30 September 2024 in IEEE Transactions on Vehicular Technology
This paper investigates the routing problem in the mega low earth orbit (mLEO) satellite constellations considering factors including distribution of the users, topology of the networks and dynamics of the inter-satellite links (ISLs). In walker-delta constellations, each satellite establishes four stable ISLs with neighboring satellites to forward data packets without relying on ground facilities. In order to minimize the delay from the source satellite to the destination satellite, the routing problem is formulated as a Markov decision problem (MDP) and the GRLR routing algorithm is proposed which integrates reinforcement learning (RL) and graph neural network (GNN) effectively. The GRLR establishes the decision network based on Actor-Critic RL framework and builds feature extraction network based on GNN to realize distributed intelligent routing decisions in mLEO constellations. Finally, the simulation experiments are carried out to illustrate that the proposed strategy exhibits rapid convergence and outperforms the baseline strategies in terms of delay and adaptability to network dynamics.
10 September 2021 in IEEE Wireless Communications Letters
This letter proposes a satellite handover strategy for the huge low earth orbit (LEO) satellite constellation. As the increasing of the constellation scale and the number of user terminals (UTs), the satellite handover becomes more complicated. Thus, we formulate the network-flows model of the satellite handover, and the handover UTs can access the satellites according to the flow matrix. In the proposed network-flows model, the weighted edges are determined by the requests of UTs and the quality of satellite services. The multiple matching between the satellites and UTs can be determined by calculating the minimum cost and maximum flow of the network-flows. Besides, we propose a modified handover strategy based on network-flows (HSNF) algorithm to enhance the algorithm performance by preventing infinite loop. Finally, the simulations show the effectiveness and superiority of the proposed strategy.
20 June 2023 in IEEE Internet of Things Journal
We investigate a computing task scheduling problem in space-air-ground integrated network (SAGIN) for Internet of remote Things (IoRT). In the considered scenario, the unmanned aerial vehicles (UAVs) collect computing tasks from IoRT devices and make offloading decisions, in which the tasks can be computed at the UAVs or offloaded to the low earth orbital (LEO) satellite. The optimization objective is to design offloading strategies for each UAV that maximum the number of tasks satisfies delay constraint and reduces the energy consumption of UAVs. We formulate this problem as a nonlinear integer optimization problem, and remodel it as a stochastic game to solve it. To copy with the dynamic and complexity environment, we propose a learning-based orbital edge offloading (LOEF) approach which can coordinate all UAVs to learn the optimal offloading strategies. This method is based on the Actor-Critic framework of multi-agent reinforcement learning, the Actor observes the environment and output the current offloading decision, the Critic evaluates the action of the Actor and coordinate the actions of all UAV to increase the reward of system. The simulation results show that the proposed offloading strategy converges fast, increases the number of satisfying the delay constraints tasks and the energy utilization of UAVs compared with the baseline strategies.
29 July 2025 in IEEE Wireless Communications Letters
This letter addressed the routing selection problem in mega low Earth orbit (mLEO) satellite constellations, where the large-scale of satellite network and highly dynamic environment result in an expansive routing space and necessitate frequent adjustments based on real-time network conditions. To tackle these challenge, the routing selection problem is modeled as an integer optimization problem and subsequently reformulated as a Markov Decision Process (MDP). A federated graph reinforcement learning-based routing strategy (FGRLR) is proposed, which combines graph neural networks (GNNs) with traditional reinforcement learning to capture dynamic changes in satellite networks, and leverages a federated learning-based training approach to accommodate diverse user demands. Simulation results demonstrate that the proposed method significantly reduces the computational complexity while maintaining superior routing performance of users.
31 March 2025 in IEEE Internet of Things Journal
Mega-constellation networks face great challenges in providing broadband reliable communication service. The main obstacle is the highly dynamic network topology. This article focuses on this issue and makes an effort to examine the topology dynamic of mega-constellation from a new perspective. Some metrics derived from complex network theory are presented to describe topology characteristics. Considering the impact of the topology dynamic on routing, the topology dynamic characteristics are then expressed by a newly defined topology correlation of the two adjacent snapshots. Three well-known mega-constellations are used as examples and compared by using these metrics. Simulation results show that fewer nodes need to update routes with the same snapshot length when there is a greater correlation between snapshots. In addition, an isochronic snapshot topology control and routing update scheme (ISTC) is proposed to mitigate the problem of the surging number of snapshots due to the complex dynamics of mega-constellations. The upper and lower bound of the snapshot length are presented. So that, local convergence of routes can be achieved and the number of routing updates are as small as possible within each snapshot. Moreover, a decreasing distance shortest distance (DDSD) access satellite selection algorithm is presented to work with the topology control scheme and reduce the number of satellite handovers. Simulation results show that these methods have advantages in coping with the dynamics of constellation networks.
16 September 2024 in IEEE Transactions on Aerospace and Electronic Systems
In this paper, an interpolation-based blind carrier frequency-offset (CFO) estimator is proposed for faster-than-Nyquist (FTN) system in low-earth-orbit (LEO) satellite channel. According to maximum-likelihood criterion, we can obtain the CFO estimation problem formulation from the joint probability density function of output samples of matched filter. After simplifying, the CFO estimation problem can be equivalently solved by nonlinear least-square approach. Based on this, we have designed the objective function for CFO estimation of FTN. The physical meaning of designed objective function can be seen as the periodogram of the data samples. The peak location of the periodogram of the data samples is corresponding to the CFO estimate of FTN. Then, the objective function is proved to be approximated as a parabolic curve. By maximizing the designed objective function, an interpolation-based blind CFO estimator is proposed without iterated procedure. Finally, simulation results demonstrated that our proposed interpolation-based CFO estimator has lower computation complexity, and outperforms traditional estimators.