S-EMG Signal Compression in One-Dimensional and Two-Dimensional Approaches

Area of Science:

• Biomedical Engineering
• Signal Processing
• Data Compression

Background:

• Surface electromyography (S-EMG) signals are crucial for neuromuscular studies but generate large datasets.
• Efficient compression of S-EMG data is essential for storage, transmission, and real-time analysis.
• Existing compression methods may not adequately address the complexity and variability of S-EMG signals.

Purpose of the Study:

• To develop and evaluate novel algorithms for compressing both 1-D and 2-D S-EMG signals.
• To improve the efficiency and effectiveness of S-EMG signal compression techniques.
• To compare the performance of the proposed methods against existing literature benchmarks.

Main Methods:

• A 1-D approach utilizes wavelet transform-based encoding with adaptive bit allocation for vector quantization and entropy coding.
• A 2-D approach segments S-EMG signals, creates a 2-D representation, and employs a high-efficiency video codec with optimized settings.
• Objective metrics and a real signal dataset were used for encoder evaluation.

Main Results:

• The proposed 1-D and 2-D S-EMG compression algorithms were successfully implemented.
• Objective performance metrics demonstrated the effectiveness of both compression strategies.
• The developed methods outperformed other efficient encoders previously reported in the literature.

Conclusions:

• The novel 1-D and 2-D algorithms offer significant advancements in S-EMG signal compression.
• These methods provide efficient solutions for handling large S-EMG datasets.
• The findings suggest potential for improved S-EMG data management in research and clinical applications.