[Hackrf-dev] Software half-duplex operation via gr-osmosdr

[Kevin Reid]

On Jun 5, 2015, at 3:01, Claus Neie <cneie at aol.com> wrote:

> interesting to read, what is the all transition time between rx and tx?  

I can't think of a way to measure the actual time from (good samples received) to (good RF transmitted) without a lot more setup and some kind of timestamped signal to receive, but the time spent by the control thread (i.e. the python code you actually see) to perform the switch, which includes the time to _construct_ the source and sink blocks and therefore reinitialize the HackRF hardware, is on average:

tx->rx: mean 59 ms
rx->tx: mean 73 ms

I suspect these numbers will be highly platform-dependent. I'm using OS X 10.9.5, and a somewhat old version of gr-osmosdr (before the multiple device support patch, which I imagine might affect the thing I'm exercising here).

-- 
Kevin Reid                                  <http://switchb.org/kpreid/>