A Novel Connectivity-Based LEACH-MEEC Routing Protocol for Mobile Wireless Sensor Network

Muqeet Ahmad, Tianrui Li, Zahid Khan, Faisal Khurshid, Mushtaq Ahmad

TL;DR

This paper introduces LEACH-MEEC, a new routing protocol for mobile wireless sensor networks (MWSNs) that improves network lifetime and data delivery by selecting cluster leaders based on both remaining energy and connectivity. The study also highlights the Reference Point Group Mobility (RPGM) model as being particularly effective for such networks.

What Problem Does This Paper Address?

Mobile Wireless Sensor Networks (MWSNs) are crucial for applications like environmental monitoring and disaster relief, but their effectiveness is often limited by the energy efficiency of their routing protocols. Traditional protocols, like LEACH and its variants, primarily focus on selecting cluster leaders (CLs) based on remaining energy. However, in dynamic MWSN environments, node mobility can lead to frequent disconnections, impacting data delivery and overall network lifetime. This paper addresses the challenge of maintaining robust connectivity and energy efficiency in MWSNs by proposing a novel CL selection mechanism that incorporates network connectivity as a key factor, alongside remaining energy.

Methods

The core of this work is the LEACH-MEEC (Low-Energy Adaptive Clustering Hierarchy-Mobile Energy Efficient and Connected) routing protocol. The protocol builds upon the LEACH framework but introduces a refined Cluster Leader (CL) selection process.

  • Connectivity Metric: LEACH-MEEC introduces a probabilistic connectivity model. It uses a binary disk sensing model to calculate neighborhood density within a defined radio range (Rmax). This density is then used to compute a probabilistic connectivity (PC) value for each node.
  • CL Selection: In the first round, a standard LEACH-like random selection is used. However, from the second round onwards, CLs are selected based on a combined metric: CVi = Erem(Ni) + PC(Ni), where Erem(Ni) is the remaining energy of node Ni and PC(Ni) is its probabilistic connectivity. This ensures that CLs are not only energy-rich but also well-connected to their neighbors.
  • Mobility Models: The performance of LEACH-MEEC is evaluated under four different mobility models: Reference Point Group Mobility (RPGM), Random Waypoint (RWP), Gauss-Markov (G-M), and Manhattan Grid (MG). The study also investigates the impact of different speeds (1.5 m/s and 7.5 m/s) for these models.
  • Statistical Analysis: Beyond simulation results, the authors employ statistical tests (ANOVA, Tukey's, LSD, and Heckman's two-stage test) to rigorously analyze the significance of their findings, particularly concerning the impact of mobility models.

Datasets

The paper does not introduce new benchmark datasets. Instead, it generates eight datasets by simulating the movement of 100 mobile sensor nodes under four distinct mobility models (RPGM, RWP, GM, MH) at two different speed levels (1.5 m/s and 7.5 m/s) within a 100x100 m² field. These datasets are used for evaluating the performance of the proposed LEACH-MEEC protocol against existing algorithms.

Key Insights and Ideas

  • Connectivity is Key for CL Selection: The most significant insight is that simply selecting CLs based on remaining energy is insufficient for MWSNs. Incorporating connectivity into the CL selection process dramatically improves network lifetime and data delivery. This is because a connected CL can efficiently gather data from its cluster members, reducing communication overhead and preventing isolated nodes.
  • RPGM Model Shines: The Reference Point Group Mobility (RPGM) model consistently shows superior performance across all evaluated metrics (alive nodes, remaining energy, packet delivery ratio, and connectivity). This is attributed to the nodes moving in groups, maintaining higher neighborhood connectivity and a more balanced energy distribution.
  • Speed vs. Efficiency Trade-off: As expected, higher node speeds generally lead to reduced energy efficiency and lower packet delivery ratios. However, LEACH-MEEC's connectivity-aware approach helps mitigate these negative impacts compared to other protocols.
  • Statistical Rigor: The extensive statistical analysis validates the simulation results, providing strong evidence for the superiority of LEACH-MEEC and the RPGM mobility model. The Heckman's two-stage test also confirms that the observed performance improvements are not due to selection bias.

Key Results

  • Number of Alive Nodes (NAN): LEACH-MEEC significantly outperforms LEACH-MAE, Mobile-LEACH, and LEACHDistance-M in terms of network lifetime, with a higher number of alive nodes throughout the simulation. The RPGM dataset consistently shows the longest network lifetime.
  • Remaining Energy (RE): The proposed LEACH-MEEC protocol maintains higher remaining energy across all nodes compared to the other algorithms, indicating better energy efficiency. Again, RPGM datasets show the highest remaining energy.
  • Packet Delivery Ratio (PDR): LEACH-MEEC achieves a higher PDR, meaning more data packets successfully reach the destination. This is attributed to the optimal CL selection, which reduces link collapses and packet drops, especially in the RPGM model where nodes are well-connected.
  • Mobility Model Performance: Statistical analysis confirms that RPGM is significantly better than RWP, GM, and MH mobility models for LEACH-MEEC, showing higher ANAN, ARE, APDR, and AC.

Limitations and Future Work

  • Limitations:
    • The study assumes homogeneous sensor nodes.
  • The simulation code is not publicly available.
  • The base station is assumed to be static and located outside the sensor distribution area.
  • Future Work:
    • Extend the research by increasing the number of mobile sensor nodes and their speeds to explore multi-hop environments.
  • Make the full source code publicly available.

Summary updated

Abstract

In mobile wireless sensor network (MWSN), the lifetime of the network largely depends on energy efficient routing protocol. In the literature, cluster leader (CL) is selected based on remaining energy of mobile sensor nodes to enhance sensor network lifetime. In this study, a novel connectivity-based Low-Energy Adaptive Clustering Hierarchy-Mobile Energy Efficient and Connected (LEACH-MEEC) routing protocol was proposed, where CL is selected based on connectivity among neighboring nodes and the remaining energy of mobile sensor nodes. Consequently, it improves data delivery, network lifetime and balances the energy consumption. We studied various performance metrics including the number of alive nodes (NAN), remaining energy (RE) and packet delivery ratio (PDR). Our proposed LEACH-MEEC outperforms all other algorithms due to the connectivity metric. Moreover, the performance of mobility models was investigated through graphical and statistically tabulated results. The results show that Reference Point Group Mobility model (RPGM) is better than other mobility models.
BibTeX
@inproceedings{ahmad2018novel,
  author = {Muqeet Ahmad and Tianrui Li and Zahid Khan and Faisal Khurshid and Mushtaq Ahmad},
  title = {{A Novel Connectivity-Based LEACH-MEEC Routing Protocol for Mobile Wireless Sensor Network}},
  booktitle = {Sensors},
  year = {2018},
  url = {https://openreview.net/forum?id=mslXZbWio5},
}
Figure