The U.K. spectrum regulator, Ofcom, has issued a commissioned report analyzing and explaining 802.11b/g/n "Wi-Fi" issues in the popular 2.4 GHz band.  The report reveals that, for the most part, overcrowding is not the issue.  Instead, interference by non-Wi-Fi devices explain a great deal about poor performance in the band. 

The 2.4 GHz band is lightly regulated in Britian (nearly identically in the U.S.) and is a popular band for unlicenced devices such as telephones, baby monitors, alarm systems, audio/video systems, security cameras, and -- of course -- Wi-Fi networks.  This regulation by lack of regulation has, very predictably, made a mess out of 2.4 GHz systems that are incompatible with one another.

For example, many baby monitors and security cameras simply broadcast a continuous signal that can be received at a remote base station.  The communication is one-way in nature, and there is no built-in capability for these cheap transmitters to check if the frequency is in use before it starts to broadcast. 

Meanwhile, our wireless networking devices both send and receive as part of their normal operation. The Wi-Fi standards all have some provision for sharing the spectrum.  This means that a Wi-Fi device will delay transmitting if it detects a nearby device broadcasting.  This works wonderfully, and we often see marketing from Wi-Fi device makers extolling the fact that their device can hit speeds of 300+ Mbps (about three times faster than common Ethernet networks).  So why, then, when we take these devices home, we can't get more than 1-2 Mbps out of them?

Interference between these two types of devices is the cause.  No amount of delayed networking transmission can compete with another frequency user that never shuts up.  Have you ever had a friend who could dominate a conversation for hours, never letting you get even a word in, edgewise?  This is the electronic version of that problem! 

Many have blamed Wi-Fi's speed disappointments on the thought that there are just too many wireless computer networks sharing the 2.4 GHz band.  Ofcom's report tends to discredit that theory, but doesn't let the "Wi-Fi congestion" alibi the hook completely -- particularly in large cities.  That said, the action item to be taken is eliminating the contention between incompatible broadcast-only devices and transmit/receive ones.

Ofcom's report does point out that some manufacturers of non-Wi-Fi products are understanding the problem, and have recently been designing their products to be more "Wi-Fi friendly" and they are labeling them as such.

The report suggests that a "2.4 GHz Friendly" sticker on the boxes of certified consumer products will help the market forces work this problem out on their own.  I have my doubts about a proactive certification scheme being enough to guide non-technical users.  Furthermore, why would the purchaser of a baby monitoring system care or even know that their product interferes?  Perhaps, however, a "Wi-Fi Unfriendly" sticker on the boxes of all unlicensed devices that don't share the channel might be more effective.  Good luck enforcing that one.

The U.K. spectrum regulator, Ofcom, has issued a commissioned report analyzing and explaining 802.11b/g/n "Wi-Fi" issues in the popular 2.4 GHz band.  The report reveals that, for the most part, overcrowding is not the issue.  Instead, interference by non-Wi-Fi devices explain a great deal about poor performance in the band. 

The 2.4 GHz band is lightly regulated in Britian (nearly identically in the U.S.) and is a popular band for unlicenced devices such as telephones, baby monitors, alarm systems, audio/video systems, security cameras, and -- of course -- Wi-Fi networks.  This regulation by lack of regulation has, very predictably, made a mess out of 2.4 GHz systems that are incompatible with one another.

For example, many baby monitors and security cameras simply broadcast a continuous signal that can be received at a remote base station.  The communication is one-way in nature, and there is no built-in capability for these cheap transmitters to check if the frequency is in use before it starts to broadcast. 

Meanwhile, our wireless networking devices both send and receive as part of their normal operation. The Wi-Fi standards all have some provision for sharing the spectrum.  This means that a Wi-Fi device will delay transmitting if it detects a nearby device broadcasting.  This works wonderfully, and we often see marketing from Wi-Fi device makers extolling the fact that their device can hit speeds of 300+ Mbps (about three times faster than common Ethernet networks).  So why, then, when we take these devices home, we can't get more than 1-2 Mbps out of them?

Interference between these two types of devices is the cause.  No amount of delayed networking transmission can compete with another frequency user that never shuts up.  Have you ever had a friend who could dominate a conversation for hours, never letting you get even a word in, edgewise?  This is the electronic version of that problem! 

Many have blamed Wi-Fi's speed disappointments on the thought that there are just too many wireless computer networks sharing the 2.4 GHz band.  Ofcom's report tends to discredit that theory, but doesn't let the "Wi-Fi congestion" alibi the hook completely -- particularly in large cities.  That said, the action item to be taken is eliminating the contention between incompatible broadcast-only devices and transmit/receive ones.

Ofcom's report does point out that some manufacturers of non-Wi-Fi products are understanding the problem, and have recently been designing their products to be more "Wi-Fi friendly" and they are labeling them as such.

The report suggests that a "2.4 GHz Friendly" sticker on the boxes of certified consumer products will help the market forces work this problem out on their own.  I have my doubts about a proactive certification scheme being enough to guide non-technical users.  Furthermore, why would the purchaser of a baby monitoring system care or even know that their product interferes?  Perhaps, however, a "Wi-Fi Unfriendly" sticker on the boxes of all unlicensed devices that don't share the channel might be more effective.  Good luck enforcing that one.