AI-Enabled Planning and Control for Aeronautical Ad-Hoc Networks

Title: AI-Enabled Planning and Control for Aeronautical Ad-Hoc Networks
Authors: Shahbazi Dastjerdi, Mohsen
Date: 2023-05-25
Abstract: In-Flight Entertainment and Connectivity (IFEC) is becoming a key trend and offering in-flight connectivity is one of the most essential demands of commercial airline passengers. A grand challenge is to provide in-flight connectivity in high altitudes and particularly in isolated locations, such as the oceans, where establishing an air-to-ground link is not possible. Moreover, the high speed and dynamic characteristics of such aircraft make this task difficult. Aeronautical Ad-Hoc Networking (AANET) intends to cope with this challenge by forming a network of airplanes having air-to-air (A2A) connections. However, the dynamic nature of such a network is likely to lead to unstable connections. The primary root cause of the majority of these stability issues is known to be the short life of A2A links which is the result of poor topology formation of aircraft. Concentrating on aircraft clustering and making them more stable can improve connection lifetime and improve the stability and performance of the network. Therefore the main objective in making AANETs feasible should be to form the topology as clusters of aircraft. With this in mind, the thesis's proposition is twofold: First, unveil the benefits of density-based clustering to improve the AANET performance. To do so, a modified DBSCAN algorithm is employed for the clustering problem that exploits several features of real flight datasets. This method also includes a weighted scheme to reflect the relative importance of each feature of the final calculation. The proposed method improved the packet delivery ratio and end-to-end latency of the state-of-the-art clustering-based AANET solutions by 51 % and 30 %, respectively. In addition, the proposed approach reduces the number of cluster changes by 22%. Second, selecting a well-connected cluster head is the next stage in enhancing connection and stability. This thesis presents a new cluster head selection technique for AANETs that calculates the Neighbor Nodes within a given distance of each node and selects the node with the most connections as the new cluster head. In instances where a cluster head cannot interact directly with another cluster, a Gateway node is chosen to facilitate connection with other clusters. According to simulations, the suggested method increases packet delivery ratio by 3, end-to-end delay by 9 and throughput by up to 10% compared to the current state of the art. In addition, the proposed method reduces cluster head replacements by 17% and increases cluster head longevity by 8%.
CollectionThèses, 2011 - // Theses, 2011 -

This item is licensed under a Creative Commons License Creative Commons