Low-density parity-check (LDPC) codes have been shown to have near Shannon-limit performance when a probabilistic decoding is used and the code length goes to infinity. On the other hand, non-binary low-density parity-check (NB-LDPC) codes outperform binary LDPC codes for short code lengths. However, the coding gain of NB-LDPC codes comes from the expense of computational complexity. \n \nIn this thesis, we consider the problem of a trade-off between reasonable complexity and good error-rate performance of decoding nonbinary LDPC codes over GF(
q). We first introduce the existing decoding algorithms of nonbinary LDPC codes and compare with decoding algorithms of binary LDPC codes. Second, low-complexity decoding algorithms of nonbinary LDPC codes is considered and two low-complexity symbol-flipping algorithms based on D-SFDP is proposed. These algorithms outperform the existing D-SFDP algorithm by improving the flipping metric. Proposed algorithms provide good trade-off between complexity and error performance.
We develop a new low complexity algorithm for decoding low-density parity-check (LDPC) codes. The developments are oriented specifically toward the low cost-yet effective-decoding of (high rate) finite geometry LDPC codes. The decoding procedure updates the hard-decision received vector iteratively in search of a valid codeword in the vector space. Only one bit is changed in each iteration and the bit selection criterion combines the number of failed checks and the reliability of the received bits. Prior knowledge of the signal amplitude and noise power is not required. An optimal mechanism to avoid infinite loops in the search is also proposed. Our studies show that the algorithm achieves an appealing performance versus complexity trade-off for finite geometry LDPC codes.
We develop a new low-complexity algorithm to decode low-density parity-check (LDPC) codes. The developments are oriented specifically toward low cost, yet effective, decoding of (high-rate) finite-geometry LDPC codes. The decoding procedure iteratively updates the hard-decision received vector in search of a valid codeword in the vector space. Only one bit is changed in each iteration, and the bit-selection criterion combines the number of failed checks and the reliability of the received bits. Prior knowledge of the signal amplitude and noise power is not required. An optional mechanism to avoid infinite loops in the search is also proposed. Our studies show that the algorithm achieves an appealing tradeoff between performance and complexity for finite-geometry LDPC codes.