Universal serially concatenated trellis coded modulations and rate-compatible high-rate LDPC codes

This dissertation presents you universal serially concatenated trellis-coded modulations (SCTCMs) that perform consistently close to the available mutual information for the periodic erasure channel (PEC), the periodic fading channel (PFC) and the 2 x 2 compound matrix channel. For the PEC and PFC, the universal SCTCMs extend the near-capacity performance of serially concatenated convolutional codes under AWGN to periodically time-varying channels. We use both maximum-likelihood decoding criteria and iterative-decoding criteria to design these universal SCTCMs. Each component of the SCTCM including the constituent codes, constellation, labeling, and interleaver are carefully chosen to achieve universality.; For the space-time channel, by de-multiplexing the symbols across the antennas, universal SCTCMs for the period-2 PFC deliver consistent performance over the eigenvalue skew of the matrix channel. Within the family of channels having the same eigenvalue skew, a time-varying linear transformation (TVLT) is used to mitigate the performance variation over different eigenvectors. Because of their consistent performance over all channels, the proposed codes will have good frame-error-rate (FER) performance over any quasi-static fading distribution.; A graph-conditioning algorithm called the approximate cycle extrinsic message degree (ACE) algorithm is used to construct high-rate (R ≥ 1/2) irregular LDPC codes. For high-rate LDPC codes, due to the large number of degree-2 variable nodes in the optimal degree distribution, it is more difficult to condition the graph. By constraining the number of degree-2 nodes, the ACE algorithm can dramatically lower the error floor with little compromise of the threshold. The same design criteria are suitable for rate-compatible applications using information-nulling.; Another rate-compatible technique uses row-combining, which combines rows of a lower-rate parity-check matrix to form one or more higher-rate parity-check matrices according to predefined combining rules. The resulting LDPC codes are called Constant Blocklength Multiple-Rate (CBMR) LDPC codes. This row-combining approach fits well with an efficient hardware architecture known as the irregular partitioned permutation (IPP) LDPC code. We identified all the constraints that the IPP codes and row-combining placed on the parity-check matrix and designed the row-combined IPP (RC-IPP) codes. The implementation issues of encoding and decoding for RC-IPP codes are exploited. As a result, the RC-IPP codes have FER as good as that of the stand-alone codes at each rate.