Cattle Collars with a Low-Cost LPWAN-Based System for Cattle Estrous Monitoring

This work proposes a prototype of LPWAN cattle collars for monitoring cattle estrus, with the aim of collecting data on cattle behavior to enhance the accuracy of determining cattle estrus. These collars consist of two main components. The first component is the transmitter, responsible for transmitting the cattle's sound data from the sound sensor (LMV324) and motion data from the gyro sensor module (MPU-9250), which are processed by the Seeeduino controller. This processing is done to refine the sensor data signal before transmitting it to the LoRa RFM95 module, which broadcasts it within the 925 MHz frequency bands to the second component. The second component is a receiver, comprising the LoRa RFM95 module and a 5 dB gain antenna, connected to the ESP32. This component serves the additional function of sending data over the internet to Google Sheets cloud. In comparison to the GM1351 standard sound level meter, the sound sensor used in the experiment exhibits an average inaccuracy of 3.84%. The data obtained from the cattle wearing the collars can be utilized to analyze their movement patterns, such as feeding, standing rest, and prone position. Furthermore, the sound data transmission to Google Sheets every five seconds was consistently successful. Consequently, valuable data regarding movements, vocalizations, and estrus behaviors can be derived from the results of the cattle collar test. This research sets the groundwork and offers significant assistance for future advancements aimed at improving the accuracy of determining cattle estrus timing.