Skip to Main Content
This paper presents a complexity reduced maximum likelihood (ML) or maximum a posteriori (MAP) based iterative MIMO detection scheme. The a priori loglikelihood ratios (LLRs) provided by the decoder have a Gaussian distribution. By using this statistical characteristics of the LLRs, we calculate the mean and variance of LLRs provided by the decoder. Subsequently, a threshold value is calculated, and on the basis of the threshold value the MIMO detector makes the decision about calculating the extrinsic LLRs or using the same LLRs values provided by the decoder for a particular symbol transmitted from each transmit antenna. Simulation results show that the proposed complexity reduction method produces nearly the same bit error rate (BER) performance as the full search MAP with about a 30% of the complexity reduction. Complexity reduction can be controlled by adjusting the parameters of the proposed method.