An improved SOVA for Turbo Code Decoding

This paper proposes an improved SOVA for turbo code decoding, and it targets the decoded state error. According to the simulation results, for 5 iterations, the performance of modification with decoding depth zeta = 12 is better than that of conventional decoder with zeta= 20. For the size and power consumption of SMU and TBU, the reduction ration can be about 20%. With 2~3 iterations, the performance of modification is equivalent to that of conventional decoder with 5 iterations. For zeta= 20 , the reduction ration of reliability values store operations and interleaving/deinterleaving is about 28% and 40%, respectively.