Optimized Energy-Efficient Cluster Routing in IoT-Enabled Wireless Sensor Networks via Mapdiminution-Based Training and Discovery Algorithm

Alaa A. Hussain, Mohamed E. Hussein, Zainab Fahad AlnaseriInternational Journal of Computational and Electronic Aspects in Engineering 2025 · DOI

TL;DR

This paper introduces MTDOA, a novel energy-efficient routing protocol for IoT-enabled Wireless Sensor Networks (WSNs). By combining a unique "mapdiminution" dimensionality reduction technique with a training-discovery optimization approach, MTDOA significantly extends network lifetime and improves data delivery compared to existing methods.

What Problem Does This Paper Address?

The proliferation of IoT devices relies heavily on Wireless Sensor Networks (WSNs) for data collection and transmission. However, WSNs face a critical challenge: limited energy resources in sensor nodes. Replacing or recharging batteries in these nodes, often deployed in remote or inaccessible locations, is impractical. This energy constraint directly impacts the network's lifespan and overall performance. Existing routing protocols, while improving, still struggle with efficient energy utilization, optimal cluster head selection, and maintaining stable network connectivity, especially in dynamic environments. This paper aims to bridge this gap by proposing a more robust and energy-aware routing strategy.

Methods

The core innovation of this paper is the Mapdiminution-Based Training-Discovering Optimization Algorithm (MTDOA). This algorithm operates in two main phases:

  • Training Phase: This phase focuses on global exploration and identifying potential cluster heads (CHs). It employs a population-based optimization algorithm where each candidate solution represents a set of CHs. A novel "mapdiminution" strategy is used here. This strategy reduces the complexity of the solution space by projecting multi-dimensional parameters (like energy, position, and density) into simplified feature vectors, preserving essential information for CH selection. This dimensionality reduction aims to speed up convergence and reduce computational overhead. The fitness function for candidate CHs considers residual energy, distance to the base station, and intra-cluster distance.
  • Discovering Phase: This phase performs local refinement on the CHs identified in the training phase. It includes:
  • Local Search: Adjacent nodes are evaluated to find better CH substitutes.
  • Route Optimization: Energy-aware cost functions are used to establish efficient routing paths from CHs to the base station, minimizing cumulative transmission energy. A greedy heuristic guides routing decisions.

The overall MTDOA process involves initialization, mapdiminution projection, training for CH selection, discovery for route establishment, cluster formation, and data aggregation/transmission, repeating until network energy depletion.

Datasets

Not discussed in the paper. The evaluation is based on simulations using established WSN configuration protocols within a simulated environment.

Key Insights and Ideas

  • Mapdiminution for Efficiency: The novel "mapdiminution" technique is a key differentiator. By reducing the dimensionality of the problem space, it allows for faster convergence and more efficient exploration of potential solutions without losing critical information for CH selection. This is crucial for energy-constrained WSNs.
  • Synergy of Training and Discovery: The two-phase approach (training for exploration, discovery for exploitation) provides a balanced optimization strategy. This allows for both broad searching for optimal CHs and fine-tuning of routing paths.
  • Adaptive CH Selection: MTDOA's adaptive CH selection, considering residual energy, proximity to the base station, and local node density, is critical for equitable energy distribution and extending network lifespan.
  • Metaheuristic Optimization for WSNs: The paper reinforces the trend of using metaheuristic algorithms to tackle complex WSN optimization problems, demonstrating their effectiveness beyond traditional methods.

Key Results

Simulations comparing MTDOA with established protocols like LEACH and DEEC showed significant improvements:

  • Network Lifetime: MTDOA demonstrated a substantially prolonged stable region. Specifically, it extended network lifetime by 25-30% compared to DEEC and over 40% compared to LEACH. The last node died (LND) in MTDOA at 2010 rounds, compared to 1585 for DEEC and 1420 for LEACH.
  • Average Residual Energy: MTDOA maintained a higher average residual energy throughout the simulation period, indicating more efficient energy utilization.
  • Packet Delivery Ratio (PDR): MTDOA achieved a PDR of 92.4%, outperforming LEACH (81.3%) and DEEC (86.7%).
  • Throughput: MTDOA also showed superior throughput, delivering 123,810 packets compared to 94,580 for LEACH and 108,340 for DEEC.
  • Convergence Speed: MTDOA converges much faster than traditional PSO and GA, attributed to the mapdiminution's dimensionality control.

Limitations and Future Work

  • Limitations: The paper acknowledges that while MTDOA shows improved performance, the trade-off between global search (exploration) and local optimization (exploitation) remains an area for further refinement. Current approaches might still be somewhat amenable to network changes and parameter configuration.
  • Future Work: The authors suggest future work will concentrate on establishing real-time closed-loop systems and exploring hybrid optimization strategies to optimize network functionality for mixed sensors in practical applications. They also aim to further enhance the algorithm's adaptability to dynamic network conditions.

Summary updated

Abstract

The combination of WSNs and IoT technologies produces rapid mass-production yet requires energy-efficient routing protocols to fulfill the estimated demands. These protocols need further development to maintain continuous sensor node connections during their operational periods. The design of WSNs encounters a significant problem because sensor nodes have limited energy capabilities. The placement of NSN in specific zones creates difficulties for standard battery replacement and maintenance operations. Network routing for these systems needs energy preservation as their fundamental construction requirement. This paper proposes a Mapdiminution-Based Training-Discovering Optimization Algorithm (MTDOA), a new protocol that aims at increasing energy efficiency for IoT based WSNs. The MTDOA protocol is based on a novel mapdiminution-based dimensionality reduction and a mixed metaheuristic optimization method. This model is intended to allow for a good trade-off between, global and local search during the optimization process in aspects such as selection of efficient heads of clusters and identification of energy optimal routing paths. Thus, these components are therefore combined in the algorithm to cut down on computational cost, ensure faster convergence time and overall lifespan of the sensor network. The MTDOA protocol uses dynamic adaptive training discovery to select cluster heads from nodes based on their energy levels to reduce base station data routing costs. Simulation experiments run in the laboratory showed MTDOA performs better than LEACH and DEEC protocols when measuring network lifetime and average residual energy together with packet delivery ratio. MTDOA methodology enables successful sustainability improvements in WSN networks through its combination of better IoT-based metrics and performance metrics.
BibTeX
@article{hussain2025optimized,
  author = {Alaa A. Hussain and Mohamed E. Hussein and Zainab Fahad Alnaseri},
  title = {{Optimized Energy-Efficient Cluster Routing in IoT-Enabled Wireless Sensor Networks via Mapdiminution-Based Training and Discovery Algorithm}},
  journal = {International Journal of Computational and Electronic Aspects in Engineering},
  year = {2025},
  doi = {10.26706/ijceae.6.2.20250403},
}
Figure