Mitigating broadcast storm problem and enhancing dissemination in swarm UAV communication

Abstract

The rapidly evolving security landscape of the 21st century has significantly increased global investment in the defense industry. The rise of asymmetric threats, the growing need for rapid decision-making, and the demand for low-cost operational solutions have all intensified interest in next-generation technologies within this field. In this context, unmanned aerial vehicles (UAVs) have emerged as flexible platforms capable of carrying out a wide range of tasks such as reconnaissance, surveillance, attack, communication, and logistics in both civilian and military scenarios. Seeking to move beyond the capabilities of individual UAVs, researchers and users have developed new approaches aimed at enabling these systems to operate in a coordinated, collective, and autonomous manner. Inspired by nature, the concept of swarm systems has facilitated more powerful and effective formations, where unmanned platforms can make decisions together, share tasks, adapt to dynamic environments, and execute missions with high flexibility. Features such as decentralized control architecture, autonomous navigation, AI-assisted decision-making, and real-time mission reconfiguration allow these systems to generate a multiplier effect in critical scenarios, including wide-area search and rescue, electronic warfare, reconnaissance, border security, and disaster response. Globally, work in this area has accelerated. The United States has conducted micro-UAV swarm trials, Germany has developed the AirShield project for information gathering in disaster zones, and China has demonstrated large-scale UAV swarm flights. These efforts indicate that swarm technology is no longer just a research concept but is evolving into a field-ready operational capability. Collectively, these developments signal that UAV swarms are poised to become one of the key defense technologies in the near future. However, realizing the full potential of these systems depends on ensuring that intra-swarm communication remains secure, efficient, and uninterrupted. Each UAV within the swarm must be capable of perceiving its environment and sharing that perception with others to support collective decision-making processes. At this point, the concept of Flying Ad Hoc Networks (FANETs) becomes central. Unlike traditional communication networks, FANETs are characterized by high mobility, decentralized control, and constantly changing topologies. In this structure, each UAV functions not only as a data-generating end node but also as an intermediate relay that forwards data to other UAVs. This enables the network to be self-organizing, distributed, and scalable. While the FANET architecture allows the swarm to share not only positional data but also speed, task status, orientation, and threat perception in real time, it also introduces significant communication challenges. In particular, when broadcast-based transmission strategies are employed, the repeated forwarding of the same message by multiple UAVs can cause substantial network congestion. This situation gives rise to three main problems, often referred to collectively in the literature as the "broadcast storm" issue: Contention occurs when multiple UAVs attempt to access the communication channel simultaneously, leading to delays and transmission conflicts. Collision results from these simultaneous transmissions interfering with each other, corrupting the data and requiring retransmission. Redundancy refers to the repeated transmission of identical messages by different nodes, wasting energy resources and reducing the efficiency of information dissemination. Addressing these three intertwined problems is crucial not only from a technical standpoint but also for maintaining the reliability, responsiveness, and energy efficiency of swarm communication systems. Developing effective communication strategies for FANETs is, therefore, indispensable for the overall success of UAV swarms. This thesis presents a two-stage solution approach aimed at mitigating the broadcast storm problem encountered during intra-swarm communication. The primary objective of the proposed system is to reduce unnecessary message transmissions in order to alleviate network load, while ensuring that critical information is delivered to all swarm members in a secure and efficient manner. In the first stage, a communication algorithm is developed to minimize redundant message forwarding within the swarm. Each UAV calculates its local situational awareness using an internal data structure referred to as the InfoList, which stores previously received messages. By keeping track of when, from whom, and what content was received, UAVs are able to evaluate the temporal and contextual attributes of incoming messages. If a message with identical content is received again within a short period, it is suppressed to prevent unnecessary retransmission. Furthermore, by estimating which neighbors have already received the message, the algorithm determines whether additional forwarding is truly needed. As a result, only essential transmissions are performed, which significantly reduces the total number of broadcasts in the network, improves energy efficiency, and helps limit contention on the communication channel. In the second stage, a dissemination algorithm is introduced to ensure that important information can be effectively propagated across the entire swarm. This mechanism enables each UAV to estimate its coverage potential based on its known set of neighbors and to evaluate which new UAVs can be reached if a broadcast is performed. Each UAV constructs a broadcast plan by analyzing its current position, previously received messages, and the known coverage status of its neighbors. This plan identifies which nodes are best suited to forward the message in a way that maximizes propagation speed while minimizing repetition. Only the selected nodes are allowed to rebroadcast, resulting in an efficient and scalable dissemination strategy. The performance of the developed algorithm was evaluated through a series of simulation scenarios conducted in the MATLAB environment. The results indicated improvements in overall network efficiency, energy balance, and the success rate of information dissemination. Especially in large-scale swarm configurations, the proposed method was observed to provide a more balanced and practically applicable communication strategy compared to existing approaches.