[Paper Review] Effective Data Aggregation Scheme for Large-scale Wireless Sensor Networks

M. Mehdi Afsar|arXiv (Cornell University)|Oct 31, 2013

Energy Efficient Wireless Sensor Networks36 references3 citations

TL;DR

This paper proposes EEMA, an adaptive hierarchical data aggregation protocol for large-scale wireless sensor networks that organizes nodes into multi-layered clusters to enhance energy efficiency and reduce routing delay. By dynamically selecting cluster heads based on residual energy, centrality, and proximity to lower-layer heads, EEMA extends network lifetime and improves scalability, with simulations showing up to 30% lower routing delay compared to flat architectures.

ABSTRACT

Energy preservation is one of the most important challenges in wireless sensor networks. In most applications, sensor networks consist of hundreds or thousands nodes that are dispersed in a wide field. Hierarchical architectures and data aggregation methods are increasingly gaining more popularity in such large-scale networks. In this paper, we propose a novel adaptive Energy-Efficient Multi-layered Architecture (EEMA) protocol for large-scale sensor networks, wherein both hierarchical architecture and data aggregation are efficiently utilized. EEMA divides the network into some layers as well as each layer into some clusters, where the data are gathered in the first layer and are recursively aggregated in upper layers to reach the base station. Many criteria are wisely employed to elect head nodes, including the residual energy, centrality, and proximity to bottom-layer heads. The routing delay is mathematically analyzed. Performance evaluation is performed via simulations which confirms the effectiveness of the proposed EEMA protocol in terms of the network lifetime and reduced routing delay.

Motivation & Objective

• Address the challenge of energy depletion in large-scale wireless sensor networks with thousands of nodes.
• Improve network scalability and reduce routing overhead in dense, large-scale deployments.
• Enhance energy efficiency through adaptive clustering and hierarchical data aggregation.
• Minimize routing delay by structuring the network into multiple virtual layers with optimized cluster-head selection.
• Balance energy load across nodes to prolong overall network lifetime

Proposed method

• Divides the sensor field into multiple virtual layers, with each layer further partitioned into clusters and super-clusters.
• Employs a hybrid metric for cluster-head (CH) selection combining residual energy, node centrality, and proximity to lower-layer CHs.
• Uses a hierarchical aggregation tree with the base station (BS) at the root and sensor nodes as leafs, enabling multi-hop data aggregation.
• Applies mathematical modeling to analyze routing delay, showing that EEMA reduces delay by eliminating intermediate forwarding delays.
• Introduces adaptive clustering to balance energy consumption and prevent early node death.
• Validated via simulations across varying network scales (300–4000 nodes) and topologies, with BS positioned centrally or off-center.

Experimental results

Research questions

• RQ1 How does a multi-layered hierarchical architecture impact energy efficiency and network lifetime in large-scale WSNs?
• RQ2 To what extent can adaptive clustering based on residual energy, centrality, and proximity improve load balancing and delay?
• RQ3 What is the effect of adding extra layers on energy consumption and scalability in dense sensor networks?
• RQ4 How does EEMA compare to flat and single-layer clustering protocols (e.g., LEACH, HEED) in terms of network lifetime and delay?
• RQ5 Can hierarchical data aggregation preserve coverage while reducing end-to-end routing delay?

Key findings

• EEMA extends network lifetime significantly more than LEACH, HEED, and DWEHC, especially in larger networks with 4000 nodes.
• The network's first node death (FND) and half-network alive (HNA) metrics show EEMA outperforms other protocols due to energy-aware CH selection.
• In dense networks (N=4000, M=2000), EEMA reduces routing delay by approximately 30% compared to flat architecture (L=1), due to optimized hierarchical routing.
• Adding extra layers improves energy efficiency in large-scale networks but increases energy consumption in smaller ones, indicating scalability benefits at scale.
• The simulation results confirm that EEMA maintains good coverage and achieves better load balancing by leveraging residual energy and spatial proximity in CH selection.
• The mathematical analysis of routing delay confirms that EEMA eliminates the need for intermediate packet waiting delays, contributing to faster end-to-end delivery.