Energy-aware task assignment and data aggregation protocols in wireless sensor networks

  • Authors:

  • Hady S. AbdelSalam;Syed R. Rizvi;Stephan Olariu

  • Affiliations:

  • Computer Science Department, Old Dominion University, Norfolk, VA;Computer Science Department, Old Dominion University, Norfolk, VA;Computer Science Department, Old Dominion University, Norfolk, VA

  • Venue:
  • CCNC'09 Proceedings of the 6th IEEE Conference on Consumer Communications and Networking Conference
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

In a typical sensor network environment, sensor energy is considered a precious resource that must be used wisely and only if necessary. Energy-unaware task assignment protocols can deplete the energy of some sensors much more than they do for others. This results in reducing network density around those heavily loaded sensors and eventually creates energy holes that isolate the network into separated islands. These problems have negative impacts on network durability and reliability. To avoid these problems, we propose and evaluate a lightweight management protocol that assigns tasks to sensors based on their energy so that energy consumption is almost even among network sensors. In addition, we propose another mechanism to aggregate data collected by sensors before sending them back to the base station. Our aggregation mechanism supports different aggregation functions including exact evaluation of the minimum, the maximum, and the logical OR and an approximation of the average of the collected sensory data. Using simulation, we compare the lifetime achieved by assigning tasks to network sensors using the proposed protocol against another energy-neutral protocol. Simulation results verified that the proposed approach increases network durability by balancing task load among sensors. Also simulation shows that, under reasonable parameters, the error in the approximated value of the average is less than %3.