Object: Using the sensor to monitor and to evaluate sports performance is a new trend in the future. The experimental results will be affected by the sampling frequency, placement of sensors, setting of the filter and instrumental correction, etc. This study is to explore the effect of different detection of position on swimming stroke. Methods: Three swimmers are participating in this research. Two inertial sensors will be placed on the right hand and forearm, upper and lower back and swimmers have to do a 50 m freestyle. We have developed a system to read the data from sensors and to measure differences between the two placement locations. Results : (a) Peak value: the sensor data from hand higher than the forearm in freestyle x-axis and z-axis. The peak of the upper back X-axis was significantly higher than the lower back (p <.05) (b) Wave range: the sensor data from hand larger than the forearm in XYZ axis. (c) Wave periods: there are no significant differences between hand-forearm and upper-lower back locations. Conclusions: The sensor data obtained from hand location show a larger amplitude and peak value, and better waveform identification. It is important to place the sensor device on the hand for analyzing swimming stroke motion. Freestyle maximum peak is to show in the z-axis. So the relevant axis should choose carefully when analyzing the waveform data. The back location can provide balance and comfort, the data of upper back had a better effect on recognition, but there may have problems of unstable device placement and water resistance; the lower back sensor was close to body center, so it can examine body acceleration, speed, rotation, and frequency.
Proceedings of The 2017 IEEE International Conference on Applied System Innovation