Optimising the Constituent Codes for Turbo Decoders

Since the introduction of turbo codes a large amount of research effort has been spent on improving turbo coding performance. One way of improving the performance is to choose the best constituent codes. Previously [1] the constituent codes have been chosen by maximising the input weight 2 effective freedistance d 2 eff . The number of nearest neighbours Neff is also an important parameter. This paper present a list of constituent codes that maximise d 2 eff and minimise the Neff . By choosing constituent codes that minimise Neff the performance of turbo codes can be improved with no increase in complexity. I. Introduction Turbo codes are powerful error correction schemes which were first proposed in [2]. Since turbo codes are a parallel concatenation of two or more convolutional codes, the characteristic of each constituent encoder as well as the interleaver is important in order to achieve good performance for turbo codes. One way to improve the performance of turbo codes is ..