Multi-UAV systems have many uses in civil applications including search and rescue, surveillance, and monitoring. In dynamic missions such as search and rescue, where multiple targets or areas of interest might need to be searched and monitored, the accuracy of target detection and information sharing among UAVs and with ground control station (GCS) play a vital role in mission success. In this work, we present a joint coverage, connectivity, and revisit time optimization framework for multi-target search missions with imperfect sensing and dynamic information exchange. We provide evolutionary algorithms (EA) and reinforcement learning (RL) based solutions to the optimization problem. Our results show that by considering cell revisit times as well as connectivity in the path planning using EA, the target detection time can be improved by 30% and we can achieve near-instant GCS inform in realistic, uncertain environments. RL-based results show worse mission time performance compared to EA methods indicating that further tuning and better representation methods might be required.

This paper presents a novel reinforcement learning framework for multi-UAV cooperative search missions with sensing imperfections. We introduce a scalable multi-agent environment that effectively models sensor uncertainty while enabling decentralized information merging between UAVs. Our reward mechanism uses reward shaping for adapting to the information flow through the mission. Using Proximal Policy Optimization (PPO), we create a training framework that achieves consistent performance across varying numbers of UAVs and targets. Our results demonstrate that the proposed method maintains high success rates even with significant sensing uncertainties, with performance improving as the number of cooperating UAVs increases. The framework shows resilience against the unlearning problem commonly encountered in uncertain environments and uses a minimal observation space. Our framework provides a foundation for deploying reliable multi-UAV search systems in real-world scenarios where sensor reliability cannot be guaranteed and computational resources are scarce.

A novel multi-agent reinforcement learning (MARL) technique utilizing Proximal Policy Optimization (PPO) is introduced to coordinate a drone team in search and rescue operations, where multiple targets with different connectivity requirements are present. The proposed approach integrates a dynamic reward system that effectively manages the trade-off between exploration, target identification, and network connectivity. The model is trained for a fixed target type and area size, but is evaluated for scenarios with different numbers of UAVs, diverse target configurations, and mission priorities, including target search and information updates, illustrating its adaptability and versatility in comparison to common methods.

Multi-UAV systems are deployed for many civil applications, such as search and rescue, surveillance, environmental monitoring, that require efficient area coverage and connectivity between UAVs and ground control station (GCS) to enable teamwork and information exchange. In dynamic missions, such as disaster management, or in case the sensors on the UAVs lead to uncertain measurements, it is desirable that whole or part of the area of interest is revisited by UAVs for information update or reliability. In this work, we propose a novel multiUAV path planning algorithm that incorporates revisit times into path decisions, without compromising coverage and connectivity requirements. We model the proposed planner by a multiobjective optimization problem and we provide an NSGA-II based solver that leads to a set of Pareto-optimal solutions. We compare our results with a benchmark path planner that only optimizes coverage and connectivity and show the tradeoffs between the proposed objective functions. We observe up to 4-fold improvement in terms of observed times between revisits.

In case of large-scale disasters such as earthquakes, where urban and public infrastructure might be destroyed, multiple unmanned aerial vehicle (UAV) or multi-drone systems can be very beneficial to rescue teams, by enabling them to carry out multiple tasks simultaneously, such as detecting, monitoring, and connecting multiple targets simultaneously. In this work, we propose an event-driven algorithm that integrates a novel connectivity-based prioritization of targets into dynamic positioning and path planning for multi-UAV systems in search and rescue (SAR) missions. Two distinct groups of UAVs are deployed. While search UAVs sense the area of interest as fast as possible, relay UAVs provide connectivity to search UAVs as well as targets depending on their priority level. Our results show that network formation and target connectivity can be improved up to 60% compared to a benchmark path planner that does not consider connectivity priorities.

This paper formulates the Sport Event Filming with Connectivity Constraints (SEF-C2) problem. The SEF-C2problem is an event coverage problem, which exploits a team of Unmanned Aerial Vehicles (UAVs) over a limited field in order to track the movements of an object (e.g., of the ball) and to deliver a video stream of the events (e.g., ball passes, goals) to the spectators meeting certain timeliness and video quality criteria. Assuming a priori knowledge of the whole sequence of actions, first a novel mathematical model is introduced that determines a sequence of movements for the UAVs, such that the timeliness of the filming is maximized and the total traveled distance is minimized. Then, dynamic, artificial potential function based, distributed UAV movement schemes that have no a priori knowledge of the sequence of game actions are proposed to optimize networking performance. Extensive simulations are used to analyze the performance in terms of video transmission quality and show that the proposed schemes outperform existing schemes.

We propose a multi-objective optimization algorithm to allocate tasks and plan paths for a team of UAVs. The UAVs must find a target in a bounded area and then continuously communicate the target information to the ground personnel. Our genetic algorithm approach aims to minimize the mission completion time, which includes the time to find the target (area coverage) and the time to setup a communication path (network connectivity). We evaluate strategies using a data mule, a relay chain, and a novel hybrid approach to communicate with the ground personnel. The algorithm can be tuned to prioritize coverage or connectivity, depending on the mission demands. Simulation results show reduced overall mission completion times (up to 65%), with more improvement as the UAV density increases.

Multicast video streaming over IEEE 802.11 is unreliable due to the lack of feedback from receivers. High data rates and variable link conditions require feedback from the receivers for link estimation to improve reliability and rate adaptation accordingly. In this paper, we validate on a test platform an application-layer rate-adaptive video multicast streaming framework using an 802.11 ad-hoc network applicable for mobile senders and receivers. Experimental results serve as a proof of concept and show the performance in terms of goodput, delay, packet loss, and received video quality.

The commercial availability of small unmanned aerial vehicles (UAVs) opens new horizons for applications in disaster response, search and rescue, event monitoring, and delivery of goods. An important building block is the wireless communication between UAVs and to base stations. Design of such a wireless network may vary vastly from existing networks due to aerial network characteristics such as high mobility of UAVs in 3D space. This paper presents experimental performance results with commercially available UAVs. First, we show throughput results for IEEE 802.11ac in a UAV setting. Second, we demonstrate that IEEE 802.11n can have much higher throughput over longer ranges than reported in [1] and [2]. Third, we analyze the fairness in a multi-sender aerial network. Fourth, we test a real-world coverage scenario with two mobile UAVs sending to a single receiver. Performance analysis considers the rate adaptation mechanism in both indoor and outdoor line-of-sight scenarios.

In this paper, we propose strategies for merging occupancy probabilities of target existence in multi-UAV cooperative search. The objective is to determine the impact of cooperation and type of information exchange on search time and detection errors. To this end, we assume that small-scale UAVs (e.g., quadrotors) with communication range limitations move in a given search region following pre-defined paths to locate a single stationary target. Local occupancy grids are used to represent target existence, to update its belief with local observations and to merge information from other UAVs. Our merging strategies perform Bayes updates of the occupancy probabilities while considering realistic limitations in sensing, communication and UAV movement - all of which are important for small-scale UAVs. Our simulation results show that information merging achieves a reduction in mission time from 27% to 70% as the number of UAVs grows from 2 to 5.

Protocols for neighbor cardinality estimation can be used by a node in an ad hoc or sensor network to obtain a fast approximation of the number of nodes in its radio range. While the performance of such estimators was studied via simulations and mathematical analysis, we show here for the first time a proof-of-concept and performance test with an implementation on low-power wireless sensor devices. We illustrate the challenges of implementing the recently proposed Multi-feedback estimator (MFE) on Z1 devices. Experimental results show that - with certain computational improvements - MFE can perform faster than simple neighbor counting for scenarios with many nodes or relaxed accuracy requirements.

Small-scale multicopters operating as autonomous teams in the air are envisioned for aerial monitoring and transport of goods in a variety of applications, including disaster management and environmental monitoring. For such applications to become reality, a high-throughput wireless network is needed. This paper presents experimental performance results with commercially available quadrocopters communicating via IEEE 802.11a. In particular, we compare the infrastructure and mesh modes of 802.11 for one-hop and two-hop communications, thus analyzing network layer versus MAC layer relaying. Results illustrate that changes are required in the mesh mode to support applications demanding high throughput with low jitter.

We specify and evaluate a protocol for cooperative relay communications in wireless networks targeted for low-budget and energy-constrained off-the-shelf hardware. The protocol located at the Medium Access Control (MAC) layer integrates radio resource reservation, relay selection, and packet flow. Performance is evaluated with different parameters, such as node density, channel coherence time, and data packet size. Higher network-wide reliability and throughput compared to noncooperative protocols can be achieved in dense networks and unreliable channels. At the same time, throughput does not degrade in sparse networks or good channel conditions.

Increasing availability of autonomous small-size aerial vehicles leads to a variety of applications for aerial exploration and surveillance, transport, and other domains. Many of these applications rely on networks between aerial nodes, that will have high mobility dynamics with vehicles moving in all directions in 3D space and positioning in different orientations, leading to restrictions on network connectivity. In this paper, we propose a simple antenna extension to 802.11 devices to be used on aerial nodes. Path loss and small-scale fading characteristics of air-to-ground links are analyzed using signal strength samples obtained via real-world measurements at 5 GHz. Finally, network performance in terms of throughput and number of retransmissions are presented. Results show that a throughput of 12Mbps can be achieved at distances in the order of 300m.

Several communication protocols and applications require a node to know how many neighboring nodes exhibiting a certain attribute it has. Conventionally, such neighbor information is obtained by explicit message exchange between nodes, which is reliable but inefficient in densely connected networks in terms of overhead and delay. An alternative approach is to perform an estimation of the neighbor cardinality using probabilistic methods. This paper pursues such an approach by proposing neighbor cardinality estimators that require no coordination among polled nodes but are based on a simple random access scheme with busy tones, where the number of empty slots is exploited to infer about the neighbor cardinality. We compare three estimators with different levels of adaptability and feedback from the query node and discuss their suitability for IEEE 802.11 and low power sensors. Performance is studied in terms of estimation accuracy and delay.

We investigate the area coverage and connectivity of an autonomous, unmanned aerial vehicle (UAV) network, whose goal is to monitor and sense a given area of interest in an efficient manner. To this end, we propose a connectivity-based mobility model that aims to sustain connectivity between the UAVs and the ground station. We compare coverage and connectivity performance of the proposed scheme with a coverage-based mobility scheme in several scenarios. Results illustrate the trade-off between achieving good spatial coverage and staying connected.

This paper proposes a probabilistic technique that enables a node to estimate the number of its neighbors that fulfill certain criteria. The technique does not require any a priori information about the network topology. Based on a query from the estimating node, the neighbors send busy tones in a sequence of time slots. Evaluating the fraction of empty slots, the querying node can infer about its neighborhood. Mathematical analysis and numerical simulations show how the estimation success and accuracy depend on the available time period. Applications of the method can be found, for example, in RFID systems and medium access protocols.

In this paper, we propose a belief-based movement approach for a network of unmanned vehicles, which is deployed to cover a given area. The spatial and temporal coverage performance of the belief-based approach is compared with legacy movement models such as random walk, random direction as well as a random sweeping model for several scenarios. Our results highlight the trade-off between how fast and how well an area can be covered and show that the proposed approach can outperform the other methods that are tested.

Cooperative relaying is a communication technique in wireless networks where neighboring nodes assist communication pairs to mitigate the negative effects of multi-path fading. The resulting performance strongly depends on the selected relays. Although a cooperative relay provides benefits to a given source-destination pair, overall network performance might be degraded due to the increased level of interference. So far almost all relay selection mechanisms consider mainly channel conditions to the potential relays. In this paper we propose a contention-based relay selection mechanism that can take into account also spatial efficiency of potential relays. For that the degree as well as relative position of the nodes are used for selection. With the proposed method a high successful relay selection probability can be achieved, while significantly reducing the amount of additional spatial resources blocked by the cooperative relay.

With the emergence of computer worms that can spread over air interfaces, wireless ad hoc and sensor networks can be vulnerable to node compromises even if the deployed network is not connected to the backbone. Depending on the physical topology of the wireless network, even a single infected node can compromise the whole network. In this work, epidemic (e.g. worm) propagation in a static wireless network is studied, where a number of infected mobile nodes are injected over the existing network. It is shown that the epidemic spread threshold and size depend on the physical topology of the underlying static wireless network as well as the mobility model employed by the infected mobile nodes. More specifically, results show that in a fully-connected static wireless network targeted attacks are more effective, whereas for a random topology random attacks can be sufficient to compromise the whole network.

The physical topology of a wireless ad hoc network has a great impact on not only the performance of the algorithms developed for such networks, but also on the robustness (i.e., vulnerability to attacks or node failures) of the network. In this work, an empirical study on the impact of mobility on the epidemic (e.g., worm) propagation is provided, where topology-dependent and independent mobility models are compared in terms of infection (compromise) propagation among nodes. Results show that if there is no node recovery (i.e., no patches), for some mobility models whole network can be compromised in a short amount of time with infection rates less than 0.01 for even sparsely-connected ad hoc wireless networks. If there is recovery an epidemic might arise depending on the infective time durations and infection rates.

While many interesting dynamic load balancing schemes have been proposed for efficient use of limited bandwidth and to increase the capacity of congested or hot spots (or cells) in wireless networks, to date, a comprehensive mathematical framework which encompasses all of these schemes does not exist. In this paper, we provide a unified mathematical framework for dynamic load balancing, which leads to closed-form performance expressions for evaluating the performance of some of the most important dynamic load balancing strategies proposed in the literature. To the best of our knowledge, this is the first generic theoretical framework that can be used to evaluate the performance of many different dynamic load balancing schemes with simple closed-form results. The accuracy of the results predicted by these analytical expressions derived from the theoretical framework is checked by comparing these results with simulation results provided in the literature for well-known schemes.

Current cellular networks are not self-organizing in nature; e.g., they are vulnerable to base station (BS) failures or malfunctions due to congestion (a large amount of traffic) or due to emergencies. To address this problem, use of fixed relay nodes (FRNs) in a geographical area (which may or may not be covered by base stations or access points) was recently proposed to introduce self-organizing features to cellular wireless networks, such as enabling traffic forwarding in case of congestion and handling base station failures. In this paper, we provide an algorithm for choosing system parameters in a joint cellular and fixed relay network such that quality of service (QoS) requirements in terms of delay are satisfied. In particular, we show how to determine the optimum number of neighbors to achieve small average path lengths, high coverage, and low vulnerability to failures.

The performance of most dynamic load balancing schemes, proposed to solve the hot spot problem in cellular systems, depends on the user distribution within a cell. In this paper, we first study the impact of user distribution on the call blocking probability performance that can be achieved by two key dynamic load balancing schemes (i.e., channel borrowing without locking (CBWL) and integrated cellular and ad hoc relay (iCAR) system (H. Jiang and S.S Rappaport, 1994)-(H. Wu et al., 2001)). We show that the achievable call blocking probability performance is location-dependent, and iCAR and CBWL schemes might not be able to achieve 2% call blocking probability for all user distributions, unless major modifications are made to CBWL or iCAR to overcome local congestion problems. Therefore, in this paper, we propose to use a joint dynamic load balancing scheme that combines the features of CBWL and iCAR schemes. Our objective is to guarantee a low call blocking probability in congested cells irrespective of user location.

Several dynamic load balancing schemes have been proposed in the literature to solve the hot spot problem in cellular networks. In O. K. Tonguz and E. Yanmaz (2003), a unified theoretical framework for the performance analysis of dynamic load balancing schemes in cellular networks have been developed, and closed-form expressions for the call blocking probability of several existing dynamic load balancing schemes have been derived. However, to the best of our knowledge, an analysis showing which one of these dynamic load balancing schemes is optimum, in terms of the achievable call blocking probability, has not been reported before. In this paper, we explore, for the first time, if a global optimum dynamic load balancing scheme exists, and we examine if and under which conditions the existing dynamic load balancing schemes can achieve optimality.

In this paper, we propose to use a dynamic load balancing scheme that utilizes fixed relay stations placed in a cellular geographical coverage area to improve the handover performance in cellular networks. To this end, we develop closed-form performance expressions and derive the handover dropping probability in terms of the main system parameters, such as the new call and handover arrival rates, the number of fixed and load-balancing channels per cell, the load balancing probability for new and handover calls, etc. Our results show that by employing a dynamic load balancing scheme, not only the new call blocking probability, but also the handover dropping probability can be improved substantially, without reserving any channels or assigning any priority to handover calls. In fact, employing dynamic load balancing without any guard channels outperforms a conventional system that employs the well-known guard channel method (GCM).

We have introduced a novel concept called "managed mobility", and addressed the mobility of the relaying devices called mobile ad hoc relaying stations (MARSs) in the integrated cellular and ad hoc relaying (iCAR) system. We anticipate that the idea of managed mobility proposed in this paper for the iCAR system as well as the mobility management strategies may also be applied in other ad hoc networks, such as the self-reconfigurable sensor network, to reduce additional node deployment cost and increase fault tolerance.

The capacity of wireless networks can be increased via dynamic load balancing/sharing by employing overlay networks on top of the existing cellular networks. One such recently proposed system is the integrated cellular and ad hoc relay (iCAR) system, where an overlay ad hoc network is employed to use the resources efficiently by dynamically balancing the load of the hot spots in the cellular network, and to provide quality-of-service to subscribers, no matter where they are located and when the request is made. It is assumed that this overlay network operates in the 2.4-GHz Industrial, Scientific, and Medical (ISM) band and, hence, the number of available ISM-band relay channels used for load balancing will be limited due to other users' interference at a given point in time. In this paper, the impact of ISM-band interference on the performance of iCAR systems, which is a representative hybrid wireless network, is studied, and it is shown that dynamic load balancing and sharing capabilities of iCAR systems are strictly dependent on the availability of the ISM-band relay channels. In addition to quantifying the impact of the number of available relay channels on the performance of iCAR systems, a simple channel assignment scheme to reduce the performance degradation due to other users' interference is also provided. Results show that this interference avoidance technique can improve the realistic performance of iCAR-like hybrid wireless networks by 12%-23% when the interferers are uniformly distributed in the ISM-band.

Many interesting dynamic load balancing schemes have been proposed for efficient use of the available frequency spectrum, and, in particular, to increase the capacity of congested or hot cells in a cellular network. We provide a simple theoretical framework which leads to closed-form expressions for evaluating the performance of some of the most important dynamic load balancing strategies proposed in the literature. To the best of our knowledge, this is the first generalized theoretical framework that can be used to evaluate the performance of many different dynamic load balancing schemes with simple closed-form results. The accuracy of the developed expressions is checked and verified by comparing their results with simulation results provided in the literature. There is an excellent match between the predictions of our theoretical framework and the simulation results.

The integrated cellular ad hoc relaying (iCAR) system is a representative heterogeneous wireless system, proposed to address the congestion problem in the wireless networks. In this paper, we present an analytic model based on Markov chains for the queuing delay performance of iCAR. Our results show that the new call requests in iCAR have a significantly lower queuing delay than that of the conventional cellular system. The analytic model developed in this paper may serve as the guideline for the delay performance evaluation of the next generation heterogeneous wireless systems.

In this paper, a simplified analysis technique for the integrated cellular and Ad hoc relay (iCAR) systems is presented. First, a simple two-cell system is analyzed using a multi-dimensional Markov-chain. The performance metric employed is the call blocking probability of each cell in the system. To this end, first a closed-form expression for the call blocking probability in the two-cell system is provided. Then, it is shown that these closed-form expressions could be used to analyze more practical systems. The accuracy of the developed simple analytical expressions is checked and verified by comparing the results predicted by these analytical expressions with simulation results. It is shown that there is an excellent match between analytical and simulation results.

Coping with congestion or hot spots is one of the common problems faced by the wireless service providers. To handle this hot spot or congestion problem in a cellular network, several dynamic load balancing schemes based on channel borrowing have previously been proposed. A recent approach to dynamic load balancing is the Integrated Cellular and Ad hoc Relay (iCAR) system. iCAR employs ad hoc relay stations (ARS's) in the cellular network to balance traffic loads efficiently and to share channels between cells via primary and secondary relaying. In this paper, it is shown that the realistic performance of ICAR systems in terms of dynamic load balancing and load sharing is heavily dependent upon the number of available ISM-band ARS channels. Results also indicate that the number of ARS channels required for dynamic load balancing is much more than the number of ARS channels required for load sharing (i.e., for bringing the call blocking probability of a hot spot to 2%).

One of the common problems faced by the wireless service providers worldwide is coping with congestion or hot spots. To handle this hot spot problem, methods that combine the existing cellular networks with ad hoc networks have been proposed. Integrated Cellular and Ad Hoc Relay (iCAR) system employs ad hoc relay stations (ARSs) within the cellular network to balance traffic loads efficiently and to share channels between cells via primary and secondary relaying. These ARSs operate in the ISM band, and therefore, do not cause interference to the cellular band. When analyzing the performance of WAR systems, there are several factors that should be taken into account These factors include the coverage area of the ARSs, the number of ARS channels, the placement of ARSs, etc. In this paper, the impact of the number of ARS channels on the performance of WAR systems is studied. To this end, a multi-dimensional Markov-chain analysis is performed for a simplified two-cell system model. Results show that, with a proper amount of ARS coverage within each cell the call blocking probabilities can be decreased significantly with a small number of channels. Results also suggest that by increasing the number of ARS channels perfect load balancing can be achieved.

A recent work explicitly models the discontinuous motion estimation problem in the frequency domain where the motion parameters are estimated using a harmonic retrieval approach. The vertical and horizontal components of the motion are independently estimated from the locations of the peaks of respective periodogram analyses and they are paired to obtain the motion vectors using a procedure proposed. In this paper, we present a more efficient method that replaces the motion component pairing task and hence eliminates the problems of the pairing method described. The method described in this paper uses the fuzzy c-planes (FCP) clustering approach to fit planes to three-dimensional (3-D) frequency domain data obtained from the peaks of the periodograms. Experimental results are provided to demonstrate the effectiveness of the proposed method.