Context and denoising based decoding of non-systematic turbo codes for redundant data

We consider strategies for universal utilization of context based redundancy left in channel coded data to improve channel decoding of non-systematic turbo codes. The statistical parameters of the sequence are extracted using such strategies and passed to an iterative decoder. Simulation results demonstrate significant coding gains obtained by utilization of this redundancy, which usually increase with well designed nonsystematic codes. Different context based strategies are compared