Statistical Characteristic of Link Quality Metrics in UAV assisted Low Power Wireless Networks

With the maturity of unmanned aerial vehicle (UAV), it has been extensively used in low power wireless networks field. The introduction of UAV will worsen the link stability in UAV assisted low power wireless networks. Therefore, choosing stable and reliable links is of critical importance for these networks. However, there is no statistical analysis of link quality metrics in existing works. Therefore, this paper analyzes the statistical characteristics of link quality metrics for UAV assisted low power wireless networks through extensive experiments. In particular, statistical characteristics of uplinks and downlinks are compared comprehensively. Experimental results show that packet reception ratio has the capability to distinguish link quality, while the average value of physical layer metrics is only applicable to identifying good links. Standard deviation of packet reception ratio is able to identify moderate links, while that of physical layer metrics cannot. However, compared with PRR, the size of time window has less influence on physical layer parameters. Therefore, when the link quality changes rapidly or it needs to be judged quickly, using physical layer metrics will be more effective in UAV assisted low power wireless networks. Moreover, compared with the downlink, fluctuations of PRR and physical layer metrics are greater in the uplink.