2026· IEEE Transactions on Wireless Communications· Vol 25, pp. 21529-21544· 0 citations· 42 references
Abstract
Satellite-airborne-terrestrial edge computing networks (SATECNs) emerge as a global solution for Internet of Things (IoT) since they can provide global coverage even in remote areas and under natural disasters. However, their dynamic and non-stationary nature makes control and resource allocation more challenging. Preserving data freshness is crucial in many IoT applications and requires timely decisions. To address these challenges, we present a knowledge-base software-defined networking architecture for satellite–airborne–terrestrial networks (KB-SAT-SDN) that enables collaboration between SDN controllers to optimize SAT configurations. A shared knowledge base (KB) is built through lifelong learning (LL) to continuously adapt and efficiently manage computing and networking resources to minimize the age of information (AoI) and energy consumption. To further accelerate learning, we exploit the heterogeneity of nodes and offloading decisions by defining different learning domains and designing a cross-domain lifelong learning (CDLL-SATECN) algorithm. With domain-specific projections, knowledge is shared between domains. Numerical results show that CDLL reduces average AoI and energy by up to 70% and converges $8\times $ faster than existing baselines. It achieves the lowest or near-lowest penalty across all RL domains, nearly halving Natural Actor-Critic (NAC)’s penalty in the most complex domains. LEO assistance lowers penalty/AoI from 61.9/49.4 to 48.5/45.1 relative to a domain without LEO, while reducing UAV energy and queues. The sensitivity analysis confirms that CDLL maintains a stable AoI–energy tradeoff over a broad range of weighting parameters.
A constrained optimization model that supports different management goals through alternative objective functions (latency-aware or power-aware) while enforcing operational constraints, including node capacities, slice-specific latency bounds, and explicit limits on VNF migrations/relocations between scheduling periods is proposed.
R. Moreno-Vozmediano, E. Huedo, R. Montero et al.· Journal of Network and Syste...· 0 citations
The Internet of Things (IoT) has grown rapidly in recent years, enabling the interconnection of a large number of heterogeneous and distributed devices. This number is expected to exceed 70 billion according to Statista. With this massive scale, fulfilling complex IoT applications that require combinations of multiple objects remains a real challenge. Moreover, several Quality of Service (QoS) requirements must be satisfied, making the problem of selecting appropriate IoT services NP-hard. In such environments, task offloading is a key mechanism to efficiently distribute computational workloads across edge, fog, and cloud resources. However, selecting the optimal offloading decision remains a difficult NP-hard problem due to system heterogeneity and conflicting objectives. In this paper, we propose a GNN-DQN-based approach for task offloading in edge–fog–cloud environments. Unlike prior GNN-DQN approaches limited to single- or dual-tier architectures, our framework explicitly models heterogeneous node types and inter-tier communication links, enabling more balanced and scalable resource allocation. Experimental results show that GNN-DQN achieves a mean latency of 2.64 s, representing improvements of 70.2% over Random, 7.8% over DQN-only, and 3.5% over Greedy. A GNN-A2C baseline is also included to broaden the comparison with a modern DRL method. Despite sharing the same GNN encoder, it underperforms GNN-DQN across all metrics, confirming the superiority of the DQN learning backbone. These results highlight the effectiveness of integrating graph-based representation with reinforcement learning, while also revealing a trade-off between latency optimization and energy efficiency.
Sirine Hakim, Sonia Yassa· International Conference on...· 0 citations
The growing demand for adaptive, resilient communication in software-defined networking (SDN)-enabled low Earth orbit (LEO) satellite networks underscores the importance of optimized SDN controller management. In particular, the overhead introduced by setup penalties and constraints on controller capacity significantly impacts key performance metrics, including satellite-to-controller delay, reassignment frequency, and the number of concurrently active controllers. These factors collectively influence the network’s ability to provide low-latency services while maintaining resource utilization in dynamic orbital environments. To investigate these dynamics, we develop a simulation framework based on OMNeT++ and INET, extending the open-source satellite simulator OS3 with enhanced satellite mobility and dynamic routing. The framework integrates SDN functionality into LEO satellite networks using the OpenFlow protocol, enabling centralized, programmable control with real-time adaptability. A central challenge in such networks is not only assigning satellites to appropriate controllers, but also dynamically activating and deactivating SDN controllers to match evolving topologies and traffic demands. To address this, we consider the baseline Dynamic Satellite-to-Controller Assignment (DSCA), the proposed Optimal Dynamic Satellite-to-Controller Assignment (Opt-DSCA), and their scalable heuristic variants (H-DSCA and H-Opt-DSCA). Opt-DSCA jointly minimizes satellite-to-controller delay and the number of active controllers, while incorporating setup penalties and activation costs to discourage unnecessary satellite migrations and redundant controller utilization. Simulation results reveal that both setup penalties and controller activation thresholds are critical determinants of system behavior across both optimization-based and heuristic schemes. Lower penalties enhance adaptability but result in more frequent reassignments and higher variability in controller state transitions. Conversely, higher penalties improve robustness by limiting controller switching, though at the cost of reduced flexibility and increased latency. These findings highlight the necessity of jointly optimizing reassignment policies and controller activation strategies to support robust, low-latency, and resource-aware SDN architectures for large-scale LEO satellite constellations.
Wafa Hasanain, Pablo G. Madoery, H. Yanikomeroglu et al.· IEEE Open Journal of the Com...· 0 citations
This article outlines the fundamental principles of the dual-layer OTA model and introduces the adaptive BH mechanism designed for time-varying topologies, aiming to provide insights for the evolution of ubiquitous non-terrestrial intelligence.
Zhendong Li, Shaojie Wang, Zhou Su et al.· 0 citations
The telecommunication industry in India is undergoing a structural change, with the deployment of the 5G network, the widespread deployment of smart devices for the Digital India initiative and the need for ultra-low latencies, high-speed services for smart cities, healthcare, agriculture and industrial automation. Traditional network management methods, which are mostly based on static configuration and human intervention are increasingly proving to be insufficient for the scale, diversity and dynamism of these networks. This paper introduces an Artificial Intelligence (AI) centric network management framework that combines 5G radio access and core capabilities, Software-Defined Networking (SDN) for centralized and programmable control and Multi-access Edge Computing (MEC) for localized and low latency processing. This framework includes a traffic prediction module (based on Long Short-Term Memory (LSTM)) and a resource orchestration engine (Deep Q-Network (DQN)), both of which allow for closed-loop self-optimizing network actions. Mininet-WiFi and Ryu SDN controller were used to simulate a representative Indian metropolitan network topology and assess the proposed framework in comparison to a baseline configuration with conventional SDN. Experimental results show significant gains in end-to-end latency (53.3 %), throughput (53.2 %), jitter (60.7 %), packet loss (72.4 %) and resource-utilization efficiency (42.6 %). The results indicate that 5G-SDN-edge continuum orchestration can positively impact the Quality of Service (QoS) in India-specific use cases such as dense urban areas and rural areas with limited infrastructure. The paper also elaborates the deployment challenges pertinent to the Indian context like spectrum availability, fiber backhaul penetration, cost of edge infrastructure, and regulatory considerations, while proposing the directions for future research like federated learning in orchestrating with privacy constraints and upcoming 6G research initiatives.
Praveen Kumar, S. Hashmi, Preeti Singhwal et al.· International journal of com...· 0 citations
With the rapid development of artificial intelligence and low-earth orbit (LEO) satellite edge computing technology, there has been a rapid increase in the demand for intelligence networks and services among users in remote areas, such as federated learning (FL). We propose a satellite aided computation FL (SACFL) system in LEO ubiquitous edge computing (UEC) networks, aiming at improving the efficiency of FL tasks in remote areas. In the considered deployment scenario, the following key factors are considered: 1) terrestrial users in remote areas; 2) offloading data to satellites for aided computation; and 3) global aggregation on satellite. However, the highly dynamic characteristics and the uneven distribution of satellite computation resources pose significant challenges to the low-delay requirements of satellite-based FL tasks. To this end, we formulate an optimization problem to minimize delay by jointly considering access selection, computation offloading, aggregation satellite (AgS) selection, and computation resource allocation. To solve the formulated problem, we propose a novel multi-agent alternating (M2A) optimization method. Specifically, three independent agents are trained alternately to make decisions on access selection, computation offloading, and AgS selection. Comprehensive simulations demonstrate that the proposed method outperforms other benchmark algorithms in terms of convergence, delay, and FL accuracy.
Junyi Yang, Yafeng Ma, Zhenyu Xiao et al.· IEEE Transactions on Cogniti...· 0 citations