Mega satellite constellations (MSCs) based on low Earth orbit (LEO) satellites and inter-satellite links (ISLs) have become increasingly important due to the seamless coverage and high throughput. Unfortunately, the communication components of satellites are susceptible to radiation-induced single event upsets (SEUs), which lead to the failure of ISLs and the decline in network throughput. In this paper, we study the impact of SEUs on network throughput and propose MSC design algorithms to enhance the throughput. To mitigate the impact of SEUs, each satellite is equipped with low-cost mitigation techniques, under which ISLs experience different levels of impairment. Furthermore, we derive the expressions of network throughput and observe the mismatch between the traffic pattern and the network topology. Based on the expressions, we develop the MSC design algorithm to address the gap for throughput enhancement. Simulation results validate the accuracy of the theoretical results, and demonstrate that the proposed algorithm can effectively enhance the network throughput by 8.42% compared to the classical topology under the impact of SEUs.

Mega-constellation networks (MCNs) based on low Earth orbit (LEO) satellites have become increasingly important due to the high throughput and seamless coverage. However, due to single-event upsets (SEUs) caused by cosmic radiation, satellites will suffer failure, which deteriorates the network throughput. This paper aims to elucidate the relationship between network throughput and the impacts of SEUs. Taking into account the long-term impacts caused by SEUs, we present the availability of satellites based on the reliability theory. Furthermore, we model the effective data rate of inter-satellite links (ISLs) and find that the upper bound of network throughput $C \propto {\left( {\frac{{1 - {e^{ - \kappa {T_p}}}}}{{\kappa \left( {{T_p} + \gamma } \right)}}} \right)^2}\sqrt {{R_o}{R_h}} $, where κ denotes the SEU rate, and Tpis the scrubbing period for SEU mitigation. Roand Rhdenote the data rates of intra-plane ISLs and inter-plane ISLs, respectively. Consequently, the throughput decline caused by SEUs can be mitigated by adjusting the structure of MCNs. Guided by the throughput upper bound, we propose an optimal throughput constellation design algorithm (OTCDA) to enhance the network throughput considering the impact of SEUs. Experimental results illustrate that the proposed OTCDA can achieve the throughput that is only 6.49% lower than the upper bound.

Beam hopping (BH) has emerged as a critical solution for interference mitigation in mega-constellation satellite networks. Traditional ground-based centralized beam scheduling methods become infeasible in mega-constellations due to prohibitive computational complexity and inadequate responsiveness to bursty traffic demands. Given the non-convex and NP-hard nature of the multi-satellite BH optimization problem, we strategically decompose it into two subproblems. Hence the two-stage on-board BH method based on collaborative satellite clusters is proposed in this paper to address the challenges for efficient BH scheduling. The pre-activated cell selection stage is designed with a mechanism for dynamic updating of cell pre-activation probability to maximize the system throughput. In the cell-satellite matching stage, load balancing across satellites is achieved by minimizing inter-satellite load disparities. Simulation results show that the average throughput could be improved by over 10% compared to the baseline. Moreover, the difference in load between satellites is significantly reduced by 26.38%.

Low Earth orbit (LEO) mega-constellations equipped with laser inter-satellite links (LISLs) is an important part of future sixth generation (6G). However, how solar background noise affects LEO mega-constellations remains an open research topic. To this regard, this paper first derives the spatio-temporal distribution of affected LISLs in LEO mega-constellations at a specific moment, based on the characteristics of the impact, such as its location and duration. This distribution is then generalized to account for all moments during the Earth's rotation, considering the positional relationship between the LEO mega-constellations and the Sun. Additionally, we define two key metrics: the maximum number of affected LISLs (MNAL) and the affected duration ratio (ADR) to quantify the impact on the constellations. Several examples are presented to show that the MNAL decreases as the phase factor increases and increases with rising inclination. The ADR, on the other hand, increases with the phase factor, but initially increases and then decreases as the inclination rises. This work offers theoretical insights that can guide the design of future LEO mega-constellations.

Mega satellite networks consisting of hybrid orbit satellites play an important role in the sixth generation (6G) wireless networks. Location management (LM) can ensure service continuity for mobile users and is one of the key technologies in mega satellite networks. Moving satellites (e.g., LEO, MEO)1 and users require repeated location updates, which creates the challenge problem of significant LM overhead. In this paper, we propose a novel dynamic hierarchical LM scheme based on the virtual attachment point (VAP). The approach utilizes MEO satellites to provide LM, and divides the direct association between the user and the satellite into two independent steps, including the association between the user and the VAP, and the association between the satellite and the VAP. With this mechanism, the user’s location no longer needs to be updated due to satellite movement. Besides, a dynamic adaptive location area (LA) scheme is proposed to update the user’s location. The scheme can reduce the update frequency of high-speed mobile users, keep the number of paging satellites controllable, and not increase drastically with the constellation scale, thus reducing the paging overhead. The simulation results demonstrate that the proposed LM technology solution can reduce the total LM overhead by at least 40.6%.

In satellite networks, information can be transmitted either directly by inter-satellite links or the movement of satellites carrying. Consequently, how to quantify network capacity, considering both the carrying and transmission capability of satellites is crucial to the deployment of satellite networks. In this paper, we define the capacity of satellite networks consisting of both, and propose a strategy to exploit the mobile carrying of satellites under the constraint of service requirements. Then, we reveal the theoretical relationship between satellite carrying and the network capacity. The theoretical analysis and simulated results show that 1) satellite carrying can improve the network capacity when the service delay constraints could be released; 2) The capacity gain from satellite carrying is influenced by network parameters, such as orbital altitude, number of satellites, and storage capacity.

With the growing demand for communications in remote and dispersed areas, Internet-of-Remote Things (IoRT) networks with joint unmanned aerial vehicle (UAV) access and geostationary orbit (GEO) satellite backhaul hold great promise to provide sufficient access services to Internet-of-Things (IoT) users and devices. As the fundamental of the performance optimization of IoRT networks, the performance analysis sheds light on the relationship between the network performance (i.e., backlog, delay, and throughput) and access scale (i.e., the numbers of UAVs and UAV users). Aiming at the challenges brought by the complex network structure (i.e., two-level queuing network along with the converged traffic), we introduce the stochastic network calculus-based min-plus convolution and the leftover service to mathematically describe the complex structure. For the analytical challenges of the continuous-time arrival process and heterogeneous two-level link capacities, we innovatively prove their supermartingale features and further derive the closed-form expressions of the network backlog and delay bounds based on the martingale theory. To pursue higher throughput while guaranteeing delay performance, we formulate a mixed-integer optimization problem of the access scale that contains a nondifferentiable variable derived from a transcendental equation. For the tractability, we propose a three-directional iterative (TDI) algorithm to search the optimal solution of the optimization problem. Simulation results verify the tightness of our performance bounds in contrast to the standard bound and the effectiveness of the proposed algorithm.

Software defined network technique is a novel approach introduced to manage low earth orbit (LEO) small satellite networks. One important challenge is the allocation of the scarce virtualized satellite network resources in space environment. We devise a virtual network functions orchestration based model to implement the virtualized resources management for LEO satellite networks. This model is formulated as an integer linear programming (ILP) problem. Further, we propose a method combining Dantzig-Wolfe decomposition, column generation and branch-and-bound algorithm for the ILP problem to attain the optimal solution. Finally, simulation results demonstrate the effectiveness and efficiency of the proposed algorithm.

The large propagation delay of underwater acoustic signals significantly affects the throughput performance of underwater communication networks. While past research focused on mitigating the impact of large propagation delays, recent work has suggested exploiting large delays. In this paper, we consider an underwater linear unicast network which employs a time- division based scheduling strategy to exploit large propagation delays to improve network throughput. We assume the protocol model in a network with partially overlapping collision domains, where the transmission range is normalized as 1 and the interference range is an integer k. We systematically discuss the throughput of the linear networks with single traffic flow, showing that the average throughput of an N-node linear network with single traffic flow cannot exceed (N-1)/k. We then propose a general transmission scheduling strategy that can achieve the throughput upper bound and also give some examples of the optimal schedules.