Reconfigurable MIMO antenna for Sub-6 GHz 5G cognitive radio with machine learning based metaheuristic optimization.

Area of Science:

• Electrical Engineering
• Wireless Communications
• Antenna Theory

Background:

• The proliferation of 5G networks and cognitive radio (CR) necessitates advanced antenna solutions.
• There is a growing demand for frequency-reconfigurable multiple-input multiple-output (MIMO) antennas for dynamic spectrum access and interference management.

Purpose of the Study:

• To present a wideband reconfigurable MIMO antenna for Sub-6 GHz 5G CR applications.
• To enable interweave and underlay operating modes for improved spectrum utilization and communication flexibility.

Main Methods:

• Frequency reconfigurability achieved using varactor diodes for continuous tuning and PIN diodes for band switching.
• Wideband operation from 2.59 to 5.09 GHz.
• Machine learning-assisted optimization using Particle Swarm Optimization (PSO), Newton Raphson Optimization (NRO), and Walrus Optimization (WO) for enhanced performance.

Main Results:

• Simulated reflection coefficient near -35 dB and peak gain of 5.6 dBi.
• Measured reflection coefficient approximately -40 dB, validating the design's practical feasibility.
• Electromagnetic simulations and measurements confirm excellent gain, isolation, and radiation performance.

Conclusions:

• The proposed reconfigurable MIMO antenna is suitable for next-generation adaptive wireless systems.
• The design offers a scalable framework for reconfigurable MIMO antennas in 5G and CR environments.
• Demonstrated ability to dynamically access spectrum while controlling interference.