Efficient data collection in wireless sensor networks using compressive sensing and clustering

Abstract

Wireless sensor networks are low cost, small multifunctional sensors and low energy networks, which are deployed densely for tracking objects, environmental monitoring or controlling the operations in industries. Due to its specific challenges and impending wide applications, the wireless sensor networks have attracted many researchers in the recent times. The crucial challenge in wireless sensor networks includes its stability and energy efficiency, since the capacity of sensor node batteries are limited and hence the replacements of such batteries are difficult. To improve the capacity of sensor node batteries and to reduce the consumption of energy in network, the idea of Compressive Sensing (CS) is used in wireless sensor networks. The compressive sensing is a new sampling theory, where, the sparse signals achieves lesser sampling rate and creates a precise reconstruction of signals through lesser linear measurements. The Compressive Sensing theory is combined with wireless sensor networks to reduce the total data transmitted. It reduces well the consumption of energy in wireless sensor network, possibly increases the lifetime of network. In order to achieve the above objective, we propose two different methods that includes: Tree Cluster Data Gathering Compressive Sensing Algorithm (TCDGCSA) and Tree Cluster Adaptive Amoeba Blockwise Compressive Sensing Algorithm (TCAABCSA). newline