Data Aggregation in Low-Power Wireless Sensor Networks With Discrete Transmission Ranges: Sensor Signal Aggregation Over Graph

Conventional wireless sensor networks (WSNs) consist of sensors with continuous transmission range, which depends on the relative positions of the transmitter and the receiver. However, sensors with different discrete transmission ranges are preferred for future generation low-power sensor networks because of certain functional advantages. The discrete transmission ranges introduce connectivity constraints in transmitting the sensor data. The proliferation of low-power WSNs has led to the explosion of the volume of the data to be processed. As the transmission of the data is the major cause of energy depletion of sensors that critically affect the network lifetime, energy-efficient aggregation of the sensor data is an important networking problem. In this work, we address the data aggregation problem in networks with sensors of discrete transmission ranges. We model the problem as a solvable integer linear program. However, this method applies only to networks of small sizes because of the hardness of the program. To solve the problem in networks of large sizes, we introduce a graphical framework that captures the characteristics of the networks with sensors of discrete transmission ranges and design a polynomial-time approximation technique to find a solution. Furthermore, we embed compression techniques based on compressed sensing (CS), which are established to yield high data compaction for temporally and spatially correlated distributed sensor data streams, and evaluate the performance of the proposed methods.

Automated summary

Conventional wireless sensor networks use sensors with continuous transmission range, but future low-power sensor networks prefer sensors with discrete transmission ranges due to certain functional advantages. The discrete transmission ranges introduce connectivity constraints in transmitting sensor data. In this study, we address the data aggregation problem in networks with sensors of discrete transmission ranges and propose a graphical framework and a polynomial-time approximation technique to solve the problem in large networks.