David Alexander Ells*, Christopher Mechefske and Yongjun Lai
DOI: 10.37421/2090-4886.2023.12.239
Wireless Sensor Networks (WSNs) can be used for machine condition monitoring to improve performance and safety. However, they present challenges with respect to energy and data management. This paper presents a novel low-power WSN and compares the performance of operating modes and data processing methods for vibration-based machine condition monitoring. The necessary software was developed to perform time and frequency analysis, and a data reduction method was proposed to reduce the data packet size. The performance of the WSN end node was then tested, and its energy consumption was compared for different operating modes. Testing showed that the end node was capable of performing basic vibration analysis. However, contrary to expectations and other reports, results showed that processing data locally to reduce the packet size consumed more energy than transmitting the raw vibration data. While the data packet was effectively reduced by 98.6 percent from 4096 bytes to 56 bytes, results showed that processing data locally consumed 8.8 to 21.4 percent more energy than transmitting the raw data.
DOI: 10.37421/2090-4886.2025.14.315
Spanish 
Chinese 
Russian 
German 
French 
Japanese 
Portuguese 
Hindi 