how-to block ads
|
rradina
join:2000-08-08
Chesterfield, MO
| reply to rahvin112
Re: Wireless Has To Improve
Hasn't the FCC been selling dedicated spectrum? Have any WISPs bought dedicated spectrum? I know it's expensive but if wireless is the path to a true third choice, why hasn't a company with cash (Google, Microsoft, Oracle, IBM) purchased some spectrum?
For that matter, why must spectrum be purchased? Why can't the government classify spectrum for a certain kind of use and then license lots of small guys to use it? Why the big money grab?
Regarding interference, at least at my parents house it isn't coming from them. Their WISP is using 900mhz band. Their phones are 5Ghz, their access point is 2.4Ghz. One of their dish network receivers has a wireless remote that transmits in the UHF band but the packet loss is consistent and doesn't seem to get better or worse with any particular activity in the house. At 2AM, the packet loss is there but the speed is better.
The packet loss is present between their router's public IP (statically configured by the ISP) and the next hop. DNS lookup identifies the next hop as owned by their ISP. This leads me to believe the packet loss occurs between in the wireless link.
Their satellite LNBs convert the Ghz satellite signals to high Mhz signals that's more suitable for coax transmission. I suppose this could be leaking and causing interference.
Of course that doesn't explain the link in the lab where I work. I don't know what band it's using and being in an office park, the interference could be coming from anywhere.
I don't doubt that wireless could be better but in my opinion, today's links cannot offer serious competition to a hard link. Perhaps dedicated frequency would eliminate these problems.
I'd agree we need a third choice but it won't be able to compete unless it's rock solid, low latency and very low packet loss. | |
rahvin112
join:2002-05-24
Sandy, UT
| reply to rradina
On the other hand, large data links are created and maintained using Microwave broadcast and towers in line of sight. You see them often, they look like they have big white cloth covered wheel barrows hanging hanging on them. Those microwave data links can be in excess of several Gb/s. The technology exists, and works well, it's only a question of the reliability of the device and the quality of the spectrum.
Currently Wisps and other wireless Internet are relegated to an unlicensed band of spectrum that is used for everything from remotes and computer mice to cordless phones and Internet. The interference you are a seeing is likely due to something YOU own. Chances are it's a cordless phone that's moved off frequency a bit and is interfering with the signal from the Wisp. (if you want to test it unplug all your cordless and wireless devices, bring up a download where you can monitor packet loss and slowly plug each item back in, one at a time. I'll bet you find the offender pretty quickly)
The point of this whitespace group is to get access to the buffer space between channels and stick a digital signal in it that is data transmission. One goal of the project is likely to get this piece of spectrum restricted to ONLY Internet use so that they can limit the interference seen in the 900mhz and 2.4ghz unlicensed bands. You might love your wireless router that operates at 2.4ghz, but tack on a 20-40 mile range, limit it's use to commercial operators (but open to any commercial operator) to limit interference by rouge consumer devices and you have the wet dream of Internet providers. Building out a wireless country wide network is FAR cheaper than building out a last mile cable based system.
If we want that 3rd provider that's essential to consumer choice and competitive pricing we are going to have to make a sacrifice. Be that power line broadband, whitespace broadband or licensed spectrum broadband, the fact is we have to get a third national provider of Internet that doesn't rely on infrastructure from the cable/bell duopoly. Unfortunately that may mean we have to accept interferance with some signal. Hopefully in the digital world the idea of interferance isn't the same idea it was with analog transmissions. | |
rradina
join:2000-08-08
Chesterfield, MO
| Our technology lab has a wireless connection and it has packet loss. The downstream speed is also far less than the upstream. (Downstream is 200 - 300Kbps and upstream is usually over 1Mbps). We have complained to the ISP and they replaced the radio and realigned the antenna. It works OK but I don't think it will ever work well enough for VOIP and gaming. The latency varies a lot and the packet loss even makes browsing sluggish as the browser waits to retry the request.
My parents recently signed up with a WISP for high-speed internet at their farm. Again, there is some packet loss.
I don't know why but my direct exposure to wireless gives me the opinion that it should only be used as a last resort. I'm most troubled by the packet loss. I don't understand why both wireless links have packet loss. Both links experience about 1% packet loss. In my opinion, that's way too high and I don't understand why the data link layer is not more robust.
I also notice more packet loss on my wireless laptop link than my hard wired links.
Unless wireless improves, I don't think it will ever be suitable as a competitive replacement to hard line connection. Perhaps if it's really cheap, folks will accept the lower quality. | |
|
