Deep learning-driven multi-user wavelet NOMA for user centric cell-free massive MIMO communications

The evolution of Cell-Free Massive MIMO (CF-mMIMO) systems in the sixth-generation (6G) communications brings notable benefits including enhanced capacity, broader coverage, and greater reliability. However, these advanced systems may be subjected to critical challenges like, exponential growth in the user connectivity, precise channel estimation, and effective mitigation of the inter-user interference. This article addresses these challenges through Deep Learning (DL) for accurate channel estimation and a robust Wavelet Transform based Non-Orthogonal Multiple Access (NOMA) scheme to mitigate the inter-user interference in a user-centric CF-mMIMO system. By eliminating the reliance on pilot-assisted channel estimation, the DL-based approach achieves higher accuracy and lowers transmission overhead in a multi-user scenario. The results highlight the superiority of DL-based channel estimation for a CF-mMIMO system employing wavelet NOMA scheme over traditional methods, showing a 17% reduction in bit error rate (BER) and a 15% improvement in achievable sum-rate.