Multi-User Scheduling Based on Improved CNN for Massive MIMO Industrial IoT Systems

The rapid development of the 5th Generation Mobile System (5G) brings many challenges. Since the number of nodes in the Internet of Things (IoT) system, 5G adopts millimeter wave and massive multiple-input multiple-output (MIMO) technology. Due to the large number of IoT nodes, a substantial computational load will be generated in the massive MIMO system when applying the traditional multi-user scheduling algorithms. In this paper, a multi-user scheduling algorithm based on the improved Convolutional Neural Networks (CNN) is proposed. This algorithm uses the predictive characteristics of the neural network to calculate the channel capacity. The improved particle swarm optimization (IPSO) algorithm is used to find the optimal network structure parameters of the CNN to acquire the CNN with high prediction accuracy and achieve low computational complexity in the scenario with massive users accessed.