Memory optimization of MAP turbo decoder algorithms

Turbo codes are the most recent breakthrough in coding theory. However, the decoder's implementation cost limits their incorporation in commercial sys tems, Although the decoding algorithm is highly data dominated, no true mem ory optimization study has been performed yet. We have extensively and syst ematically investigated different memory optimizations for the maximum a po steriori (MAP) class of decoding algorithms. It turns out that it is not po ssible to present one decoder structure as being optimal. In bet, there are several tradeoffs, which depend on the specific turbo code, the implementa tion target (hardware or software), and the selected cost function, We ther efore end up with a parametric family of new optimized algorithms out of wh ich the designer can choose. The impact of our optimizations is illustrated by a representative example, which shows a significant decrease in both de coding energy (factor 2.5) and delay (factor 1.7).