Impact of Bit-Flip Combinations on Successive Soft Input Decoding of Reed Solomon Codes

Soft Input Decoding of Reed Solomon codes using successive iterative bit-flipping was introduced earlier. The proposed method uses a concatenation of Convolutional and Reed Solomon Codes. It was shown that the proposed method not only improves the coding gain, but also helps in circumventing the miscorrections, that occur with the given probability, using the hard decision Reed Solomon decoder. In this paper, the impact of bit-flip combinations on the soft decision Reed Solomon decoder is analyzed. Increase in the bit-flip combinations results in an increase in the error locator set of the iterative decoding algorithm. It is shown that by increasing the possible bit- flip combinations using a threshold, the coding gain of the concatenated codes increases while the miscorrections rate decreases. The impact of the increase in the threshold on the decoder performance is also investigated in this paper. Simulations are performed for different code rates and an improvement in the coding gain is obvious through the simulation results presented. Moreover, the simulation statistics show a decrease in the miscorrections with an increase in the size of the bit-flip combinations.