Peter Cochrane's Blog: Wireless jelly

Written in a London coffee shop with the five-bar signal problem and dispatched via a free wi-fi node later the same day.

You know how it is: you're in a coffee shop working online and using the free wi-fi service. There are five bars of signal. You're in good shape, being creative, making progress, closing in on that deadline.

Exclusive column: The Naked CIO

See what this CIO really thinks…

The Naked CIO: The V-word

The Naked CIO: Price of panic

The Naked CIO: The skills drain needs fixing

The Naked CIO: Madness in the method

The Naked CIO: Process not bureaucracy

The Naked CIO: Is open source dead?

So, all is well - until that is, everything suddenly starts to run badly. Downloads and uploads seem slow, mouse clicks invoke a random delay, and then suddenly everything stops.

First time around you suspect your laptop is having a bad OS day. But then you notice you have no bars at all. Where did that five-bar signal go? Good question.

Unfortunately the answer is complex. For at least some of the time I think we can safely assume that the problem is plain old interference.

I mean ignition systems, pumps and motors, microwave ovens and other devices operating in, or close to, the wi-fi band, including some cordless phones.

But then there is the density of laptops in the room and wi-fi nodes in the building. On top of that, of course, other factors come in to play, such as bad power engineering and LAN cabling with uncertain connector reliability.

However, the least understood phenomenon is the way the volume of electromagnetic radiation energy changes with physical objects moving around in the space. And this includes people, laptops, metal trays and other big metal items, such as road-going vehicles.

They all change the space in an undefined and ill-understood way - in pretty much the same way we might step on a balloon full of water and change the distribution.

This all came home to me recently as I worked on a project where it was necessary to measure the far-field strength at distances of 5m to 50m.

It was like wandering around in a vast, variable density and irregular jelly. Every time you moved, so did the field. For those interested in physics, I think this is a phenomenon we could lay at the feet of Heisenberg.

Having observed this problem for some years I decided on a belt-and-braces approach in my own home.

And guess what - even in an exceptionally electrically quiet country area with almost zero road traffic, very few people and laptops, the five-bar problem still occurs from time to time.

This all gets very irritating, all the more so because I engineered filtered, no-break power into everything and exercised extreme care with wiring and how equipment was positioned.

It appears that mains- or power-feed borne interference is still getting into the environment and causing an occasional problem.

The good news is that it can only all get better as we migrate further up the spectrum from 2.4GHz and 5GHz toward 60GHz and higher.

But I suspect that future cognitive radios will almost single-handedly overcome this problem. For now I am pleased that I took the decision to invest in several kilometres of hard wiring and more than one hotspot.