Future autonomous transportation is one of the most demanding application areas in terms of connectivity, as it has to simultaneously meet stringent criteria that do not typically go hand in hand, such as high throughput, low latency, high coverage/availability, high positioning and sensing accuracies, high security and robustness to interferences, etc. In order to meet the future demands of challenging applications, such as applications relying on autonomous vehicles, terrestrial networks are no longer sufficient and are to be augmented in the future with satellite-based networks. Among the emerging satellite networks, Low Earth Orbit (LEO) networks are able to provide advantages over traditional Medium Earth Orbit (MEO) and Geo-Stationary Earth Orbit (GEO) networks in terms of signal latency, cost, and performance. Nevertheless, several challenges exist in LEO system design, which have not been fully addressed in the existing literature. In particular, the problem of LEO-system optimization of design parameters is a multi-dimensional problem with many aspects to be considered. This paper offers a comprehensive survey of the LEO-system design parameters, of the challenges in LEO system design process, and of the optimization methods for satellite communication, positioning, and sensing applications, as well as a summarizing discussion on the design considerations for LEO-based networks to support future autonomous transportation.

Low Earth Orbit (LEO) satellites are expected to improve the robustness and reliability of critical Positioning, Navigation, and Timing (PNT) services by increasing both the efficiency and diversity of Global Navigation Satellite Systems (GNSS). However, the feasibility of LEO-PNT services remains uncertain primarily due to the large number of required satellites. The deployment of a LEO-PNT system therefore depends on the feasibility of deploying many LEO satellites with characteristics strongly dependent on the performance requirements of both the payload and the platform. This work addresses the feasibility uncertainty by modeling two potential LEO-PNT payloads and satellite platform designs with different size, weight, and power (SWaP) requirements. Based on simulations and components currently available in the space market, a holistic modeling for the performance of low- and high-SWaP payload designs is evaluated with a focus on clock stability and signal quality. Additionally, the design process includes an assessment of a custom navigation antenna to ensure that the antenna can meet its stringent size and performance constraints. The feasibility of the proposed LEO-PNT constellations is evaluated based on satellite link, power, and mass budgets, as well as antenna performance. In addition, the cost of both the low- and high-SWaP satellites is estimated. This work shows that LEO-PNT payload and constellations are feasible with the technology commercially available today, and they could offer significantly stronger PNT signals compared to traditional GNSS services. The work also benchmarks the proposed payloads against commercial projects in terms of SWaP, timing and positioning accuracy. These results can assist future LEO-PNT simulations to utilize detailed information of payload components and take LEO-PNT services a step closer to realization.