23C3 - 1.5
23rd Chaos Communication Congress
Who can you trust?
| Speakers | |
|---|---|
|
Thomas Kleffel |
|
Christian Daniel |
| Schedule | |
|---|---|
| Day | 3 |
| Room | Saal 2 |
| Start time | 18:30 |
| Duration | 01:00 |
| Info | |
| ID | 1646 |
| Event type | Lecture |
| Track | Science |
| Language | English |
| Feedback | |
|---|---|
|
Did you attend this event? Give Feedback |
DVB-T - From Pixeldata to COFDM Transmission
How to build a complete FPGA-based DVB-T transmitter
As DVB-T is the key technology for terrestrial broadcasting for the next decades, this lecture tries to explain how it works: It covers the way of raw pixel data over MPEG2 video and audio encoding and via multiplexing of several streams and programs to the actual generation of the COFDM signal used for transmission. As the team has built a DVB-T transmitter, we will give a in-depth insight in how things are really done - including a demonstration of the transmitter.
- > For that we plan to stream the content of the 23C3 lectures locally on Alexanderplatz using TV channel 22. Don't forget to bring your DVB-T stick. For more information look at <a href="http://events.ccc.de/congress/2006/DVB-T">the DVB-T project description.</a>
Analog PAL based TV transmission is no more - DVB has taken over in Germany. This should be legitimate reason to take a really deep look into how the new technology works and what needs to be done to get your own transmission into the air. The base for our lecture is a standard FBAS signal coming from a video camera. From there we will go with it through all the necessary stages of encoding, framing, multiplexing and modulating.
First step is the A/D conversion of the picture and the sound. The resulting 240MBit/s stream is fed into the MPEG2 encoder where video and audio is encoded using the given parameters resulting in a bitstream of defined datarate. To achieve that, the video is divided into single blocks, movement relative to earlier and later frames is detected, and then the data is transferred into the frequency domain and then encoded. A relatively similar thing is done to audio: It is split into several frequency bands and for every band it is decided if the listener will hear this band or not using a psycho accoustical model. Then the remaining data is encoded into a bitstream as well.
After that a program clock reference is generated to enable the receiver to keep video and audio in sync while playing back the stream.
During the next stage, all video- and audio-streams are multiplexed into one big transport stream and several data tables are added. These tables define, which programs belong to this transport stream and assign video and audio streams to these programs. Electronic program guide and teletext are added and finally the COFDM modulator has the job to generate a base band signal for transmission.
Here data is scrambled and interleaved several times, two types of forward error correction are added and then the bits are distributed to the up to 8000 distinct carriers which compose the final signal.
As we have built a complete OFDM modulator we will then demonstrate how all these things work together and finally you will be able to watch our transmission using your own DVB-T stick.