An innovative Squid Game Optimizer for enhanced channel estimation and massive MIMO detection using dilated adaptive...

Area of Science:

• Wireless Communication
• Signal Processing
• Deep Learning

Background:

• Conventional signal recognition in Multiple-Input Multiple-Output (MIMO) systems faces computational challenges with increasing antennas and modulation orders.
• Deep learning offers a promising solution due to its versatility, nonlinear modeling, and parallel computation efficiency for large-scale MIMO detection.

Purpose of the Study:

• To develop a deep learning network for channel estimation and massive MIMO detection that reduces computational complexity.
• To enhance channel capacity and optimize detection performance in MIMO systems.

Main Methods:

• Channel estimation using a confusion matrix and the Modified Squid Game Optimizer (MSGO).
• Massive MIMO detection via Dilated Adaptive Recurrent Neural Network with Attention Mechanism (DARNN-AM).
• Optimization of DARNN-AM using MSGO for fine-tuning network attributes.

Main Results:

• The proposed deep learning network effectively reduces computational complexity in massive MIMO detection.
• The system demonstrates superior performance compared to existing techniques, validated through comparative analysis.
• The network achieves multi-segment mapping across various modulation schemes.

Conclusions:

• The developed deep learning approach provides an efficient solution for channel estimation and massive MIMO detection.
• The integration of MSGO and DARNN-AM optimizes performance and reduces computational load.
• This method offers a significant advancement for future high-performance wireless communication systems.