Re: [Discuss-gnuradio] Number of TX/RX channels on a USRP

[Arya Santini]

Hi Alex, List,

On Sun, Jan 23, 2011 at 3:03 AM, Alexandru Csete <address@hidden> wrote:
> On Sun, Jan 23, 2011 at 2:56 AM, Arya Santini <address@hidden> wrote:
>> Hi GNURadio,
>>
>> I'm new to the USRP and trying to understand the number of RF
>> input/output channels available (USRP1). From the GNURadio wiki
>> (http://gnuradio.org/redmine/wiki/gnuradio/UsrpFAQIntroMobo), I gather
>> we have 4 possible TX and RX channels each, since we have 4 ADC's and
>> 4 DAC's on the motherboard:
>
> Hi Arya,
>
> Thanks for reading the doc - it makes explaining easier :-)
>
>>
>> "..in principle, we have 4 input and 4 output channels if we use real
>> sampling. However, we can have more flexibility (and bandwidth) if we
>> use complex (IQ) sampling. Then we have to pair them up, so we get 2
>> complex inputs and 2 complex outputs."
>>
>> This means we could have only 2 complex TX/RX channels.
>>
>> Also from the USRP FAQ page
>> (http://gnuradio.org/redmine/wiki/gnuradio/UsrpFAQIntroFPGA), in
>> explaining the DDC implementation,  it is said:
>>
>> "Note that when there are multiple channels (up to 4), the channels
>> are interleaved. For example, with 4 channels, the sequence sent over
>> the USB would be I0 Q0 I1 Q1 I2 Q2 I3 Q3 I0 Q0 I1 Q1, …etc. In
>> multiple RX channels (1,2, or 4) , all input channels must be the same
>> data rate (i.e. same decimation ratio)."
>>
>> Do these multiple RX channels (1,2 or 4) refer to those coming out of
>> the ADC's? Then contrary to the aforementioned, how come now there are
>> 4 complex channels, I0,Q0 - I3,Q3 coming in. when it should be
>> possible to have only 2 complex RX channels? Looks like I'm
>> understanding the architecture incorrectly!
>
> This multiplexing refers to the data stream on the USB interface, i.e.
> after FPGA processing where the real signal is converted to
> quadrature/complex signal. So the 4 RX channels that use real samples
> are converted to complex signal in the FPGA.
>

Okay, there are a total of 4 pairs of IQ signals, each of the 4 real
outputs from the ADCs being converted to an IQ pair in the FPGA.

>>
>> Some help here would be greatly appreciated.
>>
>> Also, on the daughterboard, each RX port has access to 2 ADC's. So in
>> the context of having 2 RX channels possible per daughterboard (in the
>> case of Real sampling, and total 4 for the system) are the inputs
>> RX_A_A, and RX_A_B separate/independent? I mean, does the RF
>> daughterboard circuitry receive two frequencies simultaneously on a
>> single receive antenna?
>
> I am not entirely sure about this so someone may correct me if I'm wrong...
> The LFRX and BasicRX are not quadrature receivers so the use real
> sampling and conversion to quadrature signal is done in the FPGA. You
> can however also chose to input quadrature signal in which case you
> use RX_A_A and RX_A_B as I and Q input. This works in the two receiver
> configuration.

Suppose we're dealing with real signals alone, then we have 4 input
channels viz. RX_A_A, RX_A_B, RX_B_A, RX_B_B. My question is, are each
of these channels independent of each other in the sense that their
frequencies are different i.e are they 4 different RF signals? In
other words, on each daughterboard, can the circuitry be configured to
receive on 2 separate frequencies from the same antenna, and feed them
as inputs to each of the 2 ADCs its RX port is connected to?

> If you use the 4 receiver configuration then 2
> additional receivers have to be mounted on the TX_A and TX_B slots.
> Therefore, if we chose to use 4 receivers we can not have transmitters
> at the same time.
>
> Alex
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>


Thanks,

Arya