Edge Computing–Driven Framework for Scalable, Low-Latency, and Privacy-Aware Smart City Applications

Abstract

Smart city ecosystems rely on large-scale Internet of Things (IoT) deployments to support applications such as intelligent transportation, smart energy management, urban surveillance, and environmental monitoring. These applications generate continuous, high-volume, and latency-sensitive data streams that challenge traditional cloud-centric computing paradigms. Centralized architectures suffer from communication delays, scalability bottlenecks, and significant privacy risks due to the transmission of fine-grained data to remote servers. This paper presents a comprehensive edge computing–driven framework derived from doctoral research to address these limitations. The proposed framework distributes computation across device, edge, and cloud layers, enabling localized intelligence while maintaining global coordination. Extensive analytical evaluation demonstrates that the edge-enabled approach significantly reduces end-to-end latency, improves scalability under increasing device density, and mitigates privacy exposure compared to cloud-centric and fog-based architectures. The findings establish edge computing as a foundational enabler for next-generation smart city infrastructure.