"For block lengths of about 500, an IBM 7090 computer requires about 0.1 seconds per iteration to decode a block by probabilistic decoding scheme. Consequently, many hours of computation time are necessary to evaluate even a $P(e)$ in the order of $10^{-4}$."


In comparison to the early pioneers of LDPC codes, such as Robert G. Gallager, we can provide a highly parallelized GPU-Decoding framework with very high throughputs in the range of Mbit/s.

This webdemo uses a fully parallelized Sum-Product-Decoder (SPA) running on our internal GPU-Cluster. Use this webdemo to simulate your own codes in order to discover the performance of arbitrary LDPC codes.

However, even nowadays the computation complexity of such codes is very demanding and we have to limit your computation time to a maximum of several minutes.

Belief Propagation Decoder

Our decoder implements the sum-product algorithm (SPA) according to [2]. In order to achieve a maximum throughput decoding stops if all check-nodes are fulfilled.