Edge AI enabled MIMO MC-CDMA for 6G optimizing spectrum and energy efficiency with SIC and deep reinforcement...

Area of Science:

• Wireless Communication Engineering
• Artificial Intelligence in Networks
• Signal Processing

Background:

• The expansion of 6G wireless networks necessitates advanced techniques for improving spectral efficiency (SE), energy efficiency (EE), and managing interference.
• Existing systems like MIMO-OFDM and hybrid precoding face challenges in dense network environments.

Purpose of the Study:

• To propose an Edge AI MIMO MC-CDMA system integrated with Successive Interference Cancellation (SIC) and Deep Reinforcement Learning (DRL) to enhance SE and EE.
• To demonstrate the adaptive learning capabilities of DRL for optimizing network parameters in real-time at the network edge.

Main Methods:

• Implementation of an Edge AI MIMO MC-CDMA system utilizing DRL for adaptive decision-making on networking parameters.
• Leveraging Edge AI to minimize computational load and ensure rapid responses, with DRL facilitating interference cancellation.
• System simulation and performance evaluation using MATLAB.

Main Results:

• Achieved spectral efficiency of approximately 32.7 bits/s/Hz and energy efficiency of 14.8 bits/Joule, outperforming traditional systems.
• Enhanced Signal-to-Interference-plus-Noise Ratio (SINR) to 34 dB and reduced Bit Error Rate (BER) to 10^-5.
• Demonstrated effective interference management and optimized power allocation through the deep learning-based mechanism.

Conclusions:

• The proposed Edge AI MIMO MC-CDMA system with DRL offers significant improvements in spectral and energy efficiency for 6G networks.
• The system exhibits high scalability, minimal interference, and superior energy efficiency, making it suitable for ultra-high-density scenarios.
• The dynamic adaptability of the model positions it as a key approach for smart wireless communication in next-generation networks.