Patent attributes
New and improved a-posteriori decoding probabilities, decisioning metrics, and implementation algorithms for turbo and convolutional decoding to replace the probabilities and decisioning metrics currently used in the maximum likelihood ML and maximum a-posteriori MAP algorithms. A-posteriori probabilities p(x}y) replace the current ML probabilities p(y}x) wherein y is the received symbol and x is the transmitted data and the MAP a-posteriori probability p(s′,s|y) replaces the current MAP joint probability p(s′,s,y) wherein s′,s are the trellis decoding states at k−1, k and y is the observed data set y(k),k=1, 2, . . . , N. This yields a-posteriori probabilities and decisioning metrics to improve decisioning and bit error rate BER performance and to provide a new mathematical decoding framework. Complexity is the same as current implementations.
