> Martin,
>
> That is fantastic works! Thanks for putting it out there!
>
> I have a question on the channel coding work done. Was there any benchmarking
> performed on it to test the throughput speed? Also, is there any SIMD code used
> to improve the performance? If not, I wonder how difficult it might be to use
> Volk functions for that?
>
> Thanks again!
Hi Tom and everyone,
some basic benchmarks were done (if your interested in the details, see
pp59 in the included PDF. You'll find the specs of the test system
there. Figure 5.2 is memory usage in bits, Figure 5.4 is the average
time to encode one information bit in µs, Figure 5.6 is the average CPU
load when transmitting 100 kbit/s (goodput). Way more interesting is the
decoding: Figure 5.7 is the average decoding time needed to decode one
code bit. Figure 5.9 shows achievable bit rates on the given system).
As for SIMD code and Volk: Nope, we've done nothing here. But it's a
good point. In particular, the Fast Hadamard Transform was implemented
as part of this toolbox, and I guess that alone could be optimized a
great deal. I'll look into it... could be interesting. Thanks for the
feedback!
MB
--
Karlsruhe Institute of Technology (KIT)
Communications Engineering Lab (CEL)
Dipl.-Ing. Martin Braun
Research Associate
Kaiserstraße 12
Building 05.01
76131 Karlsruhe
Phone: +49 721 608-43790
Fax: +49 721 608-46071
www.cel.kit.edu
KIT -- University of the State of Baden-Württemberg and
National Laboratory of the Helmholtz Association
No comments:
Post a Comment