Low-Density Hybrid-Check Coded Superposition Mapping and its Application in OFDM and MIMO
Since Shannon’s landmark paper, many approaches have been proposed to achieve the channel capacity. In the low SNR regime, the problem has almost been solved by capacity achieving channel codes. The research on coded modulation in the high SNR regime is still under development. Among many methods in accomplishing this goal, superposition mapping is an elegant way as it does not require extra shaping to generate a Gaussian-like distributed signal. Superposition mapping has been shown to offer very close to capacity performance for the AWGN channel by combining with an irregular channel code. The aim of this thesis is to search for a code which provides stable performance for moderate sequence length and sufficient number of iterations, which is more suitable for implementation. Concerning channel coding for superposition mapping, a generalized code design has recently been proposed. The so-called low-density hybrid-check (LDHC) coding intends to contrive coding and modulation in a joint way. The LDHC coding is constructed by integrating modulation into the Tanner graph. Thus, the complete code can be obtained by taking the effects of all the components into account. In this thesis, the LDHC code design is extended to OFDM and MIMO. For OFDM, the bit loading can be realized in the graph. In case of MIMO with spatial multiplexing, the code is extended to the spatial domain. In both cases, a suitable system structure will be proposed in this thesis. It will also be shown how this novel code design improves the system performance.