Optimized Cooperative Spectrum Sensing in Cognitive Radio Sensor Networks Using Extended XGBoost Model

Cognitive radio (CR) is a transformative technology designed to address the growing demand for spectrum resources.By intelligently sensing the radio environment, CR devices identify unused spectrum bands and dynamically adapt their transmission parameters to efficiently utilize these resources.Cooperative Spectrum Sensing (CSS) is a crucial method used in cognitive radio networks to enhance spectrum sensing reliability and accuracy.By collaborating, multiple secondary users (SUs) improve their ability to detect primary user (PU) activity and identify available spectrum opportunities.In a traditional cooperative spectrum sensing paradigm, each user independently senses the spectrum and reports the local decision to the fusion center.The fusion center combines these individual decisions to make a more accurate global decision about spectrum availability.Such a static approach is unsuitable for achieving cooperative gain and adapting to a large cognitive radio sensor network (CRSN).The primary challenges are reduced sensing accuracy, limited scalability, and increased communication overhead because of unreliable or redundant local decisions.These issues necessitate the development of more adaptive and optimized methods for the efficient utilization of spectrum in dynamic environments.Thus, the proposed study presented effective Optimized cooperative spectrum sensing in CRSN (OCSS-CRSN).The proposed framework performs a clustering process using novel enhanced agglomerative hierarchical clustering (EAHC).The cluster heads are then selected using a novel chaotic pelican optimization algorithm (CPOA).Then, FC combines the local decisions made by cluster heads to reach a final decision using a novel machine learningbased extended XGBoost model (E-XGBoost).Finally, simulation results show that the proposed CSS algorithm achieves a throughput of 50.91kbps in static CRSNs and 53.33kbps in static CRSNs while maintaining CSS performance.