© 2020 IEEE.The Graph Signal Processing (GSP) is a mathematical framework that extends the Discrete Signal Processing (DSP) tools such as filtering and signals decomposition to graph data structures. In this paper, we explore the application of the GSP framework to distributed wireless sensor networks to reduce the measured noise and/or estimate the signal of missing sensors. The context is that of distributed monitoring applications, such as the monitoring of large buildings, such as hospitals, public areas and farmlands. Using simulation tools, we analysed the ability of GSP in reducing the noise and in estimating sensor's data in different WSN scenarios. We modelled the sensor networks as Graph structures and apply Graph Shift and Graph Laplacian operations on such graph signals. The analysis of obtained results shows that GSP may represent a valuable tool in the considered scenarios with outstanding performance.