Network remote identification (Remote ID) is projected to be an essential component in the unmanned aerial system (UAS) Traffic Management (UTM) for monitoring the UAS traffic. Since the data from the network Remote ID is transmitted via a cellular network, the signal quality of the cellular communication is crucial. This research aims to analyze the 4G signal quality used in the network Remote ID by conducting flight tests involving a small fixed-wing UAV. The flight data from structured and unstructured flight tests in two locations is analyzed using statistical and machine learning methods (Gaussian Process and Random Forest methods). For this analysis, the independent parameters are flight height, distance, and speed of the UAS, and the dependent parameter is the 4G signal quality represented by the signal-to-noise ratio (SNR), reference signal received power (RSRP), and reference signal received quality (RSRQ). The result reveals that the height and speed influence the 4G signal quality. The 4G signal quality up to a height of 200 m and speed of 35 m/s is at an acceptable level. The Random Forest method outperforms the Gaussian Process method, and the resulting model can be used to predict the 4G signal quality in other locations for trajectory planning purposes to minimize data losses during flight as this research's main contribution.