Differential Spatial Modulation Detection in Uplink Multiuser Massive MIMO Systems Using Optimized Shuffle Attention Convolutional Neural Network With MobileNet V1

ABSTRACT A significant interference challenge is presented by the upcoming sixth generation (6G) and beyond of wireless communication due to the increasing presence of ultra‐scale intelligent factors, such as smart vehicles and mobile robots. Managing this interference can be difficult for detection algorithms in uplink massive multiple‐input and multiple‐output (MIMO) systems, particularly when dealing with higher‐order quadrature amplitude modulation (QAM) signals. Differential spatial modulation (DSM) detection using deep learning (DL) has been a fundamental solution for effectively managing interference in 6G and enhancing the uplink performance of the MIMO system. In this paper, DSM detection in uplink multiuser massive MIMO systems using optimized shuffle attention convolutional neural network with MobileNet V1 (SMD‐MIMO‐SACNN‐MNV1) is proposed. The proposed SACNN‐MNV1 technique is used to detect the DSM in uplink multiuser massive multiple‐input multiple‐output (M‐MIMO) systems. Artificial lizard search optimization algorithm (ALSOA) is utilized to enhance the SACNN‐MNV1 detector that detects the DSM precisely. The proposed SMD‐MIMO‐SACNN‐MNV1 method is executed in Python. The proposed method achieves 23.54%, 22.65%, and 23.18% higher spectral efficiency when compared with existing techniques respectively.