Citation

Barua, Hrittik Raj and Dey, Arpan and Khan, Mohammad Rakibul Hoque and Nandy, Oishi and Dewanjee, Joy (2026) Artificial Neural Network Based Optimization of a Miniaturized 28 GHz Antenna for 5G Communication. In: 2026 IEEE 2nd International Conference on Quantum Photonics, Artificial Intelligence & Networking (QPAIN), 16-18 April 2026, Chittagong, Bangladesh.

Text 13.pdf - Published Version Restricted to Repository staff only Download (697kB)

Abstract

This paper presents an Artificial Neural Network (ANN) aided framework for the optimization of a compact 28 GHz microstrip patch antenna designed for fifth generation (5G) communication applications. The antenna is designed on a Rogers RT5880 substrate (εr=2.2, thickness =0.508mm) with a compact size of 9×8mm2. An ANN regression model is established to efficiently predict the reflection coefficient by accurately correlating the geometrical parameters of the antenna. The model is trained and tested using the CST simulation results and the prediction errors are low with the training and testing MSE, RMSE of 0.8,0.9 and 0.2,0.5 respectively and the R2 value is 0.9 which shows high learning efficiency and generalization. The optimized antenna demonstrates superior impedance matching with a minimum S11 of -56.36 dB at 28 GHz, a gain of 6.79 dBi and a radiation efficiency of 85%. The proposed ANN aided framework offers an efficient and scalable solution for high performance mmWave antenna design in 5G communication.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	mmWave, 5G communication, ANN, MSE
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	31 Jul 2026 07:20
Last Modified:	31 Jul 2026 07:20
URII:	http://shdl.mmu.edu.my/id/eprint/16431

Downloads

Edit (login required)