Optimization and Algorithm Research of Intelligent Monitoring System Combining Artificial Intelligence and Internet of Things

Abstract

As smart city construction becomes an accelerating global trend, with the aim of enhancing urban living quality, ensuring sustainability, and enabling efficient urban management, IoT technology, a core infrastructure, is increasingly used in intelligent monitoring systems. However, existing systems often face issues like low resource utilization, high latency, and poor energy efficiency when handling large - scale data transmission and complex computing tasks. These problems impede the seamless operation of crucial smart city services such as real - time traffic control, environmental monitoring for pollution prevention, and efficient waste management. To address these challenges, this article proposes an optimized solution. First, a multimodal data fusion framework is designed to efficiently process IoT - collected data. Next, an LSTM network is employed to capture temporal dependencies. To further optimize resource scheduling, the PPO (Proximal Policy Optimization) algorithm is used for adaptive resource allocation. Finally, by leveraging edge computing, tasks are distributed to edge nodes, reducing network latency. Experimental results show our method significantly reduces response time and improves throughput in large - scale data processing. For example, with 20,000 data entries, the optimized response time is 0.28 seconds. Additionally, resource utilization and energy efficiency are notably enhanced, demonstrating the superiority of the proposed approach in resource scheduling, real - time processing, and system stability for the successful development of smart cities.