Efficient Fixed-Point Implementation of a SC-MMSE Turbo Equalizer

Turbo equalization is a promising approach to deal with the inter-symbol interference when transmitting over frequency selective radio channels. The iterative exchange of soft information between the equalizer and the decoder improves the decision reliability and hence reduces the bit error probability compared to conventional receivers. However, the processing of real-valued soft information with floating point precision causes high memory consumption and complex arithmetic logic units. In this paper, the effects of a fixed-point implementation of linear minimum mean square error (MMSE) based turbo equalizers with soft cancellation (SC) are studied. The applied soft-input soft-output (SISO) algorithms are analyzed concerning the value range and efficient fixed-point representations are derived. Results show, that even for small word lengths of 8 bit, the performance loss is negligible.