Real-time kinematic (RTK) and precise point positioning (PPP)-RTK techniques are essential for safety-critical applications like structural health monitoring and open-pit mining operations, among others. These applications often operate in remote areas lacking the necessary traditional communication infrastructure for high-precision global navigation satellite systems (GNSS). This article explores non-terrestrial networks (NTN) as a viable solution, specifically using low Earth orbit (LEO) satellite constellations to deliver RTK and PPP-RTK correction data through NTN. We evaluate the convergence time of these techniques and the required link capacity through real-world experiments, relating these findings to actual LEO satellite communication access duration and intervals. Our results demonstrate that these communication periods provide sufficient time for achieving convergence in high-precision GNSS navigation solutions.
The evolution of Non-Terrestrial Networks (NTN) will require some conceptual changes in the communication paradigm established in common cellular and low-power networks. This will be paramount in sparse Low Earth Orbit (LEO) satellite constellations that cannot ensure service continuity. Limited satellite access time for terrestrial devices and the typical Internet of Things (IoT) duty-cycled application will require coordination mechanisms, so that desired sleep and wake-up periods can be synchronized with satellite availability. Towards this coordination, this work first characterizes the access time of LEO satellites for terrestrial devices. It then evaluates the model proposed by 3GPP for orbit determination, the Simplified General Perturbation 4 (SGP4), in terms of accuracy and energy saving from the low power point of view. From the evaluation, a novel method is proposed based on relaxing the periodic updates of satellite orbital information required by SGP4 to increase battery life of IoT-NTN devices. The method is enhanced with a function that matches the uncertainty in the access estimates to a wake-up guard time, improving and guarantying access success. Finally, the proposal is validated using data from a real communication link between a terrestrial device and the Enxaneta LEO CubeSat (3B5GSAT).
Neuromorphic systems based on hardware neural networks (HNNs) are expected to be an energy and time-efficient computing architecture for solving complex tasks. In this paper, we consider the implementation of deep neural networks (DNNs) using crossbar arrays of memristors. More specifically, we considered the case where such devices can be configured in just two states: the low-resistance state (LRS) and the high-resistance state (HRS). HNNs suffer from several non-idealities that need to be solved when mapping our software-based models. A clear example in memristor-based neural networks is conductance variability, which is inherent to resistive switching devices, so achieving good performance in an HNN largely depends on the development of reliable weight storage or, alternatively, mitigation techniques against weight uncertainty. In this manuscript, we provide guidelines for a system-level designer where we take into account several issues related to the set-up of the HNN, such as what the appropriate conductance value in the LRS is or the adaptive conversion of current outputs at one stage to input voltages for the next stage. A second contribution is the training of the system, which is performed via offline learning, and considering the hardware imperfections, which in this case are conductance fluctuations. Finally, the resulting inference system is tested in two well-known databases from MNIST, showing that is competitive in terms of classification performance against the software-based counterpart. Additional advice and insights on system tuning and expected performance are given throughout the paper.