Project Size: 4/5
My Responsibly: Thesis Idea, Research, Hypothesis, Testing, Presenting, Software Development.
Description: A sound such as a gunshot can be found using the speed of sound and trigonometry but requires distance. I aimed to prove that the intensity (volume) of a could be used to find a sound in a shorter distance such as within a building.
My new hypothesis became to find a sound source within a room with no prior information on that sound source. The challenge here was most techniques used to find a sound source used accurate measurement of when the sound was heard at a few different locations but this technique wasn't useful within a room. I decided to use sound intensity instead.
The dissipation of sound as it crosses a distance is well understood but I wanted to use the intensity to get distance, that was simple enough when I knew the starting intensity and the received intensity, In ideal conditions I could find the distance. But I wanted to find the sound source using three sound levels and with no information about the sound source, that Included it's intensity. I didn't know how to find it but I was certain I had all the information required to find the source.
The dissipation of sound as it crosses a distance is well understood but I wanted to use the intensity to get distance, that was simple enough when I knew the starting intensity and the received intensity, In ideal conditions I could find the distance. But I wanted to find the sound source using three sound levels and with no information about the sound source, that Included it's intensity. I didn't know how to find it but I was certain I had all the information required to find the source.
So I set about modifying a room by sealing the windows and vents and padding the walls and reducing as much sound reflection as possible and mapping a grid onto the floor. I then began testing to find the sound source with known intensity and distances, etc.
I then had the Idea about how to find the source with no prior information, I figured I couldn't directly look for the source but I could hypothesis an arbitrary decible level and probably disprove it by showing said sound couldn't have reached my 3 sound level meters at the levels received. I could produce a small set of positions and decibel levels going through the set of possible decibel levels and either disproving or plotting the position for each decibel level. This left a usual one position and decible level that could be possible, sometimes there was a second position well outside the testing area.
Throughout this process I was writing a proof of concept program in C# allowing someone to enter sound values and positions and calculate the sounds origin position and intensity. Although it wasn't used in the end the program also could show frontier transformations from a sound source like a microphone or bluetooth headset and keep track of the sound level in arbitrary units that I hoped to find a decible conversion for.
Forward
This thesis will cover the techniques used to determine the distance a sound is from aknown point, it will then go on to explain how this knowledge can be used to then plotthe point at which the sound source is located.A number of techniques exist to find a sounds origin but none of these work on a smallscale or with inexpensive technology. Many of these techniques use precise recording ofthe received time of a sound at multiple points and knowledge of the speed of sound todetermine a sounds point of origin. The problem with these techniques is that they donot work well for distances under hundreds of meters and the precise timing involvedrequires precise specialised equipment.My hypothesis is that dissipation of sound; the manner in which a sound becomes lessloud over distance can be used to find a sounds source. The dissipation of sound iscalculable and with three or more sound levels at a number of positions a sound sourceof known intensity can be found, at least in ideal conditions. I propose that the soundsources initial intensity need not be known in order to find its position and that by aprocess of elimination the original intensity and location can be found.Finally these assumptions will be tested in a variety of environments with a variety ofsound frequencies to conclude a margin of error for each.
