Muhammad Nizam, Ramzi Adriman, Yunida Yunida, Nasaruddin Nasaruddin
The evolution of wireless communication technology is inextricably linked to the growing demand from consumers to connect multiple devices to the network. Massive multiple input multiple output (MIMO) is a new advancement compared to conventional communication systems, which use tens to hundreds of interconnected antennas that serve multiple user terminals at the same time and frequency. 5G's massive MIMO system performance degrades as a result of its extensive use of antennas; consequently, it is characterized by high complexity and wasteful energy consumption. The sum rate is an essential system performance parameter in 5G massive MIMO. Deep learning (DL) is utilized to advance and enable the maximum potential of 5G massive MIMO in order to increase the sum rate. DL can process, generate, and provide sophisticated analysis on virtually any given task, with concise and verifiable outcomes. Consequently, this paper analyzes the 5G massive MIMO sum rate using a deep neural network (DNN). The proposed DNN technique employs seven hidden layers per node (512, 256, 128, 128, 32, 5, and 6) and Relu activation. The aggregate rate was obtained by simulating the 5G massive MIMO system model with the DNN. The simulation results demonstrated that the DNN algorithm was able to produce a high sum rate value and is superior to conventional 5G massive MIMO. © 2023 IEEE.
Department of Electrical Engineering and Computer, Universitas Syiah Kuala, Banda Aceh, Indonesia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock