Enhanced 6G MIMO Signal Detection With Dynamic Sparse Region‐Based Convolutional Neural Network With White Shark Optimization

Abstract

Multiple‐input multiple‐output (MIMO) technology is emerging as a promising choice for upcoming sixth‐generation (6G) and beyond‐fifth‐generation (B5G) networks due to the advancements in AI. However, the complexity of MIMO systems, which involve numerous antennas, results in a challenging detection process with high computational demands. To address the interference challenges inherent in large‐scale MIMO technology, this research proposes a 6G massive MIMO signal identification algorithm that integrates a Dynamic Sparse Region‐based Convolutional Neural Network (DSR‐CNN) with White Shark Optimization (WSO) into conventional MIMO receivers. The DSR‐CNN dynamically refines proposal boxes and features, enhancing detection accuracy by effectively handling complex signal patterns. The WSO optimizes the weight parameters within the R‐CNN, improving the algorithm's convergence toward optimal solutions. This hybrid methodology significantly reduces bit error rates (BER) and enhances detection performance, making it ideally suited to the advanced requirements of 6G networks. The proposed approach, DSR‐CNN‐WSO, not only improves accuracy and efficiency but also seamlessly integrates with existing MIMO systems, offering a robust solution for future wireless communication technologies. The proposed method achieves significantly lower BER (0.01) and Mean Square Error (MSE) (0.1), indicating more reliable bit detection and more accurate signal detection.