Multi-dimensional systematic block codes

This paper is concerned with a family of Mmulti-dimensional systematic block (MDSB) codes, which can be decoded iteratively. Simulation results of MDSB codes based on Single-parity-check (SPC) and Hamming codes are presented for an additive white Gaussian noise channel. It is shown that performance improves and the code rate decreases with the increasing dimensionality. Convergence characteristics of iterative decoding are described in detail. Simulation results show that these systematic codes perform better than 2-dimensional counterparts, but have a disadvantage of error floor although they can provide near Shannon capacity's limit performance. MDSB codes offer a good performance versus complexity tradeoff if they use mixture of SPC and Hamming codes as their component codes.