2.4 ghz wireless.....?!

[Ait'al]

Is the only problem with 2.4ghz wireless the channel width? If so why has it not been fixed and why do they keep selling it without doing something about it since it makes it unusable and creates the very mess by allowing the channels to be to close to start. It's as big of a design flaw as you can have and these people should be held to the standard of engineers.... Especially since they insist on continually selling and making known faulty designs. And they push the products out vigorously might I add! Am I missing something?

[Mutant]

The reason 2.4 Ghz is used is down to the fact it is generally license-free and that is because it is poor at transmitting longer ranges (mainly due to the fact water absorbs it; Microwave ovens often use 2.4 Ghz)

All the "better" communication frequencies are generally paid for by corporations (think cell phones) or reserved for the military / Government.


802.11a/n uses 5 Ghz too. which adds nicely to the bandwidth.

[Ailos]

Originally Posted by Ait'al

Is the only problem with 2.4ghz wireless the channel width? If so why has it not been fixed and why do they keep selling it without doing something about it since it makes it unusable and creates the very mess by allowing the channels to be to close to start. It's as big of a design flaw as you can have and these people should be held to the standard of engineers.... Especially since they insist on continually selling and making known faulty designs. And they push the products out vigorously might I add! Am I missing something?

As a real-life engineer, I can't help but take that statement a bit personal. So here comes a rant:

Engineering decisions like "what spectrum bandwidth to use for the next-gen wireless" take years to make and are never light. A shortlist of reasons why it takes so long:

• The design has to be practical. We could use X-Rays as our transmission wavelengths - I mean, they would have great penetration through buildings and great range on top of the massive bandwidth. But then we'd all have cancer.
• The design has to be universal. It's not a standard if it's not... standard. At its heart, this means that a part made in Germany will be compatible with a part made in England and could be attached to a board made in China so that it can be sold in the US for $20. This often means going for the "lowest common denominator" - a perfect example for this is that here in NA we use 120 V while EU and China use 220, but not with the same socket shape. And yet my laptop's charger works everywhere. That wasn't an accident.
• Engineering decisions are always trade offs. You can't have your cake an eat it, too, isn't a cliché, it's an annoying reality for engineers. When we design things, we have to balance manufacturing costs with consumer desires, and most of the time this is done by only keeping the features that are necessary RIGHT NOW. Also, if we completely changed the frequencies on the products you buy every year, that means your laptop would only have wireless connectivity for a year, after which, you'd have to buy a new adapter. By extension, this would apply to the wired connection, too.

Things that engineers design and build are rarely designed to stand forever. Structures, vehicles, and equipment have to be maintained. Computer software has to be updated. And eventually, when the silicon chips powering your internet connection can't keep up with demand, they are replaced and recycled.

Wireless 802.11a/b doesn't provision for very high bandwidths or protection against interference because it came out more than a decade ago... when most people didn't even have cell phones. A lot of things have changed since then, and the standards have kept up. 802.11g introduced higher speeds, ability to much more effectively secure your signal, and some anti-interference protection. 802.11n improved on all that while also improving much of the nitty-gritty stuff that's necessary to make sure your data goes where it needs to.

Are there engineering challenges left to solve? Always. Which ones are being solved right now? All of them, in some shape or another. So next time you're frustrated with a piece of technology that you own, I have a sure-fire way that you can solve your problem: read the fucking manual. We don't write those things to be artistic. We write them to tell you how to protect your signal against interference.

[Rbstr]

Originally Posted by Ait'al

Am I missing something?

In short: Yes

[Ait'al]

All I'm asking is what was the problem at the time. I'm behind on things and I wanted to know (or I simply can't remember at the moment which is common for me)...

[Electrofreak]

Originally Posted by Ait'al

All I'm asking is what was the problem at the time. I'm behind on things and I wanted to know (or I simply can't remember at the moment which is common for me)...

The problem was that a limited range of bandwidth was allocated for the purpose. Just about every scrap of available radio bandwidth has been claimed. 2.4 GHz is what was made available.

When we're talking about any FDM radio standard, the problem is that to carry more data, you need more radio bandwidth, or wireless channel width. But if you use more bandwidth, you have less channels that you can transmit different streams of data over.

Then take into consideration that the error-correction and transmission protocols involved in modern wireless standards use roughly half of the total throughput for overhead. So when you see your laptop connected at 54 Mbps on a wireless G connection, understand that a little less than half of that, around 22 Mbps, can actually be used for raw data. For Wireless B, you can get around 6 Mbps of raw throughput out of an 11 Mbps connection.

2.4 GHz wireless is split into 11 channels, but each channel bleeds into the adjacent 2 channels, and so channels 1, 6, and 11 are the most that can be used simultaneously without introducing radio interference. If the width of the band was reduced to be only as wide as a channel, the throughput of the wireless stream would be very low. Using overlapping channels was the compromise that needed to be made.

Hopefully this explains the issue further. It's not that there was any design flaw with 2.4 GHz wireless, it's just that there was a very limited space within which the engineers could work, and they picked a compromise that allowed up to 3 2.4 GHz wireless networks to overlap while providing the greatest throughput possible.

Understand that it was back in 1999 that wireless A and wireless B specifications were launched. Wireless A provided greater throughput than B (54 Mbps vs 11 Mbps for B) on the 5 GHz range, but also had range problems as higher frequencies tend to have trouble penetrating solid objects. At that time, 54Mbps was a huge amount of bandwidth and largely unecessary, while range was more important. Keep in mind it was being used primarily by businesses in the early years. Wireless G, launched in 2003 (yes, I'm having to check for the dates, I don't remember those ) bumped up the throughput capable on the 2.4 GHz range while maintaining a reasonable usable range. This made it become immensely popular as wireless went mainstream in the home networking market.

Of course, as soon as everyone in apartment complexes started using 2.4 GHz wireless, the problem of radio interference became very evident, and increasing bandwidth requirements as the internet turned into a media source meant that the 2.4 GHz range couldn't just be split up into narrower channels to allow more overlapping wireless networks. Wireless N was drafted (and years passed with draft Wireless N products being sold to consumers) and finally ratified. Wireless N uses the 5 GHz range but manages to maintain good useable range, fairly high throughput (by using multiple streams referred to as MIMO and expanded channel sizes) along with allowing enough overlapping channels (though the number depends upon whether 20 or 40 MHz channels are being used) to make it viable in dense residential environments.

So, if you managed to make it through all that, it's about as much explanation as I can offer. I may not always be using the exact proper terms as I'm an enterprise WAN data technician and wireless is just something I know a little about so that I can touch upon it if I need to when I'm providing a consultation.

TL;DR - There is no flaw with 2.4 GHz wireless, only a necessary compromise made a decade ago before residential use was typical.

[Rbstr]

2.4 ghz was pretty much only open because of microwave ovens. They needed a way to make an exception for the 2.45ghz emissions from just any old person, instead of the licensing you need in other parts of the spectrum. Plus they made that part of the spectrum rather crap before better error tolerance was around.

There are a couple others. http://en.wikipedia.org/wiki/ISM_band