A comprehensive evaluation of an inertial sensor for High-Intensity shocks and vibration analysis in a real tennis context

The use of inertial sensors (IMUs) in sport performance scenarios has increased in recent years. However, it is necessary to understand the limits of use of these systems and how far they can go in terms of measuring intense shocks and mechanical stimuli, such as impacts in racquet sports. This study evaluates the accelerometer embedded in an IMU for measuring high-intensity shocks and vibration characteristics in a real tennis game context. Two nonprofessional tennis players performed free rallies using a racquet set up with the IMU and a piezoelectric accelerometer used as the reference system. The agreement between the two systems was assessed by comparing shock characteristics in baseline impacts. Linear Regression, Wilcoxon Signed-Rank Test, Intraclass Correlation Coefficient, and Spearman’s correlation were calculated. The presence of systematic errors was found in the IMU, due not only to the lower sampling rate but also to the exceeding of the sensor’s measurement range (IMU saturated for impacts exceeding 200 g). Excluding peak amplitude, in all the other parameters evaluated, highly significant correlations were found for all parameters (ρ = 0.46–0.96; p < 0.001). Linear regression analysis showed slopes ranging from 0.467 (ti) to 1.001 (SC30), indicating variable proportional bias. These findings suggest that, albeit with systematic measurement error, the IMU can be used in tennis contexts to (noninvasively) obtain information indicative of the mechanical shock to which the athlete is subjected.