Area/power efficient implementation of a Log-MAP decoder for Turbo Codes through memory optimization

Turbo-codes have generated a great deal of interest since their discovery in 1993. Turbo decoders employing the Log-MAP algorithm, not using the sliding window technique, require large amount of memory as either the forward or reverse state metrics have to be stored for the whole block being processed, before the log likelihood decisions are obtained. In this paper, the authors´ present analysis and results towards reducing the impact of area overhead while achieving low power consumption constraint in a Log-MAP decoder implementation for Turbo codes which employs a reduced complexity technique of the forward recursive computation of reverse state metrics when compared to a conventional Log-MAP decoder implementation. It has been observed that along with a power reduction of 35%, there is also a 34% reduction in area consumption in spite of the introduction of an additional block for the forward calculation of the reverse state metrics. This reduction in area is due to the enormous reduction in the memory storage.