Abstract
Drone technologies have attracted the attention of many researchers in recent years due to their potential opportunities. Fleets of drones integrated with widely available relatively short-range communication technologies have various application areas such as wildlife monitoring, disaster relief, and military surveillance. One of the major problems in this manner is maintaining the connectivity of the drone network. In this chapter, the authors study the connectivity management issues in drone networks. Firstly, movement, communication, and channel models are described by the authors, along with the problem definition. The hardness of the problem is investigated by proving its NP-Hardness. Various algorithms proposed to solve the connectivity management problem and their variants are evaluated in detail. Lastly, for future directions, the authors present mathematical methods to solve the emerging problem in drone networks.
TopIntroduction
One of the most progressive technologies related to computer science and engineering and affected our daily lives dramatically is drone technology. Recently, various research institutes, universities, and companies have been dealing with drone projects for solving timely problems around the world. These studies can be investigated in a wide range of application areas such as disaster management, smart farming, military surveillance, wildfire tracking, civil security, photogrammetry, and power line inspection. A drone is a type of unmanned vehicle that is an aircraft without a human pilot, and it is generally composed of a flight controller, a battery, a transceiver, camera(s), a GPS module, and various sensors such as collision avoidance, accelerometer, and altimeter. Drones are unmanned aerial vehicles (UAVs) forming unmanned aircraft systems together with ground-based controllers and data transmission systems, which provide communication between these two parties. UAVs can be controlled remotely and manually by a human operator, whereas they can perform fully autonomous missions provided by onboard circuits that are integrated into the drone in advance.
The transmission capabilities and energy resources of the drones manufactured under the constraints of current technologies are generally limited for applications requiring a long lifetime. In many scenarios, a fleet of economic drones integrated with the communication devices such as IEEE 802.11 is designated instead of drones that are capable of communicating over long distances. An example network of drones is given in Figure 1.a, where 6 drones are connected in an ad hoc manner without any predefined infrastructure, and the transmission range of the base station only covers one of these drones. The use of a drone fleet or swarm can be much more advantageous in applications such as agriculture, meteorology, disaster detection, and search and rescue operations. For example, as given in Figure 1.b, the delivery of service among users who can be mobile in nature (military units on a battlefield, relief teams, or people using the internet service during a disaster) can be achieved with a drone swarm. As another example given in Figure 1.c, a network of these flying devices are assigned to accomplish a search and rescue operation. In this manner, drone swarms can scan the search area in detail and carefully, and they can send the required information to the base station over the backbone of drones after finding the target.
Figure 1. Drone Networks and Their Applications
One of the most important requirements of multiple drone systems is initiating and maintaining communication between drones. If all drones in a swarm can access a base station or a central satellite system in a single hop manner, the communication between them can be done utilizing these infrastructures without providing multi-hop ad hoc network infrastructures (Bekmezci et al.,2013). However, this type of communication between drones can be very costly, even impossible in most scenarios. For this reason, in many applications, the communication between the control center and the remote drones should be provided by the intermediate drones. To cover a large area with drones and reach much further distances from the control center, the drones must have a reliable and continuous connection. Therefore, maintaining the connectivity between drones is one of the most important requirements in such systems (Sahingoz, 2014). In multiple drone-based systems, the number of studies conducted in the literature on maintaining connectivity between drones is limited and is emphasized by reputable publications that this problem is open to research (Erdelj&Natalizio,2016; Zhao et al.,2018).
Key Terms in this Chapter
Connectivity Restoration: Re-connecting a disconnected ad-hoc network by either moving the nodes to desired positions or manipulating the link characteristics.
Area Coverage: Sensing and collecting meaningful data from the entirety or parts of a given mission area. Most drone applications involve means of area coverage.
Communication Model: A collection of rules of a set of nodes' behavior regarding establishing and maintaining their communication. It may enforce specific technologies and/or include allowable data rates, modulations, encoding, and link characteristics such as symmetricity, synchronization, ranges, etc.
Obstacle Avoidance: Drone (as solo or as in swarms) ability to autonomously avoid colliding into any observed or unexpected objects that exist in their route to an arbitrary destination. Such a feature may comply with the mission requirements and should not cause termination of any task unless it is inevitable.
Path Loss: Attenuation and degradation of radio signals used due to the distance between communicating parties, signal's properties (e.g., frequency) and the environmental conditions.
Channel Model: A collection of empirical and/or theoretical relations that define the characteristics of the communication model, including but not limited to the path loss, shadowing, multi-path fading, noise, and Doppler effects.
Movement Model: A collection of rules and activities that apply to a swarm of unmanned aerial vehicles. It defines how the swarm collaboratively move to accomplish the given tasks. For instance, it may be enforced to visit (cover) a portion of an area exactly k times.
K-Connectivity: Also known as k -vertex-connectivity. A connected graph's ability to remain connected after removal of any fewer than k vertices.
Drone: An electric-powered rotating-wing unmanned aerial vehicle that also has a means of computing and sensing capabilities.
Neighborhood: For an arbitrary node, the set of surrounding nodes that it is directly connected to. A k -neighborhood may also include indirectly connected neighbors at most at its k -hop distance.