discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] Quesion on AUX ADC of AD9862


From: Eric Blossom
Subject: Re: [Discuss-gnuradio] Quesion on AUX ADC of AD9862
Date: Tue, 16 Jun 2009 10:52:41 -0700
User-agent: Mutt/1.5.18 (2008-05-17)

On Tue, Jun 16, 2009 at 09:15:34PM +0800, Su Jinzhao wrote:
> Hi,
> 
> I want to read RSSI value using analog AUX ADC, I think we can read RSSI
> through AUX_ADC_A1 pin of AD9862, I looked through the code
> of usrp_read_aux_adc in file
> /root/gnuradio/gnuradio-3.1.3/usrp/host/lib/legacy/usrp_primes.cc,

If you are using usrp_standard_rx, or gr.usrp_source_*, you can do
this much more simply by using:

  adc_val = u->read_aux_adc(which_side, which_adc)

http://gnuradio.org/doc/doxygen/classusrp__basic__rx.html
http://gnuradio.org/doc/doxygen/classusrp__standard__rx.html

You didn't mention which daughterboard you are using.  Not all
daughterboards have an analog RSSI output.

> my question is:
> 1.The AD 9862 specification says: "The AUX SPI can be enabled and configured
> by setting register AUX ADC CTRL. Setting register use pins high enables the
> AUX SPI port.", but in the above code AUX_ADC_CTRL_AUX_SPI was not set high(
> In my opinion, aux_adc_control should be added with | = AUX_ADC_CTRL_AUX_SPI
> because usrp_9862_write and usrp_9862_read all need SPI operation, so we
> should enable AUX API), Is this a mistype or another method ?

The code is correct.  There are a couple of ways to use the AUX ADCs.
We use them by polling, not over the AUX SPI bus.

> 2.stefan.bruens in the list had said that the low 16 bits is the RSSI value,
> I want to know why did the above code combine the high and low 16-bit and
> format the data into 12-bit ?

This is a digital RSSI, determined post high-speed ADC.  You can look
at the computation here:

  http://gnuradio.org/trac/browser/gnuradio/trunk/usrp/fpga/sdr_lib/rssi.v
  

> 3. Even with the RSSI value (int type), what should I do to convert that
> value into dBm unit ?

You would need to calibrate the entire signal path.  Left as an exercise...

Eric




reply via email to

[Prev in Thread] Current Thread [Next in Thread]