[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Discuss-gnuradio] GSoC: mid-term update (DRM transmitter)

From: wunsch
Subject: [Discuss-gnuradio] GSoC: mid-term update (DRM transmitter)
Date: Sun, 1 Jul 2012 12:12:50 +0200
User-agent: SquirrelMail/1.4.19

Hi all,

as the mid-term evaluations happen next week, I want to give you an
insight into  what I have done until now and what will be done until the
end of GSoC (and later on).

In the first phase of GSoC, I concentrated on reading and understanding
the DRM Standard [1], involving source encoding (AAC), channel encoding
(punctured convolutional codes), interleaving (time and frequency
domain) as well as modulation and transmission (QAM/OFDM).

Having all this read and understood, I started developing a DRM
transmitter and its inverse counterpart (for unit testing) in Matlab. A
functional demo can be downloaded at github [2]. The transmitter uses a
very common configuration and produces a standard compliant DRM signal
(.wav format). As input, any wav file can be used (although it should be
sampled at 24kHz).

I successfully tested this demo transmitter with a USRP1 and Basic TX
daughterboard as transmit frontend against a commercially available DRM
receiver [3].

At the moment, I'm porting everything I did in Matlab to Gnu Radio and
adding support for other transmission configurations. All blocks will
have GRC bindings and the flow graph will look almost exactly as in the
standard (page 15) making the functionality intuitively clear and
ensuring high modularity.

For communication between the blocks, I use vectors instead of streams
as I always have fixed block lengths that are processed in each step
from block to block. My impression was that this makes the information
flow clearer as there is always one block of information (with varying
length), that is processed and handed along the flow graph.

A more detailed description of my design including a flow graph with
data types and exemplary vector lengths can be found on github(folder
'misc') [2].

Especially at this point I would really appreciate feedback from the
community about what you think about this concept or simply any general
ideas or hints which you think might be helpful for me. I already
encountered the problem of suboptimal buffer allocation.

As time has shown that the initial timeline for the project was a bit
optimistic, we decided to concentrate on the transmitter as final scope
of my GSoC project. But as this is only a part of my bachelor thesis I
am confident to be able to develop this further (meaning the receiver)
after GSoC has ended. I will notify the community about any new additions.

As a conclusion I want to say that I am really happy to be part of the
Gnu Radio community, that I already have learned a lot and that I hope
to make a valuable contribution to the Gnu Radio project.

Best regards,

[1] ETSI ES 201 980 v3.1.1 www.drm.org/uploads/files/es_201980v030101p.pdf
[2] https://github.com/fewu/gnuradio_drm
[3] NEWSTAR DR111 http://www.cdnse.com/products/dr111

reply via email to

[Prev in Thread] Current Thread [Next in Thread]