MF-OAMP-Based Joint Channel Estimation and Data Detection for OTFS Systems

To overcome the low spectral efficiency issue caused by large guard space adopted in conventional channel estimation schemes for orthogonal time frequency space (OTFS) systems, superimposed pilot-data placement schemes without the use of guard space have been considered in the literature. However, it is still an open problem for efficient joint channel estimation and OTFS symbol detection based on superimposed pilot-data placement schemes. In this work, an orthogonal approximate message passing (OAMP) based joint channel estimation and detection algorithm with the aid of mean-field (MF) message passing is proposed. In particular, soft information corresponding to the effective channel matrix and data symbols are fed OAMP-based symbol detector and channel estimator via MF, respectively. Thanks to the MF message passing rules, enhanced error performance is brought by iterative information exchange between the OTFS detector and the channel estimator. Numerical results demonstrate that the proposed iterative algorithm can efficiently detect the data symbols without the use of guard space.