Self-Powered Cattle Behavior Monitoring System Using 915 MHz Radio Frequency Energy Harvesting

Abstract

The accurate monitoring and management of dairy cattle behavior are critical for improving farm productivity as well as animal welfare and health status. In this paper, we present a self-powered dairy-cattle-behavior monitoring system that harnesses 915 MHz radio-frequency (RF) energy harvesting and bidirectional long short-term memory (Bi-LSTM) networks. The system aims to enable continuous and real-time monitoring of cattle behaviors while eliminating the need for battery replacements. By harvesting RF energy from the surrounding electromagnetic radiation, our system achieves long-term, self-sustainable operation, reducing maintenance efforts and costs. The Bi-LSTM network effectively captures the temporal dependencies and patterns in the collected sensor data, enabling accurate behavior recognition and prediction. Experimental results demonstrate the effectiveness of the proposed system in accurately classifying cattle behaviors, with an overall accuracy of 96.79%. Compared with traditional manual observation methods and battery-dependent systems, our self-powered monitoring system offers enhanced automation, improved welfare monitoring, and increased operational efficiency. The combination of RF energy harvesting, and Bi-LSTM networks affords a promising approach for self-powered and intelligent dairy-cattle-behavior monitoring, facilitating optimized management practices in the dairy industry.

Keywords

• Cattle monitoring system
• radio-frequency energy harvesting
• bi-directional long short-term memory
• one-dimensional convolutional neural network deep learning