Receiver Architectures for MIMO-OFDM Based on a Combined VMP-SP Algorithm

Iterative information processing, either based on heuristics oranalytical frameworks, has been shown to be a very powerful tool forthe design of efficient, yet feasible, wireless receiverarchitectures. Within this context, algorithms performingmessage-passing on a probabilistic graph, such as the sum-product(SP) and variational message passing (VMP) algorithms, have becomeincreasingly popular.In this contribution, we apply a combined VMP-SP message-passingtechnique to the design of receivers for MIMO-ODFM systems. Themessage-passing equations of the combined scheme can be obtainedfrom the equations of the stationary points of a constrainedregion-based free energy approximation. When applied to a MIMO-OFDMprobabilistic model, we obtain a generic receiver architectureperforming iterative channel weight and noise precision estimation,equalization and data decoding. We show that this generic scheme canbe particularized to a variety of different receiver structures,ranging from high-performance iterative structures to low complexityreceivers. This allows for a flexible design of the signalprocessing specially tailored for the requirements of each specificapplication. The numerical assessment of our solutions, based onMonte Carlo simulations, corroborates the high performance of theproposed algorithms and their superiority to heuristic approaches