Convergence of iterative decoding for fixed-point implementations
暂无分享,去创建一个
In this paper, we analyze the fixed-point implementation of the Turbo Decoder algorithm. Fixed-point numbers have a finite dynamic range and are quantized, which often results in a performance loss. Although the performance loss can be significant for 3 or 4-bit fixed-point numbers, we show that it is possible to minimize the loss by configuring the algorithm to reduce the required dynamic range at any one point in the data flow. Indeed, for the iterative Turbo Decoder, the dynamic range is the most critical performance factor.
[1] John Cocke,et al. Optimal decoding of linear codes for minimizing symbol error rate (Corresp.) , 1974, IEEE Trans. Inf. Theory.
[2] Sergio Benedetto,et al. A soft-input soft-output maximum a posteriori (MAP) module to decode parallel and serial concatenated codes , 1996 .