A Machine Learning Approach for SNR Prediction in 5G Systems.

2019 
Channel State Information (CSI) feedback from the User Equipment (UE) on the uplink (UL) channel is an integral part of the 5G-NR standard. It allows next-generation Node-B (gNB) to obtain the information about the channel impairments, and schedule appropriate radio resources to the UE in order to maintain the required Quality of Service (QoS). Errors in the channel estimation by the UE increases erroneous downlink transmission and subsequently inefficient spectral use. There is an inherent tradeoff between CSI feedback periodicity from the UE and the accuracy and relevance of channel estimation within the periodicity by gNB. Since the CSI depends on the received Signal-to-Noise Ratio (SNR) at the UE. Hence, the goal of this work is to design and develop suitable methodologies to estimate a CSI parameter namely, Channel Quality Indicator (CQI) by predicting the SNR using the state-of-the-art Machine Learning (ML) techniques. In this paper, we present an experimental framework to predict CQI using SNR for different environmental scenarios and UE characteristics like delay profile and speed respectively. The experimental results show better performance in terms of CQI to SNR mapping, and error in prediction over current techniques.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    13
    References
    6
    Citations
    NaN
    KQI
    []