Reconstruction for Rolling Bearing Vibration Signals Integrating 5G-NR-LDPC Codes and Weighted Compressed Sensing

Accurate reconstruction of vibration signals is essential for effective fault diagnosis of rolling bearings. However, existing methods often struggle to achieve a balance between high compression and effective signal reconstruction. To tackle this challenge, we propose a novel algorithm known as the 5G-WCS algorithm, which integrates 5G New Radio Low-Density Parity-Check Codes (5G-NR-LDPC) with weighted compressed sensing (CS). In this study, a weighted matrix is constructed based on the sparsity coefficients of the signal. This weighted strategy significantly improves compressed sensing's ability to capture critical information. During the signal observation stage, we use the parity-check matrix of the 5G-NR-LDPC code for efficient sampling and compression, leading to effective signal compression and hardware implementation. Simulation results validate the effectiveness of the proposed 5G-WCS algorithm, demonstrating its capability to achieve desirable quality of signal reconstruction while maintaining high compression of rolling bearing vibration signals. This hardware-friendly scheme presents an efficient solution for industrial signal processing and mechanical fault diagnosis, showcasing significant potential for real-world applications.