Seeing is Better Than Hearing: Sensing-Assisted Efficient Routing Scheme for UAV Network

The geographic routing protocol is pivotal in flying ad-hoc networks as it eliminates the need for sharing and storing extensive topology information. However, traditional methods that rely solely on Hello packets for neighbor discovery and routing decisions often suffer from low efficiency. This paper introduces a sensing-assisted efficient geographic routing scheme (SAEGRS) that utilizes computer vision technology to detect and maintain neighbor information. The scheme generates a unique physical identity (P-ID) based on relative entropy and node feature popularity, reducing the need for frequent beacon exchanges by correlating IP addresses with P-IDs. Additionally, SAEGRS integrates link quality, neighbor reliability, and distance to the destination in the relay selection process. Real-world experiments confirm the effectiveness of the proposed sensing method, while simulation results demonstrate that SAEGRS significantly reduces network overhead and latency without compromising packet delivery rate.