Slashdot Mirror
Has 2.4 GHz Reached Maximum Capacity?
An anonymous reader writes "There's been a lot of talk lately about the concept of Personal Area Networks. At CES Intel and Connectify both released software that turns Windows laptops into Access Points for file transfers, wirelessly syncing pictures from cameras, and Internet sharing. This is good, maybe great, if you're a road warrior, but what about the rest of us holed up in apartment buildings and small neighborhoods? We already have to deal with the wireless chatter of the 50 or so other Linksys routers in the vicinity. What will happen when every laptop also acts as a software router? To add fuel to the fire, Intel and Netgear also announced the Push2TV device that allows you to stream your display, including Netflix videos straight to your television. Isn't this going to kill lower powered 2.4 GHz devices, like Bluetooth mice and headsets? When does the 2.4 GHz band collapse completely? Why can't we push all this short range, high bandwidth stuff onto 5 GHz?"
If you're in a place with too much WiFi noise, try using the Japanese-only channel 14 - it doesn't overlap with any other channels, and you're pretty much guaranteed to be the only person using it. DD-WRT and other alternative firmwares will broadcast on 14, and Windows just requires a simple registry hack to receive on 14. Macs, I believe, can connect without hassle. It's technically illegal, but the likelihood of being caught is pretty much nil. I've still never found an explanation for what this frequency is used for in the US, if anything.
Yup, none of the access points are actually smart enough to switch channels and optimize frequency/power usage. So you end-up with the problem I encountered this December, when a neighbor across the way got a wireless router, and suddenly my internet stopped working. You couldn't even *see* my access point anymore, it was just overpowered.
My access point configuration was set to "Auto," but this just meant it kept trying to use channel 1 like an idiot. So I forced it to use channel 6, and the problem was solved.
But this isn't how things should be. The devices and the protocol should be smart enough to optimize spectrum, both by analyzing the noise at various frequencies and choosing the band with the least automatically, and by playing nice with other devices and dialing-down the power to that needed by the connected device furthest from the access point.
Too bad the above is just a pipe dream. I can't imagine how bad it is living in dense residential/apartments, where these users still don't know how to configure things, but there are 2 dozen within range instead of 5.
Man is the animal that laughs.
And occasionally whores for Karma.
Boom! Headshot.
You'll be comforted to know that ZigBee devices, and other devices that conform to the IEEE 802.15.4 standard, do this today, and have done so since 2003.
The second part of your wish, that devices dial-down their power to the minimum needed, is certainly possible, but requires that the receiving device send an RSSI (Received Signal Strength Indication) or, better, an RQI (Received Quality Indication) value back to the transmitting device (perhaps as part of the ACK), and that the transmitting device be capable of lowering its transmit power in the first place (it's an extra expense). I don't know of any protocol that does this (other than CDMA protocols, for a different reason), although I expect newer protocols will -- but to save power (extend battery life), not to reduce spectral pollution.
The problem is dumb receivers, not lack of spectrum.
Channel capacity is determined by the MIMO form of Shannon's Theorem. Add more antennas and smarter processing in the receivers, and the capacity in a channel with lots of multipath (eg. in an apartment) increases approximately linearly.
The problem is that most hardware is a decade (or more) behind state of the art, and that people aren't prepared to pay for a more complex, and costly receiver.
The theoretical max for wireless G is 27.4 Mbps. Reading a DVD at 1X is 10.5 Mbps. As I understand it, when you are transmitting from one wireless device to another, you are talking to the AP which then retransmits to the other device, effectively doubling your data.
So, starting with 10.5Mbps then doubling the data and adding overhead, you are dangerously close to the theoretical 27.4Mbps max.
Copyright 2010. All rights reserved. This comment may not be copied in any way including, but not limited to caching.