Delay-Aware Energy-Efficient Routing towards a Path-Fixed Mobile Sink in Industrial Wireless Sensor Networks

Shaobo Wu, Wusheng Chou, Jianwei Niu, Mohsen Guizani

TL;DR

This paper introduces DERM, a novel routing algorithm for industrial wireless sensor networks (WSNs) with a mobile data collector (sink) that follows a fixed, uncontrollable path. DERM effectively balances energy efficiency and strict delay requirements by intelligently routing data to dynamic "destination regions" and employing a lightweight location calibration and a fault-tolerant "track routing" mechanism.

What Problem Does This Paper Address?

Industrial WSNs are increasingly incorporating mobile elements, such as robots or vehicles, to collect sensor data. While mobility can boost network performance, existing routing strategies struggle when the mobile sink's path is fixed and its movement is unpredictable. These scenarios often fail to guarantee timely data delivery (delay constraints) while simultaneously conserving energy. Traditional methods either prioritize low latency at the cost of high energy consumption (immediate multi-hop) or offer energy savings but with potentially unacceptable delays (rendezvous-based). This paper tackles the challenge of achieving both energy efficiency and strict delay guarantees in such path-fixed, uncontrollable mobile sink environments.

Methods

DERM is built upon three core components:

  • Delay-Aware Geographic Routing (EO-DRA): This is the primary routing protocol. Nodes estimate the sink's current location and a "destination region" (an area accessible to the sink within the delay constraint). Data is then forwarded greedily towards this region using a novel routing metric, remaining ETX (rETX), which considers both link quality and the distance to the destination region. A modified perimeter routing approach is used to handle communication voids.
  • Lightweight Sink Location Calibration: To combat inaccuracies in estimating the sink's location due to its unpredictable movement, DERM employs a calibration method. Nodes only broadcast updated location information when the estimated route length dilation exceeds a predefined threshold, significantly reducing control overhead compared to continuous broadcasting.
  • Track Routing: This fault-tolerant mechanism addresses residual location errors or unexpected sink movements. If a packet reaches a rendezvous node (RN) but the sink hasn't arrived by the deadline, track routing is activated. It uses a "greedily advance, discreetly step back" strategy to guide the packet along the sink's path, ensuring it's delivered before its deadline.

Datasets

Not discussed in the paper. The evaluation is based on simulations using the ns-2 simulator with randomly generated network topologies.

Key Insights and Ideas

  • Dynamic Destination Regions: Instead of targeting a single point, DERM routes data to a "region" that the sink will traverse within the delay constraint. This flexibility is crucial for path-fixed mobile sinks.
  • rETX Metric: The novel routing metric effectively balances link quality with the predicted path to the dynamic destination region, leading to energy-efficient routing.
  • Proactive, Low-Overhead Calibration: The threshold-based location calibration is a clever way to maintain accurate sink location information without overwhelming the network with control messages.
  • "Greedily Advance, Discreetly Step Back": The track routing mechanism's strategy is an intuitive and effective way to recover from missed deadlines and ensure reliable delivery.

Key Results

  • Energy Savings: DERM significantly reduces transmission costs compared to Multihop (13%-44%) and DC-Rendezvous (26%-38%).
  • Delay Guarantee: DERM consistently meets the delay constraints, unlike the uncontrollable delay of the traditional Rendezvous approach.
  • Reduced Overhead: The location calibration overhead in DERM is about 40% less than Multihop.
  • Robustness: DERM demonstrates effectiveness in realistic scenarios where sink movement deviates from estimations, maintaining high delivery ratios and on-time delivery.

Limitations and Future Work

  • Multiple Mobile Sinks: The current work focuses on a single mobile sink. The authors plan to extend the approach to handle multiple path-fixed sinks.
  • End-to-End Optimization: Future work will involve more comprehensive optimization considering end-to-end delay, transmission cost, and the total amount of collected data.
  • Real-world Deployment: While simulations are extensive, real-world validation would be a valuable next step.

Summary updated

Abstract

Wireless sensor networks (WSNs) involve more mobile elements with their widespread development in industries. Exploiting mobility present in WSNs for data collection can effectively improve the network performance. However, when the sink (i.e., data collector) path is fixed and the movement is uncontrollable, existing schemes fail to guarantee delay requirements while achieving high energy efficiency. This paper proposes a delay-aware energy-efficient routing algorithm for WSNs with a path-fixed mobile sink, named DERM, which can strike a desirable balance between the delivery latency and energy conservation. We characterize the object of DERM as realizing the energy-optimal anycast to time-varying destination regions, and introduce a location-based forwarding technique tailored for this problem. To reduce the control overhead, a lightweight sink location calibration method is devised, which cooperates with the rough estimation based on the mobility pattern to determine the sink location. We also design a fault-tolerant mechanism called track routing to tackle location errors for ensuring reliable and on-time data delivery. We comprehensively evaluate DERM by comparing it with two canonical routing schemes and a baseline solution presented in this work. Extensive evaluation results demonstrate that DERM can provide considerable energy savings while meeting the delay constraint and maintaining a high delivery ratio.
BibTeX
@inproceedings{wu2018delay,
  author = {Shaobo Wu and Wusheng Chou and Jianwei Niu and Mohsen Guizani},
  title = {{Delay-Aware Energy-Efficient Routing towards a Path-Fixed Mobile Sink in Industrial Wireless Sensor Networks}},
  booktitle = {Sensors},
  year = {2018},
  url = {https://openreview.net/forum?id=SXThX9BSVA},
}
Figure