Minimizing the Deployment Cost of UAVs for Delay-Sensitive Data Collection in IoT Networks


In this paper, we study the deployment of Unmanned Aerial Vehicles (UAVs) to collect data from IoT devices, by finding a data collection tour for each UAV. To ensure the `freshness' of the collected data, the total time spent in the tour of each UAV that consists of the UAV flying time and data collection time must be no greater than a given delay B, e.g., 20 minutes. In this paper, we consider a problem of deploying the minimum number of UAVs and finding their data collection tours, subject to the constraint that the total time spent in each tour of any UAV is no greater than B. Specifically, we study two variants of the problem: one is that a UAV needs to fly to the location of each IoT device to collect its data; the other is that a UAV is able to collect the data of an IoT device if the Euclidean distance between them is no greater than the wireless transmission range of the IoT device. For the first variant of the problem, we propose a novel 4-approximation algorithm, which improves the best approximation ratio 4 4/7 for it so far. For the second variant, we devise the very first constant factor approximation algorithm. We also evaluate the performance of the proposed algorithms via extensive experiment simulations. Experimental results show that the numbers of UAVs deployed by the proposed algorithms are from 11% to 19% less than those by existing algorithms on average.


Computer Science

Publication Status

Early Access


The work of Sajal K. Das was supported in part by NSF Grants under Award CCF-1725755, Award CNS-1818942, Award SCC-1952045, and Award SaTC-2030624.

Keywords and Phrases

Approximation Algorithms; Approximation Algorithms.; Data Collection; Dispatching; Minimum Cycle Cover with Neighborhoods; Minimum Numbers of UAV Deployments; Mobile Data Collection; Monitoring; Multiple UAV Scheduling; Sensors; Unmanned Aerial Vehicles; Wireless Communication

International Standard Serial Number (ISSN)

1558-2566; 1063-6692

Document Type

Article - Journal

Document Version


File Type





© 2021 Association for Computing Machinery (ACM), All rights reserved.

Publication Date

13 Nov 2021