How to Deal with the Noise Floor

Last post I talked about how the noise floor is a very succinct way of measuring all the interference in a given frequency range. In other words, if the wireless environment is tough and challenging, the noise floor will tell you just how challenging it is.

So how do wireless protocols deal with that incessant interference?

Get a Room

The first way is to pick a frequency range, or spectrum, that has as low a noise floor as possible. This requires choosing spectrum in which no other technologies are allowed to operate.

This is like having a private room to talk with your friend where you and your designated friends are the only ones allowed in. As with all things, there’s a catch. The private room costs billions of dollars and it’s called licensed spectrum.

Get Loud

The second way wireless protocols overcome the noise floor is to increase the signal strength. This is the method I mentioned in my first post.

If you’re at a party and the music is loud and the crowd is rowdy, to talk to your friend, you simply yell. As long as you yell louder than the ruckus, you’ll be heard. This approach also has a catch: it’s exhausting.

For partiers, it requires a lot of energy to sustain a conversation. For wireless solutions, it’s taxing as well. Transmitting louder requires more power, and for battery-powered devices that can be a real problem. And sometimes the music is simply too loud and your friend never hears you. The same happens in wireless solutions and messages simply get drowned out by the interference and don’t get through.

Go Stealth

The third approach to dealing with the noise floor is perhaps the cleverest and most fascinating of them all. Some wireless protocols simply blend in with the noise. They embrace, no, they literally become the noise.

The way it works is the signal is scrambled in a way that makes it look like noise, even though it isn’t. You take the message and a unique string of characters (or gold code) and run them through a special mathematical function. The function then scrambles the message in a way that, to anyone receiving it on their end, it looks just like noise. It’s randomized, scrambled, such that it can be unscrambled on the other end only if you have that code.

So if you are some random person looking at the frequency range on which that message has been sent — and who doesn’t love doing that in their spare time right? — all you’ll see is static. The noise floor will have raised just a hair, almost imperceptibly.

The catch for this approach is that it takes time to decode the signal. So if you have to scramble the message into a really long string to get the message through, then it will take longer to get through.

Paradoxically, because the signals look like noise to each other as well, you can actually send many, many messages at the same time on the same frequency without them interfering with one another. I know, it blew my mind the first time as well. So while each individual message may take longer, you can send an entire batch of message simultaneously and ultimately get more messages through over the same time period than by the other means.

The Noise Floor - Next Time

As with most things, there are tradeoffs and I started to discuss some of these already. In my next post, I’ll go into more detail on these tradeoffs and what they mean for those wanting to use them.