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 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.

With the emergence of parallel computing systems and distributed time-sensitive applications, it is urgent to provide statistical guarantees for age of information (AoI) in wireless cyber-physical systems (WCPS) with diverse parallelisms. However, most of the existing research on AoI have tended to focus on serial transmission, and the AoI performance of multi-stage parallel systems remains unclear. To help address these research gaps, in this work, we set out to investigate the age of event (AoE) violation probability in a three-stage WCPS with diverse parallelisms such as fork-join and split-merge. We analyze both transient and steady-state characteristics of AoE violation probability (AoEVP). Using these characteristics, we transform the AoEVP minimization problem into an equivalent minimization problem. Moreover, we develop a queuing model to capture the queue dynamics under the max-plus theory of stochastic network calculus (SNC) approach. Based on the max-plus model, we derive a closed-form Chernoff upper bound for the equivalent problem by applying the union bound and the Chernoff inequality. Furthermore, we characterize the service process for different parallelisms applicable to each stage. By solving the Chernoff upper bound with the service moment generation functions (MGFs), we obtain heuristic update period solutions for minimizing the AoEVP of three-stage WCPS. Simulation results validate our analysis and demonstrate that our heuristic update period solutions are near optimal for minimizing the AoEVP of three-stage WCPS with diverse parallelisms. Through the theoretical framework, we further provide insights into how such networks can be designed for better age performance.

Unmanned aerial vehicle (UAV) is expected to bring transformative improvements to the integrated sensing and communication (ISAC) systems, due to its high flexibility, high autonomy, large coverage and strong adaptability to various terrains. Sensory data is gathered by sensing UAVs (SUs) from the coverage area and then relayed to the corresponding fusion center UAVs (FCUs). Afterwards, terrestrial base stations receive the sensory data from FCUs in such air-ground networks. However, due to complex task execution environment and transmission environment, it is challenging to capture the network-layer performance of the sensory data transmission and evaluate the trade-off relationship between sensing and communication. In this work, we model and analyze the network-layer delay violation for an ISAC UAV network to address this challenge. Specifically, the UAV formation is distributed according to a Poisson cluster process (PCP). Then, the successful sensing probability is derived, with which the sensory data traffic can be captured. Under the sensory data flow, the delay violation probability is calculated for the two-stage sensory data transmission queue by exploiting stochastic network calculus (SNC). Furthermore, a delay minimization problem is proposed to reveal the trade-off relationship between sensing and communication under the power allocation strategy. Based on the long-term network-layer queue backlog evaluated, we are devoted to analyze the delay violation probability under an emergency that results in the antenna misalignment for one typical sensing UAV during a certain period. The steady-state and transient analysis for the ISAC UAV network not only illustrate the trade-off relationship between sensing and communication for the network, but also provide insights for on-demand power allocation, network deployment, control module provisioning and sensory data flow control under certain performance requirements.

Given the network and the time-triggered flow requests of a Time Sensitive Network (TSN), configuring the gate control lists (GCL) of IEEE 802.1Qbv for the ports of each node can be formed as a Job Shop Scheduling Problem, which is NP-hard. At present, most of the existing heuristic solutions for such problems consider scenarios where all given traffic flows can be scheduled. In order to solve the undetermined flow scheduling problem in scenarios no matter whether the flows can be scheduled or not, we propose to maximize the remaining time in conjunction with optimizing the network utilization instead of only minimizing the flowspan. Though the new problem is still NP-hard, it is a unified framework capable of covering general scenarios. On the basis of the new framework, we propose a novel Mixed initial population Genetic Algorithm (MGA) to solve the problem. Extensive simulation evaluation shows that MGA performs better and faster in different network scenarios while other methods prevails only in specific scenarios. This feature makes the method attractive in realistic TSN scheduling applications for in most cases it is hard for users to properly classifying the problem.

Wireless networks in next generation will provide users ubiquitous computing services with low delay by devices at the network edge, namely mobile edge computing (MEC). The intensive computation tasks can be partially offloaded to the MEC server via the wireless link and then processed through the MEC computation resources to cater for the delay demand. A parallel computation process is formed in the MEC network consists of local computation at MEC users (MUs) and MEC computation at MEC servers. However, the fluctuating wireless channel environment, changeable spatial distribution of MUs and the randomness of MEC servers’ locations make it hard to characterize and guarantee the end-to-end quality of service requirements. In this work, we are devoted to analyze and optimize the overall delay bound for MEC networks under two orthogonal frequency division multiple access (OFDMA) strategies via stochastic network calculus (SNC). Specifically, Poisson cluster process is utilized to capture the randomness of MEC servers’ and users’ spatial locations and to derive the Laplace transform of interference suffered by an MU of interest. The upper bounds for the delay violation probability of two OFDMA strategies are established by exploiting SNC with the Mellin transform of signal-to-interference ratio. Furthermore, we propose an optimal task offloading scheme by minimizing the overall delay, which balances the local computation delay and MEC delay.

Unmanned aerial vehicles (UAV) wireless communications have attracted great interests in 5G networks due to its high mobility, on-demand deployment and low cost. However, it arises new serious concerns about the malicious eavesdropping attacks against UAV communications. In this paper, we study the UAV transmission with a wiretap Rayleigh fading channel, over which UAVs transmit data to a target receiver in an unsafe environment with multiple eavesdroppers. A secure transmission scheme is proposed for satisfying various performance requirements including secrecy and transmission latency, considering the unavailability of wiretappers' instantaneous channel state information (CSI). In particular, secrecy performance is measured by the derivation of secure transmission probability (STP) by physical layer security (PLS) technique. A novel stochastic-network-calculus (SNC) approach is proposed to analyze the service capability of the wiretap channel and as well calculate the latency bounds, whilst obtaining the internal relationship between secrecy and latency. Simulation results verify the theoretical performance bounds, which provide a guidance for designing secure transmission strategies with various performance requirements.

At existing cloud computing environment, virtual machine allocation mechanisms have been actively applied to resources management for processing dynamic workload in large-scale data center. In this paper, we address the problem of how many virtual machines should be allocated to a server, so that a given percentile of the execution time of a job is bounded by a predefined value. We first consider the case of “dedicated CPU”, whereby we calculate the CPU allocated to a single VM. Subsequently, we extend the analysis to the case where the CPU is shared equally by different groups of VMs. In this case, we calculate how many VMs need to be allocated in each group.

Stochastic network calculus is the probabilistic extension of deterministic network calculus for offering stochastic performance guarantee in packet networks. However, it has been shown that the attempts to achieve a scalable stochastic performance bound only have limited success. In this paper, we propose a novel stochastic network calculus approach to perform scalability analysis for the end-to-end delay. With the assumption of independent arrival and services across multiple network nodes, a scalable delay bound is derived in linear way with its moment generating functions. After that, we investigate the tightness of the bound for further improving its scalability using Doob's maximal inequality on a suitable martingale construction. Compared with previous conclusions, numerical results validate that our approach improves the scalability of stochastic delay bound in terms of the linear scaling and tightness. The results help make an important step forward towards improving scalability for delay analysis of realistic SDN.

Stochastic network calculus is the probabilistic ex-tension of deterministic network calculus for offering stochastic performance guarantee in packet networks. However, it has been shown that the attempts to achieve a scalable stochastic performance bound only have limited success. In this paper, we propose a stochastic network calculus approach to perform scalability analysis for end-to-end delay. With statistical independence assumptions on arrival and services across multiple network nodes, we present a linear delay bound with its moment generating functions. For further improving the scalability of the bound, the tightness of the bound is investigated using Doob’s maximal inequality on a suitable martingale construction. Numerical results validate that our solution improve the scalability of stochastic delay bound in terms of the linear scaling and tightness. The results are applicable to general computer network to determine a path that meets the delay requirements.

In order to improve the utilization of the fading channel of the high-rate IEEE 802.11n wireless networks, an adaptive rate control and frame length adjustment scheme (i.e., ARCLA) is proposed in this paper. In ARCLA, the subframe error rate (SFER) of the aggregate MAC protocol data unit (A-MPDU) is used for the estimation of the link quality, and the channel air time is fairly allocated over the data transmissions based on the channel coherent time to ensure the uniformity of the link quality during one frame transmission. According to that, the data rate and the data frame length are adaptively adjusted in two-dimensions to fit the fluctuant channel condition. Furthermore, an analytical framework based on the semi-Markov process is formulated to evaluate the performance of our scheme. Simulations results verify the theoretical analysis and show the superiority of ARCLA in respect to throughput performance.