Could transmitter and receiver alternately work in time on the same USRP1 with LFTX/RX?

Dear all,

I’m developing a radar transmitter and receiver on USRP1 and LFTX/RX.
The transmitter and receiver are working fine separately on two USRP1
with TX and RX daughter board plugged in. I’m trying to make the
transmitter and receiver alternately work in time on the same USRP1. For
instance, transmitting signal for 0.52ms then disabling TX and enabling
RX simultaneously for signal reception for 1ms. Then this process will
repeat over and over again.

Is it possible to get the TX and RX wok on the same USRP1 motherboard
with both RX and TX? The reason why I’m asking this question is that it
seems impossible to call usrp_source or usrp_sink block every
millisecond to enable TX or RX, since usrp_source loads FPGA image only
with TX path enabled and usrp_sink does only with RX path enabled, if my
reading of the source code usrp_standard.cc services me right. In
addition, loading the whole FPGA image takes longer time than 1ms, which
will cause USRP overrun problem if shifting TX to RX or RX to TX every
millisecond as my experiments showed me. Therefore, I’m wondering if
there is an approach to allow the TX and RX to alternately work in time
on the same USRP1 board rather than calling usrp_source or usrp_sink to
make this TX/RX altering. Simply, is there any mean to shift TX and RX
back and forth every two milliseconds and repeat over and over again by
loading FPGA image only once for the whole process.

I will really appreciate if any suggestion on how to solve this problem.

Wish you a great day!

Thanks,
Yan

On Tue, 2011-02-08 at 13:29 -0500, Yan N. wrote:

Dear all,

I’m developing a radar transmitter and receiver on USRP1 and LFTX/RX. The
transmitter and receiver are working fine separately on two USRP1 with TX and RX
daughter board plugged in. I’m trying to make the transmitter and receiver
alternately work in time on the same USRP1. For instance, transmitting signal for
0.52ms then disabling TX and enabling RX simultaneously for signal reception for
1ms. Then this process will repeat over and over again.

Is it possible to get the TX and RX wok on the same USRP1 motherboard with both
RX and TX? The reason why I’m asking this question is that it seems impossible to
call usrp_source or usrp_sink block every millisecond to enable TX or RX, since
usrp_source loads FPGA image only with TX path enabled and usrp_sink does only
with RX path enabled, if my reading of the source code usrp_standard.cc services
me right. In addition, loading the whole FPGA image takes longer time than 1ms,
which will cause USRP overrun problem if shifting TX to RX or RX to TX every
millisecond as my experiments showed me. Therefore, I’m wondering if there is an
approach to allow the TX and RX to alternately work in time on the same USRP1
board rather than calling usrp_source or usrp_sink to make this TX/RX altering.
Simply, is there any mean to shift TX and RX back and forth every two milliseconds
and repeat over and over again by loading FPGA image only once for the whole
process.

You might consider just running TX and RX simultaneously instead of only
enabling RX after the TX pulse. This has the ancillary benefit of giving
you a precise time reference for the outgoing pulse, removing any
uncertainty in the delay time between turning off TX and turning on RX.

–n

This forum is not affiliated to the Ruby language, Ruby on Rails framework, nor any Ruby applications discussed here.

| Privacy Policy | Terms of Service | Remote Ruby Jobs