Soft Decision Decoding vs. Hard Decision Decoding with Viterbi Algorithm

In this paper, we investigate the performance comparison between soft decision decoding and hard decision decoding when transmitted symbols undergo corruption in an AWGN channel. Our analysis reveals that soft decision decoding significantly outperforms hard decision decoding in reducing frame error rate (FER) and minimizing erroneous frames, as HDD incurs higher quantization loss compared to SDD. The implementation utilizes Viterbi decoding algorithm where branch metrics are computed differently: HDD uses Hamming distance for binary decisions while SDD employs Euclidean distance calculations with multi-bit quantization (typically 3-8 bits). Our research demonstrates that soft decision decoding achieves superior performance under AWGN channel conditions, which is crucial for enhancing data transmission reliability and reducing bit error rates. The Viterbi decoder implementation features path metric updates and traceback operations common to both approaches, with SDD requiring additional quantization levels and soft-input handling.