International Peer-Reviewed JournalOpen AccessISSN 2456-8880
irejournals@gmail.com+91-7433024337

Home / Current Issue / Paper 1711914

1711914PublishedVol 9 · Issue 5

Machine Learning Based Classification of Electromagnetic Interference Using Synthetic Signal Features

Shanija S R Gopika K Sittalatchoumy R

Subject area: Science,Engineering and Technology  ·  Area of research: Electromagnetic Interference

DOI: https://doi.org/10.64388/IREV9I5-1711914

Abstract

Electromagnetic interference (EMI) poses significant challenges in modern electronic and communication systems by degrading signal integrity and overall system performance. This project presents a machine learning-based classification approach for identifying and distinguishing between different types of EMI using synthetic signal features. A dataset containing 600 samples was created, consisting of 100 clean signals and 500 EMI-contaminated signals generated from five sources like motor, switching, lighting, Wi-Fi, and inverter. Key signal features such as mean, variance, root mean square (RMS), dominant frequency, first harmonic, and second harmonic were extracted to characterize each signal. Two machine learning algorithms such as Decision Tree and SVM were trained and tested using an 80:20 ratio for training and testing, respectively. The classification performance was evaluated using confusion matrices and classification reports. The Decision Tree model achieved an accuracy of 58%, while the SVM attained 98% accuracy, demonstrating its robustness and superior generalization ability. Additionally, a user-interactive interface was developed with a drop-down menu enabling users to select a signal type and obtain real-time classification results. The proposed system provides an efficient and automated method for EMI source identification, which can aid in EMI mitigation and system reliability enhancement.

How to cite this paper

Shanija S R, Gopika K, Sittalatchoumy R "Machine Learning Based Classification of Electromagnetic Interference Using Synthetic Signal Features" Iconic Research And Engineering Journals Volume 9 Issue 5 2025 Page 608-617 https://doi.org/10.64388/IREV9I5-1711914
Shanija S R, Gopika K, Sittalatchoumy R "Machine Learning Based Classification of Electromagnetic Interference Using Synthetic Signal Features" Iconic Research And Engineering Journals, vol. 9, no. 5, Nov. 2025, doi: https://doi.org/10.64388/IREV9I5-1711914
Shanija S R, Gopika K, Sittalatchoumy R (2025). Machine Learning Based Classification of Electromagnetic Interference Using Synthetic Signal Features. Iconic Research And Engineering Journals, 9(5). doi: https://doi.org/10.64388/IREV9I5-1711914
Shanija S R, Gopika K, Sittalatchoumy R "Machine Learning Based Classification of Electromagnetic Interference Using Synthetic Signal Features" Iconic Research And Engineering Journals, vol. 9, no. 5, Nov. 2025. Crossref, https://doi.org/10.64388/IREV9I5-1711914
@article{1711914,
      author = {Shanija S R, Gopika K, Sittalatchoumy R},
      title = {Machine Learning Based Classification of Electromagnetic Interference Using Synthetic Signal Features},
      journal = {Iconic Research And Engineering Journals},
      year = {2025},
      volume = {9},
      number = {5},
      pages = {608-617},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1711914.pdf},
      abstract = {Electromagnetic interference (EMI) poses significant challenges in modern electronic and communication systems by degrading signal integrity and overall system performance. This project presents a machine learning-based classification approach for identifying and distinguishing between different types of EMI using synthetic signal features. A dataset containing 600 samples was created, consisting of 100 clean signals and 500 EMI-contaminated signals generated from five sources like motor, switching, lighting, Wi-Fi, and inverter. Key signal features such as mean, variance, root mean square (RMS), dominant frequency, first harmonic, and second harmonic were extracted to characterize each signal. Two machine learning algorithms such as Decision Tree and SVM were trained and tested using an 80:20 ratio for training and testing, respectively. The classification performance was evaluated using confusion matrices and classification reports. The Decision Tree model achieved an accuracy of 58%, while the SVM attained 98% accuracy, demonstrating its robustness and superior generalization ability. Additionally, a user-interactive interface was developed with a drop-down menu enabling users to select a signal type and obtain real-time classification results. The proposed system provides an efficient and automated method for EMI source identification, which can aid in EMI mitigation and system reliability enhancement.},
      month = {November},
      doi = {https://doi.org/10.64388/IREV9I5-1711914}
  }