Slashdot Mirror
How Reliable is 900Mhz Wireless Internet?
amrust asks: "I live in an area that currently is not being served by broadband in any way. Local ISP's are discussing bringing Wireless Internet into our area, and for some of the more difficult to reach places, they have mentioned 900Mhz being used. I was wondering how reliable/secure the 900Mhz spectrum is for Wireless Internet, and if anybody has some comments on experiences with 900Mhz wireless internet that they can pass on?"
I don't have any personal experience with 900mhz, but i do know that it is the same frequency that cordless phones use. Some phones today use 2.4ghz and they cause interference with .11b and g (my friend's neighbor has a phone that is so powerful it knocks my friend offline everytime the neighbor turns it on.) There are many more 900mhz phones that 2.4ghz phones, so I'd assume the problem is worse.
However, you are in a rural area, so maybe the congestion on that band will be somewhat less (We live in a major metro area).
-Ryan
AUWYHSTOT (Acronyms are Useless When You Have to Spell Them Out Too)
Just picking up from the first page yields some tidbits. Range, reliability; Interference.
Nice troll.
But I'll bite.
First of all, its the 2.4Ghz range.
Second, security through obscruity is complete rubbish, 'no one other than you and the intended recipient will be able to read the message' is a load of, quite frankly, bollocks.
Mod this down.. no need to give trolls karma.
Most WISP's use 900mhz where there is a substantial amount of tree cover. 900mhz can penetrate the tree canopy much easier than 2.4ghz because of the wavelength.
2.4ghz has a problem with dense water, and in tree leaves their is significant amount to cause multipath reflection. Multipath reflection is caused when the wavelength of any given radio signal is short enough to be relected by the water molecules.
At 900mhz the wavelength is much greater than 2.4ghz so it doesn't have as much difficulty.
Platinum Networks Hosting www.platinum-networks.com
Bandwidth IS directly proportional to frequency. Baud Rate (in the scientific sense) is exactly twice the frequency. @ 900 MHz, the bandwidth is 900 MHz, the maximum switching rate (BAUD) is 1800 MHz. tha Shannon Channel capacity, assuming 60 dB signal-to-noise, is 900e6log_2(1 + 1,000,000) = 17.9 Gb/s. obviously, this is extremely high. it is technically possible to hit this raw rate of data transfer. and it is impossible to exceed this value without data compression
When they say "transmit at 9OO MHz" of course you can not send any real data if you transmit ONLY at 900MHz. Because that would be a single unvarying siqn wave at 900 MHz.
Instead, you are allowed to vary around 900 MHz. I think the standard 900 MHz is really 900 MHz to 914 MHz.
So 14 MHz is the WIDTH OF THE BAND, otherwise known as BANDWIDTH.
At any particular center frequency you can set a wide or narrow range around it.
I'm not even a ham or anything, come on guys I learned this by osmois somehow. You should know this.
I listened to Bill Joy talk at a conference in Aspen where he went into fair detail about the 900MHz system they had installed there long before 802.11b came about (if anyone remembers seeing a story about an Aspen cabby with a totally connected neon taxi cab, he was using this system for his net access).
... for long distance penetration through anything other than air, 2.4GHz plain sucks (I'm using a WISP now and occasionally when strong winds hit I have to give up working for awhile ... shame). Mr. Joy did mention that they weren't able to get more than a few hundred Kb/s out of their system, but that was acceptable for a few dozen people sharing a single T1 and the system sounds like it was rock solid stable.
... though since their 5GHz access point doesn't have trees between it and myself and it is far less war driven I will probably switch to that anyway.
Anyway
If you only have one choice (or if you consider a normal dialup line a choice maybe 2) then that is definitely not a bad one. I think I would actually prefer my 2.4GHz ISP to offer 900MHz as an option. Instead they're switching to 5GHz for high-end use
It is more productive to voice thoughtful opinions (reply) than to judge (moderate) others.
If the bandwidth is 14MHz then the center frequency above and below which the carrier will vary would probably be 907 MHz. However 14Mhz is wide enough for about 2 and 1/2 regular television channels, so for a telephone call, which only needs 0.0027 Mhz of bandwidth, I expect that only a small part of that 14 MHz is used with the rest available for other cordless phones in the same area.
I see even classic Slashdot is now pretty much unusable on dial up anymore.
I would recommend digging a little deeper to see what standards are being considered, then visiting this very informative page at IBM.com pertaining to current and emerging wireless standards. Once you know the exact standard the ISP's are considering, you should be able to have your questions concerning security and reliability answered on the IBM link provided.
Beware blue cats moving at
[disclaimer: the following is based on research, not personal experience]
So, I've been looking at using 900MHz for some low bandwidth stuff. Primarily because 2.4GHz can be a PITA to deploy - near line of sight (over distance), trees, rain etc.
Anyway, it appears that the products available round 900MHz have taken a completely different, and much more "agricultural" approach to the problem. Basically they're 900MHz FM radios with a modem bolted on. No clever multipath resolution, no time division, no orthogonal frequency division (outside that which the modem does). Consequently the amount of bandwith available is about seven eigths of fuck all - a claimed 115Kbit (being the limit of the RS232 connectors they use) - with something rather worse being the practicality of situation.
Security is kinda interesting though. Quite a lot frequency hop across a range of channels within the spectrum. It appears that you need to get both radios, from the same manufacturer, and put them on the same hopping scheme before they'll talk.
Dave
I write a blog now, you should be afraid.
Ricochet ran a well regarded service over 900mhz. They claimed 1 mile range from base stations, w/o line of site. So it is certainly possible to build a reliable service.
Urban environments have a fair amount of 900mhz interference, in a rural setting you are probably much better off. Heck, w/ all the 2.4ghz phones being sold I wonder if 900mhz is clearing up some.
There's a whole lot of cheap equipment around. I got 3 900mhz DEC Roamabout pcmcia nics for $12 (inc shipping). They are 2mbit and the linux driver worked perfectly.
Many houses have old directional antennas for UHF TV that aren't used anymore because of cable / satellite. I've found a fair amount of these range up to 900mhz. So maybe you've already got an installed antenna and cable run in your house. If not you probably know someone who has one they aren't using or can pick one up for cheap.
The important factor here is how efficiently the signal can pass through a medium, and how directional the signal is. We don't live in a vacuum (if you didn't know this, you might need to get out once in a while :P), and therefore any signal sent has to travel through any objects in its path, including the air. Not sure exactly off hand which law of physics covers this, but lower frequencies lose less energy when passing through a medium, and are also less directional. Think about sound here, a low frequency (bass), say 40Hz will have relatively no problem passing through an object, like a car driving in front of your house at 2:00A.M., but a high frequency (treble), for example 10kHz is highly directional and has great difficulty traveling through objects.
I believe this is due to the fact that to pass a signal through a medium, the particles in the medium must be excited to the same frequency as the signal, and it takes significantly less energy to excite a material at a lower frequency than it does at a higher one. Higher frequencies give higher fidelity, but are less resistant to signal loss.
--That's the point of being root, you can do anything you want, even if it's stupid.
Bandwidth IS directly proportional to frequency
...and WOW! I can crank up my 1200baud packet system to 3600baud by just switching from 2M to 70cm. Thank you, oh great radio genius!</SARCASM>
What?
<SARCASM>Wow! I can do morse code at only about 8wpm (which equates to about 40 baud) on 80M. Does that mean that I can do 856wpm on 70cm?
Sorry, but that's such an offensively ignorant statement from a member of a very technically-savvy group that I have to go off on it a bit. Frequency occupied by a signal indicates how much data could be fit in it. Depending on your modulation scheme, you'll reach some significant fraction of that bandwidth. In a perfect propogation environment, with perfect components on both ends, that fraction can asymptotically approach 1. For relatively-reliable real-world links, you'll be running well below 50% of the theoretical throughput, to make it easier for the receiver to discriminate the actual states of the transmitter. Ethernet is an example of a protocol expecting near-perfect transmission. Ever had a 100MBps link fail back to 10 because of crappy terminations, or had any link degraded? That's because individual packets are simply not getting resolved by the receivers, due to substandard signal-to-noise ratio - noise leaking into the cable and/or signal leaking out.
That's why 802.11 links drop in speed, so the reciever can sort of go "Is that bit off or on? It sounded on, but it's really not much different from the static. Well, it keeps looking like it's on. Yes, that's definitely '1'", while at higher speeds, there's less chance to discrimnate the signal from the noise.
Have you ever worked in a really loud factory? You can't understand anything coming over the PA system, but the signal tones sent over the same system at the same volume in the same environment are perfectly understandable.
Now, if systems used everything from DC to their operating frequency, a 2.4Ghz system would have about 2.5x the potential bandwidth as a 900Mhz system. In fact, however, the systems use a slice of spectrum in the range of their stated frequency. I don't know the amount of spectrum used by these systems, but I know some things about the frequencies and the environment they work in.
First: In general, path loss increases with frequency. My beam has nearly 3 times the gain on 70cm as it does on 2M, but using the same power, I can work stations much farther away with 2M than 70cm.
Second:there are specific absorptions in different frequency ranges. Since the early 1970s, at least, many households have made use of the strong water absorption band around 2.4Ghz to heat food in microwave ovens. That means two things A)2.4Ghz performance is significantly affected by humidity. B)There are ubiquitous high-power ISM devices in the band. Why do you think most 802.11b gear has a setting to work around microwave ovens?
In the end, you're probably going to be using a system with throughput capability many times larger than the amount you are to receive, which will allow for big sags in performance to not hurt you unless they coincide with peaks in demand. I would recommend going for the one with the smaller collision domain. I bought into Sprint Broadband Direct, which uses large swaths of licensed spectrum (no significant interference). This system operates from single central systems covering vast geographic areas - mine is a mountain 16 miles away, and i'm near the fringe. That's 804 square miles. Half of it is blocked by the mountains behind it, but the sector it does cover is heavy population (north of Denver). That means that even though the RF transit time to the tower is insignificant, My unit has to wait in line for a long time for its turn to transmit, giving me ping times ranging from 100ms (quite acceptable) to 5 seconds (no interractive work over VPN, I can tell you). Mesa networks has come into the area now, using 802.11B and gain antennae. Base stations are on poles (cell sites, phon
On the contrary, as has been pointed out, frequency is not bandwidth.
Actually, it is possible to approach the Shannon channel capacity the way you calculated it, but impossible in practice to achieve it. Technical possibility is not the same as theoretical maximum throughput.