Federated Learning (FL) enables the collaborative training of models by users distributed across various locations, transforming traditional data sharing into model sharing. This paradigm holds the promise of facilitating the development of safe, reliable, and accurate driving models within Internet of Vehicles (IoV), with its performance contingent upon the stability of the training process. However, traditional FL relies on a central server for aggregation, which is susceptible to malicious attacks. Moreover, limited communication resources prevent the inclusion of all users in the training process. To resolve issues related to reliability and resource utilization, this paper proposes a reliable Rotating Server Federated Learning (RSFL) algorithm to enhance the security and efficiency of FL. Specifically, we first consider the vehicular topology and participation in FL during their transition, and introduce a server rotation algorithm that incorporates a weighted sum of multiple factors including model training activity, vehicle credibility, speed stability, and distance to augment system security. Additionally, addressing the limitation of server channel resources that can impede FL efficiency, this paper proposes a method to select high-quality users for channel resource allocation by comprehensively considering participation latency, contribution, energy, and channel state during the FL process. This optimizes resource usage at the FL server side and constructs an efficiency-maximization problem for FL to improve the convergence rate. Simulation results confirm that the proposed RSFL algorithm can significantly enhance the security and system efficiency of FL.

In the Internet of Vehicles (IoV), centralized transmission of vehicle data poses significant privacy risks. Applying federated learning (FL) to vehicular networks enables collaborative model training while preserving data privacy. However, due to heterogeneous computational capabilities across devices, some users may drop out due to insufficient computing resources, resulting in slow convergence and degraded model accuracy. Furthermore, the traditional sequential transmission of model parameters creates communication bottlenecks and increases training latency, particularly when dealing with a large number of participating vehicles. Over-the-air computation (AirComp) provides an innovative solution by exploiting the natural super-position property of wireless channels to enable simultaneous transmission and aggregation of signals from multiple users, significantly reducing communication rounds and system latency compared to conventional sequential approaches. In order to solve the above problems, this paper combines AirComp with FL to form an adaptive OA-FL algorithm that addresses computing resource heterogeneity and improves communication efficiency through parallel signal processing. Specifically, we first adaptively adjust the number of local iterations based on the assessment of users’ available remaining computational resources to improve user participation rates. Subsequently, we formulate and solve an optimization problem to minimize model parameter distortion caused by AirComp aggregation, thereby reducing communication overhead while accelerating model aggregation and enhancing model accuracy. Simulation results demonstrate that the proposed OA-FL method effectively reduces client unavailability, improves FL training efficiency, and achieves faster convergence compared to conventional FL approaches.

Federated learning (FL), as a distributed machine learning paradigm, facilitates collaborative training without sharing raw data and holds promise for effective application in the Internet of Vehicles (IoV) for tasks, such as traffic flow prediction and driving behavior analysis. However, the efficiency of FL systems relies on the integrity of the local dataset and the level of user contribution. Vulnerabilities to attacks by malicious users and suboptimal aggregation methods can compromise system performance. To address these issues, this article proposes a blockchain-based FL secure aggregation algorithm to bolster FL robustness. Specifically, in the absence of a centralized trust authority in the IoV, we establish a hierarchical blockchain-empowered IoV reputation management framework that leverages smart contracts to create a trustworthy environment for reputation sharing. Additionally, a lightweight consensus protocol tailored for blockchain efficiency is proposed, thus facilitating a flexible and effective implementation of FL in the IoV. Furthermore, we introduce a reputation-based model selection evaluation scheme and, based on this, a robust FL secure aggregation algorithm. This novel reputation assessment strategy mitigates the effects of interaction uncertainties and integrates a broader spectrum of IoV-specific reputation determinants, thereby enhancing the precision of model selection. The simulation results validate the proposed framework’s superiority in terms of robustness, adaptability, and security.

With the rapid evolution of vehicular network technology, the integration of Machine Learning (ML) with Connected and Autonomous Vehicles (CAVs) presents both remarkable opportunities and formidable challenges. This paper addresses the crucial need for efficient ML model training in the context of Federated Learning (FL) within vehicular networks. Recognizing the limitations imposed by the tradeoff between the high energy cost at the local level with the performance problem at the global level, we propose an innovative approach that harmonizes cost optimization with strategic model selection. Our strategy primarily focuses on optimizing energy consumption during model training and updating at the vehicle end, thereby resolving the prevalent issue of limited end-user participation in FL due to high energy demands. Additionally, we introduce an advanced model selection method, prioritizing local model uploads and adaptively allocating bandwidth to clients with more extensive training data. This method enhances the efficiency and reliability of model updates, ensuring robust global model performance. We validate our approach through extensive simulations, demonstrating not only improved learning performance but also a significant reduction in energy consumption among participating clients.

To investigate the diversified applications in vehicular networks, artificial intelligence, intelligent edge computing, and vehicular networks are combined. By offloading computation tasks to devices close to vehicles, Vehicular Edge Computing (VEC) has emerged as a new computing paradigm to tackle the problem. Most existing VEC methods simply slice the application into subtasks for offloading purposes without considering the dependencies between subtasks. In practice, the dependency information is critical to the efficiency of offloading strategies. If a subtask requires the computation result of another subtask, the latter has to be processed before the former is finished. In this paper, we propose a deep reinforcement learning based offloading strategy for multi-vehicle collaboration VEC, with task dependency taken into account. With the proposed strategy, we formulate the offloading problem as an Markov Decision Process (MDP) and use the Sequence-to-Sequence (S2S) neural network to represent the policy/value function of the MDP. Furthermore, we train the S2S neural network to obtain the appropriate offloading policy using the Proximal Policy Optimization (PPO) technique. Our simulation results indicate that, by considering task dependencies during offloading, the proposed strategy outperforms existing methods in effectively reducing task offloading latencies.

The ever-growing Internet of Things (IoT) provides a powerful means for complex and changeable environmental monitoring. Directional sensor networks (DSNs), as a typical architecture of IoT, can efficiently facilitate various digital and intelligent IoT applications. In the DSNs, due to the asymmetry in coverage focus and diversity in detection angle of the directional IoT sensors, how to enhance the coverage performance with the limited sensors becomes a new challenge. To this end, we develop a novel sensor redeployment scheme based on the minimum exposure path (MEP) to optimize the coverage performance of the DSNs. Specifically, we first propose a minimum exposure path searching algorithm based on the particle swarm optimization (MEP-PSO) algorithm with the target of obtaining the MEP in the DSNs. With this algorithm, the traditional MEP problem can be analyzed and simplified by conducting the grid discretization and building the weighted undirected graph. Then, an MEP-based coverage optimization (MEP-CO) algorithm is proposed to determine the optimal deployment locations and the dispatch sensors so that the IoT sensors can be dynamically redeployed to achieve the coverage optimization. After that, we derive the formula for the coverage upper bound (CUB) and develop a CUB algorithm to provide a benchmark for evaluating the effectiveness of different coverage optimization algorithms. Simulation results demonstrate that the proposed coverage optimization scheme can significantly promote the minimum exposure value (MEV) and coverage ratio of the monitoring area compared with the existing algorithms.

The dissemination of traffic incident information (TII) will greatly help to decrease fuel consumption and congestion under future Internet of Vehicles (IoV) environments. Compared with the current semitargeted dissemination strategies of TII that focus on communication performance, we propose a complete targeted-dissemination strategy by jointly considering the impact of the dissemination on the route planning of connected vehicles and the communication performance of information dissemination in the IoV environment. This strategy further alleviates the considerable challenges caused by the increasing number of vehicles and limited radio resources in dissemination, while reducing the additional fuel consumption caused by excessive and meaningless dissemination by selectively distributing traffic information to connected vehicles. Specifically, the proposed strategy consists of a radio resource allocation strategy guaranteeing communication quality and a traffic-influencing targeted dissemination strategy selecting the targets. Simulation results validate the effectiveness of the proposed strategy in ensuring communication performance, reducing total cost, and decreasing the carbon dioxide emission rate.

Random access systems are potential for Internet of Things in the future wireless communication network for its operational simplicity. Irregular repetition slotted ALOHA (IRSA) system is one of the high-efficiency random access systems. In this paper, performance analysis of the irregular repetition slotted ALOHA systems with short-packet, i.e. finite-blocklength, transmission for the quasi-static Rayleigh fading channel is given. A cumulative distribution function of signal-to-interference power ratio (SIR) is derived and thus a closed-form expression of an average packet error probability (PEP) at the SIR with short packets for the Rayleigh fading channel is given. The closed-form expression makes it possible to optimize the degree distributions at a specific blocklength in the sense that the systems give the maximum system load.

Wireless body area networks (WBANs) can be widely used in wireless medical, motion detection etc. However, the existence of interference results in the increase in energy consumption and delay. To mitigate interference, nodes are prevented from using the same or similar spectrum resources by adopting multi‐channel media access control (MAC) protocols. Here, the authors propose a multi‐channel MAC protocol towards energy efficiency and low interference (EIMAC). Firstly, the states of each channel are clarified by the channel mapping mechanism. A novel channel selection strategy, considering the unfairness between high or low priorities, then is carried out. After that, considering the characteristics of node including residual energy, user priority, and data volume, the authors propose a low energy consumption enabled transmission mechanism. Lastly, the authors utilise a novel collision avoidance mechanism to reduce the collision probability of packets. Numerical results show that EIMAC significantly enhance the performance of WBANs in terms of delay, throughput, and energy consumption.

Due to the highly dynamic network feature, Vehicular Ad Hoc Networks (VANETs) suffer from the frequent link breakage and low packet delivery rate, which pose challenges to design routing protocols. To address this issue, we propose a Fuzzy Logic Routing Based on Forwarding (FLRBF) optimization scheme depending on receiving nodes for optimizing forwarding broadcast packets. We first calculate and record the distance factor and time delay from one source node to the destination node. Via the above information, we define a forwarding probability value for each node by the proposed fuzzy logic system based on high priority routing algorithm. Motivated by the defined forwarding probability values, nodes set timers and forward packets to achieve balancing broadcast efficiency, network throughput and average end-to-end delay. In the end, we conduct simulations to verify the performance of FLRBF. Results demonstrate that the proposed protocol performs well in terms of packet delivery ratio, end-to-end delay and overhead.

Resource management is a crucial task in vehicular ad hoc networks (VANETs) due to the existence of various resources, such as text, audio and video. However, the highly dynamic network feature and the limited memory of the local server pose challenges to resource management, which not only lead to the failure of resource presentations to users, but the transmission of the invalid fragment data would also result in the significant waste of precious bandwidth and memory. To address the issue, our paper proposes a Fuzzy Logic based Resource Management scheme (FLRM) under fog computing platform in VANETs. In the scheme, we first gather and record the request time and download time for each resource by the designed Vehicle to Infrastructure (V2I) communication mode. Depending on the above information, we define a survival time for each stored resource by the proposed fuzzy logic based popularity evaluation algorithm. Motivated by the defined survival time, the local server can update the resource list in real time. In the end, we conduct simulations to verify the performance of FLRM. Results demonstrate that the proposed scheme performs well in terms of throughput, which increases the user experience with fresh resources.

As a technical method over VANETs, video delivery has the potential power to enhance the application experience associated with traffic safety, management and infotainment. The experience of user is seriously affected by the quality of the video display. Therefore, Quality of Experience (QoE) enhancement for video delivery in VANETs is an important issue. However, the high mobility and dynamic nature of VANETs cause the dynamic topology which poses a significant challenge for video delivery. In this paper, we propose a user-oriented cluster-based solution called CDOV (Cluster and Dynamic Overlay based video delivery over VANETs), which combine the novel clustering algorithm and the dynamic overlay structure into a novel structure. Simulation results show that CDOV scheme for video delivery over VANETs significantly reduces the startup delay and increases the delivery rate compared with the gossiping-based scheme.

The internal collision managing mechanism in Enhanced Distributed Channel Access (EDCA) makes it more difficult to successfully access the channel for the data with low priority. To address this issue, a novel Mechanism to solve the Internal Collision problem (MIC) is proposed to improve the performance of EDCA. The main concept of MIC is to decrease the backoff time of the Access categories (ACs) with low priority. In this way, MIC can improve the performance of the low priorities without compromising that of the high priorities. A Markov chain model is used to analyze the performance of MIC under saturated conditions. The correctness of the analysis results are verified by simulating MIC with NS-2. Both the analysis results and simulation results confirm that MIC outperforms EDCA in terms of the average access delay.

Multiple input multiple output (MIMO) transceivers can potentially increase overall network throughput via spatial reuse of spectrum by allowing multiple simultaneous communications. Efficient use of MIMO technique in wireless ad hoc networks still meets challenges especially for upper layer protocols. In this paper we propose a cross-layer strategy of allowing the data stream and routing control stream transmitting concurrently with different antennas of the MIMO system. Implementing on the ad hoc on-demand distance vector (AODV) protocol, simulation results show that the strategy of queuing different type of packets obviously decreases the duration of route establishment, especially when the network is suffering heavy traffic load.

The traditional routing protocols for mobile ad hoc networks (MANET) are not appropriate for wireless sensor networks (WSNs). The reason is the large amount of overhead produced may exhaust the energy of the network and shorten its lifetime, especially in the process of maintaining routes. In this paper, considering a large scale WSN, we propose a simple but efficient AODV-based protocol, called AODV with Route Identification and RREP Capturing (AODV-RIRC). With route identification the source nodes don't have to build a new route. And after capturing the RREP the intermediate nodes build routes to the destination silently. The simulation results verify its efficiency in dealing with events in a large scale WSN which consists of 1000 nodes at most. According to the results, AODV-RIRC reduces the control overhead significantly, saves more energy and shortens the route building time relative to the other two AODV-based protocols.

Abstract The IEEE 802.11 wireless local area network (WLAN) media access control (MAC) specification is a hybrid protocol of random access and polling when both distributed coordination function (DCF) and point coordination function (PCF) are used. Data traffic is transmitted with the DCF, while voice transmission is carried out with the PCF. Based on the performance analysis of the MAC protocol for integrated data and voice transmission by simulation, this paper puts forward a self‐adaptive transmission scheme to support multi‐service over the IEEE 802.11 WLAN. The simulation results show that, on the premise of satisfying the maximum allowable delay of packet voice, the self‐adaptive transmission scheme can improve the data traffic performance and increase the WLAN capacity through dynamic and appropriate adjustment of the protocol parameters. Especially, voice traffic is sensitive to delay jitter, and the self‐adaptive scheme can effectively decrease it. Finally, it is worth noting that the adaptive scheme is easy to be realized, whereas no change in the MAC protocol is needed. Copyright © 2006 John Wiley & Sons, Ltd.

The use of smart antennas in extending coverage range and capacity of wireless LANs dictates the employment of novel media access control (MAC) schemes, with which the access point (AP) provides access to users by learning their spatial signatures. This paper puts forward an exact approach to analyze the performance of the schemes in terms of required time so that the AP becomes aware of user locations and grants access to the system. The numerical results show that the scheme which distinguishes the users residing in or out of broadcasting coverage range of the AP can provide rapider media access. In addition, the simulation results verify the theoretical approach.

This paper studies the performance of the IEEE 802.11 standard MAC protocol for integrated data and voice transmission with the DCF (distributed coordination function) and the PCF (point coordination function). By simulation, we evaluate the network performance for various protocol parameters, especially, the delay jitter for voice traffic. The main factor to influence delay jitter is given. Numerical results show that it is important to choose appropriate parameters and compromise the number of voice stations and the data traffic throughput to get the enhanced performance of IEEE 802.11. The performance of protocol in theory is derived and is verified by the simulation results.

The IEEE 802.11 standard MAC is a hybrid protocol of random access and polling when both DCF (distributed coordination function) and PCF (point coordination function) are used. On the base of the performance analysis of the MAC protocol for integrated data and voice transmission by simulation, this paper puts forward a selfadaptive transmission scheme to support multiservice over the IEEE 802.11 WLAN. The simulation results show that, on the premise of satisfying the maximum allowable delay of packet voice, the self-adaptive transmission scheme can improve the data traffic performance and increase the WLAN capacity through dynamic and appropriate adjustment of the protocol parameters. Especially, the scheme is easy to be realized for no change in the MAC protocol is needed.

The IEEE 802.11 standard MAC is a hybrid protocol of random access and polling when both DCF (distributed coordination function) and PCF (point coordination function) are used. This paper evaluates the performance of the MAC protocol for integrated data and voice transmission with the two access mechanisms. By simulation, we evaluate the network performance for various values of the protocol parameters. Especially, voice traffic is sensitive to delay jitter and here we point out the main factors to influence it. Numerical results show that it is important to choose appropriate parameters and we should compromise the number of the voice stations and the data traffic throughput to get the enhanced performance of IEEE 802.11. Finally, the performance of protocol in theory is derived and is verified by the simulation results.