In urban areas, the extensive use of Unmanned Aerial Vehicles (UAVs) promises to revolutionize many industries but will increase the number of UAVs inhabiting a given airspace drastically. A necessity for this is the establishment of an Unmanned Aerial Traffic Management (UTM) system that enables the safe and reliable operation of UAVs. One way to establish this is to form an ad-hoc network between UAVs, operators and ground stations and distribute movements and commands throughout the entire network. Recently, Rate Decay Flooding (RDF) was proposed to realize such a network by gradually delaying packets the further they travel and thereby reducing the traffic density. Furthermore, Network Coding (NC) is a novel approach to reduce the traffic as different pieces of information can be encoded into a single transmission. In this work, we present a combination of both approaches. For this we propose a coding policy that determines when to use NC based on the contents of packets waiting to be forwarded. Further we implement a Bayesian online estimator for the current packet loss rate, which is a crucial component of the coding policy and thereby allowing the protocol to adapt to changing network sizes. The protocol is evaluated in an open-source simulator. Here we show that the coding policy successfully determines high-gain coding actions and reacts to larger networks with higher packet loss, by reducing the share of coded transmissions. Using our approach networks 95 % larger than previously possible with RDF are supported by reaching 21.2 % more receivers per transmission.

L-band Digital Aeronautical Communications System (LDACS) is the selected Air-to-Ground (A2G) technology for future aeronautical communications and a proposed candidate for Air-to-Air (A2A) links. Geographic greedy routing in sparse LDACS A2A networks, typical during gradual system deployment, frequently encounters local minima, necessitating backup mechanisms that are inefficient. Previous research has primarily focused on refining backup mechanisms, neglecting the root cause of geographic greedy routing failures. This paper investigates why geographic greedy routing performance deteriorates in sparse network scenarios, where failures occur more frequently than in dense deployments. We introduce a novel metric to quantify the quality of hop-by-hop forwarding decisions in geographic greedy routing. Furthermore, we develop a second-order absorbing Markov chain model to predict the success ratio and hop stretch factor. The model is validated through Monte-Carlo simulations over the French airspace with varying LDACS equipage fractions, achieving an average difference from simulation results of less than 3.4% for the success ratio and 1.5% for the hop stretch factor. The proposed model demonstrates high accuracy and can be generalized to evaluate other geographic routing protocols. Consequently, the outcomes provide valuable insights toward designing optimized geographic greedy routing protocols.

Advances in mechanical capabilities and mass manufacturing of Unmanned Aerial Vehicles (UAVs) are driving their application in various fields from precision agriculture to infrastructure monitoring and on-demand parcel delivery. Especially in urban areas it is projected that large amount of UAVs will inhabit the airspace. To facilitate the safe and reliable operation of large-scale urban UAV deployments, an Unmanned Aerial Traffic Management (UTM) system is required. Such a system needs to be aware of all movements within the airspace to control and monitor urban UAV operations. One way to realize this is the establishment of an ad-hoc network, which UAVs use for network-wide dissemination of their positions. Recently, Rate Decay Flooding (RDF) has been proposed as a tailor-made protocol to realize such a system. Although RDF has been proven to be efficient in supporting UTM applications in larger networks than ordinarily possible, much of its success relies on the proper selection of protocol parameters. In this work, we propose a reinforcement-learning framework that automatically adapts the configuration of RDF to its perceived environment. We utilize deep contextual bandits as a light-weight, but effective method to capture the non-linear relationship between the perceived environment and the achieved performance. We name this extension Dynamic Rate Decay Flooding (dynRDF). In a simulation study, we show that this solution is effective in finding optimal configurations for RDF for varying network sizes. To achieve this, only 2.7 % of all possible configurations had to be explored. Allowing dynRDF to also take the local UAV density into account, a performance gain of more than 12 % is achieved in a relevant composite metric capturing both the timely dissemination of position updates to nearby UAVs and reliable network-wide dissemination.

The emergence of the L-band Digital Aeronautical Communications System (LDACS) presents a significant opportunity for enabling Air-to-Air (A2A) communication to accommodate the growing number of aircraft. However, it requires overcoming significant Medium Access Control (MAC) delays and enhancing connectivity in sparse networks. Geographic greedy routing, commonly used in Aeronautical Ad-hoc networks, utilizes position information to eliminate the need for topology discovery. Yet, its efficacy declines as network density decreases. With the gradual introduction of aircraft equipped with LDACS, it becomes crucial to improve greedy forwarding performance. This research investigates Greedy-k, a greedy forwarding variant using k-hop neighborhood information, to boost sparse network performance. We introduce a method to minimize beacon size by transmitting a subset of k-hop neighborhood data that fits within an LDACS time slot. We derived the subset size analytically and evaluated the performance through simulations benchmarked against the conventional Greedy-1. Our results indicate that the proposed approach achieves up to 13% higher Packet Delivery Ratio (PDR) than Greedy-1, while capturing additionally 70.1% and 34.6% of 2ndand 3rdorder neighbors, respectively.

The use of Unmanned Aerial Vehicles (UAVs) has been proposed for numerous applications from recreational photography to commercial deliveries and infrastructure monitoring. But their effective deployment depends on the establishment of an Unmanned Aerial Traffic Management (UTM) which in turn requires reliable communication protocols to ensure safe and efficient operations. Recently, Rate Decay Flooding (RDF) was proposed as a novel protocol to enable position sharing among UAVs, realizing one of the major UTM applications. Another application is the provisioning of a redundant control connection which allows for point-to-point communication between a UAV and its operator in case their proprietary primary connection breaks. Every protocol used for this must harmonize well with RDF so that both applications can be realized at the same time. In this work we propose Spotlight Flooding (SLF) and its enhanced version SLF+, which build on top of RDF to realize a fast and reliable point-to-point connection between any two nodes in the network. These protocols are evaluated in an open-source simulator considering two scenarios. First, we consider a single UAV in distress using SLF/SLF + to explore the performance of our novel protocols for networks of up to 525 UAVs. The results show, that even for large networks, SLF+ achieves a Packet Delivery Ratio (PDR) above 90 % at an end-to-end delay of less than 60 ms. Additionally, we evaluate a scenario of 300 UAVs with an increasing share of UAVs using SLF/SLF+. Again, even if half of the UAVs in the scenario use SLF/SLF+ a PDR above 90 % is achieved at a delay of 142 ms.

While the capabilities of Unmanned Aerial Vehicles (UAVs) lay the foundation for exciting new applications, many communication challenges remain unsolved. To make UAV operation safer and more reliable, the establishment of an Unmanned Aerial Traffic Management (UTM) is necessary. The most important functions of such a system are Airspace Monitoring (AM) and Communication Aided Detect and Avoid (CADA). To realize CADA, UAVs must timely communicate their position to other UAVs nearby to avoid collisions. For AM, positions must be forwarded to a ground station e.g. through an ad-hoc network. In this work, we evaluate whether these two functions can be realized for dense urban UAV networks in a decentralized way by using position-flooding. To measure this, we select three flooding protocols from the literature. Further, we propose Rate Decay Flooding (RDF) as a novel flooding protocol, that leverages the fact, that position updates become less urgent the further they have travelled. RDF reduces the traffic density by lowering the data rate with which traffic is forwarded on every hop. All four protocols are evaluated in an open-source simulator. The results show that RDF can support position-flooding in networks more than twice the size compared to the next best protocol. In doing so, it generates 52% less data traffic and thereby allows for the concurrent use of the ad-hoc network for other use-cases such as remote control. Using RDF improves the position uncertainty of neighbors at a distance of 2500m by 180m.

As the global air-traffic rises, current communication systems for aircraft like satellites and ground based infrastructure will reach their capacity limits. Especially for rural and oceanic regions, the establishment of an Aircraft Adhoc Network (AANET) is a promising solution. In this paper, a tailor-made routing protocol (Link Duration Based AODV) for communication of aircraft on trans-oceanic routes is proposed. It uses an estimation of the expected residual path duration to make informed routing decisions and establish especially long-lasting routes. Two alternative strategies to estimate route durations were designed: In the deterministic strategy (AODV-LD-D), the fact that aircraft mostly exhibit uniform motion is leveraged to estimate link durations from their position and velocity. The stochastic strategy (AODV-LD-S) employs the known and distinct distribution of link durations experienced by oceanic flights. Both strategies were evaluated on real-world aircraft traces over 72 days in simulation. It is shown that both strategies were able to improve the average route lifetime from around 800 s to over 1100 s with the stochastic strategy and even around 1500 s with the deterministic strategy. The increased route lifetime decreases the number of necessary control packets by as much as 44.2 ±8. 5%. Further, it was shown, that AODV-LD achieves the same packet delivery ratio as AODV in all scenarios. Both strategies increase the average hop count of routes by roughly one and therefore increase the average End-to-End delay by around 8ms which is the expected transmission delay for a single hop. Comparing the two strategies, it was observed that AODV-LD-D required a significantly higher amount of control data compared to AODV-LD-S.

Aircraft Ad-hoc Networks (AANETs) are capable not only to allow low-latency Internet connectivity on aircraft but also to transmit safety-critical data. An aircraft in distress could transmit important information to another nearby aircraft in range before it crashes on the ground. In this paper the link probability for aircraft crossing the North Atlantic Corridor is investigated. An analytical model for the nearest neighbor distribution is proposed taking the shape and dynamic position of North Atlantic flight corridors into account. A uniform distribution of nodes is assumed in a bounded model area. Simulation results and empirical aircraft position data captured over 10 weeks are analyzed and used for verification of the analytical model. The link probability is derived from the nearest neighbor distribution and the model performance is evaluated. Aircraft sticking to North Atlantic flight tracks bias the assumed uniform distribution over the area. Therefore the results differ for east-and westbound flights. A link probability of more than 99% can be achieved if the communication range is greater than 250km and at least 40 aircraft are located within the model area. The average error of the model predicting the required communication range for a 99% link probability amounts to 8.94%.