Abstracts:This index covers all papers that appeared in JCN during 2025. The Author Index contains the primary entry for eachitem, listed under the first author's name, and cross-references from all coauthors. The Title Index contains paper titles for each Division in the alphabetical order from No. 1 to No. 6. Please refer to the primary entry in the Author Index for the exact title, coauthors, and comments / corrections.

Abstracts:Video streaming has become one of the primary contributors to global Internet traffic, posing significant challenges to network infrastructures. Software-defined networking (SDN) offers a promising solution for managing such dynamic and bandwidth intensive services by enabling centralized control and realtime adaptability. However, decoupled decision making fails to account for the interplay between server workload and link congestion, often leading to suboptimal resource allocation. To address this issue, this paper presents a graph neural network (GNN)-based client-server-link (G-CSL) prediction model designed to optimize video streaming performance in SDN environments. G-CSL utilizes a machine learning model in conjunction with a GNN-based link estimation model. The machine learning predicts the video streaming server CPU utilization, which serves as input to the GNN model as node embeddings for link estimation between the client and server. For load forecasting, two machine learning and two deep learning models are evaluated, with random forest (RF) outperforming its counterpart. For the link estimation task, both traditional and GNN-based models are considered. GraphSAGE outperforms its counterparts by accurately estimating the existence of a link between the client and the video streaming server. A lightweight neighbor score heuristic then assigns each request to the least loaded server over the highest confidence path, maximizing a composite utility of computational headroom and bandwidth. An ablation study of the GraphSAGE model is presented highlighting the importance of architectural components, including batch normalization, bilinear decoders, temporal features, and threshold-based edge masking, in enhancing model robustness. The proposed model is evaluated under realistic video streaming scenarios involving 10,000 requests and compared with baselines. Experimental results show that G-CSL has achieved a 61% reduction in request drop rate, maintains an average delay of 22 ms per request, and improves system utility by 23%, demonstrating its effectiveness in balancing computational and bandwidth resources.

Abstracts:In integrated sensing and communication (ISAC) system, achieving signal-level integration is essential but challenging, as direct amalgamation of communication and sensing signals confronts inherent heterogeneity, engendering substantial complexity in both software and hardware. This paper presents a strategy for enhancing ISAC systems by implementing a semantic-level slicing approach to address key issues as resource utilization, real-time capability, and computational complexity in ISAC systems. A bidirectional mapping mechanism is introduced, combined with edge intelligence, that enhances system predictability, autonomy, and reduces task completion costs by slicing and allocating semantic resources in cloud-edge environments. Additionally, the proposed communication-sensing complementary strategy leverages semantic fusion to enable efficient and adaptable execution of communication and sensing tasks at the resource level, resulting in enhanced task flexibility. The simulation results show that the proposed bidirectional mapping and communication-sensing complementary method significantly improves the signal-to-interference-noise ratio of the traditional system by about 55% in medium and low dynamic environments.

Abstracts:The performance of Internet services heavily relies on efficient traffic management mechanisms. This paper investigates the impact of imposing a local bottleneck at the access networks' side compared to the conventional approach of traffic shaping by Internet service providers (ISPs) through first-in-firstout (FIFO) queues. The proposed local bottleneck strategy aims to enhance Internet performance by reducing latency, mitigating congestion, isolating distinct traffic flows, and enabling quality of service (QoS) differentiation. Through extensive experimentation on various home and office network setups, including fiber to the cabinet (FTTC), fiber to the home (FTTH), and fixed wireless access (FWA), we demonstrate the efficacy of the local bottleneck approach in delivering consistent and high-quality Internet performance, especially in congested environments. The results reveal that the traditional ISP bottleneck struggles to maintain a high-performance standard under congested conditions, highlighting the need for innovative traffic management techniques.

Abstracts:The global expansion of Internet connectivity is a well-documented trend, especially in urban and developed regions. However, many rural and low-income populations still face limited or no Internet access, where the absence of Internet connectivity impedes crucial services such as remote healthcare, emergency assistance, distance learning, and personal commu nication. In this context, the main challenge for the research community is to expand digital coverage in rural areas, thus providing a better quality of life and service in the area. Existing solutions often focus on optimizing isolated metrics (such as latency, energy consumption, or throughput) limiting their flexibility and real-world applicability, which reduces their applicability and flexibility, respectively. This paper addresses this digital divide by proposing an innovative strategy using unmanned aerial vehicles (UAVs) to create a network and deliver digital services to remote rural areas without Internet access. The strategy involves breaking down Internet of things (IoT) applications into microservices and deploying them via UAVs. This approach achieves both reduced latency and lower energy consumption, improving the quality of service for latency-sensitive applications. The focus is particularly crucial for applications with stringent requirements, such as those related to remote health care or emergency services. Simulations conducted in a realistic scenario validate the efficacy of the proposed solution. The results showcase a notable reduction in energy consumption and latency associated with UAVs while handling such requests.

Abstracts:Benefit to the advantages of flexibility and low-cost deployment, employing unmanned aerial vehicle (UAV) as the aerial base station (ABS) constitutes a promising technology to support the multi-user access. However, facing the challenges of both mobile ABS and ever-increasing users, it is crucial to design the UAV trajectory in the dynamic environment, in order to providing communications services for multiple mobile users with fair consideration. Therefore, in this paper, we propose a fairly non-orthogonal multiple access (FNOMA) scheme for UAV communications network, which enables the ground mobile users to be accessed to the ABS by sharing the same spectrum resources with high fairness. For the sake of optimizing the attained system throughput, a novel greedy genetic algorithm assisted Q-learning (GGA-Q) method is conceived, where the UAV trajectory is elaborately designed by jointly considering dynamic user grouping and power allocation. Our simulation results indicate that the proposed UAV trajectory planning algorithm based on FNOMA scheme can significantly improve the fairness level while enhancing system throughput.

Abstracts:Effective integration of available resources within edge nodes is essential to improve the performance of vehicular edge computing (VEC) to support various randomly offloaded tasks with limited computing capacity and constrained energy. This paper presents an intelligent adaptive resource integration strategy for VEC with energy harvesting. Service caching, task migration and resource allocation are jointly employed to accommodate the temporally and spatially varying computing demands. The optimization to minimize the long-term average task execution time under energy constraint is formulated as Markov decision processes and solved with a parameterized deep Q-network based learning algorithm. This algorithm employs a centralized training and distributed execution framework, where a parameter network and an action network respectively handle continuous and discrete decisions, effectively tackling the hybrid action space challenges in problem solving. Simulation results demonstrate that the proposed algorithm not only achieves faster convergence but also significantly improves system performance compared to benchmarks.

Abstracts:The increasing demand for massive device access in the Internet of things (IoT) necessitates the use of non-orthogonal multiple access (NOMA) technology and grant-free (GF) transmission to enhance spectrum efficiency. Semi-grant-free (SGF) transmission has garnered significant research attention in recent years because it enables grant-free (GF) users to share resource blocks with grant-based (GB) users, thereby enhancing spectrum efficiency in large-scale device access scenarios. This study focuses on distributed GF transmission and integrates it with the sparse code multiple access (SCMA) scheme to address the low-latency needs of large-scale short-packet transmission devices while simultaneously meeting the transmission rate requirements of GB users. The proposed SGF-SCMA scheme, based on a user hierarchical strategy, significantly enhances spectral efficiency and reduces delay overhead for GF users compared to existing schemes. The successful transmission probability of SGF-SCMA scheme is derived, and the proposed control transmission factor is optimized to improve the throughput in massive user access scenarios. The solution of the derived successful transmission probability was showed through simulation, and the crosstalk problem of the SCMA system was analyzed. Considering the delay of GF users, the throughput can be improved by 20% to 43% compared with the existing SGF scheme.

Abstracts:In recent years, symbiotic radio systems have garnered significant attention from academia and industry for addressing the challenges of spectrum scarcity and energy consumption in large-scale Internet of Things (IoT) deployments. At the same time, due to the wide distribution of IoT devices, some remote areas cannot be covered by mobile networks. Introducing unmanned aerial vehicles (UAV) into wireless networks can improve network coverage performance and increase spectrum utilization. Therefore, this paper proposes a joint optimization algorithm for channel allocation, reflection coefficients and UAV's position for a UAV-assisted symbiotic radio system consisting of multiple primary users (PUs) and multiple backscatter devices (BDs). Under the constraints of energy and the quality of service (QoS) of the primary transmission system, the optimization problem of sum rate maximization in the backscatter communication system is constructed. The Kuhn-Munkres (KM) algorithm is used to solve the channel optimal matching problem. Based on the block coordinate descent (BCD) algorithm, the non-convex problem is decomposed into three subproblems: transmission power, BDs' reflection coefficients and UAV's position. The transmission power subproblem is solved in two cases where the number of BDs is less than/greater than the number of PUs, and the expression of the optimal solution of the reflection coefficient is derived. The Nelder-Mead algorithm is used to solve the UAV's position subproblem. Finally, the global optimal solution is obtained through global iteration. Simulation results demonstrate that the proposed algorithm achieves strong convergence and significantly enhances the sum rate of backscatter communication in UAV-assisted symbiotic radio systems.