An Energy-Efficient Region Source Routing Protocol for Lifetime Maximization in WSN

Chuan Xu, Zhengying Xiong, Guofeng Zhao, Shui YuIEEE Access 2019 · OpenReview

TL;DR

The ER-SR protocol maximizes Wireless Sensor Network (WSN) lifetimes by combining dynamic "source routing nodes" with a path-finding algorithm inspired by ant colonies. By mathematically deriving an "effective distance" for data hops and distributing routing responsibilities to high-energy nodes, the protocol significantly reduces energy waste and prevents individual nodes from burning out prematurely.

The Challenge of WSN Longevity

Wireless Sensor Networks (WSNs) serve as the "eyes and ears" of the Internet of Things (IoT), often deployed in hard-to-reach or hostile environments to monitor physical conditions. As shown in the IoT architecture (Figure 1), the sensing layer is the foundation of data collection, but it is constrained by limited battery power.

IoT Architecture and the Sensing Layer (WSN)

The biggest energy drain in a WSN is communication. Traditional routing protocols often fall into two traps:

  1. Flooding: They blast discovery messages across the entire network to find a path, wasting massive amounts of energy on overhead.
  2. Static Clustering: They rely on fixed "cluster heads" to manage traffic. These heads quickly run out of juice, creating "hotspots" and causing the network to fail even if most nodes still have power.

The ER-SR Protocol

The proposed Energy-Efficient Region Source Routing (ER-SR) protocol operates in three distinct phases to balance the energy load across the network while finding the most efficient physical path for every bit of data.

1. Distributed Energy Region Algorithm

Instead of fixed cluster heads, the network dynamically selects Source Routing Nodes (SRNs). A node becomes a candidate if its residual energy is higher than the average of its neighbors. A timer mechanism ensures that SRNs are spread out evenly, preventing "isolated nodes."

2. Region Division

Common nodes join a region based on energy efficiency. If a node sees multiple SRNs, it calculates a probability based on the SRN’s distance and remaining energy, ensuring it doesn't just pick the closest node, but the "healthiest" one. This regional organization helps manage the network topology effectively (Figure 2).

Network Region Division Model for Energy Balancing

3. Effective Distance-Based Ant Colony Optimization (ACO)

This is the core innovation. The authors derive a mathematical "effective distance" ($d_c$)—the theoretical "sweet spot" for hop length that minimizes total energy consumption based on radio dissipation models. They then use an ACO algorithm (where virtual "ants" leave pheromone trails on good paths) to find a global route where each hop stays as close to $d_c$ as possible.

Key Insights and Ideas

  • The "Effective Distance" ($d_c$) Sweet Spot: The authors prove that there is an optimal distance for a wireless hop. If a hop is too short, too many intermediate nodes consume reception energy; if it's too long, the transmission power required increases exponentially.
  • Signaling vs. Data Separation: By using SRNs to calculate the path and then embedding that path in the packet header (source routing), intermediate nodes don't need to perform complex calculations. They just "follow the map" in the header.
  • Avoiding Local Optima: Standard ACO can get "stuck" on a decent path that isn't the best. The authors implemented local and global pheromone update strategies to ensure the "ants" keep exploring until they find the truly optimal global path.

Performance Results

Simulation results using NS3.28 demonstrated that ER-SR outperformed established benchmarks (ER-RPL, MSGR, and PRD) across all major metrics:

  • Network Lifetime: ER-SR extended the time until 20% of nodes die significantly, especially as node density increased.
  • Energy Consumption: ER-SR showed a drastically lower total energy cost. For 300 nodes, ER-SR consumed roughly 2.1J total, while the standard ER-RPL consumed nearly 2.5J.
  • Packet Delivery Ratio (PDR): ER-SR maintained a PDR of ~98% in dense networks.
  • Delivery Delay: By optimizing the number of hops and reducing signaling interference, ER-SR achieved the lowest end-to-end latency (approx. 90ms vs. 130ms for ER-RPL).

Limitations and Future Work

While ER-SR shows great promise for stationary sensors, the current study assumes nodes do not move. In many IoT applications like vehicular networks, nodes are mobile. The authors acknowledge that extending ER-SR to mobile networks and performing a deeper analysis of the computational overhead on SRNs are necessary next steps for real-world hardware implementation.

Summary updated

Abstract

As the sensor layer of Internet of Things (IOT), enormous amount of sensor nodes are densely deployed in a hostile environment to monitor and sense the changes in the physical space. Since sensor nodes are driven with limited power batteries, it is very difficult and expensive for wireless sensor networks (WSNs) to extend network lifetime. In order to achieve reliable data transmission in WSNs, energy efficient routing protocol is a crucial issue in extending the network lifetime of a network. However, traditional routing protocols usually propagate throughout the whole network to discover a reliable route or employ some cluster heads to undertake data transmission for other nodes, which both require large amount energy consumption. In this paper, to maximize the network lifetime of the WSN, we propose a novel energy efficient region source routing protocol (referred to ER-SR). In ER-SR, a distributed energy region algorithm is proposed to select the nodes with high residual energy in the network as source routing node dynamically. Then, the source routing nodes calculate the optimal source routing path for each common node, which enables partial nodes to participate in the routing process and balances the energy consumption of sensor nodes. Furthermore, to minimize the energy consumption of data transmission, we propose an effective distance-based ant colony optimization algorithm to search the global optimal transmission path for each node. Simulation results demonstrate that ER-SR exhibits higher energy efficiency, and has moderate performance improvements on network lifetime, packet delivery ratio, and delivery delay, compared with other routing protocols in WSNs.
BibTeX
@inproceedings{xu2019energy,
  author = {Chuan Xu and Zhengying Xiong and Guofeng Zhao and Shui Yu},
  title = {{An Energy-Efficient Region Source Routing Protocol for Lifetime Maximization in WSN}},
  booktitle = {IEEE Access},
  year = {2019},
  url = {https://openreview.net/forum?id=MWBiWw1rtKc},
}
Figure