Machine Learning for Relaying Topology: Optimization of IoT Networks With Energy Harvesting

Abstract

In this paper, we examine Internet of Things (IoT) systems related to smart cities, smart factories, connected cars, etc. To support such systems in a wide area with low power consumption, energy harvesting technology utilizing wireless charging infrastructure is necessary for the longevity of networks. Considering that the position and amount of energy charged for each device could be unbalanced according to the distribution of nodes and energy sources, maximizing the minimum throughput among all nodes has become an NP-hard challenging issue. To overcome this challenge, we propose a machine learning based relaying topology algorithm with a novel backward-pass rate assessment method to present proper learning direction and an iterative balancing time slot allocation algorithm which can utilize a node with sufficient energy as the relay. To validate our proposed scheme, we conducted simulations on our established system model; thus, we confirm that the proposed scheme is stable and superior to conventional schemes.

Keywords

• Unsupervised learning
• variational autoencoder
• IoT network
• TDMA system
• energy harvesting
• relay