Improving the constituent codes of turbo encoders

The performance of turbo codes can be improved by choosing the best constituent encoder. Previous lists of the best constituent codes were found based on maximising the input weight, two effective free distance d/sub 2/. We show that minimising the number of nearest neighbours N/sub 2/ is also an important requirement. If there are multiple codes having the same d/sub 2/ and N/sub 2/, we show that the minimum distance d/sub min/ can be a useful criteria to identify the best code. A list of generator polynomials suitable for turbo codes are presented in this paper which satisfy the criteria mentioned above. The previously published results are also extended to include larger memory.

[1]  Daniel J. Costello,et al.  A distance spectrum interpretation of turbo codes , 1996, IEEE Trans. Inf. Theory.

[2]  D. Divsalar,et al.  On the Design of Turbo Codes , 1995 .

[3]  Andrew J. Viterbi,et al.  Error bounds for convolutional codes and an asymptotically optimum decoding algorithm , 1967, IEEE Trans. Inf. Theory.

[4]  Robert H. Deng,et al.  Trellis-coded multidimensional phase modulation , 1990, IEEE Trans. Inf. Theory.

[5]  Sergio Benedetto,et al.  Unveiling turbo codes: some results on parallel concatenated coding schemes , 1996, IEEE Trans. Inf. Theory.

[6]  Sergio Benedetto,et al.  Design of parallel concatenated convolutional codes , 1996, IEEE Trans. Commun..

[7]  A. Glavieux,et al.  Near Shannon limit error-correcting coding and decoding: Turbo-codes. 1 , 1993, Proceedings of ICC '93 - IEEE International Conference on Communications.