Electric Field Detection System Based on Denoising Algorithm and High-Speed Motion Platform

Qi Liu, Zhaolong Sun, Runxiang Jiang, Jiawei Zhang, Kui ZhuSensors 2022 · DOI

TL;DR

This paper introduces a novel electric field detection system for underwater targets mounted on a high-speed boat. By analyzing and effectively denoising low-frequency electric field noise using an improved ICEEMDAN algorithm, they significantly enhance detection range and accuracy, achieving a 53.1% increase in detection distance.

What Problem Does This Paper Address?

Underwater target detection is crucial for marine monitoring and defense. While autonomous underwater vehicles (AUVs) and other mobile platforms are increasingly used, detecting targets passively remains challenging, especially with the noise generated by the platform itself, particularly at high speeds. Traditional acoustic detection methods can be limited, and electric field detection, while promising due to the low-frequency nature and long transmission distance of electric fields emitted by targets, suffers from significant noise interference from the moving platform. This noise can obscure target signals, leading to detection failures, especially at higher speeds. The paper aims to overcome this by developing a robust denoising strategy for electric field signals collected from a high-speed mobile platform.

Methods

The core of this work lies in developing and applying an advanced denoising algorithm. The researchers first built an electric field detection system mounted on a high-speed boat. They then conducted tests to analyze the characteristics of the low-frequency electric field noise generated by the boat's movement at various speeds.

The key innovation is an improved denoising algorithm based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN). This algorithm was enhanced and combined with an adaptive thresholding technique. Here's a breakdown:

  • ICEEMDAN: This technique decomposes the noisy signal into several intrinsic mode functions (IMFs), which represent different frequency components. ICEEMDAN is known for its ability to handle non-stationary signals and reduce mode mixing issues common in simpler EMD methods.
  • Adaptive Thresholding: After decomposing the signal into IMFs, the researchers analyzed the correlation coefficients between each IMF and the original signal. This analysis helped identify which IMFs contained genuine target information versus noise. An adaptive threshold was then used to screen and select the most relevant IMFs.
  • Signal Reconstruction: Finally, the selected IMFs were recombined to reconstruct a denoised signal, effectively removing the platform-induced noise while preserving the target's electric field signature.

Datasets

The study utilized data collected from an offshore sea test. A simulated radiation source (MMO electrode) was used as the target. The data consisted of electric field measurements from sensors mounted on a high-speed boat moving at various speeds (5, 10, 15, 20, and 25 knots). The tests were conducted in an area with a water depth of approximately 5 meters and a seawater conductivity of 2.8 S/m. The paper does not mention the use of pre-existing benchmark datasets for training or evaluation; the focus is on experimental data collection and analysis.

Key Insights and Ideas

  • Platform Noise is Significant: High-speed boat movement introduces substantial low-frequency electric field noise, particularly affecting sensors at the stern. This noise is non-stationary and can completely mask target signals at higher speeds.
  • ICEEMDAN is Effective for Non-Stationary Noise: The choice of ICEEMDAN proved crucial for decomposing the complex, non-stationary noise signals into meaningful components.
  • Adaptive Thresholding is Key: Simply applying ICEEMDAN wasn't enough. The adaptive thresholding based on correlation coefficients between IMFs and the original signal was essential for distinguishing target signals from noise components.
  • Bow Placement Advantage: Sensors placed at the bow generally experienced lower noise levels, suggesting optimal sensor placement is important for this detection method.
  • Improved Detection Range: The denoising algorithm not only cleans the signal but directly translates to a significant increase in the effective detection range of the electric field.

Key Results

The proposed denoising algorithm demonstrated significant improvements:

  • Enhanced Detection Ratio: The line spectrum detection ratio for the longitudinal component at the bow (Ex1) increased from 0% to 20.9% at 20 knots after denoising.
  • Increased Detection Distance: The maximum detection range was improved from an original 853 m to 1306 m, representing a 53.1% increase.
  • Effective High-Speed Detection: The algorithm successfully enabled target detection even at high speeds (up to 25 knots), where previous methods failed. For example, at 20 knots, the detection distance expanded from 0 m to 821 m for Ex1.
  • Superior to Standard ICEEMDAN: The proposed adaptive algorithm outperformed a standard ICEEMDAN approach, especially at higher speeds, indicating the value of the adaptive thresholding and IMF screening.

Limitations and Future Work

  • Limitations:
    • The study primarily focused on a simulated radiation source. Real-world target signatures might introduce further complexities.
  • While the algorithm improved detection, the paper doesn't explicitly detail the performance on very faint or distant targets in highly noisy environments beyond the tested speeds.
  • The effectiveness of the adaptive threshold might depend on the specific noise characteristics encountered.
  • Future Work:
    • The authors suggest improving the line spectrum detection method to further expand the target detection range.
  • They also propose conducting a rigorous analysis of the system's capacity for detecting unknown targets, implying a need for more generalized detection capabilities.

Summary updated

Abstract

Effective denoising can ensure fast and accurate target detection. This paper presents an electric field measurement system based on a high-speed motion platform, which was built to analyze the characteristics of low frequency electric field noise. An offshore test has shown that it is possible to detect a low-frequency electric field using a high-speed motion platform. Low frequency electric field noise was then collected to analyze its characteristics in terms of time and frequency domains. Based on the noise characteristics, complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) was improved and combined with an adaptive threshold algorithm for denoising and reconstructing target containing noise signals. As revealed in the results, the proposed algorithm achieved highly effective denoising to overcome the line spectrum detection failure resulting from a high-speed motion platform. The detection range had also been improved from the original 853 m to 1306 m, a 53.1% increase.
BibTeX
@inproceedings{liu2022electric,
  author = {Qi Liu and Zhaolong Sun and Runxiang Jiang and Jiawei Zhang and Kui Zhu},
  title = {{Electric Field Detection System Based on Denoising Algorithm and High-Speed Motion Platform}},
  booktitle = {Sensors},
  year = {2022},
  doi = {10.3390/s22145118},
}
Figure