[Paper Review] A Testbed of Magnetic Induction-based Communication System for Underground Applications

Xin Tan, Zhi Sun+1 more|arXiv (Cornell University)|Mar 9, 2015

Underwater Vehicles and Communication Systems28 references19 citations

TL;DR

This paper presents a laboratory-scale testbed for magnetic induction (MI)-based wireless communication in underground environments, validating MI as a viable alternative to traditional EM wave communication. The system demonstrates significant improvements in signal strength and reliability—especially in high-moisture soils—through MI waveguides and 3D MI coils, achieving low packet error rates across varying soil conditions and orientations.

ABSTRACT

Wireless underground sensor networks (WUSNs) can enable many important applications such as intelligent agriculture, pipeline fault diagnosis, mine disaster rescue, concealed border patrol, crude oil exploration, among others. The key challenge to realize WUSNs is the wireless communication in underground environments. Most existing wireless communication systems utilize the dipole antenna to transmit and receive propagating electromagnetic (EM) waves, which do not work well in underground environments due to the very high material absorption loss. The Magnetic Induction (MI) technique provides a promising alternative solution that could address the current problem in underground. Although the MI-based underground communication has been intensively investigated theoretically, to date, seldom effort has been made in developing a testbed for the MI-based underground communication that can validate the theoretical results. In this paper, a testbed of MI-based communication system is designed and implemented in an in-lab underground environment. The testbed realizes and tests not only the original MI mechanism that utilizes single coil but also recent developed techniques that use the MI waveguide and the 3-directional (3D) MI coils. The experiments are conducted in an in-lab underground environment with reconfigurable environmental parameters such as soil composition and water content. This paper provides the principles and guidelines for developing the MI underground communications testbed, which is very complicated and time-consuming due to the new communication mechanism and the new wireless transmission medium.

Motivation & Objective

• To address the critical challenge of high path loss and signal attenuation in underground wireless communication due to soil and water absorption.
• To validate theoretical models of magnetic induction (MI) communication in realistic underground environments with variable soil composition and water content.
• To design and implement a reconfigurable in-lab underground testbed for empirical evaluation of MI-based communication systems.
• To compare the performance of conventional single-coil MI, MI waveguide, and 3D MI coil systems under controlled environmental conditions.
• To provide practical design guidelines and performance benchmarks for future MI-based underground wireless sensor networks (WUSNs).

Proposed method

• Developed a reconfigurable in-lab underground environment using a 255 cm long tank filled with 980,000 cm³ of sand and a dynamic water cycle to control soil water content (VWC).
• Designed and fabricated MI coils for three configurations: single-coil (original MI), MI waveguide (with relay coils), and 3D MI coils (omnidirectional coverage).
• Utilized Universal Software Radio Peripherals (USRP) for signal generation, transmission, and reception, enabling flexible and accurate measurement of communication parameters.
• Implemented a modular testbed architecture integrating signal generation, transceiver units, and environmental control to enable repeatable and scalable experiments.
• Measured key performance metrics including path loss, bandwidth, and packet error rate (PER) across varying communication distances and VWC levels.
• Applied a theoretical channel model to compare experimental results and validate the accuracy of prior MI communication models in real-world conditions.

Experimental results

Research questions

• RQ1 How does magnetic induction (MI) performance vary with soil water content (VWC) in a controlled underground environment?
• RQ2 To what extent does the MI waveguide improve received signal strength and communication range compared to conventional single-coil MI?
• RQ3 Can 3D MI coils achieve robust, omnidirectional communication regardless of angular misalignment between transmitter and receiver?
• RQ4 How do experimental results compare with theoretical predictions of MI channel capacity and path loss in underground media?
• RQ5 What are the practical limitations of current MI coil fabrication (e.g., hand-wired vs. PCB-based) in real-world underground communication systems?

Key findings

• The MI waveguide significantly increased received signal strength compared to the original single-coil MI system, enabling longer communication ranges.
• In high-moisture soil (20% VWC), path loss increased substantially, but the MI waveguide maintained strong signal reception, demonstrating resilience to high absorption.
• The 3D MI coil achieved consistently low packet error rate (PER) across all angular orientations (0° to 180°), eliminating the 90° blind spot observed in conventional 1D MI coils.
• The original MI system exhibited high PER (up to 100%) when coils were orthogonal (90°), especially in wet soil, highlighting the need for 3D coverage.
• Bandwidth remained relatively stable across varying VWC and distances, indicating that signal strength—not bandwidth—was the dominant factor limiting channel capacity.
• The testbed confirmed that MI-based communication is feasible and superior to EM wave-based systems in high-loss underground environments, particularly in wet soils with high electrolyte content.