Departamento de Computação, Universidade Federal de Ouro Preto, Rua Diogo de Vasconcelos, 122, Bairro Pilar, Ouro Preto 35400-000, Brazil.
Departamento de Tecnologia em Engenharia Civil, Computação, Automação, Telemática e Humanidades, Universidade Federal de São João Del Rei, Campus Alto Paraopeba-C.A.P, Rod.: MG 443, KM 7, Ouro Branco 36420-000, Brazil.
Sensors (Basel). 2020 Dec 5;20(23):6954. doi: 10.3390/s20236954.
The benefits of using mobile sinks or data mules for data collection in Wireless Sensor Network (WSN) have been studied in several works. However, most of them consider only the WSN limitations and sensor nodes having no more than one data packet to transmit. This paper considers each sensor node having a relatively larger volume of data stored in its memory. That is, they have several data packets to send to sink. We also consider a drone with hovering capability, such as a quad-copter, as a mobile sink to gather this data. Hence, the mobile collector eventually has to hover to guarantee that all data will be received. Drones, however, have a limited power supply that restricts their flying time. Hence, the drone's energy cost must also be considered to increase the amount of collected data from the WSN. This work investigates the problem of determining the best drone tour for big data gathering in a WSN. We focus on minimizing the overall drone flight time needed to collect all data from the WSN. We propose an algorithm to create a subset of sensor nodes to send data to the drone during its movement and, consequently, reduce its hovering time. The proposed algorithm guarantees that the drone will stay a minimum time inside every sensor node's radio range. Our experimental results showed that the proposed algorithm surpasses, by up to 30%, the state-of-the-art heuristics' performance in finding drone tours in this type of scenario.
在无线传感器网络 (WSN) 中，使用移动水槽或数据骡子进行数据收集的好处已经在一些研究中得到了研究。然而，它们大多数只考虑了 WSN 的限制和传感器节点没有超过一个数据包要传输。本文考虑每个传感器节点在其内存中存储了相对较大的数据量。也就是说，它们有几个数据包要发送到 sink。我们还考虑了具有悬停能力的无人机，如四旋翼飞机，作为移动收集器来收集这些数据。因此，移动收集器最终必须悬停以确保所有数据都将被接收。然而，无人机的电源有限，限制了它们的飞行时间。因此，必须考虑无人机的能源成本，以增加从 WSN 收集的数据量。这项工作研究了在 WSN 中确定最佳无人机巡回飞行以进行大数据收集的问题。我们专注于最小化从 WSN 收集所有数据所需的总体无人机飞行时间。我们提出了一种算法，用于在无人机移动过程中创建传感器节点的子集来发送数据，并相应地减少其悬停时间。所提出的算法保证了无人机将在每个传感器节点的无线电范围内停留最短的时间。我们的实验结果表明，在所研究的这种情况下，所提出的算法在找到无人机巡回飞行方面的性能超过了最先进启发式算法的性能，最高可达 30%。
