>From a CPU and thus, SDR _software_ throughput, point of view, the E310 certainly isn't more powerful than a modern laptop + B2xx; latency might be a bit shorter, though, but in the end, you're making the most important point here:
If you need to eradicate the nondeterministic behaviour of general purpose operating systems running on general purpose computing hardware (such as Linux or Windows on a PC or the Zynq's ARM), you can only do two things: 1. Have enough leeway for timing, so that you're *always* faster than you need to be, or 2. doing the timing-sensitive stuff directly on dedicated realtime environments, which in this case means the FPGA. Cheers, Marcus On 03/29/2017 10:27 PM, Saulo Queiroz wrote: > An interesting question is whether more powerful SDR (e.g. E310) > should manage to do that (DATA+SIFS+ACK). > RFNoC may facilitate the FPGA design... > > 2017-03-29 17:19 GMT-03:00 Bastian Bloessl <[email protected] > <mailto:[email protected]>>: > > Hi, > > On 03/29/2017 01:11 PM, sumitstop wrote: > > Hi, > > > > With gr-ieee 802.11, I know that SIFS cannot be respected becz > of USB delays > > (B210), but what we can also try is that : sending ACK packet > before all the > > layer-2 retries are exhausted ! Instead of sending ACK withing > 10 microsecs. > > > > In this way, the throughput will be down for sure but the device can > > communicate with a commercial 802.11 AP. > > > > I was wondering if anyone tried like this and I can get some > heads up. > > > > It's not only about sending fast enough, but mainly about sending > at the > right point in time (i.e. immediately after a frame). Since the SDR > doesn't do channel sensing on the FPGA, this is at least not > straightforward. > > Best, > Bastian > > _______________________________________________ > Discuss-gnuradio mailing list > [email protected] <mailto:[email protected]> > https://lists.gnu.org/mailman/listinfo/discuss-gnuradio > <https://lists.gnu.org/mailman/listinfo/discuss-gnuradio> > > > > > -- > Saulo Jorge bq > - "Beware of bugs in the above code; I have only proved it correct, > not tried it." > Donald Knuth. > > > _______________________________________________ > Discuss-gnuradio mailing list > [email protected] > https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
_______________________________________________ Discuss-gnuradio mailing list [email protected] https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
