Based on a finite constellation interference assumption, a coded modulation approach to dirty paper coding is presented in this thesis. Two dirty paper coding schemes based on probabilistic shaping are proposed and their achievable rates are analyzed. The predicted coding gains are verified by numerical simulations where shaping is realized by a many-to-one mapping. By considering the finite constellation interference as the result of interference quantization, the proposed coding schemes apply also to the case of continuous interference.     «

Based on a finite constellation interference assumption, a coded modulation approach to dirty paper coding is presented in this thesis. Two dirty paper coding schemes based on probabilistic shaping are proposed and their achievable rates are analyzed. The predicted coding gains are verified by numerical simulations where shaping is realized by a many-to-one mapping. By considering the finite constellation interference as the result of interference quantization, the proposed coding schemes apply...     »