SQID: A deep learning and network design synergy for next-generation IoT resource allocation management

Abstract

The exponential growth of mobile broadband and Internet of Things (IoT) devices has pushed traditional IoT models to their operational limits, necessitating more efficient data management strategies. This research introduces the SQID framework, a solution that integrates advanced techniques, including Sierpinski triangle design (STD) for network optimization, quantum density peak clustering (QDPC) for intelligent device clustering, and improved deep deterministic policy gradient (IDDPG) for deep learning-driven traffic prediction. By utilizing STD to optimize device communication, the framework applies the QDPC algorithm to efficiently cluster devices, ensuring balanced packet distribution and minimizing latency. Additionally, IDDPG enhances network performance by enabling accurate traffic prediction and resource allocation, optimizing data transmission. Extensive simulations reveal that SQID outperforms existing methods in critical metrics such as time efficiency, latency reduction, throughput maximization, and packet loss. These results indicate that SQID has the potential to significantly improve data management in IoT networks, paving the way for next-generation IoT advancements.