Investigating the tradeoffs between spatial granularity and energy requirements in wireless sensor networks

In some wireless sensor network applications like precision agriculture, the network area is divided into a number of well-defined regions (spatial granules) and for each spatial granule a separate measurement is made. In performing the task of collecting the data pertaining to these measurements, there is an inherent tradeoff between number of spatial granules and minimum energy requirements of sensor nodes deployed in the area. In this paper, through a linear programming (LP) framework, we investigate the impact of spatial granularity of measurements on the energy requirements of sensor network. Once redundancy is defined in this context as the duplication of data collected for each granule, our LP model also allows us to determine almost achievable performance benchmarks in idealized yet practical settings which are achievable when redundancy is totally eliminated.