Shima Pakdaman Tirani - Department of Electrical Engineering, Islamic Azad University, Najafabad Branch, Najafabad, Iran
Avid Avokh - Department of Electrical Engineering, Islamic Azad University, Najafabad Branch, Najafabad, IranDigital Processing and Machine Vision Research Center, Islamic Azad University, Najafabad Branch, Najafabad, Iran

This paper addresses the problems of energy consumption, energy consumption balancing, and Compressive Sensing (CS) recovery performance in wireless sensor network. We propose an efficient method named ‘‘Compressive Data Gathering using Sector-based Random Walk (CDG-SRW)’’. The proposed scheme uses the CS theory along with the sector-based random walks to transmit CS measurements from sensors to the sink node. This routing method is presented to decrease the overlap of random walk paths with each other and improve the quality of the measurement matrix. Furthermore, we provide a formulation to theoretically study average and minimum number of hops for a random walk path. Numerical results demonstrate that the proposed algorithm reduces energy consumption in comparison with M-CSR by ۷۲ % and also decreases energy consumption variance compared to it by ۴۵ % for an average number of CS measurements. Based on the simulation results, the proposed CDG-SRW algorithm also illustrates a performance improvement in comparison to M-CSR in terms of reconstruction error.

Average number of hops, compressive sensing, random walk, wireless sensor network