Forum: GNU Radio Question about latency

Announcement (2017-05-07): www.ruby-forum.com is now read-only since I unfortunately do not have the time to support and maintain the forum any more. Please see rubyonrails.org/community and ruby-lang.org/en/community for other Rails- und Ruby-related community platforms.
Matigakis E. (Guest)
on 2009-04-01 15:53
(Received via mailing list)
Hello everyone,

I am trying to implement some relaying schemes in gnuradio which require
fast response from
the relay. I only know how to use flow graphs at the moment so I made my
code with flowgraphs and
the delay from receiving a packet until I retransmit it is about 10ms.
I'm using USRP1 and the
sampling rate is 2MSPS and I use the autotransmit capability of USRP.

First of all I'm trying to understand where the delay comes from. I'm
sure the delay is not due to
processing time of the packet because the processing is very simple. I
don't know about latencies in
the kernel though. I read in a paper  that the time it takes for data
to go from user space to kernel
is only 200us.

Secondly I would like to know if using mblocks has any chance of
reducing this delay. By the way is there
any documentation for the mblocks?

Thanks
Manolis
Firas A. (Guest)
on 2009-04-02 14:05
(Received via mailing list)
Hi,



Hello everyone,

I am trying to implement some relaying schemes in gnuradio which require
fast response from
the relay. I only know how to use flow graphs at the moment so I made my
code with flowgraphs and
the delay from receiving a packet until I retransmit it is about 10ms.
I'm
using USRP1 and the
sampling rate is 2MSPS and I use the autotransmit capability of USRP.

First of all I'm trying to understand where the delay comes from. I'm
sure
the delay is not due to
processing time of the packet because the processing is very simple. I
don't
know about latencies in
the kernel though. I read in a paper  that the time it takes for data
to go
from user space to kernel
is only 200us.

Secondly I would like to know if using mblocks has any chance of
reducing
this delay. By the way is there
any documentation for the mblocks?

Thanks
Manolis



1) The following page may be useful:
http://gnuradio.org/trac/wiki/UsrpFAQ/Latency


2) Did you tried to increase sampling rate up to 4MHz ?

3) How did you measured the delay ?


Best Regards,

Firas

--
View this message in context:
http://www.nabble.com/Question-about-latency-tp228...
Sent from the GnuRadio mailing list archive at Nabble.com.
Mattias K. (Guest)
on 2009-04-02 19:30
(Received via mailing list)
> First of all I'm trying to understand where the delay comes from.
Hi,
I'm not entirely sure, but I think there is a sleep in the (FlexRF)
daughter- board drivers... Try grepping "usleep".

BR.
//Mattias
Mohd adib S. (Guest)
on 2009-04-04 18:40
(Received via mailing list)
i have some Question.. in the
http://gnuradio.org/trac/wiki/UsrpFAQ/Latency
it state that the data from usrp to usb will be send only if it is more
than
512byte right? so what happen if my data is less?

i try to send data using benchmark_tx and benchmark_rx.. most of the
data at
the back of my packet is loss.. for example if i send 100 bytes i only
manage to receive 76 byte.. the loss data is always the last bytes.. if
this
cause by the minimum data that USB required?

my 2nd Question is how to disable the switching between tx and rx in
rfx2400? (You can do the test with RFX2400 but you have to disable auto
T/R
switch to use TX/RX sma for tx and RX2 for RX [after using select
antenna
command). ] how to do this? what is the command that i have to type?
thanks
=)

adib
--
View this message in context:
http://www.nabble.com/Question-about-latency-tp228...
Sent from the GnuRadio mailing list archive at Nabble.com.
This topic is locked and can not be replied to.