Complexity Bounds of LDPC codes for Parallel Channels
Youjian Liu, Jilei Hou, Vincent K. N. Lau · 2004
We study complexity bounds for parallel channels. The complexity is defined as the number of ones in a parity check matrix per information bit. For message passing decoding, this complexity indicates computation required per iteration. We demonstrate that there is an optimization problem, analogous to power water-filling in parallel channels, that computation resource can also be allocated according to channel qualities through ��Ø � �ÐÐ�Ò � in a parity check matrix. The optimization is illustrated by the complexity bounds. Some previous bounds for a single channel are generalized to parallel channels. We show that codes designed for parallel channels can also be used for a single channel with transmitter side information and with i.i.d. states. I.