AI-Enhanced Pilot-Assisted Channel Estimation and Signal Detection for 6G with Zero CSI

This paper proposes a feed-forward neural network-based artificial intelligence (AI)-enhanced system for pilot-assisted channel estimation and signal detection in sixth-generation MIMO wireless communication, operating without channel state information (CSI). Least square estimation is employed to determine channel characteristics using pilot symbols. These characteristics are then utilized by the AI model to estimate the channel, which, combined with received data symbols, enables signal detection via AI model. The proposed model has shown some promising results for the future research in MIMO technology. Overall, this paper stands as a comprehensive guide for the future developments in channel estimation and signal detection at zero CSI.