Joint Terminal Identification and Channel Estimation of Asynchronous Grant-Free Access for LEO Satellite-IoT Network

In this paper, the joint active terminal identification (ATI) and channel estimation (CE) problem is investigated for the asynchronous grant-free random access system in the context of low-earth orbit (LEO) satellite Internet of Things (IoT) network. An asynchronous grant-free model is established to characterize the implications of delay and Doppler caused by LEO satellites. Concurrently, a generalized approximate message passing-aided structured joint detection (SJD) scheme with Rician parameters learning (RPL) is proposed for joint ATI and CE. The prior mean and variance of the Rician channel are treated as hyperparameters and updated via the expectation-maximization algorithm. Furthermore, to alleviate the modeling mismatch, we develop an off-grid model following Taylor expansion, accompanied by a mismatch error parameter learning (MPL) framework to boost the accuracy of joint detection. Simulation results demonstrate that the proposed RPL-SJD scheme outperforms existing approaches in terms of normalized mean square error with 4dB and activity detection error rate with 7dB.