18 ms·

Starlink doesn’t change anything, it’s not for dense coverage. Not too much overlap with WISP. Revolutionary in other ways though. The main problem with this w

by seaorg 5y ago

Starlink doesn’t change anything, it’s not for dense coverage. Not too much overlap with WISP. Revolutionary in other ways though. The main problem with this way of transmitting is that you are on unregulated parts of the radio spectrum. So there’s nothing stopping someone from jamming your equipment, grinding everything to a halt, whether it’s on purpose or by accident. Maybe in practice it’s rare, but it’s too big a vulnerability for me to invest in such a business. I was wondering about laser modems. It could never be jammed without some kind of expensive and conspicuous physical intervention. I would feel much more comfortable investing in that even if lasers were more expensive or a pain in the ass. Not that it matters. 5G is going to put all these guys out of business.

grahamburger 5y ago
(Author here) > The main problem with this way of transmitting is that you are on unregulated parts of the radio spectrum. Traditionally this is true and is still mostly true today, but a ton of WISPs picked up PAL licenses in the CBRS band and are using that. Also, as you mentioned, in practice "jamming" equipment is uncommon (and very illegal!) Also I always recommend that folks use licensed or FCC registered links for their wireless backhaul, which limits the amount problems you'll get from interference. > I was wondering about laser modems. It could never be jammed without some kind of expensive and conspicuous physical intervention. I would feel much more comfortable investing in that even if lasers were more expensive or a pain in the ass. I've used some of these device (called "open air optical.") They're ok for very short range connections, but over anything more than a few hundred yards the performance of wireless is much better. > Not that it matters. 5G is going to put all these guys out of business. As a WISP you have the option of deploying 5G networks. It's expensive, but might be worth it. Many, many WISPs (myself included), operate fixed 4G LTE networks, and 5G is the next step. I haven't dabbled in 5G yet, but I have sales people calling me somewhat regularly ready to sell the gear. To the extent that 5G will live up to it's promises, WISPs are well positioned to take advantage of it.
- delabay 5y ago
  Just because you mentioned 5G CBRS: Freedomfi (https://freedomfi.com/ https://freedomfi.com/) is selling hardware later this year which will allow you to become your own cell tower for 5G CBRS bands, based on Facebook's magma project. You can backhaul a 5G modem/antenna, offload to a major MNO/MVNO, and get paid $.50 per gb consumption, paid in Helium HNT crypto (yes, crypto). Its still a big work in progress but could totally unlock private deployments of 5G networks and make them almost as easy as setting up wifi.
  - grahamburger 5y ago
    This is cool, thanks! I'll check it out and probably add a link on startyourownisp.com. A few of the other companies that I know are doing fixed 4G LTE right now are Blinq and BaiCells. BaiCells, at least, is also moving toward 5G.
    - delabay 5y ago
      Baicells will be a big partner with FreedomFi and probably a preferred antenna/modem vendor for the 5g small cells which CBRS is good for.
kalleboo 5y ago
> 5G is going to put all these guys out of business Does 5G make much of a difference here? From what I've read, in the the bands that give you reasonable range (mmWave doesn't work at these distances), 5G at the very most gives you like 20% more bandwidth than 4G LTE. That doesn't seem game-changing.
- kube-system 5y ago
  In most of the places where I know people who use a WISP, they barely have 4G coverage, and when they do, it's not a strong signal to say the least. Most of the places (at least in the US) where 5G is going to compete are probably places that already have cable.
  - kalleboo 5y ago
    Exactly. 5G doesn't focus on increasing coverage but mostly latency. The mmWave stuff with the 1 Gbps+ speeds has a coverage of around a block, so that will only be used in super-dense areas. From the competition section of the site, the sweet spot for WISPs seems to be just under suburban, where there's enough density but not quite cable coverage (or the local cable monopoly isn't giving their network enough TLC)
    - grahamburger 5y ago
      > Exactly. 5G doesn't focus on increasing coverage but mostly latency. This is my opinion as well - 5G will improve latency and also capacity (meaning carriers won't have to be as stingy with tethering plans) but not necessarily top speeds or coverage area.
    - bcrl 5y ago
      mmWave won't penetrate a door or a window or a tree. It really doesn't work in rural areas at all which is where WISPs make sense. And WISPs don't make sense if the local geography is rocks, hills and trees that make it impossible to get a clear line of sight.
- SilverRed 5y ago
nszceta 5y ago
I am already working full time remotely in the middle of the woods where there is "no cell service" getting 6 mbps up and down on an unlimited plan with a super high gain antenna aimed perfectly at the tower through several miles of trees. Once 5G modems come down in price my throughput will only increase. Coverage is already amazing on the under 700 MHz LTE bands even with carriers that traditionally have had a reputation for terrible coverage - even when the tower cells are aimed in the wrong direction! (They are aimed at the town and highway nearby, away from me.) High gain antennas are the most important component of my stack.
- adrianpike 5y ago
  Which antenna are you running? I've got a proxicast yagi that's good but I'm always interested in expanding my toolchest.
  - nszceta 5y ago
    11 dBi Yagi Antenna for TV White Space (470-862 MHz) is my choice for extreme penetration. It's several feet long. I don't trust the specs of the parabolic antennas on the market. They are supposed to be better but I have heard from multiple reviewers on multiple products that their specs are wrong and they simply don't deliver high gain at low frequencies when advertised to be tuned for 4G LTE. I think it is because to cover 4G LTE bands fully you need to support a ridiculously wide range of frequencies from 600 to 2200 MHz. My low frequency LTE yagi is awesome for difficult wooded areas. It can shoot through at least 3 miles of trees. I think your Proxicast peaks at 11 dB gain and you need to check which frequencies that's valid for. With such a large frequency range spec'd on that you can't be sure you have high gain on all frequencies. It can drop off drastically on important bands. That's why I have a dedicated yagi for low frequencies and a similar antenna to what you have for ~2 GHz bands. I found through experience the best proxy for correct antenna aim is an upload speed test. And the best way to determine aim is by using cellmapper.net and calculating a compass bearing from your location, then using a compass (app) to aim your antenna. Not the more high tech metrics reported by your modem like RSSI and link quality. They can be very misleading in general and especially when the link is idle.
Railander 5y ago
> So there’s nothing stopping someone from jamming your equipment Nearly a decade at a small/medium ISP in Brazil (where illegal competitiveness isn't uncommon) and I've never seen this happening to us (or if it did, we never found out). This is because jamming the spectrum requires a very powerful antenna that will _also_ interfere with every other antenna in the region, including probably their own. Instead of that, it's way easier to just grab some scissors and cut off your competitor's wires (which has in fact happened to us on numerous occasions). So while there are anticompetitive practices I don't think this is one you'd ever have to worry about.

by seaorg 5y ago

grahamburger 5y ago

(Author here) > The main problem with this way of transmitting is that you are on unregulated parts of the radio spectrum. Traditionally this is true and is still mostly true today, but a ton of WISPs picked up PAL licenses in the CBRS band and are using that. Also, as you mentioned, in practice "jamming" equipment is uncommon (and very illegal!) Also I always recommend that folks use licensed or FCC registered links for their wireless backhaul, which limits the amount problems you'll get from interference. > I was wondering about laser modems. It could never be jammed without some kind of expensive and conspicuous physical intervention. I would feel much more comfortable investing in that even if lasers were more expensive or a pain in the ass. I've used some of these device (called "open air optical.") They're ok for very short range connections, but over anything more than a few hundred yards the performance of wireless is much better. > Not that it matters. 5G is going to put all these guys out of business. As a WISP you have the option of deploying 5G networks. It's expensive, but might be worth it. Many, many WISPs (myself included), operate fixed 4G LTE networks, and 5G is the next step. I haven't dabbled in 5G yet, but I have sales people calling me somewhat regularly ready to sell the gear. To the extent that 5G will live up to it's promises, WISPs are well positioned to take advantage of it.

delabay 5y ago
Just because you mentioned 5G CBRS: Freedomfi (https://freedomfi.com/ https://freedomfi.com/) is selling hardware later this year which will allow you to become your own cell tower for 5G CBRS bands, based on Facebook's magma project. You can backhaul a 5G modem/antenna, offload to a major MNO/MVNO, and get paid $.50 per gb consumption, paid in Helium HNT crypto (yes, crypto). Its still a big work in progress but could totally unlock private deployments of 5G networks and make them almost as easy as setting up wifi.
- grahamburger 5y ago
  This is cool, thanks! I'll check it out and probably add a link on startyourownisp.com. A few of the other companies that I know are doing fixed 4G LTE right now are Blinq and BaiCells. BaiCells, at least, is also moving toward 5G.
  - delabay 5y ago
    Baicells will be a big partner with FreedomFi and probably a preferred antenna/modem vendor for the 5g small cells which CBRS is good for.

kalleboo 5y ago

> 5G is going to put all these guys out of business Does 5G make much of a difference here? From what I've read, in the the bands that give you reasonable range (mmWave doesn't work at these distances), 5G at the very most gives you like 20% more bandwidth than 4G LTE. That doesn't seem game-changing.

kube-system 5y ago
In most of the places where I know people who use a WISP, they barely have 4G coverage, and when they do, it's not a strong signal to say the least. Most of the places (at least in the US) where 5G is going to compete are probably places that already have cable.
- kalleboo 5y ago
  Exactly. 5G doesn't focus on increasing coverage but mostly latency. The mmWave stuff with the 1 Gbps+ speeds has a coverage of around a block, so that will only be used in super-dense areas. From the competition section of the site, the sweet spot for WISPs seems to be just under suburban, where there's enough density but not quite cable coverage (or the local cable monopoly isn't giving their network enough TLC)
  - grahamburger 5y ago
    > Exactly. 5G doesn't focus on increasing coverage but mostly latency. This is my opinion as well - 5G will improve latency and also capacity (meaning carriers won't have to be as stingy with tethering plans) but not necessarily top speeds or coverage area.
  - bcrl 5y ago
    mmWave won't penetrate a door or a window or a tree. It really doesn't work in rural areas at all which is where WISPs make sense. And WISPs don't make sense if the local geography is rocks, hills and trees that make it impossible to get a clear line of sight.
SilverRed 5y ago

nszceta 5y ago

I am already working full time remotely in the middle of the woods where there is "no cell service" getting 6 mbps up and down on an unlimited plan with a super high gain antenna aimed perfectly at the tower through several miles of trees. Once 5G modems come down in price my throughput will only increase. Coverage is already amazing on the under 700 MHz LTE bands even with carriers that traditionally have had a reputation for terrible coverage - even when the tower cells are aimed in the wrong direction! (They are aimed at the town and highway nearby, away from me.) High gain antennas are the most important component of my stack.

adrianpike 5y ago
Which antenna are you running? I've got a proxicast yagi that's good but I'm always interested in expanding my toolchest.
- nszceta 5y ago
  11 dBi Yagi Antenna for TV White Space (470-862 MHz) is my choice for extreme penetration. It's several feet long. I don't trust the specs of the parabolic antennas on the market. They are supposed to be better but I have heard from multiple reviewers on multiple products that their specs are wrong and they simply don't deliver high gain at low frequencies when advertised to be tuned for 4G LTE. I think it is because to cover 4G LTE bands fully you need to support a ridiculously wide range of frequencies from 600 to 2200 MHz. My low frequency LTE yagi is awesome for difficult wooded areas. It can shoot through at least 3 miles of trees. I think your Proxicast peaks at 11 dB gain and you need to check which frequencies that's valid for. With such a large frequency range spec'd on that you can't be sure you have high gain on all frequencies. It can drop off drastically on important bands. That's why I have a dedicated yagi for low frequencies and a similar antenna to what you have for ~2 GHz bands. I found through experience the best proxy for correct antenna aim is an upload speed test. And the best way to determine aim is by using cellmapper.net and calculating a compass bearing from your location, then using a compass (app) to aim your antenna. Not the more high tech metrics reported by your modem like RSSI and link quality. They can be very misleading in general and especially when the link is idle.

Railander 5y ago

> So there’s nothing stopping someone from jamming your equipment Nearly a decade at a small/medium ISP in Brazil (where illegal competitiveness isn't uncommon) and I've never seen this happening to us (or if it did, we never found out). This is because jamming the spectrum requires a very powerful antenna that will _also_ interfere with every other antenna in the region, including probably their own. Instead of that, it's way easier to just grab some scissors and cut off your competitor's wires (which has in fact happened to us on numerous occasions). So while there are anticompetitive practices I don't think this is one you'd ever have to worry about.