Franklin
Edge Deep Learning and Computational Offloading in IoT Systems examines the integration of deep learning, edge computing, and computational offloading for intelligent Internet of Things environments. The book explores how machine learning workloads can be processed closer to IoT devices to reduce dependence on centralized computing resources and support efficient data processing. It introduces key concepts in edge intelligence, deep neural networks, IoT architectures, distributed computing, and computational offloading. Attention is given to the relationship between device capabilities, workload distribution, communication requirements, latency, energy consumption, and computational resources. The discussion considers approaches for assigning computational tasks between IoT devices, edge nodes, and other computing resources while supporting the execution of deep learning workloads. It also addresses important considerations in system design, including resource management, processing efficiency, network constraints, and performance evaluation. The book provides a technical foundation for students, researchers, computer engineers, network professionals, and practitioners working in edge computing, artificial intelligence, distributed systems, and IoT technologies.
