For me WiFi is my occupation and my hobby. I get a thrill that most people would find unbearable out of debugging wireless network problems and analyzing layer 2 management frames. But I sometimes forget that most folks just want their WiFi to work. Take note that my B-in-law is no slouch at solving networking issues. As a developer of WebBased applications via Linux and as the maintainer of a (as he puts it), "3-tier, cross-platform application development framework, written in Python atop the wxPython GUI toolkit" I am pretty sure he can handle a couple of wireless packet issues. Here is the email:
I feel for him. This is a common problem. 2WIRE and other manufacturers of consumer grade access points don't really get WiFi. If they did they wouldn't have the devices auto-choose interfering channels like channel 5 or 9. I do not mind that they sample the air and select an unused channel but please select a non-overlapping channel instead. I also have had my share of issues in my 120 year old home in San Francisco. Normal 802.11b/g doesn't like old fashioned plaster and lathe walls. It also doesn't like the 40-60 APs on my block all vying for the same airspace and sure as heck hates my 15 year old 1200 watt microwave oven.Bruce, Of late, it seems all my neighbors are getting 2Wire/AT&T DSL wifi routers.I think their signals are interfering with mine. What channel should I put myself on given this list I just got from kismet, although I've seen the number of devices at almost twice this list:Channel AvgSignal SSID 1 22 2WIRE365 1 18 2WIRE341 5 19 2WIRE248 6 18 pete 6 18 2WIRE675 9 28 2WIRE219 I've also seen devices on channels 3, 10, and 11. I'm currently on channel 8 (previously on 3) and 8 seems better, but still not great. I'm going to be wiring fast ethernet wired connections to [name omitted]'s office and my office, but it would be nice to be able to wander around the house without the connection dropping sporadically (I watch my wifi signal go from 5 bars to 0 bars, stick there for 5 seconds, and go back to 5 bars, and 10-20 seconds later, the same thing happens). Then several hours may pass with no incident. Maybe someone's microwave or cordless phone is interfering? Any ideas other than putting ugly repeaters all over the house? Thanks!
Then I read about one of my favorite technology writers, Andy Intakho suggesting we go back to wires. I saw the first post at Glenn Fleishman's WiFiNetNews site and followed the link to Andy's. It is an amusing read but I am a little perplexed. Andy does not share the steps he tried to solve the issue. He doesn't share the equipment he was using that gave him such frustrations. We have no sense of why he made what is, IMHO, a rash decision to abandon the future for the past.
Glenn Fleishman has it right when he suggests,
"Andy, maybe you need a working 802.11n router and some modern hardware? Or maybe your apartment building is simply being bombarded by untoward RF interference.As a former Apple loyalist, I would have suggested the Apple Airport Extreme with both 2.4gHz and 5gHz 802.11n (and a ton of other high-end features). It is easy to setup, easy to use and goes really fast!
Don’t get me wrong: I like my copper Ethernet wiring, too, especially when I’m moving big files around my network. But with Draft N, I’m more likely to have a gating factor at my Internet gateway or a particular computer’s ability to shoot files over a given protocol than I am by the network’s raw speed."
However, I also see both Andy's point and that of my Brother-in-law. It shouldn't be this hard. The average home user should not have to worry about "RF Interference" this and "Channel Congestion" that. It should just work.
So whose responsibility is this? The manufacturers? The IEEE? The WiFi Alliance guys?
Maybe I am limiting my future as a WiFi engineer but I would like to see it fixed. If a groundswell of disaffected WiFi users starts now, with WiMAX on the horizon, 3g-4g cellular becoming really popular and iPhone users now having the same bandwidth as the DSL in my last apartment, we might see the sky falling after all.