Robust Design for Integrated Satellite–Terrestrial Internet of Things

Abstract

In this article, we investigate an integrated satellite–terrestrial (IST) Internet-of-Things (IoT) network, where satellite networks provide wireless services to the IoT devices uncovered by terrestrial cellular networks, so as to realize seamless coverage across the world. In order to support massive access of a large number of IoT devices, the low Earth orbit (LEO) satellites and the terrestrial base station (BS) both adopt power-domain nonorthogonal multiple access (NOMA) techniques designed according to channel state information (CSI). Due to the limited capacity of feedback links, there exists channel uncertainty at the LEO satellites and BSs, resulting in internetwork interference inevitably. In order to guarantee the performance of the IST-IoT network, we formulate a optimization problem to minimize the total transmit power of the IST-IoT network subject to outage probability constraints on the rates for the IoT devices associated to both satellite and terrestrial networks. Since the original problem is mathematically intractable, we utilize a series of transformations to obtain an approximate equivalent convex problem. Then, we propose an iterative penalty function (IPF)-based algorithm to jointly design robust beamforming for the satellite and BSs. Finally, we provide extensive simulation results to reveal the impacts of important system parameters and confirm the effectiveness of the proposed robust design algorithm for IST-IoT networks.