Implementation and Analysis of Channel Assessment for AMC in 5G Networks Using Deep Learning methods

An essential component of 5G is channel estimation. Both OFDM and AMC require CE to achieve optimal performance. We present a deep learning-based approach to determine channels in OFDM networks that addresses a variety of channel models. Here, we investigate the deployment of deep Learning for CE in OFDM 5G, distinctively under the non-orthogonal user access technique called Cooperative Non-Orthogonal Multiple Access (CNOMA). Accurate channel estimation (CE) in Orthogonal Frequency Division Multiplexing (OFDM) 5G systems is critical for reliable data transmission. A number of methods are used to estimate channels, including Decision-Directed Channel Estimation (DDE), Pilot-Assisted Channel Estimation (PACE), and Blind Channel Estimation (BCE). The pivotal role of this work is to evaluate the efficacy of DL-based CE compared to the basic PACE approaches. We investigate the performance of the proposed approach under different channel models, including Rayleigh fading, Rician fading, and multipath propagation.