Random Access With Massive MIMO-OTFS in LEO Satellite Communications

This paper considers the joint channel estimation and device activity detection in grant-free random access systems, where a large number of Internet-of-Things devices intend to communicate with low-earth orbit satellite sporadic way. In addition, massive multiple-input multiple-output (MIMO) orthogonal time-frequency space (OTFS) modulation is adopted combat dynamics terrestrial-satellite link. We first analyze input-output relationship single-input single-output OTFS when delay Doppler shift both exist, then extend it MIMO-OTFS. Next, by exploring sparsity delay-Doppler-angle domain, two-dimensional pattern coupled hierarchical prior sparse Bayesian learning covariance-free method (TDSBL-FM) developed for estimation. Then, active are detected computing energy estimated channel. Finally, generalized approximate message passing algorithm combined convolution (ConvSBL-GAMP) proposed decrease computations TDSBL-FM algorithm. Simulation results demonstrate that algorithms outperform conventional methods.