Mathematicians have proposed improvements to cloaking technology to hide objects which emit their own electromagnetic radiation.

In October, a transatlantic team of scientists demonstrated the world's first working invisibility cloak, which uses metamaterials to bend electromagnetism around an object as if it weren't there.

However, the sums say the effect would be ruined by electromagnetic radiation from inside the cloak - so an object like a mobile phone, computer or aircraft could not be hidden. Analysis also showed that with current technology, to a person inside the cloak its inner surface would appear mirrored and they would not be able to see out.

A group led by Professor Allan Greenleaf at the University of Rochester, New York, said on Tuesday they had solved these problems with the invisibility cloak by revisiting the 19th century work of James Clerk Maxwell. Maxwell's four equations describe how electromagnetism interacts with matter.

The team propose two possible solutions to the twin issues; to line the inner surface with conductive materials tuned to the geometry of the cloak, or to "match" the outer metamaterial surface with an inner one. They say Maxwell's equations back up the fix.

With current metamaterials, devices are only able to provide a cloaking effect for a very limited wavelength range, as Greenleaf points out. He said: "For example, an object could be rendered invisible at just a specific wavelength of red; it would be visible in nearly every other color."

The new findings do seem to bring a functioning invisibility cloak nearer however. David R. Smith, a physicist at Duke University, part of the team which built the first cloak, said: "Allan has been looking at the problem much more generally, and deriving the conditions for when true invisibility is or is not possible. We are very interested in what he and his colleagues come up with!"

We're sure the news will also be read with great interest in the corridors of the DARPA, which funds Smith's work. ®