Turbo Decoder Visualization

Turbo codes use two parallel convolutional encoders and iterative belief propagation. Watch extrinsic information flow between the two decoders, each refining its beliefs over multiple iterations.

0
Iteration
BER
Avg |LLR|
3.0
Eb/N0 (dB)
3.0 dB
12

Each arrow represents extrinsic LLR information flowing from one decoder to the other, updating bit probability estimates.