Improving the signal subtle feature extraction performance based on dual improved fractal box dimension eigenvectors

Area of Science:

• Signal Processing
• Fractal Geometry
• Data Analysis

Background:

• Traditional fractal box-counting dimension algorithms struggle with subtle feature extraction in radiation source signals.
• Limitations in existing methods necessitate advanced techniques for accurate signal analysis.

Purpose of the Study:

• To propose a dual improved generalized fractal box-counting dimension eigenvector algorithm for enhanced radiation source signal recognition.
• To overcome the limitations of traditional methods in capturing subtle signal characteristics.

Main Methods:

• Signal preprocessing and Hilbert transform to obtain instantaneous amplitude.
• Extraction of improved fractal box-counting dimension for both instantaneous amplitude and original signal as dual eigenvectors.
• Utilizing grey relation algorithm for radiation source signal recognition with multi-dimensional eigenvectors.

Main Results:

• The proposed dual improved algorithm demonstrates superior extraction of subtle signal distribution characteristics compared to traditional and single improved methods.
• Experimental results indicate a better recognition effect for radiation source signals.
• The algorithm exhibits good real-time performance.

Conclusions:

• The dual improved generalized fractal box-counting dimension eigenvector algorithm effectively captures subtle signal features.
• This advanced algorithm offers improved accuracy and efficiency in radiation source signal recognition.
• The method provides a robust solution for analyzing complex signal distributions.