Near perfect decoding of LDPC codes

Cooperative optimization is a new way for finding global optima of complicated functions of many variables. It has some important properties not possessed by any conventional optimization methods. It has been successfully applied in solving many large scale optimization problems in image processing, computer vision, and computational chemistry. This paper shows the application of this optimization principle in decoding LDPC codes, which is another hard combinatorial optimization problem. The cooperative optimization algorithm is much simpler in computation than the classical sum-product algorithm. Compared to the sum-product algorithm, our algorithm lowered the error floor more than one order of magnitude in decoding an experimental LDPC code for China´s HDTV