1 October 2018 in IEEE Wireless Communications Letters
In this letter, we propose a transceiver resource allocation scheme based on the time-expanded graph (TEG) for the satellite networks. Due to the time-varying topology of the satellite networks, the design of the efficient transceiver resource allocation requires multiple nodes to make joint decision across time. To overcome this problem, the distribution of the transceiver resources is first depicted by the TEG. Then, we add some virtual nodes and edges in the TEG to represent the resources, and obtain the modified TEG, which can transform the nodes’ transceiver resources into the link constraints. Finally, the joint resource allocation will be implemented through solving the max-flow problem over the modified TEG. Numerical simulation results will also be presented to verify our proposed scheme.
20 April 2020 in IEEE Transactions on Vehicular Technology
Due to the capability of simultaneously delivering the same information to multiple users, the multicast transmission has been widely applied over the satellite networks. Furthermore, more and more information needs to ensure the transmission delay within a predefined bound. However, the satellite networks possess the time-varying characteristics and limited energy, which causes challenges for the low delay multicast design. In this paper, we investigate the energy-efficient delay-bounded (EEDB) multicast problem to accomplish the multicast transmission within the given delay bound, but to consume energy as less as possible. Specially, with the help of the time-expanded graph (TEG), we construct the multicast time-expanded graph (MTEG) through introducing some auxiliary vertices and arcs to accurately depict the across-time multicast transmission and the “wireless multicast advantage”. Then, we formulate the EEDB multicast problem as one combinatorial optimization and propose an iterative heuristic algorithm to achieve the suboptimal multicast scheme. Simulation results are also presented to verify our proposed algorithm.
29 May 2019 in IEEE Communications Letters
Non-orthogonal multiple access (NOMA) can dramatically improves the spectral efficiency, while the cell-edge user's performance will be fiercely degraded due to inter-cell interference in the multi-cell NOMA networks. In this letter, we invoke interference alignment (IA) to resist the performance loss of the cell-edge users and propose a cooperative transmission scheme for the multi-cell NOMA networks. Then, we formulate an optimization problem in terms of joint multi-cell power allocation to seek an optimal transmission scheme. To solve the challenging nonconvex problem, the tight lower bound function and the logarithmic change of the variables are employed to convert the prime problem into a convex optimization problem, consequently obtaining a suboptimal solution through iteratively updating the lower bound. Finally, we demonstrate the efficacy of the proposed scheme through the numerical simulations.
25 November 2025 in IEEE Communications Letters
The low cost, compact design, and flexible mission capabilities of microsatellites have accelerated the growth of the microsatellite industry. However, their limited weight and volume budgets constrain the number of available onboard transceivers, posing a fundamental challenge known as the Contact Plan Design (CPD) problem, which involves scheduling inter-satellite communications. While several approaches, such as MILP and heuristic methods, have been proposed to address the CPD problem, these methods either incur high computational complexity or suffer from suboptimal end-to-end delay, making them unsuitable for online, time-critical missions in microsatellite networks. To overcome these limitations, we propose an efficient graph-based online contact plan design scheme. We first construct a novel graph representation, referred to as the contact plan graph, that unifies the modeling of transceiver availability and constraints, as well as storage and communication resources in the satellite network. This formulation transforms the online CPD into a pathfinding problem, significantly reducing computational overhead. Building on this, a customized shortest-path algorithm is developed, tailored to transceiver constraints such as half-duplex operation and single transceiver limitation. Simulation results under realistic satellite network settings demonstrate that our method achieves substantial improvements of up to 29% in end-to-end delay and over $6400\times $ in runtime compared to state-of-the-art baselines.
18 June 2026 in Science China Information Sciences