Non-Orthogonal Multiple-Access Strategies for Direct-to-Satellite IoT Networks

Direct-to-Satellite IoT (DtS-IoT) has the potential to support multiple verticals, including agriculture, industry, smart cities, and environmental disaster prevention. However, device transmissions are affected by intrinsic factors such as satellite trajectory, limited visibility time windows, collision messages and signal reception quality. This work introduces two DtS-IoT schemes using power domain Non-Orthogonal Multiple Access (NOMA) in the uplink with either fixed (FTP) or controlled (CTP) transmit power. We consider that the IoT devices use LoRa Modulation to transmit data packets to the LEO satellite in orbit, equipped with a Successive Interference Cancellation (SIC)-enabled gateway. We also assume the IoT devices are empowered with a predictor of the satellite orbit. Using real geographic location and trajectory data, we evaluate the performance of the average number of successfully decoded transmissions, goodput (bytes/lap), and energy consumption (bytes/Joule) as a function of the number of network devices. Numerical results show the trade-off between goodput and energy efficiency for both proposed schemes. Comparing FTP and CTP with ALOHA for 100 devices, we find goodput improvements of 28% and 18%, respectively. For the same number of devices, we also achieved improvements of 234% and 207% when the proposed strategies are compared with Max Angle, a method that transmits at the maximum elevation angle only. Notably, CTP effectively leverages transmission opportunities as the network size increases, outperforming the other strategies. Moreover, CTP shows the best performance in energy efficiency.