UAV Network for Surveillance of Inaccessible Regions with Zero Blind Spots

Unmanned aerial vehicles (UAVs) or drones are an important component of 5G and beyond 5G communications network. In this paper, we have proposed the deployment of UAV network for video surveillance over an inaccessible region that lacks cellular infrastructure. The remote surveillance of such a region is facilitated by streaming the surveillance video via the ground control station (GCS) connected using a 5G backhaul. We have developed the UAV wireless network and obtained the optimal topology of UAV grid (with the the least number of drones) for such a surveillance requirement. We achieve zero blind spots during surveillance by obtaining a suitable topology of UAV grid with minimum number of UAVs using the concept of Field of View (FOV) of the on-board camera. A wireless network for the UAV grid is set up using Raspberry Pi mounted on each UAV. The inter-UAV distance in the network has been fixed based on acceptable packet loss and Quality-of-Experience (QoE) for video streaming. The results achieved from the experimental setup showed that the QoE for the video streaming falls below acceptable limits at 4% packet loss. Hence, the inter-UAV distance should be such that the packet-loss does not exceed the threshold limit of 4%. The paper also brings out the necessary conditions to avoid overlap in coverage area between neighbouring UAVs, thereby, optimising the number of drones to cover an area of interest.