From those spectrum results I'd say you were doing pretty good. Even if the vertical scale is linear you have a factor of 10 more "signal" than noise. As to what's in the actual samples, acoustic noise, echos, and system noise could all be culprits. The usual electret mic is good up to around 15Khz and you probably don't have an input low-pass filter, so you may be seeing aliasing as well.
You'd probably have to try the whole thing in an anechoic chamber to get a clearer idea. You could also try setting up regular mics next to each node and comparing the signals, and swapping the nodes around to see how much noise moves with the mote itself. MS Omar Bouzid wrote: > Hi Michael, > > I mean by "coherent signals" that received signals are uncorrelated > with each other (shapes and amplitudes of received signals are > different although they were captured by sensor nodes which were > positioned exactly at the same distance from a sound source and under > the same conditions). > > My sampling rate is 4kHz and I was trying to capture, for example, a > 1kHz and 1.2kHz sine waves (see attached file). As you can see from > the attached file, in both cases the captured signals have different > shapes and this is my question why? > > Thanks, Omar > >> -----Original Message----- From: Michael Schippling >> [mailto:sc...@santafe.edu] Sent: 08 August 2010 17:22 To: Omar >> Bouzid Cc: 'Tinyos-help@millennium.berkeley.edu' Subject: Re: >> [Tinyos-help] Why my received signals incoherent? >> >> Please, can you define "coherent signals"? >> >> What is your sample rate? If you are trying to sample sound you >> will get very curious results if you are below the Nyquist >> frequency of your signal. >> >> MS >> >> Omar Bouzid wrote: >>> Hi all, >>> >>> I'm using three sensor nodes (MICAz + MTS310), positioned at the >>> same >> distance from a sound source, to sample a click signal at 4kHz.The >> sensor nodes start sampling simultaneously once they receive a >> start command from a base station and a buffer is used to hold 1200 >> samples in each sensor node. When the buffer is full, each sensor >> node will start sending their sampled data to the base station >> which will forward them to a PC. They send their own sampled data >> serially to the base station. When I plot the received signals I >> expected to get coherent signals but this was not the case although >> the experiments were conducted in the same environment and under >> the same conditions. So, what could be the reasons for such >> incoherent signals? >>> I appreciate any help in advance, >>> >>> Omar >>> >>> _______________________________________________ Tinyos-help >>> mailing list Tinyos-help@millennium.berkeley.edu >>> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos- >>> >> help _______________________________________________ Tinyos-help mailing list Tinyos-help@millennium.berkeley.edu https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
