Flexible Deep Learning in Edge Computing for Internet of Things

Deep learning is a promising approach for extracting accurate information from raw sensor data from IoT devices deployed in complex environments. Because of its multilayer structure, deep learning is also appropriate for the edge computing environment. Traditional edge computing models have rigid characteristics. Flexible edge computing architecture solves rigidity in IoT edge computing. Proposed model combines deep learning into edge computing and flexible edge computing architecture using multiple agents. Since existing edge nodes have limited processing capability, we also design a novel offloading strategy to optimize the performance of IoT deep learning applications with edge computing. FEC architecture is a flexible and advanced IoT system model characterized by environment adaptation ability and user orientation ability. In the performance evaluation, we test the performance of executing deep learning tasks in FEC architecture for edge computing environment. The evaluation results show that our method outperforms other optimization solutions on deep learning for IoT