An apparatus for performing a wireless communication includes a communication interface configured to measure uplink (UL) Sounding Reference Signals (SRSs) transmitted from a mobile client device, and at least one processor configured to buffer a number of uplink (UL) SRS measurements derived from UL SRS transmissions of the mobile client device, the number of UL SRS measurements exceeding a threshold, extract features from UL SRS measurements, obtain a machine learning (ML) classifier for determining a category to be used for estimating mobility associated with the mobile client device, and determine the category of the mobile client device by applying the extracted features to the ML classifier. Methods and apparatus extract the features of either a set of power spectrum density measurements or a set of pre-processed frequency domain real and imaginary portions of UL SRS measurements and feed the features to an AI classifier for UE speed estimation.