Unmanned Aerial Vehicle (UAV) networks, renowned for their rapid deployment and high mobility, are well-suited for emergency communications that require swift responses to dynamic tasks. However, complex operational environments and frequent target movements cause service clusters to move dynamically, leading to constant changes in communication demands and service types, as well as highly dynamic distributions of service source and destination clusters, which can result in service interruptions and delays. To address these challenges, this paper proposes a dynamic task UAV network slicing architecture, comprising multiple UAV task clusters and a UAV physical infrastructure, designed to meet diverse service requirements. A slice resource reconfiguration model is developed to assess communication terminal issues caused by dynamic tasks, considering factors such as service interruptions, traffic rate, routing, node allocation, and spectrum allocation. The model aims to minimize service interruptions while considering constraints on limited spectrum, node resources, and flow capacities. To solve it, an algorithm based on the Deep Deterministic Policy Gradient (DDPG) is introduced. Simulation results show that this algorithm effectively reduces service interruptions, decreases reconfiguration needs, and enhances resource utilization in dynamic environments.

The novel notion of Edge Computing Power Networks (ECPN) has recently been proposed to provide highly flexible matching strategies among edge servers and user devices to facilitate seamless computing power and network architectures. However, recent work about ECPN is still in its infancy, and almost all work has ignored the tidal phenomenon of computing power requests from mobile user devices, including temporal characteristic of the quantity and spatial characteristic of the distribution in different periods of one day, which leads to the energy waste of idle edge servers for always keeping active. To deal with this problem, we propose the ECPN model in this article, taking into account the tidal phenomenon of mobile user devices to develop energy-efficient computing power matching strategies. Specifically, we formulate the optimization problem that encompasses computing power allocation and task matching for user devices as well as dynamic on-off for edge servers. Our main objective is to maximize the quality of service (QoS) for user devices while minimizing the energy consumption for task execution with respect to resource limitations and task requirements. To solve the formulated problem efficiently, we propose a distributed algorithm based on matching theory to determine the optimal computing power allocation, task matching and on-off strategies. Extensive numerical results show that the proposed scheme can reduce energy consumption while ensuring the QoS.

The sixth-generation (6 G) networks aim to achieve ubiquitous intelligent connectivity while ensuring extremely low latency, reducing energy consumption, and enhancing privacy protection. Mobile edge computing (MEC) offers an effective solution to reduce latency and energy consumption by leveraging resources near end devices for task offloading. However, MEC faces significant challenges in meeting the requirements of 6 G networks, including limited computational resources, high mobility, and strict data privacy demands. Efficiently allocating edge resources while preserving privacy has become a critical issue for realizing the objectives of 6 G networks. In this paper, we propose a privacy-preserving edge computing power network (EdgeCPN) model that jointly leverages the computing resources of edge computing nodes and protects sensitive computing power information through differential privacy methods. In addition, we propose a task matching problem that aims to minimize the privacy-budget-weighted energy consumption while ensuring privacy protection and meeting task requirements. We propose a dynamic graph-based multiagent reinforcement learning (MADRL) algorithm to find the optimal strategy for task matching and computing resource allocation with privacy protection. The results show that our proposed task matching model with energy and privacy tradeoffs can minimize the energy consumption in the matching process while ensuring privacy, and the algorithm can find the optimal strategy for task matching efficiently.

Aerial edge computing (AEC) networks, which employ multiple unmanned aerial vehicles (UAVs) as mobile edge computing servers, have emerged as a promising solution to provide computation offloading services, especially in scenarios where the coverage of existing infrastructures is limited for wireless networks. Recently, there has been a growing focus on leveraging UAVs with diverse computing capabilities to enhance the performance of AEC networks through cooperative computing. However, heterogeneity and cooperation introduce a higher degree of coupling between trajectory planning and computing offloading strategy for AEC networks. In particular, the joint decision-making in an AEC network need to balance minimizing the distance between UAVs and access users and enabling collaborative offloading. And this must be done while considering the time-varying computation requirements and the long-term impact on system performance. To address the aforementioned challenges, we formulate an optimization problem to design a joint dynamic cooperative computation offloading and trajectory optimization scheme for the AEC network. The complexity arises from the problem’s nature as a mixed-integer nonlinear program. To tackle this challenge, we propose a multi-agent deep reinforcement learning algorithm based on QMIX. We leverage both theoretical analysis and an action branching architecture to reduce the complexity of our proposed deep reinforcement algorithm. Simulation results demonstrate a substantial performance improvement over the benchmarks, affirming the effectiveness of our complexity reduction approach.

Federated Learning (FL) is an emerging distributed learning architecture that allows multiple agents to share knowledge in machine learning. However, in the scenario with mobile agents, the mobility of agents significantly affects the learning performance. Besides, frequent exchange of local models could result in high communication overhead. In this paper, we propose a Mobility-aware Federated Reinforcement Learning (MFRL) framework. In MFRL, we model the influences of agent mobility on communication quality and data correlation, and devise a mobility-aware node selection algorithm, so as to accelerate the training procedure and improve the learning performance, taking learning quality, wireless channel quality, and data correlation of agents into consideration. A knowledge distillation (KD) based model compression method is integrated into the MFRL to reduce the communication overhead as well as accelerate the inference process. Finally, taking deep reinforcement learning (DRL) based collision avoidance of intelligent vehicles as a study case, the effectiveness of MFRL is verified. Numerical results demonstrate that the proposed MFRL can accelerate the training process and improve the learning performance.

The rapidly growing number of Internet-of-Things (IoT) devices poses new challenges for spectrum management in future wireless communication networks. It is critical to achieve efficient and dynamic spectrum management in the sixth-generation (6G) wireless communication networks era. To tackle the challenges of managing a large-scale IoT network with heterogeneous devices, we propose a directed acyclic graph (DAG) blockchain-enhanced user-autonomy spectrum sharing model. As the proposed consensus rule is closely related to system utility, the swarm intelligence of users gradually reaches the point of convergence in the process of blockchain consensus. We analyze the effect of the tip selection method of the DAG blockchain on spectrum allocation utility. A dynamic tip selection method is proposed to enhance the global utility, which is related to the spectrum supply–demand. In addition, the ring signature technique is utilized to realize privacy protection during the sharing process. Simulation indicates that the proposed tip selection method achieves a 10% enhancement in terms of the global utility. Furthermore, significant reductions in administrative expense and reliability improvement are demonstrated by simulation results. The stability of the tip number in the proposed model has been proved theoretically, which is also validated by simulation experiments.

The emergence of Intelligent Connected Vehicles (ICVs) shows great potential for future intelligent traffic systems, enhancing both traffic safety and road efficiency. However, the ICVs relying on data driven perception and driving models face many challenges, such as the lack of comprehensive knowledge to deal with complicated driving context. In this paper, we investigate cooperative knowledge sharing for ICVs. We propose a secure and efficient blockchain based knowledge sharing framework, wherein a distributed learning based scheme is utilized to enhance the efficiency of knowledge sharing and a directed acyclic graph (DAG) system is designed to guarantee the security of shared learning models. To cater for the time-intense demand of highly dynamic vehicular networks, a lightweight DAG is designed to reduce the operation latency in terms of fast consensus and authentication. Moreover, to further enhance model accuracy as well as minimizing bandwidth consumption, an adaptive asynchronous distributed learning (ADL) based scheme is proposed for model uploading and downloading. Experiment results show that the DAG based framework is lightweight and secure, which reduces both chosen and confirmation delay as well as resisting malicious attacks. In addition, the proposed adaptive ADL scheme enhances driving safety related performance compared to several existing algorithms.

Energy saving for mobile devices is considered to be one of prospective benefits of mobile edge computing (MEC) networks, where computation-intensive tasks can be offloaded from the mobile devices to their associated MEC servers for execution. Extra energy consumption for data migration should therefore be less than the energy consumption for local execution. However, in multi-cell MEC-assisted networks, due to both the presence of co-channel interference and the latency requirement of each offloading task, power control is tightly coupled with computation offloading, which becomes an obstacle to achieve the aim of energy saving. In this paper, we develop an analytic model to decouple power control and computation resource allocation from each other, in which the transmission power can be considered as a solution to a set of linear equations with a coefficient matrix depending on the computation resource budget. Based on this analytic foundation, we show that with a fixed offloading decision, the joint power control and computation resource allocation problem is invex, which ensures that every KKT (Karush–Kuhn–Tucker) stationary point of the problem must be a global minimizer. Moreover, we deduce a criterion for energy-efficient offloading decision making from the partial derivative of the total energy consumption of mobile devices with respect to the computation resource budget. Finally, we propose a heuristic algorithms to jointly optimizing power and computation resource allocation, and offloading decision. The numerical results demonstrate the optimality and efficiency of our proposed algorithm.

Knowledge sharing in IoV shows great potential for future vehicular networks. Vehicles, platoons and even traffic infrastructures can exchange the driving experiences or sensing data to facilitate intelligent transportation applications such as autodriving and traffic analysis. However, it is challenging for vehicular knowledge-sharing systems to address the issues brought by information security and vehicular mobility. Although blockchain technology shows defensibility in dealing with trust issues, it is difficult to be applied in large-scale vehicular networks due to the computation consumption of mining process and frequent synchronization of ledger. In this paper, we propose a directed acyclic graph (DAG) enabled knowledge-sharing framework in which vehicular knowledge is encapsulated as a site in the DAG. A new tip selection algorithm (TSA) and a fast authentication scheme for cross-regional vehicles are designed to reduce computation and storage expenditure. Simulation results show that the proposed DAG framework can achieve a higher knowledge sharing quality and lower authentication latency compared with traditional DAG systems.

With the rapid growth of the network applications, more services with diverse QoS requirements have emerged. Efficient routing technique plays a vital role in supporting the diversified serveries in dynamically changing wireless multi-hop networks. To this end, we propose the reinforcement learning empowered QoS-aware adaptive Q-routing (RL-QAQ) algorithm, so as to provide discriminated transmission for different services with various QoS requirements as well as reduce the delivery delay and routing overhead. In the proposed RL-QAQ algorithm, an adaptive probability is devised to optimize the exploration strategy to reduce the overhead of acquiring the network status. Besides, the QoS-aware reward function and Q-tables for the different services are devised to support multi-QoS transmission requirements. Simulation results demonstrate that the proposed RL-QAQ algorithm can adaptively adjust the routing policy according to the varying network environment to meet the transmission requirements of different services with low delivery delay and routing overhead.

Intelligent Transportation System has emerged as a promising paradigm providing efficient traffic management while enabling innovative transport services. The implementation of ITS always demands intensive computation processing under strict delay constraints. Machine Learning empowered Mobile Edge Computing (MEC), which brings intelligent computing service to the proximity of smart vehicles, is a potential approach to meet the processing demands. However, directly offloading and calculating these computation tasks in MEC servers may seriously impair the privacy of end users. To address this problem, we leverage federated learning in MEC empowered internet of vehicles to protect task data privacy. Moreover, we propose optimized learning cooperation schemes, which adaptively take smart vehicles and road side units to act as learning agents, and significantly reduce the learning costs in task execution. Numerical results demonstrate the effectiveness of our schemes.

Internet-of-Vehicles (IoV) has the potentials of enhancing automatic driving in various transportation environment. However, there is very little investigation on quantifying the potential influence of automatic driving applications with the road efficiency in IoV. This paper studies the connection of safety distance to the road congestion under different IoV resource conditions. We propose an elastic wave equation model to reveal the relation between safety distance and road congestion. It can be found that the propagation speed of road congestion is largely affected by the safety distance. To recoup the efficient road safety and alleviate road congestion, an optimization problem is formulated with cooperative communication and computing via platoons that aims to minimize the total safety distance. Since the optimization is a complicated 0-1 programming problem, we propose a practical resource allocation algorithm and solve the problem through Lagrangian relaxation. Simulation experiments show that the proposed algorithm leads to near-optimal results with low complexity but no overhead of vehicular information exchange.

Road accidents and traffic congestion are two critical problems for global transport systems. Connected vehicles (CV) and automated vehicles (AV) are among the most heavily researched and promising automotive technologies to reduce road accidents and improve road efficiency. However, both AV and CV technologies have inherent shortcomings, for example, line of sight sensing limitation of AV sensors and the dependency of high penetration rate for CVs. In this paper we present a cooperative connected intelligent vehicles (CAV) framework. It is motivated by the observation that vehicles are increasingly intelligent with various levels of autonomous functionalities. The vehicles intelligence is boosted by more sensing and computing resources. These sensor and computing resources of CAV vehicles and the transport infrastructure could be shared and exploited. With resource sharing and cooperation CAVs can have comprehensive perception of driving environments, and novel cooperative applications can be developed to improve road safety and efficiency (RSE). The key feature of the cooperative CAV system is the cooperation within and across the key players in the road transport systems and across system layers. For example, the various levels of cooperation include cooperative sensing, cooperative RSE applications and cooperation among the vehicles and among the vehicles and infrastructure. We will present the potentials that could be brought by cooperative CAV, the roadmap for research and development, the preliminary research results and open issues.

This paper studies the problem of optimal power allocation towards maximizing the throughput of point-to-point wireless communication systems with energy harvesting. A novel energy-transferring approach is proposed to analyze the throughput maximization problem with causality constraints, in which we study the transfer energy rather than the water level widely used in the existing literature. The proposed approach simplifies the power allocation as a linear function with respect to only two transfer energy variables, i.e., the energy transferred from the previous epoch and the energy transferred to the next epoch. Moreover, we prove that all the positive transfer energy variables can be determined by solving a set of linear equations with a special coefficient matrix derived from the KKT conditions for the dual problem. Based on the energy-transferring approach, we propose an iterative algorithm to obtain the optimal solution with a much lower complexity compared to those existing algorithms based on the directional water-filling structure results. Numerical studies verify the analytical results as well as the effectiveness of the proposed algorithm.

The mixture of users with different carrier aggregation (CA) capabilities presents new challenges to optimize the performance of the next generation wireless networks. This paper focuses on the joint resources allocation for OFDMA-based multi-carrier system. Distinguished from many existing methods, our approach deploys the joint dynamic spectrum resources assignment and adaptive power allocation technologies for the carrier-aggregated systems. A low complexity suboptimal algorithm, named as the joint CC, RB and power allocation (JCRPA) algorithm, is proposed in this paper. The algorithm combines the dynamic component carrier (CC) and resource block (RB) assignment, as well as adaptive power allocation iteratively. In contrast to the conventional static CC assignment, a novel suboptimal dynamic CC and RB assignment algorithm is designed to maximize network utility. Simulation results demonstrate that JCRPA is able to improve the system performance in terms of network utility, average throughput and fairness.

The mixture of users with different bandwidth capability presents new challenges to optimize the transmission performance of the next generation wireless networks. This paper focuses on resource allocation for pervasive wireless networks with component carrier (CC) aggregation. Distinguished from many existing methods that decompose the resource optimization problem into two sequence steps, i.e., CC scheduling and resource block (RB) assignment on each carrier, we develop a novel joint CC and RB allocation algorithm to maximize network utility, namely Minimizing System Utility Loss (MSUL) algorithm. Numerical results indicate that MSUL is able to improve the system performance in terms of network utility, average throughput and fairness compared with the algorithms optimizing CCs and RBs allocation separately.