With an increasing energy demand, the appliancespecific energy consumption information can help in the residential sector to identify their saving potential. Thereby, algorithms allow to disaggregate the whole-house data into its major constituents. For this purpose, the data are measured by a smart meter under the concept of the Internet of Things that encompasses all the embedded devices and networks that are natively IP-enabled and Internet-connected. In this context, this paper presents the initial steps of a new Non-Intrusive Load Monitoring (NILM) technique based on the analysis of the waveform with a sliding window. NILM is a method for energy disaggregation from a single point of measurement. The data used can be acquired from the main electrical panel outside of the building or the residence, hence It is considered to be nonintrusive as the method avoids any equipment installation inside the customer's property. In this way, the initial approach of this technique searches for elevator patterns in the power signal of the existing Smart Grid of the Swiss Federal Institute of Technology in Lausanne (EPFL), Switzerland.