Adaptive Stream Processing Framework for Smart City IoT Networks Using Self-Powered Photodetector-based Data Collection

This paper presents the Adaptive Resource Allo-cation for Stream Processing (ARASP) framework, a novel approach integrating self-powered photodetector-based sensors with intelligent stream processing for smart city IoT networks. The framework achieves 94.8 % reduction in processing latency (45±3ms) compared to traditional systems while improving resource utilization by 89.2% (94.2% efficiency) a nd energy sustainability by 96.8 % across 14 international deployments. Our photodetector network, operating in the 200-980nm spectrum, demonstrates 85 % energy harvesting efficiency through a dual-mode system combining photovoltaic and pyro-phototronic ef-fects. The adaptive resource allocator employs machine learning algorithms that reduce operational costs by 47 % while main-taining performance stability across diverse environmental con-ditions, from −30°C to +48°C. Comparative analysis with existing frameworks shows ARASP outperforms current solutions by 45% in latency reduction, 35 % in resource optimization, and 28 % in adaptation capabilities. Six-month stability testing confirms consistent performance across variable workloads, establishing ARASP as a viable solution for sustainable smart city infras-tructure with significant improvements i n both computational efficiency and environmental adaptability.