Improved communication in contexts with limited resources is made possible by cognitive radio sensor networks (CRSNs), which merge the flexibility of cognitive radio (CR) technology with the data-gathering efficiency of wireless sensor networks (WSNs). Reliable network performance and energy conservation are two sides of the same coin in CRSNs. Low-energy adaptive clustering hierarchy (LEACH) and similar algorithms use too much energy since they do not take CRSNs’ dynamic spectrum access capabilities into consideration. This research presents a low-energy adaptive uneven clustering hierarchy (PSO-LEAUCH) technique that is optimized for CRSNs using Particle Swarm Optimization. The suggested method takes channel availability into account and makes use of uneven clustering to maximize energy distribution across cluster heads (CHs). When compared to LEACH, LEAUCH, and CogLEACH algorithms, PSO-LEAUCH considerably enhances network longevity, energy efficiency, and load balancing, according to the simulation results. Thanks to these improvements, PSO-LEAUCH is now a viable option for CRSNs looking for a dependable and energy-efficient communication solution.