An Opportunistic Routing in Energy-Harvesting Wireless Sensor Networks With Dynamic Transmission Power

Hui Cheng, Cong Wang, Xinming ZhangIEEE Access 2019 · OpenReview

Abstract

In energy-harvesting wireless sensor networks (EH-WSNs), nodes can harvest energy from the environment. However, the energy harvested is usually insufficient, time-varying, and unbalanced. To make full use of the energy harvested from the environment, each node can dynamically adjust its transmission power, which affects the associated link quality. Consequently, the link quality between two nodes is dynamic. For the forwarding node, the transmission cost of it and its one-hop candidate relay node may change at any time. In this scenario, for the design of opportunistic routing, the size, members, and priorities in the relay set need to be updated dynamically. In this paper, we propose an opportunistic routing in EH-WSNs with dynamic transmission power (ORDTP). ORDTP helps the sending node collect information of neighbor nodes in time, accurately estimate the transmission cost, and dynamically update the size, members, and priorities in the relay set. Simulation results show that ORDTP effectively reduces delay, improves the transmission success rate, and reduces the packet redundancy rate.

Introduction

Traditional Wireless Sensor Networks (WSNs) are often constrained by the finite battery life of their sensor nodes. Energy-Harvesting WSNs (EH-WSNs) offer a sustainable alternative by drawing power from environmental sources such as solar, wind, or thermal energy. However, this transition introduces significant complexity: energy availability is highly unpredictable and varies across the network.

To achieve perpetual operation, nodes must adhere to the Energy Neutral Operation (ENO) principle, ensuring that energy consumption does not exceed energy acquisition. This requires nodes to constantly adjust their transmission power based on their current energy reserves and predicted harvesting rates. These fluctuations in power directly impact link quality; a node that is a reliable relay at high power may become unreachable when it scales back power to conserve energy. Existing Opportunistic Routing (OR) protocols, which broadcast packets to a set of potential relays, often fail to account for these rapid, power-driven topological changes. This paper introduces ORDTP to address these dynamic challenges.

The ORDTP Framework

ORDTP (Opportunistic Routing with Dynamic Transmission Power) is designed to manage the uncertainty of EH-WSNs through three integrated mechanisms: multi-stage relay selection, cost optimization via a new metric, and redundancy control.

Multi-Stage Relay Set Selection

Unlike static routing protocols, ORDTP builds its relay set dynamically through a three-step filtering process. First, it applies a geographical filter to identify nodes physically closer to the sink, preventing routing loops. Second, it evaluates link quality by calculating the Packet Acceptance Rate (PAR) based on the current transmission power of the nodes. Finally, it performs cost optimization to prioritize the most efficient relays.

Relay set selection and prioritization in ORDTP

As shown in the figure above, the sending node (S) identifies a set of candidate nodes within its transmission radius ($d_{tr}$). Nodes are categorized as actual relay nodes (b, c, e), unsuitable candidates (d), or other nodes. These relay nodes are then assigned priorities (1, 2, 3) to coordinate the forwarding process.

Expected Dynamic Transmission Cost (EDTT)

A core innovation of ORDTP is the Expected Dynamic Transmission Cost (EDTT) metric. In opportunistic routing, a node may wait to see if a higher-priority neighbor has already forwarded a packet before acting itself. Traditional metrics often ignore this "wait cost."

Timing diagram of communication and wait costs in ORDTP

EDTT explicitly accounts for both the physical communication time and the inherent wait time in opportunistic schemes. As illustrated in the timing diagram, the total transmission cost for a node includes the active data transmission period and the wait cost incurred while monitoring higher-priority relays. By minimizing EDTT, ORDTP significantly reduces end-to-end latency.

Relay Circle and Power Management

To prevent network flooding, ORDTP utilizes a "Relay Circle" data structure to manage packet redundancy. It calculates a looping round value, $k$, based on the success probability of the relay set. If the set is small, $k$ increases to ensure delivery; if the set is large, $k$ decreases to prune redundant transmissions.

Furthermore, nodes utilize the Accurate Solar Energy Allocation (ASEA) model to predict future energy harvesting. This allows them to set the highest possible transmission power that still satisfies the ENO-Max condition, maximizing link quality without risking energy depletion.

Performance Evaluation

The researchers evaluated ORDTP using the OMNet++ (Castalia) simulation environment, comparing it against state-of-the-art protocols: TCOR (Transmission Power Control-based OR) and COOR (Cross-layer Optimized OR). The simulation involved nodes deployed in a 200m x 200m area with solar energy profiles ranging from 20mW to 50mW.

  • End-to-End Delay (EED): ORDTP maintained a stable delay of approximately 75ms even as node density increased, whereas TCOR and COOR saw delays climb to 130ms.
  • Packet Delivery Ratio (PDR): In dense network scenarios, ORDTP achieved a success rate of nearly 95%, outperforming its competitors by 5-10%.
  • Packet Redundancy Ratio (PRR): Despite the broadcast nature of opportunistic routing, the Relay Circle logic effectively kept redundancy lower than TCOR and COOR by pruning unnecessary retransmissions.

Conclusion

ORDTP provides a robust solution for the dynamic and unpredictable nature of Energy-Harvesting WSNs. By treating link quality as a variable of the energy budget and incorporating wait costs into routing decisions, the protocol achieves superior performance in terms of latency and reliability. Future work may involve adapting the ASEA model for more erratic energy sources and testing the protocol in mobile sensor environments.

Summary updated

Abstract

In energy-harvesting wireless sensor networks (EH-WSNs), node can harvest energy from the environment. However, the energy harvested is usually insufficient, time varying and unbalanced. In order to make full use of the energy harvested from the environment, each node can dynamically adjust its transmission power which may affect the associated link quality. Thus, the link quality between two same nodes is dynamic. For the forwarding node, the transmission cost of it and its one hop candidate relay node may change at any time. In this scenario, for the design of opportunistic routing, the size, members and priorities in the relay set need to be updated dynamically. In this paper, we propose an opportunistic routing in EH-WSNs with dynamic transmission power (ORDTP). ORDTP can help the sending node collect information of neighbor nodes in time, accurately estimate the transmission cost, and dynamically update the size, members and priorities in the relay set. The simulation results show that ORDTP can effectively reduce the delay, improve the transmission success rate, and reduce the packet redundancy rate.
BibTeX
@inproceedings{cheng2019opportunistic,
  author = {Hui Cheng and Cong Wang and Xinming Zhang},
  title = {{An Opportunistic Routing in Energy-Harvesting Wireless Sensor Networks With Dynamic Transmission Power}},
  booktitle = {IEEE Access},
  year = {2019},
  url = {https://openreview.net/forum?id=nYi2DQh7vm},
}
Figure