Graph Partitioning for Iterative Signal Reconstruction in Wireless Sensor Network

Abstract

One of the important applications of graph signal processing (GSP) in wireless sensor network (WSN) is reconstruction of the bandlimited graph signals from local measurements. The reconstruction may be performed by iterative algorithm. Graph partitioning plays an important role in the signal reconstruction using iterative algorithm. In this paper, we propose a new partitioning approach to improve the convergence rate of the signal reconstruction. The proposed partitioning method relaxes calculating the maximum bandwidth of graph signal to be reconstructed.