This thesis addresses the problem of reliable communication in swarm unmanned aerial vehicle (S-UAV) networks operating in crisis scenarios where communication infrastructure may be damaged and cluster heads (CHs) may become non-functional. Existing cluster-based routing schemes often suffer from communication disruption, packet loss, and network partitioning when CH failures occur, reducing their effectiveness in emergency and disaster-response environments. To address these limitations, this research proposes a progressive family of clustering protocols designed to improve communication reliability and resilience. First, the Redundant Weighted Cluster Routing Protocol (RWP) introduces weighted CH selection based on inter-node distance, cooperation-based rewarding index, and relative velocity, while incorporating a redundant CH mechanism to mitigate single-point failures. Second, the Multi-Redundant Weighted Cluster Routing Protocol (MRWP) extends the approach by integrating residual energy into the clustering decision and supporting multiple redundant CHs to enhance network survivability. Third, the Dynamic Multi-Redundant Weighted Cluster Routing Protocol (DMRWP) introduces adaptive weight adjustment that dynamically modifies clustering priorities according to network conditions, improving performance under changing environments. Finally, the Multi-Agent Dynamic Multi-Redundant Weighted Cluster Routing Protocol (MADMRWP) expands the framework to heterogeneous crisis infrastructures by enabling cooperation among UAVs, vehicles, and terrestrial communication agents. 3 The proposed protocols were evaluated through extensive simulations using communication metrics including packet delivery ratio, end-to-end delay, packet loss, forwarding efficiency, throughput, recovery time, and energy efficiency. Results demonstrate that the proposed schemes significantly improve network resilience and communication reliability in the presence of non-functional nodes and damaged infrastructures. The inclusion of redundancy mechanisms reduces communication disruption, dynamic adaptation enhances performance stability, and multi-agent collaboration improves data delivery and recovery capabilities in complex crisis environments. The research contributes a comprehensive clustering framework for reliable multi-agent crisis communications and demonstrates the effectiveness of redundancy, adaptive clustering, and heterogeneous agent cooperation in ensuring successful data transmission under adverse conditions.
Main Files
Full text Description: PhD_Thesis_Grace_Khayat.pdf (pdf)
Book Reader
