Estimating Channel Quality Indicator in 5G NR V2X Using Deep Learning

The 5th generation of vehicular to everything (V2X) technology has been developed to handle the expanding current traffic congestion and the demand for data flow, which necessitate exceptionally high throughput and low latency to sustain a stable connection. The channel estimate is crucial to upholding this criterion, which aids the user equipment (UE) in selecting the necessary modulation and coding scheme to handle the rapidly shifting V2X environment. In this research, a potential method for predicting channel quality indicator (CQI) using deep learning models that were built by utilizing a real-world 5G environment dataset is proposed. These models are based on long short-term memory (LSTM) and gated recurrent unit (GRU), which use the RSSI, SNR, and velocity of the vehicle to estimate the CQI values. The outcomes of these approaches indicate the possibility of an effective approach for maximizing the use of radio resources in a real-world V2X context.