Topology-Independent MAC Performance for Long-Distance UAV Swarms: Why p-Persistent Outperforms Random Backoff

Applications for intelligent cooperative Unmanned Aerial Vehicle (UAV) swarms are rapidly expanding. Efficient and reliable communication is critical for realizing this swarm intelligence, especially in remote areas lacking infrastructure where ad hoc networking is a prevalent approach. However, in such long-distance scenarios, significant propagation delays pose a fundamental challenge to Medium Access Control (MAC) protocols like carrier sense multiple access with collision avoidance (CSMA/CA). This paper theoretically compares random backoff and p-persistent to determine the optimal strategy for these conditions. We present analytical models for both strategies. The model for random backoff reveals its optimal performance is dependent on network topology, making it ill-suited for dynamic swarms. In contrast, our model for p-persistent yields an optimal transmission probability that is independent of the network topology. Simulation results validate our models, showing p-persistent achieves significantly higher throughput (over 40% improvement in an 80-node swarm). We conclude that the topology-independent characteristic of p-persistent makes it a more feasible, more robust, and superior solution for long-distance, dynamic UAV swarm networks.