MLSR: a novel routing algorithm for multilayered satellite IP networks

The increasing demand for ultra-low-latency and high-throughput communication necessitates network architectures capable of ensuring reliable real-time connectivity. To address this requirement, integrated architectures combining terrestrial networks (TN) and non-terrestrial networks (NTN)—particularly hierarchical multi-layer satellite systems composed of geostationary Earth orbit (GEO), medium Earth orbit (MEO), and low Earth orbit (LEO) satellites—have attracted considerable attention. Such hierarchical systems offer global coverage, low latency, and high capacity. However, their highly dynamic topology, fast-moving satellites, and fluctuating link quality pose significant challenges to establishing efficient end-to-end routing paths. Traditional routing methods based on static paths or centralized control lack the adaptability required for such environments. To address these challenges, this paper introduces a neural hierarchical reinforcement learning (HRL)-based routing algorithm for satellite networks. The framework exploits the layered structure of GEO, MEO, and LEO satellites, enabling gateway-assisted routing and adaptive path selection. Control responsibilities are distributed across orbital layers: the GEO layer allocates hop-count budgets, the MEO layer selects feasible LEO paths, and the LEO layer forwards data packets. When inter-satellite links are unavailable, terrestrial gateways provide alternative routes. Routing decisions further account for satellite-to-satellite and satellite-to-ground link quality, with an end-to-end delay-based reward function capturing transmission, processing, and propagation effects. Simulation results show that the proposed neural hierarchical framework, supported by integrated terrestrial and non-terrestrial networking, achieves faster convergence, greater route stability, and better adaptability than conventional routing algorithms and HRL approaches.

This work develops an analytical framework for downlink low Earth orbit (LEO) or medium Earth orbit (MEO) satellite communications, leveraging tools from stochastic geometry. We propose a tractable approach to the analysis of such satellite communication systems, accounting for the fact that satellites are located on circular orbits. We accurately incorporate this geometric property of LEO or MEO satellite constellations by developing a Cox point process model that jointly produces orbits and satellites on these orbits. Our work contrasts with previous modeling studies that presumed satellite locations to be entirely random, thereby overlooking the fundamental fact that satellites are jointly positioned on orbits. Employing this Cox model, we analyze the network performance experienced by users located on Earth. Specifically, we evaluate the no-satellite probability of the proposed network and the Laplace transform of the interference created by such a network. Using it, we compute its SIR (signalto-interference) distribution, namely its coverage probability. By presenting fundamental network performance as functions of key parameters, this model allows one to assess the statistical properties of downlink LEO or MEO satellite communications and can thus be used as a system-level design tool to operate and optimize forthcoming complex LEO or MEO satellite networks.

Aiming at the problems of high dynamic topology and limited on-board resources in Low Earth Orbit (LEO) satellite networks leading to poor performance of traditional routing algorithms, this paper combines the centralized control idea of Software-Defined Networking (SDN) with the collective learning ability of the Particle Swarm Optimization (PSO) algorithm to propose an adaptive potential field routing algorithm based on PSO. This algorithm models the network as a dynamic potential field, defines potential energy based on the cumulative congestion degree to guide data packet transmission; adaptively learns and optimizes link cost weights through the PSO mechanism, achieving dynamic evolution of the path evaluation criterion; and adopts a probability-based multi-path forwarding mechanism to achieve load balancing. Simulation results show that compared to traditional algorithms such as Dijkstra and ACO (Ant Colony Optimization), this algorithm significantly improves performance in terms of average end-to-end delay, network throughput, and load balancing, especially suitable for dynamic traffic scenarios.

This paper proposes a double-layered satellite network based on the rateless code, which can globally distribute the information with a few satellites. The satellite network is composed of Geostationary Earth Orbit (GEO) and Low Earth Orbit (LEO) satellites. The GEO satellites which cover the low and middle latitudes serve as the backbone network, whereas the LEO satellites are considered the enhanced network to make up for the shortage of the backbone network. Due to the interruption tolerance of the rateless code, only a few satellites are employed to achieve worldwide information distribution in the design of the satellite network. The coverage performance and the average elevation angle of the user are simulated, and results demonstrate that the network can achieve globally information distribution.

Due to the relative orbiting of Low Earth Orbit (LEO) satellite and Medium Earth Orbit (MEO) satellite, the inter-layer links (IIL) of LEO/MEO Double-Layered Satellite Networks (DLSN) have to switch dynamically, resulting in dynamic changes of satellite network topology, which has an impact on the data transmission performance of DLSN. In this paper, the IIL topology of LEO/MEO DLSN is analyzed and simulated. Firstly, the visibility calculation method and establishment strategy of IIL between LEO and MEO satellite as well as the calculation and optimization of system snapshots is analyzed. Then, simulation analysis regarding the number and duration of system snapshots, IIL switching time for LEO, the number of switching IIL in each system snapshot, the number of IILs that each MEO needs to establish are presented. Finally, the future research trends of DLSN IIL topology are discussed, such as how to reduce the maximum IIL number for MEO and so on.

Low Earth Orbit (LEO) satellite communication networks have become an important means to provide internet access services for areas with limited infrastructure. Compared with the Geostationary Earth Orbit (GEO) satellites, the LEO satellites have limited on-board communication caching and calculating resources. Furthermore, the distribution of traffic requests is dynamically changing and uneven due to the relative movement between the LEO satellites and the ground. Therefore, how to schedule the multi-dimensional resources is an important issue for the LEO satellite communication networks. Beam-hopping is an efficient approach to improve the resource utilization by dynamically allocating time, power, and frequency according to the traffic requests. This paper proposes an efficient multi-dimensional resource allocation mechanism for beam-hopping in LEO satellite networks, which simultaneously satisfies the GEO interference avoidance. First, we construct the beam-hopping model of LEO satellites, and formulate the resource optimization problem. Second, we provide the weighted greedy strategy to determine the illumination pattern. In order to reduce the search space, the cells are clustered to non-interference clusters. Then, an improved genetic algorithm is provided to jointly allocate the communication resources. Finally, we construct various simulations to evaluate our proposed mechanism. Compared with the random-BH, polling-BH and traditional genetic algorithm, our algorithm achieves better performance in terms of both system throughput, access success rate, average delay and fairness between cells. The performance improvement is more significant in scenarios where traffic demand is unevenly distributed.

The improvement of BDS Observation Geometry with LEO constellations in Orbit Determination

Current low earth orbit (LEO) satellite network display poor power efficiency, running network devices at full capacity all the time regardless of the traffic matrix and the distribution of the population over the Globe. Most of the research on energy efficiency of LEO satellites has focused on component level or link level. Therefore, this kind of research is not holistic to try to look at the satellite system as a single node. To enhance the energy efficiency. The solution should exploits multipath routing and load balancing. LEO network is overprovisioned, and hence selectively shutting down some satellite nodes and links during off-peaks hours seems like a good way to reduce energy consumption. In this paper, we exploit the fact that due to geographical and climatic conditions, some satellite links are expected to be loaded with data while others remain unused. Our approach is to power down satellite nodes and links during period of low traffic, while guaranteeing the connectivity and QoS. Finding the optimal solution is NP-problem and therefore, we explore in this work two heuristic algorithms. We evaluate our heuristics on a realistic LEO topology and real traffic matrices. Simulation results show that the power saving can be significant.

We study the lightpath provisioning problem of data relay services (DRSs) in a geostationary Earth orbit (GEO) satellite network. In DRS, a low Earth orbit (LEO) satellite user (e.g., remote sensing satellite) delivers the image data to a dedicated ground station (GS) through a GEO network. The LEO satellite needs to circle the Earth several rounds in order to fulfill the transmission, which results in it being connected to different GEO satellites at different times. This phenomenon reveals a fundamental difference between DRS and the services in a terrestrial network, where the source and destination nodes are typically fixed. The lightpath provisioning problem of DRSs is challenging, as we should jointly address routing, bandwidth assignment, and the time schedule of the lightpaths for each DRS. In this paper, we first devise the problem mathematically as a mixed integer linear program (MILP) and then propose a heuristic algorithm called data relay request provisioning as the MILP does not scale well. In the numerical analysis, we demonstrate the effectiveness of the heuristic algorithm and provide a sensitivity analysis on several parameters of our problem.

Today most operational Earth observing satellites reside in low Earth orbits (LEO) at less than 1,000 km altitude, and in geostationary Earth orbits (GEO) at ~35,800 km altitude. These orbits have been the venues of choice for observations, albeit for very different reasons. LEO provides high spatial resolution with low temporal resolution while GEO provides for low spatial resolution, but high temporal resolution. NOAA utilizes both venues for their environmental satellites. The NOAA Polar-orbiting Operational Environmental Satellites (POES) reside in LEO Sun synchronous orbits at approximately 830 km in altitude, as do the Defense Meteorological Satellite Program (DMSP) satellites of the Department of Defense. In the near future the POES and DMSP satellites will be merged into a new satellite system referred to as the National Polar-orbiting Operational Environmental Satellite System (NPOESS). The NOAA Geostationary Operational Environmental Satellite (GOES) system, as the name specifies, resides at the other preferred observational venue of GEO. The Jet Propulsion Laboratory (JPL), under contract to NOAA, has been studying the characteristics of medium Earth orbits (MEO), at altitudes between 1000 and 35,800 km, as an observation venue to answer the question as to whether MEO might capture the attributes of the two traditional venues. This on-going study initially focused on determining the optimal altitude for MEO observations, through numerous trade studies involving altitude, instrument complexity, coverage, radiation environment, data temporality, revisit time, data rates, downlink requirements and other parameters including cost and launch complexity. Once the optimal altitude of 10,400 km had been determined the study proceeded to explore single through multiple MEO satellite constellation performance capabilities using two instrument types, a visible through infrared (IR) imager and IR sounder as the satellites’ payload. The MEO performance capabilities were compared to comparable LEO and GEO satellite constellation capabilities. This portion of the study concluded that indeed for global coverage a constellation of satellites operating in the MEO venue could capture the attributes of those operating in the LEO and GEO venues. Three 8-satellite constellations configurations - Walker, ICO, and Equatorial-Polar (EP) - then were studied to develop more constellation coverage statistics including robustness to individual satellite failure. That study phase concluded that the EP constellation was superior to both the ICO and Walker configurations. The study is presently examining if, and to what extent, the equatorial portion of the EP constellation might provide substantive supplemental data to that collected by the NPOESS and GOES satellite constellations.

The rapid expansion of low Earth orbit (LEO) satellite networks to improve global internet access has raised concerns about their environmental impact compared to terrestrial infrastructure. This study conducts a panel lifecycle assessment (LCA) and econometric analysis to compare the carbon intensity of LEO satellite and terrestrial fibre-optic networks across 100 countries from 2005 to 2023. Using ISO 14040-compliant LCA, we find LEO systems emit 0.1–0.5 kg CO₂e/GB, significantly higher than terrestrial systems (0.05–0.20 kg CO₂e/GB), with these emissions generated primarily during the launch (40%) and operational (35%) phases. Fixed-effects regression reveals that internet usage increases CO₂ emissions per capita (β = 0.1435, p < 0.01), while the share of renewable energy reduces emissions (β = –95.1288, p < 0.05), with LEO penetration having only a minimal impact (β = 0.0012, p = 0.332). Advanced two-stage least squares, Arellano-Bond generalized method of moments, Moran’s I (I = 0.12, p = 0.08), and random forest analyses confirm these findings, highlighting renewables and internet usage as key drivers of emissions. The findings suggest that emissions reporting frameworks and renewable energy integration may play an important role in aligning digital connectivity expansion with climate goals. These results highlight the need for sustainable digital infrastructure to balance global connectivity with planetary boundaries.

Recent research indicates that self-similarity is a common phenomenon in wired network. However, there are no sufficient studies that focus on the characteristics in wireless network, especially in low earth orbit (LEO) satellite network. This paper presents a practicable model based on analysis of aggregate self-similar traffic and impact on performances of LEO satellite network, and then validate that the traffic in LEO satellite network is self-similar. Then, the quality of service (QoS) measures under self-similar traffic such as queue length and buffer size are discussed.

According to previous studies, the UDLs (user data links) or ISLs (intersatellite links) switch and failure can result in the packet reordering in Low Earth Orbit (LEO) satellite networks. The packet reordering can severely deteriorate the performance of transmission in LEO satellite networks, especially at very high levels of packet reordering. In order to improve the TCP (Transmission Control Protocol) performance in LEO satellite networks, this paper proposes a modified TCP variant named TCP-D-SACK, which is based on the TCP-SACK. Our results indicate that the new TCP variant can improve the throughput.

Reinforcement learning based dynamic distributed routing scheme for mega LEO satellite networks

The current constellation of the BeiDou Navigation Satellite System (BDS) consists of five geostationary earth orbit (GEO) satellites, five inclined geosynchronous satellite orbit (IGSO) satellites, and four medium earth orbit (MEO) satellites. The advantage of using GEO satellites to monitor the ionosphereis the almost motionless ionospheric pierce point (IPP), which is analyzed in comparison with the MEO and IGSO satellites. The results from the analysis of the observations using eight tracking sites indicate that the ionospheric total electron content (TEC) sequence derived from each GEO satellite at their respective fixed IPPs is always continuous. The precision of calculated vertical TEC (VTEC) using BDS B1/B2, B1/B3, and B2/B3 dual-frequency combinationsis compared and analyzed. The VTEC12 precision based on the B1/B2 dual-frequency measurements using the smoothed code and the raw code combination is 0.69 and 5.54 TECU, respectively, which is slightly higher than VTEC13 and much higher than VTEC23. Furthermore, the ionospheric monitoring results of site JFNG in the northern hemisphere, and CUT0 in the southern hemisphere during the period from 1 January to 31 December 2015 are presented and discussed briefly.