Energy Heterogeneous Wireless Sensor Network Clustering Mechanism in Farmland

Dayong Liu, Chunjiang Zhao, Huarui Wu, Li Fei-FeiThe Open Automation and Control Systems Journal 2014 · DOI

Abstract

Solar-powered nodes employed in farmland wireless sensor networks offer significant energy advantages but come with high implementation costs. To balance cost and performance, networks often employ a mix of solar-powered and battery-powered nodes. This paper presents an improved clustering algorithm based on the LEACH protocol designed for such heterogeneous environments. The algorithm classifies nodes into three types based on their energy characteristics and prioritizes solar-powered nodes for the energy-intensive role of cluster head through a two-stage election process. Simulation results demonstrate that this approach improves solar energy utilization, reduces node mortality, and significantly extends the overall network lifecycle.

Introduction

Monitoring large-scale farmland requires wireless sensor networks (WSNs) capable of long-term operation in remote areas. Traditional battery-powered networks are limited by finite energy supplies, leading to "coverage holes" as nodes fail. While solar-powered nodes provide a renewable alternative, their high cost and dependence on weather conditions make a purely solar network impractical. This research addresses the challenge of energy heterogeneity: how to effectively coordinate a hybrid network of solar and battery nodes to maximize lifespan while minimizing costs.

Proposed Clustering Mechanism

The proposed algorithm enhances the Low Energy Adaptive Clustering Hierarchy (LEACH) protocol. While standard LEACH selects cluster heads (CHs) randomly, this improved version uses a "solar-first" priority system to ensure that nodes with renewable energy carry the heaviest data-routing burdens.

Node Classification and Election

At the beginning of each round, the sink node categorizes nodes based on their energy source and current levels:

  • Quasi-cluster head solar nodes: Solar nodes with energy above a specific threshold ($E_{sthr}$).
  • Downgrade solar nodes: Solar nodes with insufficient energy (e.g., during nighttime or rainy weather).
  • Battery nodes: Nodes with a non-renewable, decreasing energy supply.

The election process occurs in two stages. First, the algorithm attempts to fill the required number of CH positions using only the "Quasi-cluster" solar nodes. If the number of available solar nodes is insufficient to meet the network's needs, a second election is held among the battery-powered nodes to fill the remaining slots.

Figure 4: Flowchart of the proposed clustering algorithm illustrating the node classification and two-stage election process.

Simulation and Performance Analysis

The algorithm was tested via MATLAB simulations in a 100m × 100m field with 100 nodes (20% solar, 80% battery). The simulation included environmental modeling of day/night cycles and a multi-round "rainy" period where solar charging was unavailable, forcing solar nodes to rely on their internal reserves.

Key Results

The improved algorithm was compared against the standard LEACH protocol over 5,000 rounds.

  • Energy Efficiency: The hybrid approach successfully offloaded the most energy-intensive tasks to solar nodes during sunny periods, preserving the limited reserves of battery nodes for when they are most needed.
  • Residual Energy: As shown in the simulation results, the improved algorithm maintained a significantly higher average residual energy across the network compared to LEACH throughout the entire lifecycle.
  • Network Lifetime: By prioritizing renewable energy, the algorithm delayed the "first node death" and kept approximately 40% of the network operational by round 5000, whereas the standard LEACH network had almost entirely failed by that point.
Figure 5: Comparison of average residual energy over 5,000 simulation rounds, showing the improved performance of the proposed algorithm over standard LEACH.

Conclusion

By intelligently managing energy heterogeneity, the proposed clustering mechanism offers a robust solution for farmland monitoring. Prioritizing solar-powered nodes for high-energy roles ensures that battery resources are conserved for periods of low solar availability. Future research will focus on optimizing node placement and adapting the algorithm for mobile sensor nodes in smart agriculture applications.

Summary updated

Abstract

Solar-powered nodes employed in farmland wireless sensor network have energy advantages but at the same time have the problem of high cost. Wireless sensor network employs both solar-powered nodes and battery powered nodes which can reduce cost and prolong network lifetime. This paper presents an algorithm based on LEACH to study the clustering process of farmland environment monitoring sensor network which contains both solar-powered nodes and battery-powered nodes. This algorithm classifies nodes into three types based on the nodes characteristics first and the nit selects cluster nodes based on the comparison of the number of cluster head nodes and the solar powered nodes. The algorithm in this paper gives priority to utilize the solar-powered battery by two elections in one round. Simulation result shows that this algorithm can improve the utilization of solar energy, decrease node death and finally prolong the network lifecycle.
BibTeX
@article{liu2014energy,
  author = {Dayong Liu and Chunjiang Zhao and Huarui Wu and Li Fei-Fei},
  title = {{Energy Heterogeneous Wireless Sensor Network Clustering Mechanism in Farmland}},
  journal = {The Open Automation and Control Systems Journal},
  year = {2014},
  doi = {10.2174/1874444301406010374},
}
Figure