On the Relation of MIMO APP Detection and SIMO Maximum Ratio Combining

In classical non-iterative MIMO detection a soft-output MIMO detector computes likelihoods of the transmit bits being 1 or 0, given the received symbol vector. In iterative MIMO detection-decoding, detection performance is improved by exploiting apriori information about bit probabilities from the decoder. Transmit bits are viewed as random variables and the optimum detector performs Bayesian updating of transmit bit probabilities to compute the aposteriori probabilities (APP). We show that with growing apriori knowledge APP and max-log-APP MIMO detector performance increases up to the performance of maximum ratio combining (MRC) for SIMO transmission of BPSK modulation, when transmitting with the same energy per symbol. This is an upper bound for detector performance in iterative detection-decoding (turbo MIMO receiver).