A 6db change in signal level is one bit position in a binary value -- 2 times the voltage (note that this is 4x the power if driven into the same load resistance, which is why +3db is twice the perceived power). If you have a 10 bit converter you can resolve 10 double-ings of voltage or 60 db. Given that there is (a lot) of noise, your low end response is probably reduced by a (at least) couple bits, so the final dynamic range you can measure is closer to 8 bits or 48 db, with fairly bad resolution at the low end.
I think I've done that right anyway... MS Rogério De Pieri wrote: > Dear MS > > Thanks for you grat help. > > I just can't understood about the range at the end of your message. > > Clould you give-me more details? > > Thanks for all > > PS: Sorry for my bad english > > RP > > On Sun, Apr 20, 2008 at 6:06 PM, Michael Schippling <[EMAIL PROTECTED] > <mailto:[EMAIL PROTECTED]>> wrote: > > SPL is measured as an average of the instantaneous values you get from, > for instance, a microphone. What you need to do is sample the mic at > some > reasonable rate (making some assumptions about the frequency range > you are > interested in). In T1 there is a HighFrequencySampling demo app that > sorta > does this. > > Then the (sl)easiest thing to do would be to look for max values > over a moving > window of some large number of samples. The real way to do it would > be to > estimate a zero crossing point -- probably at about 1/2 the ADC > range (what > you get from a mote ADC is fundamentally an AC signal with a fixed > DC offset), > do a full-wave bridge around that zero crossing, and then an RMS > average of > the result (again with a moving window of some size which will > capture the > lowest frequency of interest). > > In either case you should probably do the SPL conversion on the mote and > only send the results because it markedly reduces the bandwidth of the > data you need to send. > > In order to get an engineering measurement in dBm you need to > calibrate your averaged data against a known source and perhaps make > some adjustments for > the non-flat frequency response of the microphone, then convert the > measurements to a log scale. Note that you will have some noise > floor that > makes the lower couple of bits insignificant, and that the max you > can get is > 10 to 12 bits (for Atmega and MPS converters). This means 8-10 > significant > bits where each bit is 6dB (a doubling of sound pressure), so your > measurement > range is at best somewhere between 48 and 60 dBm. Hopefully that is > sufficient > for your system... > > MS > > Rogério De Pieri wrote: > > Hi dear all > > I am working on an acoustic project within tinyos-2.x > environment. The goal of the project is to take a view of the > distribution of SPL (Sound Pressure Level) in an ambience. > > > I took just a few of steps on that, and a this time I'm looking > for documentation about how to take datas from acoustic/sound > sensor from sensorboards. > > Another problema I know I will have is to turn de sound or > acoustic data in SPL or dB. > > > Does anybody caan help me? > > Thanks for all > > > -- > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Tinyos-help mailing list > [email protected] > <mailto:[email protected]> > > https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help > > > -- > Platform: WinXP/Cygwin > TinyOS version: 1.x, Boomerang > Programmer: MIB510 > Device(s): Mica2, MicaZ, Tmote > Sensor board: homebrew > > > > > -- > > > ------------------------------------------------------------------------ > > _______________________________________________ > Tinyos-help mailing list > [email protected] > https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help _______________________________________________ Tinyos-help mailing list [email protected] https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
