In this work, random telegraph noise (RTN) associated to discrete current fluctuations in the high resistive state of Ni/HfO2-based RRAM devices is investigated. For this purpose, a dedicated software tool has been developed to control the instrumentation and to perform successive and smart RTN measurements in the time domain. After data acquisition, the advanced Weighted Time Lag (WTL) method is employed to accurately identify the contribution of multiple electrically active defects in multilevel RTN signals. Finally, the internal dynamics of trapping and de-trapping processes through the defects close to the filamentary path and its dependence on voltage and time are analyzed.