[Tinyos-2.0wg] Concerns about ADC in 0.1mV

[Prabal Dutta]

Wouldn't the Vref be associated with the sample and not the ADC
interface since it can change between calls?  Joe's approach might
work with some caveats: the getRef call could be valid until the
associated readDone event returns.  That is, getRef would have to be
called from inside the readDone event handler.
Otherwise, Vref for the particular call could be lost forever.

- Prabal

On 7/26/06, Joe Polastre <joe at polastre.com> wrote:
> Why not a left-aligned 16-bit value with an optional "getRef" function
> which is a 16-bit signed value that will return the ref in millivolts
> or -1 if the ref isn't known?
>
> -Joe
>
> On 7/26/06, Prabal Dutta <prabal at cs.berkeley.edu> wrote:
> > I think the real problem is making sure you know the value of Vref
> > used for the A/D conversion.  If Vref is static, then that's great and
> > an ADC count will suffice.  However, if Vref can change, then the
> > conversion factor needs to be sent along with the ADC counts.  This
> > approach allows engineering units to be recovered from the data and
> > also minimizes conversion computation on the mote.  However, it might
> > (a) not be supported in all platforms (if Vref can't be compared to a
> > bandgap reference) and (b) could require two ADC conversions -- one
> > for the signal and one for Vref.
> >
> > - Prabal
> >
> > On 7/26/06, Martin Turon <mturon at xbow.com> wrote:
> > >
> > > I totally agree with Gil here.  Typically we have seen deployments where
> > > the motes send raw samples in fixed point, and the conversion happens on
> > > the gateway or client application.  There are a few benefits to this:
> > >
> > >        - less power/speed hit on the mote hardware
> > >        - reliable data -- no doubt about whether mote conversions are
> > > correct
> > >        - standard data -- fixed point is more standard than floating
> > > point
> > >
> > > Left shifting the samples is fine, as is providing knowledge of the
> > > reference voltage, but actual conversions should probably be left out of
> > > the ADC interfaces.
> > >
> > > Martin
> > > __________________________________________________
> > > Martin Turon  |  Crossbow Technology, Inc.
> > >
> > > -----Original Message-----
> > > From: tinyos-2.0wg-bounces at Mail.Millennium.Berkeley.EDU
> > > [mailto:tinyos-2.0wg-bounces at Mail.Millennium.Berkeley.EDU] On Behalf Of
> > > Gilman Tolle
> > > Sent: Wednesday, July 26, 2006 12:44 PM
> > > To: tinyos-2. 0
> > > Subject: [Tinyos-2.0wg] Concerns about ADC in 0.1mV
> > >
> > > Upon first read, the proposal to return ADC HIL values in 0.1mV
> > > sounded great. After thinking it through a bit more, I have a few
> > > concerns.
> > >
> > > We're going to lose information during the conversion and rounding to
> > > the nearest 0.1mV, simply because we don't have enough decimal places
> > > to maintain the actual voltage. The uV proposal would certainly help
> > > with this problem, at the cost of engaging 32-bit math and doubling
> > > the size of the samples. Even with uV, we'll still hide information
> > > about the precision of the ADC.
> > >
> > > This proposal assumes people are going to want to translate from
> > > voltage into engineering units on the mote. So far we haven't seen
> > > much of a demand for this, likely resulting from the heavy math needed
> > > to do these conversions. The closest need we've seen so far is
> > > converting the output of the battery voltage sensor from ADC units to
> > > voltage, to be used in a check before performing flash memory
> > > operations. For that purpose, TinyOS 1.x only performed the conversion
> > > when reading from a battery-voltage HIL called VoltageC, not for every
> > > ADC operation.
> > >
> > > For high rate sampling, performing the conversion for every sample
> > > when using ReadStream will impose some non-trivial CPU performance
> > > penalty. This feels analogous to the earlier discussion about
> > > left-shifting the read results. For something like audio data, the
> > > processing algorithms may not even need true voltage, if fractions of
> > > full-scale is enough and full-scale is known.
> > >
> > > At least no information is lost with left-shifted ADC readings, and
> > > the differences in ADC precision are still abstracted away. Although,
> > > we still pay the cost of the left-shift, and actual precision must be
> > > exposed through another interface to support the processing steps that
> > > care.
> > >
> > > We should check with some sensing domain experts to see whether the
> > > rounding error, information hiding, and conversion costs would be
> > > acceptable.
> > >
> > > Gil
> > > _______________________________________________
> > > Tinyos-2.0wg mailing list
> > > Tinyos-2.0wg at Mail.Millennium.Berkeley.EDU
> > > https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-2.0
> > > wg
> > >
> > > _______________________________________________
> > > Tinyos-2.0wg mailing list
> > > Tinyos-2.0wg at Mail.Millennium.Berkeley.EDU
> > > https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-2.0wg
> > >
> > _______________________________________________
> > Tinyos-2.0wg mailing list
> > Tinyos-2.0wg at Mail.Millennium.Berkeley.EDU
> > https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-2.0wg
> >
>