrum efficiency, which is the "practical implementation" of all this. It really helps to understand how spectrum is a limited public resource. For example, a carrier with 20MHz of spectrum each way and a tower covering 1km2 in 3 sectors could offer at most 60Mbps down over 3G. If you have 6000 customers in that same area (remember that some cities have densities in the 10+k/km2), that means ~10Kbps per customer on average, which is enough for 1.5 GB/month/customer.

by acd 11y ago

dijit 11y ago

LiFi is such a sketchy idea to me, due to how IrDA has been in the past. Maybe somehow I'm the last generation who remembers really shitty remote controls for televisions. :\

ucaetano 11y ago

The bandwidth doesn't depend on the frequency you're occupying, but on the amount of spectrum available: you "usually" get in the order of 1 bps for every Hz of spectrum available for mobile: a 20Mz chunk of spectrum will give you ~20Mbps, no matter if it is 700MHz or 5 GHz. (Please notice that YMMV, LTE gets several times that, illustrative, bla bla bla, etc.). Higher frequencies have awful penetration and range, that's why today you define who wins in the mobile game by the amount of 700MHz and 800MHz spectrum they own. In other words, lower frequency spectrum is (within certain limits) always better.

Armisael16 11y ago
That may be true at present, but higher frequency systems do ideally allow for higher data rates. EDIT: No, I'm wrong.
- gsteinb88 11y ago
  That's simply false. What is true is that per-band allocations tend to be larger at higher passband frequencies, but the width of the baseband in absolute terms is what determines information capacity. See: The Shannon limit
- ucaetano 11y ago
  Not really, the bandwidth doesn't depend on the frequency, but on the how much spectrum you have. The frequency impacts in other ways (noise, penetration, etc.) but not in the amount of raw data you can put through a channel.
rayiner 11y ago
Cellular network capacity is a function of channel width, modulation, and spatial reuse. Because there are many more low-frequency channels of a given width than high frequency ones, it's much easier to justify very wide channels at higher frequencies. The better propagation of lower frequencies is also a double edged sword. It's much easier to build a very dense cellular network if cell sites far apart can't hear each other (causing co-channel interference). Also, because higher frequencies propagate more directionally, it's also easier to take advantage of spatial reuse techniques.
- LeifCarrotson 11y ago
  > there are many more low-frequency channels of a given width than high frequency ones, it's much easier to justify very wide channels at higher frequencies. Right conclusion, wrong (mis-stated?) premise. There are more high-frequency channels of a given width than low-frequency ones. 700-800 MHz is a whopping 12.5% change, and it gets you 100 MHz to work with, but 5.725-5.875 is only a 2.5% change and gets you 150 MHz to work with.
  - rayiner 11y ago
    Yeah, meant that the other way around.
- ucaetano 11y ago
  > because higher frequencies propagate more directionally Again, right conclusion but wrong premise. The don't propagate more directionally, but as they have limited penetration and high attenuation, they are more useful for point-to-point with line-of-sight than for broadcast.
auganov 11y ago
Anyone got a resource that explains the relationship between spectrum/frequency/modulation-tech/etc and bandwidth in an understandable way for someone not-so-physics/signal-processing-literate? My [uninformed] intuition tells me that you can have theoretically infinite bandwidth over any spectrum or modulation scheme, but the cost/size/energy-requirement quickly gets prohibitively high because of the "sensitivity" such a device would have to have?
- samcheng 11y ago
  That's why they call it "bandwidth" - the width of the spectrum band is what determines the information throughput. Wikipedia might work as a primer. See Nyquist & Shannon: https://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem#Sampling_of_non-baseband_signals https://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampli... For example, in order to sample the FM radio signals in the frequency range of 100-102 MHz, it is not necessary to sample at 204 MHz (twice the upper frequency), but rather it is sufficient to sample at 4 MHz (twice the width of the frequency interval).
  - typon 11y ago
    When I first learned this it blew my mind at it's implications and then obviousness. Shannon was a once-in-a-century genius...in a century of many once-in-a-century geniuses.
  - ucaetano 11y ago
    Also take a look at spectrum efficiency, which is the "practical implementation" of all this. It really helps to understand how spectrum is a limited public resource. For example, a carrier with 20MHz of spectrum each way and a tower covering 1km2 in 3 sectors could offer at most 60Mbps down over 3G. If you have 6000 customers in that same area (remember that some cities have densities in the 10+k/km2), that means ~10Kbps per customer on average, which is enough for 1.5 GB/month/customer.