Echoes And An Algorithm Reveal Shape Of A Room
redOrbit Staff & Wire Reports – Your Universe Online
Researchers have developed a new algorithm that makes it possible to map the shape and dimensions of a room using just a few microphones and a snap of the fingers.
Developed at the Ã‰cole polytechnique fÃ©dÃ©rale de Lausanne (EPFL) in Switzerland, the algorithm models a room using echoes produced from sound, similar to the way in which bats and dolphins emit sounds in order to navigate. This ability, known as echolocation, is even possessed by some blind people.
The researchers used their algorithm to successfully build a full 3D image of a room using four microphones to record echoes bouncing off of walls.
But the EPFL algorithm goes further than the echolocation found in nature in that is able to distinguish between stronger and weaker echoes, and can determine whether the echoes have bounced one or more times around the room.
Walls made from different materials reflect sound differently, but it was not the amplitude of the sound wave that the algorithm was measuring, but rather the differing arrival times between the echoes. The same result could therefore be achieved using any sound, the researchers said.
“Our software can build a 3D map of a simple, convex room with a precision of a few millimeters,” said Ivan Dokmanic from the EPFL, lead author of the study.
“Each microphone picks up the direct sound from the source, as well as the echoes arriving from various walls.”
“The algorithm then compares the signal from each microphone. The infinitesimal lags that appear in the signals are used to calculate not only the distance between the microphones, but also the distance from each microphone to the walls and the sound source.”
The new algorithm offers many promising applications, Dokmanic said.
For instance, architects building a concert hall might know the specific acoustics they want the new building to have. Using this new algorithm they could now simply “plug” certain echoes into a computer and “get the kind of space needed to produce those echoes,” Dokmanic said in an interview with BBC News.
Virtual reality could also benefit from using accurate audio to correspond to visual images.
“If your eyes see and your ears hear something that doesn’t correspond, your brain is confused and disorientated. If you want to start a virtual design using sound, you could use this to create spaces that are realistic with respect to the sounds, because you would know the echoes.”
Dokmanic also sees potential applications in the field of audio forensics, where audio is used as evidence in crime investigations. For example, a sound recording obtained in an unknown place could provide clues as to the space it was recorded in. But further research is needed before this could happen, Dokmanic said.
The study was published last week in the journal Proceedings of the National Academy of Sciences.