Modern communication depends on wireless networks, which allow data to be transmitted across multiple devices and sites. Still, efficiently passing data through such networks is a major problem due to constantly evolving network structures, changing signal conditions, and poor routing. Current methods for routing, frequently do not manage to optimize network operation, which results in higher latency and lower throughput. Also, limited adaptive systems to address real-time network alterations contribute to reduced data performance. This paper responds to these challenges by developing a new approach called Graph-Theoretic Wireless Network Optimization (GTWNO). Graph theory and discrete structures are employed by the proposed method to present wireless networks as graphs, treating access points as nodes and signal links as weighted edges. The model improves routing by reducing delay, response time, and energy use, and by increasing the throughput of the network. Precise network modeling is achieved by the optimization process through the use of graph construction, topology preprocessing, and temporal data aggregation. Experimental data indicate that GTWNO is superior to Model A, Model B, and Model C in improving wireless network performance. The model results in a 6.8-second reduction in end-to-end delay, a 3.2-second improvement in response time, a 1.1-second drop in waiting time, a 5.2 Mbps rise in throughput, and energy consumption of only 0.95 J. The results confirm that GTWNO is effective at improving wireless network performance and deliver a reliable system for data routing and transmission.