A PERFORMANCE IMPROVEMENT TECHNIQUE USING MACHINE LEARNING IN A SIMULATION-BASED 5G-SMALL CELL MOBILE COMMUNICATION SYSTEM

[ 30 Jun 2021 | vol. 14 | no. 2 | pp. 13-20 ]

About Authors:

Yoon-Hwan Kim1, Sang-Hyun Bae2 and Tae-Yeun Kim3*
-1Department of computer Science & Statistics, Chosun University, Gwang-Ju, Korea
-2Department of computer Science & Statistics, Chosun University, Gwang-Ju, Korea
-3National Program of Excellence in Software center, Chosun University, Gwang-Ju, Korea

Abstract:

The 5G mobile communication system is aiming for high-speed data transmission, low latency, and acceptance of many devices compared to 4G-LTE systems. To implement this, 5G communication uses a high frequency band and inevitably suffers from high path loss. Small cell technology, which has been introduced to overcome these shortcomings, has advantages such as coverage extension and shadow area elimination as a small, low-power base station, but in order to solve the effects of interference and deviation of access devices due to large number of small cell deployments. Algorithm development for service methods is inevitably required. In this paper, a decision about the small cell connection was made using a machine learning algorithm, and the performance improvement of the machine learning algorithm was confirmed by comparing the application of the 5G macro system and the small cell, and the application of the machine learning algorithm. As a result of the application of dual training methods in machine learning training, the improvement of machine learning algorithm.

Keywords:

Arsenic, Concentration, Water masses, Techniques

 

About this Article: