Mutual-Interference-Aware Throughput Enhancement in Massive IoT: A Graph Reinforcement Learning Framework

As the number of devices increases dramatically in the Internet of Things (IoT), features of dense deployment of massive devices generate mutual interference in communication overlapping areas, which will impose an imperative challenge on spectrum resource allocation. To handle this challenge, this article proposes a mutual interference-aware throughput enhancement scheme. For the mutual interference among multiple IoT devices, this scheme first builds an interference hypergraph model to quantify the impact of the mutual interference for each device. According to the main goal of the spectrum resource allocation, this article formulates a graph reinforcement learning (GRL) framework, whose action space is multidimensional discrete, and the reward function is designed to enhance the throughput and mitigate the impact of interference. Then, a graph convolutional network-double dueling deep Q-network-based spectrum resource allocation algorithm is developed upon the proposed GRL framework to extract the mutual interference information from the hypergraph model, and then achieve a dynamic resource allocation for massive IoT. Simulation results prove that the proposed GRL algorithm effectively improves the network throughput compared to the comparison algorithms.

I. Introduction

With the number of wireless devices and traffic demands increasing dramatically, massive Internet of Things (IoT) [1] is poised to become indispensable in the future smart cities and other domains [2], [3], [4], which are envisioned to seamlessly connect a large number of IoT devices across diverse environments. These significant challenges arise when massive devices communicate through a single access point (AP) [5], such as a base station (BS) [6], including substantial energy consumption and heightened latency. To handle these challenges, device-to-device (D2D) communication has been considered one of the most promising technologies in the massive IoT to realize direct data transmission among devices in proximity to each other [7]. It also allows cooperation between machine-type devices and user equipment. In general, a large number of IoT devices are densely deployed close to each other with the characteristics of overlapping coverage among multiple devices’ communication ranges in massive IoT [8], [9], [10]. Meanwhile, due to the limited available spectrum resources for data transmission, intensive mutual interference will inevitably be caused when a D2D receiver receives interference from multiple co-channel D2D transmitters in the overlapping coverage area [11]. With an enormous increase in the number of devices, the limited available spectrum resources have become more strained [12], and this mutual interference will further deteriorate the entire network performance. Therefore, in this case, how to allocate limited available spectrum resources to ensure the entire data transmission for massive devices is of great importance in such D2D communications underlaid massive IoT with mutual interference.