On Gallager bounding technique with applications to turbo-like codes over fading channels

New rigorous analytical upper bounds on the block and bit error probabilities of ML decoded block codes are presented. These bounds are applied to turbo-like codes over fully interleaved fading channels, demonstrating a good match to practical efficient iterative decoding algorithms. The proposed improved bounding technique is a modified and generalized version of the previously introduced bounds for the AWGN channel by Duman and Salehi (see IEEE Trans. Communications, vol.46, p.717-23, 1998), which are based on modifying the classical Gallager technique