Federated learning (FL) is a privacy-preserving technique that enables local model training on devices without raw data sharing. However, a critical challenge in FL lies in the communication requirement of uploading the trained models to servers, which can be hindered by interference from ambient devices, particularly in unreliable wireless environments. To address this, hierarchical FL (HFL) introduces an additional intermediate layer where the edge server performs work aggregation from the devices nearby, aiming at reducing the communication load and improving the efficiency of model training. However, existing approaches suffer from two critical limitations. First, they fail to fully quantify the impact of device competition-induced interference on transmission performance, which leads to unacceptably high upload latency and low success upload probability (SUP). Second, they lack a targeted optimization strategy to balance model accuracy and transmission efficiency under dynamic interference conditions. To address these critical limitations and mitigate their adverse impacts on FL performance, we take these gaps as the core motivation of our work and propose a targeted solution. Specifically, we first model the network as a two-layer binomial point process (BPP), which allows us to analyze the network-layer performance and calculate the SUP for the trained model. Based on this model, we propose optimizing cluster selection to balance accuracy and latency, thereby enhancing overall FL performance. We formulate this optimization as a Markov decision process (MDP) and solve it using a twin-delayed deep deterministic policy gradient (TD3)-based cluster selection algorithm (CS-TD3). In addition, to guarantee network-layer performance and enhance the efficiency of HFL, we employ an experimental exhaustive search algorithm to find the best solution within a limited range. The experimental results show that our algorithm overperforms other commonly-used algorithms in terms of HFL accuracy and model transmission latency, achieving a 10.95% improvement over the other methods.

Integrated sensing and communication (ISAC) is emerging as a key technology and research focus for future vehicular communication networks (VCN). It achieves efficient reuse of wireless infrastructure and spectrum resources through the collaborative design of sensing and communication functionalities. However, this integration leads to inevitable mutual interference, and the complexity of channel conditions further complicates the coordination of these functionalities. This paper primarily focuses on the joint sensing-communication performance analysis of ISAC-enabled VCN. Specifically, we model the spatial distribution of roads using a Poisson line process, while the locations of vehicles and roadside units (RSUs) are represented by two one-dimensional Poisson point processes. We characterize the dynamic interference distribution caused by RSUs during the sensing and communication phases and calculate the probability of successful perception (PSP) for a typical pair. Furthermore, for this typical pair, we meticulously derive the communication coverage probability based on the derived PSP for such a pair. To provide a detailed analysis of the interaction between communication and sensing functionalities, we evaluate their trade-off relationship and derive the joint probability of ISAC coverage. Moreover, we perform comprehensive simulations to verify the theoretical results. Additionally, the numerical results demonstrate how different parameters impact the network performance, providing guidance for network deployment and resource allocation under certain performance requirements.

Federated Learning (FL) is an approach that allows nodes with limited resources to collaborate without the need to share their data. However, the computational performance of local devices, the the non-independent and identically distributed (Non-IID) nature of data, and limited communication resources inevitably reduce the model's convergence speed and accuracy. It is urgent to obtain FL models with superior performance under Non-IID data. In this paper, we propose FedCGSU, a federated learning method for client grouping based on similar uncertainty in local client distributions. FedCGSU achieves grouping by leveraging the similarity in local distributions, which reduces the impact of weight divergence among clients with different distributions. It then combines loss values and local data volumes in the aggregation method, which reduces the impact of computationally weak devices on the global convergence speed. Experiments using three public datasets are conducted, demonstrating that the FedCGSU approach exhibits outstanding performance in improving accuracy and accelerating convergence speed.

Federated edge learning (FEL) has become a research hotspot to relieve the computational burden on servers and protect users’ data privacy. In an FEL system, adjusting the client selection and resource allocation scheme can reduce the energy consumption and improve the learning accuracy. However, obtaining a high-learning accuracy and low-energy consumption are primary challenges for FEL when the channel state information (CSI) is imperfect and the resources are dynamic. With this concern, to balance the learning accuracy and energy consumption, we formulate a joint client selection and dynamic resource allocation problem for FEL with imperfect CSI. The optimization problem is formulated as a Markov decision process (MDP) that defines the state space, action space and reward function. To cope with traditional optimization algorithms’ inefficiency in solving the formulated problem, a deep reinforcement learning (DRL)-assisted method is used. We use a Softmax deep double deterministic policy gradient (SD3) framework to train the model. Furthermore, a novel SD3-based FEL algorithm (FL-SD3) is proposed for client selection and dynamic resource allocation. Simulation results show that the proposed FL-SD3 improves the success rate by 11.6%, whilst the accuracy-to-energy (AE) gain is improved by 70.1% compared with some existing methods.

In this article, we investigate the wireless power transfer and energy-efficiency (EE) optimization problem for nano-empowered vehicular networks operating over the terahertz band. The nano-sensors in air can harvest energy from a power station and then can transmit the trace information to the micro-device under reconnaissance vehicular scenarios. Hence, by considering the properties of the terahertz band, we develop a long-term EE optimization problem. Furthermore, with the help of the equivalent transformation method, we converted the EE optimization problem into a series of energy-efficient resource allocation problems over the time slots. Each reformulated optimization problem becomes a mixed integer nonlinear programming (MINLP) over a time slot. Hence, to obtain the sub-optimal solution of the reformulated optimization problem, we developed a Quantum-behaved Particle swarm-based EE Optimization (QPEEO) algorithm. Furthermore, by exploiting the special structure of the reformulated problem, we propose an Improved Discrete Particle swarm-based EE Optimization (IDPEEO) algorithm. The proposed IDPEEO algorithm handles the problem's constraints effectively, and greatly reduces the search space and the convergence time. Our simulation results validate the theoretical analysis of the proposed scheme.

A Tactile Internet-based nanonetwork is an emerging field that promises a new range of e-health applications, in which human operators can efficiently operate and control devices at the nanoscale for remote-patient treatment. A haptic feedback is inevitable for establishing a link between the operator and unknown in-body environment. However, haptic communications over the terahertz band may incur significant path loss due to molecular absorption. In this paper, we propose an optimization framework for haptic communications over nanonetworks, in which in-body nanodevices transmit haptic information to an operator via the terahertz band. By considering the properties of the terahertz band, we employ Brownian motion to describe the mobility of the nanodevices and develop a time-variant terahertz channel model. Furthermore, based on the developed channel model, we construct a stochastic optimization problem for improving haptic communications under the constraints of system stability, energy consumption, and latency. To solve the formulated nonconvex stochastic problem, an improved time-varying particle swarm optimization algorithm is presented, which can deal with the constraints of the problem efficiently by reducing the convergence time significantly. The simulation results validate the theoretical analysis of the proposed system.

In this letter, we investigate the dynamic rate allocation and forwarding strategy adaption scheme for wireless networks with potential selfish nodes. Aided by an incentive mechanism, we develop a stochastic differential equation (SDE) to portray the dynamic node selfishness in terms of node's energy resource and incentives. Then, a stochastic optimization model is employed to maximize the average network utility while bounding the node selfishness. Based on the continuous-time Lyapunov optimization theory, we solve the optimization problem and propose a dynamic rate allocation and forwarding strategy update (DRAF) algorithm to accommodate the dynamic network state. We further analyze the tracking errors between the output of DRFA algorithm and the optimal solution. Then, an adaptive-compensation rate allocation and forwarding strategy update (ACRAF) algorithm is designed, which iterates only once when network state changes. Finally, we provide a sufficient condition that the ACRAF algorithm asymptotically tracks the moving equilibrium point with no tracking errors. Simulation results validate the theoretical analysis.