Information-geometric Analysis of Iterative Receiver Structures

This thesis gives an introduction to basic concepts of information geometry, a theory that uses notions of differential geometry to describe information and probabilities. These concepts are applied to the analytic description of an iterative receiver for bit interleaved coded modulation (BICM). The operation of the receiver’s subblocks are recognized as implicit projections of input probability distributions on certain sub-manifolds, which for example represent the code. Furthermore, the effects of log-likelihood ratio (LLR) clipping onto the receiver performance are studied using information geometry. Again, clipping has an interpretation as a projection. Finally numerical simulations of a BICM system show the performance improvements due to iterative decoding and the effects of LLR clipping.