Related works of UAV communication in disaster events

The UAV communication and network status are quite a significant area of study during all three life cycles of the disaster. This subsection introduces the network and communi-cation in some of the related works in Table 2.1 and what the target research field. The unique features of the UAV employment in the network have been considered in the [Luo, 2015a]. Authors in work [Luo, 2015a] proposed a distributed gateway selection algorithm with dynamic network partition by taking into account the application characteristics of UAV networks, and cloud framework as Figure 2.2 demonstrates. In the proposed algo-rithms, the influence of the proposed work asymmetry information phenomenon at UAVs’

control is overcome by separating the network into several subsystems. When the number of gateways can be controlled entirely according to the system requirements. In circumstantial, authors fixed the durability of UAV network, creating a network partition model, and design a distributed gateway selection algorithm. The experiments were executed in a simulator, and the results of using the proposed system revealed that the proposed system is faster and has more stable topology compared to other systems [Luo, 2015a]. The involvements of this system can be in three life cycles of the disaster, and the main field is UAV-aided communi-cation systems, as demonstrated in Table 2.1.

Figure 2.2:The UAV cloud framework. [Luo, 2015a].

A multi-UAV distributed decisional architecture has been developed in the framework of the AWARE Project [Maza, 2011]. A set of UAVs platforms and Wireless Sensor Networks have been used to confirm the proposal of the designed system in disaster management and public safety applications. Various elements and scenarios where the multi-UAV mis-sions were executed in real flight experiments. The mismis-sions were surveillance with multiple UAVs, sensor deployment, and fire threat verification. Nevertheless, the authors proposed some critical issues in multi-UAV systems, such as distributed task allocation, conflict resolu-tion, and plan refining. These issues have been solved during the execution of the missions.

The AWARE project aimed to contribute to the post-disaster life cycle and the main field target of researchers are monitoring, information, situational awareness, and also UAV com-munication control and network systems [Maza, 2011].

In post-disaster life cycle, authors in the study of [Neto, 2012] proposed a modular em-bedded architecture. The study is structured from three various stages: an emem-bedded sys-tem, communications links, and navigation system. Several flight experiments were con-ducted in an environmental disaster. In the mountain terrain, a pilot assigned scanning missions for mudslides regions. The experiments revealed the benefit of using UAV to scan such disaster incidents. Moreover, the study covers the communication part where UAV is hard to communicate with GCS in the mountain terrain environment. The communica-tions system has been designed to match the dynamic agents’ environment. The protocol considers any communications interface.

The contribution of work [Neto, 2012] is in the post-disaster life cycle, and the main fields are damage assessment and some contributions in UAV communication control and network.

In wireless sensor networks is possible to experience faults for many reasons, which in-clude broken links or even the nodes have been damaged via a disaster such as an earth-quake in post-disaster. These faults can give increase to critical problems if WSNs do not have a reconfiguration mechanism. Many wireless sensor networks are designed to recog-nize disasters are used in places with a history of disaster. The damaged node can leave a part of the system until it recovers from that failure. Authors in work of [Ueyama, 2014]

proposed a solution by employing UAVs to reduce the problems arising from defects in a sensor network when observing natural disasters like floods and landslides. UAVs can be transported to the site of the disaster to mitigate problems caused by defects (refer to Fig-ure 2.3). All experiments conducted with UAVs and with a WSN-based prototype for flood detection.

Figure 2.3: When the WSN is in normal operation mode, the mes-sages with information about the monitoring of the urban river are transmitted over multihop to the sink node: 2.3 a) the sink node sends all the information to a central processing unit through an In-ternet connection; 2.3 b) when a failure occurs in a sensor node, pre-vious nodes cannot transmit their data to the sink node. [Ueyama,

2014].

To establish an emergency communications system during unexpected events of natu-ral disasters, authors in work [Tuna, 2014] proposed the value of teamwork of UAVs. Their proposed system is in the post-disaster life cycle, focusing on communication and infrastruc-ture of the UAV network. Every UAV in the team has an onboard computer which runs three main subsystems responsible for end-to-end communication, formation control, and autopi-lot system. The embedded onboard processor and the small UAV with a low-level controller cooperate to accomplish the purpose of providing local communications infrastructure. In the study mentioned above, the subsystems running on each UAV were explained and eval-uated by simulations and field tests using autonomous helicopter UAV frameworks. The study used simulations to address the efficiency of the end-to-end communication subsys-tem using UAV. Also, the field tests evaluate the accuracy of the autopilot subsyssubsys-tem. The results showed that the proposed system could be deployed in case of disasters to establish an emergency communications system. In Figure 2.4 illustrates the system specification for the multi-UAV teamwork system that proposed by work [Tuna, 2014].

Figure 2.4: System specification for the UAV-team coordination [Tuna, 2014].

In work [Morgenthaler, 2012] proposed a developed system called a "UAVNet" which is a structure of autonomous deployment of a flying Wireless Mesh Network using small quadrocopter-based UAVs as Figure 2.5 demonstrates. The flying UAVs are automatically interconnected to each other and building an IEEE 802.11 set of local area network protocols and wireless mesh network (see Figure 2.6). The developed software includes basic func-tionality to control the UAVs and to set up manage and monitor a wireless mesh network.

The experiments of flying UAVNet have revealed that the prototype proposed system can significantly improve network performance in the scenario of disaster events.

Figure 2.5:UAV framework [Morgenthaler, 2012].

Figure 2.6:UAVNet prototype [Morgenthaler, 2012].

One possible way to determine the end of the endurance of the wireless sensor network during the disaster occurrence life cycle is to set a threshold for the number of disconnections among the sensor nodes in the impacted area. When it passes this threshold, the wireless

sensor network becomes incapable of providing the quality of services. Disconnections iso-late sensors or group of sensors which cannot send the collected data, thus developing a sparse nonfunctional wireless sensor network. Although some of its isolated or grouped sensors remain operational, providing services are not guaranteed. Authors in work [Mar-inho, 2013] found a method to overcome such a deficiency which provides an alternative stable connection through other types of nodes to maintain the communication between isolated parts of a disconnected network. The method proposed multiple cooperative in-put multiple-outin-put (MIMO) techniques to support communication among static sensors in scattered wireless sensor networks and a relay network composed UAVs having the wire-less sensor networks connected, thus extending their endurance (refer to Figure 2.7). The study used simulations top-reform, and acquire results through real flight experiments and highlighted the benefits of this designed system in their work[Marinho, 2013].

Authors used simulations to perform, and acquire the results that highlight the benefits of the designed system, as shown in Figure 2.8 and Figure 2.9 with no MIMO techniques and with MIMO techniques, respectively. Their work presents an approach that merges cooperative MIMO techniques and relay networks of mobile nodes to support connectivity in a sparse wireless sensor network.

Figure 2.7:Cooperative MIMO communication between clusters of sensors and a mobile sensor [Marinho, 2013].