A soft-output detector for the golden code

Soft-output detection of the golden code is an important but computationally difficult task. We propose a low-and fixed-complexity soft-output detector for the golden code that uses linear equalization to simplify the task of finding a list of candidate values for one pair of information symbols, and then - for each pair on the list - it uses decision-feedback equalization to find candidate values for the remaining pair of information symbols. We propose a simple ordering algorithm that exploits the golden code´s structure to ensure that the overall algorithm performs well. Numerical results indicate that the proposed algorithm is significantly less complex than previously reported algorithms, yet performs comparably.