Ad Hoc Networks最新文献

Latency is a critical aspect for a broad spectrum of applications that relies on the internet, such as, voice over IP (VoIP) or teleconferencing, and the lack of ultra-fast and highly reliable communications is prominent in rural areas even in mature economies. Our proposal focuses on optimizing the deployment of microservice-oriented architectures (MSA) in computing and routing enabled unmanned aerial vehicles (UAVs). For that matter, an information system which gathers all the information of the flying ad hoc network (FANET) is developed. From there, we propose multiple approaches, based on integer linear programming (ILP) and heuristics, to tackle the minimization of end-to-end latency by deploying multiple instances of microservices in the UAVs that are close to the users that make use of them. Extensive experiments based on network emulation prove the performance of our ILP formulation of the problem and address the optimality gap between the ILP-based approach and the heuristics ones, which are highly scalable and usable in real-time for large-scale scenarios.

In the digital era, our lives are intrinsically linked to the daily use of mobile applications. As a consequence, we generate and transmit a large amount of personal data that puts our privacy in danger. Despite having encrypted communications, the DNS traffic is usually not encrypted, and it is possible to extract valuable information from the traffic generated by mobile applications. This study focuses on the analysis of the DNS traffic behavior found in mobile application traces, developing a methodology capable of identifying mobile applications based on the domains they query. With this methodology, we were able to identify apps with 98% accuracy. Furthermore, we have validated the effectiveness of the characterization obtained with one dataset by identifying traces from other independent datasets. The evaluation showed that the methodology provides successful results in identifying mobile applications.

Ensuring secure authentication between participating entities in VANETs has emerged as a critical challenge. Most of existing schemes mainly consider authentication issue in single administrative domain and suffer from various limitations that include privacy-preserving and malicious entity tracking. This paper proposes a double-layer blockchain-assisted conditional privacy-preserving cross-domain authentication scheme (DBCPCA) that leverages blockchain technology and certificate-less signatures to address these challenges. In DBCPCA, the upper-layer blockchain is used in cross-domain authentication by sharing inter-domain information among multiple different administrative domains. The lower-layer blockchain is employed in intra-domain authentication. In DBCPCA, we also introduce an anonymity mechanism to protect the real identity of a vehicle while enabling the system to trace a malicious vehicle, thereby addressing conditional privacy-preserving concerns. In addition, a security analysis of the proposed scheme demonstrates that it can meet our specified security objectives. Finally, we make a detailed experimental comparison with the most relative solutions such as BCPPA and BCGS. The results show that the DBCPCA scheme reduces the time cost by at least 66.68 % compared to the BCPPA scheme during the signature generation phase. During the signature verification phase, the DBCPCA scheme reduces the time cost by at least 62.39 % compared to the BCGS scheme.

The operations of unmanned aerial vehicles (UAVs) are susceptible to cybersecurity risks, mainly because of their firm reliance on the Global Positioning System (GPS) and radio frequency (RF) sensors. GPS and RF sensors are vulnerable to potential threats, such as spoofing attacks that can cause the UAVs to behave erratically. Since these threats are widespread and potent, it is imperative to develop effective intrusion detection systems. In this paper, we propose an image-based intrusion detection system for detecting GPS spoofing cyberattacks based on a deep learning methodology. We combine convolutional neural networks with Principal Component Analysis (PCA) to reduce the dimensionality of the dataset features, data augmentation to increase the size and diversity of the training dataset, and transfer learning to improve the proposed model’s performance with limited data to design a fast, accurate, and general method. Extensive numerical experiments demonstrate the effectiveness of the proposed solution carried out using benchmark datasets. We achieved an accuracy of 100% within a running time of 120.64 s at 0.3529 ms latency and a detection time of 2.035 s in the case of the training dataset. Further, using this trained model, we achieved an accuracy of 99.25% within a detection time of 2.721 s on an unseen dataset that was unrelated to the one used for training the model. In contrast, other models, such as Inception-v3, showed lower accuracy on unseen datasets. However, Inception-v3 performance improved significantly after Bayesian optimization, with the Tree-structured Parzen Estimator reaching 99.06% accuracy. Our results demonstrate that the proposed image-based intrusion detection method outperforms the existing solutions while providing a general model for detecting cyberattacks included in unseen datasets.

A Roadside Unit (RSU) serves as essential infrastructure in Vehicular Ad Hoc Networks (VANETs) that supports the goals of Intelligent Transportation Systems (ITS) by providing safety services, shared storage, and enhanced internet connectivity to vehicular users, drivers, and pedestrians. Additionally, the efficiency of VANETs, concerning network service utility and latency, depends on the relative positioning of these RSUs within the network topology. Most existing RSU deployment approaches deal with a single objective, either enhancing network service utility or minimizing the latency. For instance, some studies suggest deploying RSUs in high-traffic road segments that enhance network service utility but lead to higher latency. Conversely, some suggest deploying the RSUs in low-traffic road segments that minimize the network latency, but there will be low network service utility. Hence, there exists a trade-off between these two conflicting objectives in VANETs, and none of the studies address both objectives simultaneously. To achieve the balance between these two objectives, this paper proposes a Multi-Objective UAV assisted RSU Deployment (MOURD) scheme that leverages the Unmanned Aerial Vehicles (UAVs) for VANET efficiency. The MOURD scheme statically places RSUs in high-traffic road segments and dynamically dispatches the UAVs in low-traffic road segments to facilitate seamless network coverage and minimize the overall network latency. The simulation results on the road network of Delhi, India, demonstrate the effectiveness of the proposed MOURD scheme compared to other benchmark RSU & UAV deployment approaches. MOURD scheme outperforms on an average of 17.42%, 13.29%, 15.67% and 6.23% in terms of vehicle connection time, packet delivery ratio, throughput, and latency, respectively.

With the rapid growth in access demand for Internet of Things (IoT) devices, effective utilization of spectrum resource has become a key challenge to ensure reliable communications. Traditional dynamic spectrum access methods are inefficient when there are too many device accesses, channel reductions, and channel quality deterioration. In this paper, we propose a dynamic spectrum access method based on a fusion algorithm of graph neural network (GNN) and deep Q network (DQN), improving spectrum access efficiency while maintaining a good access success accuracy. Compared with traditional DQN, the computation time can be reduced by over 35%. Our approach first uses GNN to interact with the environment and predict the state of the IoT spectrum environment. Subsequently, automatic learning and optimization of spectrum access policies are achieved by selecting the mobile IoT user’s actions based on these predicted states using the DQN’s target network, experience playback, and reinforcement learning techniques. Simulation results show that the system model based on the proposed method can operate with better efficiency than the conventional method while maintaining a good channel access rate and channel quality.

Opportunistic Mobile Ad Hoc Networks (MANETs) offer versatile solutions in contexts where the Internet is unavailable. These networks facilitate the transmission between endpoints using a store-carry-forward strategy, thereby allowing information to be stored during periods of disconnection. Consequently, selecting the next hop in the routing process becomes a significant challenge for nodes, particularly because of its impact on Quality of Service (QoS). Therefore, routing strategies are crucial in opportunistic MANETs; however, their deployment and evaluation in real scenarios can be challenging. In response to this context, this paper introduces a monitoring software-driven tool designed to evaluate the QoS of routing algorithms in physical opportunistic MANETs. The implementation and its components are detailed, along with a case study and the outcomes provided by an implementation of the proposed solution. The results demonstrate the effectiveness of the implementation in enabling the analysis of routing protocols in real scenarios, highlighting significant differences with simulation results: mobility patterns in simulations tend to be inaccurate and overly optimistic, leading to a higher delivery probability and lower latency than what is observed in the real testbed.

The dynamic nature of the atmosphere, especially wind gust, poses a crucial challenge to efficient and real-time drone operations. This article presents a novel MQTT based software-defined drone network for trajectory correction of drone flights in gusty wind conditions using Glowworm Swarm Optimization (GSO). By imposing the GSO to the software-defined drone network, our proposed model SoftWind has optimized the navigation and control capabilities of drones by correcting the trajectories in a gusty wind environment. We have analyzed the trajectories and convergence of drones due to wind gusts. As wind disturbances affect the trajectories of drones, we have corrected it by our trajectory correction model and evaluated the direction of the drones must fly to mitigate the wind gust and the resultant velocity compared to the no-wind environment. This study analyzed the trajectories of 100 drone flights due to various wind gust lengths (i.e., 40 m, 10 m, 6 m, and 3 m) for a fixed gust amplitude of 15 m/s and various gust amplitude (i.e., 0 m/s, 5 m/s, 15 m/s, and 40 m/s) for a fixed gust length 5 m. We observed that all the drones are converged to a single point due to low gust length (≤ 5 m) and high gust amplitude (≥ 35 m/s). It is also found that the direction of the drone must fly 28.87°. East of South to mitigate the effect of wind gusts having 10 m gust length and 15 m/s gust amplitude and the resultant velocity of the drone is 22.38 m/s. The result shows that SoftWind reduces the convergence time by 26 %-54 % as compared to other existing models.

This paper introduces a new priority-aware routing protocol for mobile Ad-hoc networks to be utilized in emergencies, which is based on AODV. Mobile Ad-hoc networks find extensive use in various domains including military operations, environmental monitoring, healthcare, disaster response, smart transportation systems, unmanned aerial vehicles, and smart homes. During emergencies, communication can be severely restricted or even impossible due to the congestion of physical communication channels and unexpected technical failures in the infrastructure. Mobile Ad-hoc networks offer a solution to maintain continuous and reliable communication under such challenging conditions. In emergency scenarios, it is crucial for any node in the network to promptly deliver urgent messages to the intended destination, especially when certain nodes require ongoing active communication. The proposed routing protocol effectively addresses this requirement through its priority-aware mechanisms. The protocol ensures that nodes not involved in emergency tasks select the least congested route to prevent any delays or disruptions in the transmission of critical emergency data. This approach guarantees seamless communication for emergency nodes while allowing non-emergency nodes to communicate with each other as well. The study proposed in this paper introduces a new priority-aware routing protocol based on AODV for mobile Ad-hoc networks in emergencies. The packet transmission ratio of emergency nodes within the network is improved while maintaining the overall network performance unaffected. The adoption of proposed mechanisms to enhance performance does not necessitate an expansion in the size of data and control packets. These mechanisms do not inflict any supplementary latency or incur packet loss expenses on the network. The proposed protocol has been implemented and evaluated using ns-3 simulation software across various emergency scenarios. The results show that emergency nodes using the proposed protocol, achieve better packet delivery ratios compared to the original AODV, DSR, P-AODV, and AOMDV protocols, with improvements of 10.8%, 15.9%, 6.2%, and 5.9% respectively. This improvement in the packet delivery ratio for emergency data traffic is achieved without causing any disruptions in the overall network communication flow.

This paper proposes a new method for predicting downlink traffic volume in mobile networks, aiming to minimize overprovisioning while meeting specified service-level agreement (SLA) violation rates. We introduce a multivariate and multi-step prediction approach and compare four machine learning (ML) architectures: long short-term memory (LSTM), convolutional neural network (CNN), transformer, and light gradient-boosting machine (LightGBM). Our models predict up to 24 steps ahead and are evaluated under both single-step and multi-step conditions. Additionally, we propose parametric loss functions to adhere to SLA violation rate constraints.

Our results emphasize the importance of using parametric loss functions to meet SLA constraints. We discovered that LSTM when paired with our custom multivariate feature sets, outperforms the transformer architecture in short-term forecasting up to 4 h ahead. For these short-term predictions, we demonstrate that methods based on domain knowledge, like our custom feature sets combined with simpler models such as LSTM, surpass more complex models like transformers. However, for long-term forecasting (8 to 24 h ahead), transformers outperform all other models.