Efficient LLR Approximation and Receiver Design for Coded Constant Envelope OFDM

In the field of constant envelope orthogonal frequency division multiplexing (CE-OFDM), the use of phase modulation heralds a paradigm with an extremely reduced peak-to-average power ratio (PAPR) of zero. However, the non-linear characteristics inherent to phase modulation pose a formidable challenge to the integration of modern channel coding strategies aimed at improved bit error rate (BER) results. To address this issue, we present an efficient low-complexity log-likelihood ratio (LLR) computation strategy based on the traditional phase receiver, anchored in an effective approximation to the phase noise variance. Furthermore, this foundation paves the way for an innovative receiver architecture, namely the near-maximum likelihood (NML) receiver, designed to improve decoding performance without incurring significant computational overhead. Comparative analyses of encoded CE-OFDM across different receiver models not only confirm the accuracy and superiority of the proposed LLR estimation, but also underscore the dual advantages of our proposed NML receiver in optimising BER and detection complexity, positioning CE-OFDM as an optimal candidate for scenarios requiring high-efficiency power amplifiers (PAs).