An in-depth assessment of the physical layer performance in the proposed B5G framework

The introduction of fifth-generation (5G) technology marks a significant milestone in next-generation networks,
offering higher data rates and new services. Achieving optimal performance in 5G and beyond 5G (B5G)
systems requires addressing key requirements like increased capacity, high efficiency, improved performance,
low latency, support for many connections, and quality of service. It is well-known that suboptimal network
configuration, hardware impairments, or malfunctioning components can degrade system performance. The
physical layer of the radio access network, particularly channel estimation and synchronization, plays a crucial
role. Hence, this paper offers an in-depth evaluation of the 5G Physical Downlink Shared Channel (PDSCH),
along with its related channel models such as the Clustered Delay Line (CDL) and the Tapped Delay Line
(TDL). This work assesses 5G network performance through practical and IA-based channel estimation and
synchronization techniques, and anticipates numerologies for B5G networks. Extensive simulations leveraging
the Matlab 5G New Radio (NR) toolbox assess standardized channel scenarios in both macro-urban and indoor
environments, following configurations set by the 3rd Generation Partnership Project (3GPP). The numerical
results offer valuable insights into achieving the maximum achievable throughput across various channel
environments, including both line-of-sight (LoS) and non-line-of-sight (NLoS) conditions. The throughput
comparisons are performed under assumptions of ideal, realistic, and convolutional neural networks (CNN)-
based channel estimation with both perfect and realistic synchronization conditions. Importantly, the study
pinpoints certain physical layer elements that have a pronounced impact on system performance, providing
essential insights for devising effective strategies or refining CNN-based methods for forthcoming mobile B5G
networks.