To search, Click below search items.


All Published Papers Search Service


Signal Reconstruction through Compressive Sensing and Principal Component Analysis in Wireless Sensor Networks


Kia Jahanbin1* and Saeed Mehrjoo2


Vol. 17  No. 6  pp. 147-154


With the emergence of Wireless Sensor Networks (WSNs), Data Acquisition (DAQ) and signal reconstruction have been considered as the main area of interest in the IT research. In this paper, using an external server connected to the internet, the adaptive acquisition framework and WSN of signal reconstruction (AAR framework with D-PCA) along with the combination of the distilled sensing algorithms have been taken into account. Furthermore, for monitoring, Distributed Principal Component Analysis (D-PCA), data collection and signal reconstruction of WSNs are also considered. The results of the simulation show that using the adaptive algorithms of the Compressive Distilled Sensing in the signal sampling is more significant than the non-adaptive compressed Sensing algorithms. The former can solve the scalability problem and it also leads to the increase of the quality of signal sampling in WSNs. Moreover, by exploiting the algorithm of D-PCA for designing the Sparse Dictionary i.e. Ψ matrix in the server, the measurements with greater sparse have been transferred to the server which leads to a more exact reconstruction. In reconstructing the acquired signals, especially the sparse signals or signals with temporal correlation, the proposed framework in this work is very effective. The presented method decreased the number of samples and improved the signal reconstruction error smaller than 5 × 10-6.


Compressive sensing, compressive distilled sensing distilledsensing:distributed principal component analysis wireless sensor network