Simplified Method for Log-Likelihood Ratio Approximation in High-Order Modulations Based on the Voronoi Decomposition

摘要

In modern communication systems, high-order modulation schemes have been widely employed to improve the spectral efficiency. In this context, soft decision methods are usually preferred, due to their superior performance over hard decisions. Soft decoders use metrics such as the log-likelihood ratios (LLRs) to measure the reliability of their decisions. Typically, the complexity of the exact channel LLR calculation increases exponentially with the modulation order of the system. The most well-known LLR approximation, the Max-Log-MAP symbol-to-bit demapper, still demands a large number of operations. In this paper, we present a novel simplified scheme that greatly reduces the complexity of the LLR demapper at high-order PSK and APSK modulations compared to the Max-Log-MAP soft-demapper. Moreover, it is shown that the proposed method has a negligible overall system performance loss compared to the same method.