An Unmanned Aerial Vehicle (UAV) network specifies a novel type of Mobile Ad hoc Network (MANET) in which drones serve as nodes and facilitate the retransmission of messages to their final destinations. Aside from its military application, it has recently begun to seep into the civilian sector. Similar to MANET and vehicular ad hoc networks, Flying Ad hoc Networks (FANET) are a subset of ad hoc networks. An FANET is different because it is founded on UAVs. Due to the characteristics of this sort of network, which is defined by a highly changing topology in a 3D environment, we must employ an adjusted configuration method to ensure good routing performance. Therefore, to deal with this problem, a technique that responds to any change in topology by always finding the best route is required. In this work, we propose a new protocol based on the hybrid optimization of the 2-opt heuristic and Honey Badger Algorithm (HBA), called HB-AODV. In order to locate its prey, a badger must move slowly and continuously while using scent markers and mouse-digging skills to catch it. In other words, the most efficient routes in terms of the number of hops are identified. Several simulations were conducted via the 3D version of Network Simulator (NS-2) on different deployment strategies. In comparison to AODV, DSDV, and AntHocNet, the obtained results demonstrated the proposed scheme’s good performance in terms of quality of service metrics.
1 April 2019 in International Journal of Applied Metaheuristic Computing
A mobile ad hoc network (MANET) is an autonomous system of mobile hosts (nodes) connected by a wireless link that forms a temporary network without the aid of any established infrastructure or centralized administration. The main problem of MANETs is the design of routing protocols that allow for communication among the hosts. The dynamic nature of such networks makes this problem especially challenging. The routing problems in ad hoc networks are due to their unpredictable and dynamic nature and the few resources (speed and autonomy). Therefore, bio-inspired algorithms are widely used to design adaptive routing strategies for MANETs. In this study, the authors propose a new approach based on the bee colony for the resolution of the routing problem in MANETs. The implementation (simulation) of the method is realized by Matlab, and the authors select Random WayPoint as mobility model. To validate the work, the authors compare the proposed approach with the AODV routing protocol in terms of the Quality of Service parameters, namely, End-to-End Delay, Packet Delivery Ratio and the Normalized Overhead Load.
2 April 2023 in Engineering, Technology & Applied Science Research
Association rule methods are among the most used approaches for Knowledge Discovery in Databases (KDD), as they allow discovering and extracting hidden meaningful relationships between attributes or items in large datasets in the form of rules. Algorithms to extract these rules require considerable time and large memory spaces. This paper presents an algorithm that decomposes this complex problem into subproblems and processes items by category according to their support. Very frequent items and fairly frequent items are studied together. To evaluate the performance of the proposed algorithm, it was compared with Eclat and LCMFreq on two actual transactional databases. The experimental results showed that the proposed algorithm was faster in execution time and demonstrated its efficiency in memory consumption.
The proliferation of medical imaging data, coupled with stringent privacy regulations, necessitates scalable and reliable classification methods to address the challenges of big data. Motivated by the critical need for accurate pneumonia detection from Chest X-rays (CXR) across diverse clinical settings, this research confronts the five fundamental challenges of big data: volume, velocity, variety, value, and veracity. In such scenarios, traditional centralized methods are often constrained by data heterogeneity, computational bottlenecks, and privacy risks. To overcome these constraints, we propose a Federated Learning (FL) system that distributes model training across multiple clients, thereby ensuring data privacy while efficiently managing large-scale datasets. Our method leverages transfer learning by fine-tuning a pre-trained VGG11 model and employs FedProx regularization to mitigate client drift arising from non-Independent and non-Identically Distributed (non-IID) data distributions. Furthermore, we introduce an innovative data partitioning technique that simulates real-world conditions by generating imbalanced label distributions with a Dirichlet process and injecting Gaussian noise to mimic image quality variations. By enabling distributed local training and dynamic learning rate adjustments, our approach effectively manages high-volume, high-velocity data while preserving data privacy. Experimental results demonstrate that our proposed method efficiently aggregates diverse and noisy client updates while achieving competitive performance in pneumonia classification.