Research on robustness of MEMS-based wearable sensors

Wearable electronics have been fashionable in recent years due to the rapid development of Micro-electromechanical Systems (MEMS) technology. Accelerometer, gyroscope and magnetometer are the most popular MEMS-based sensors, and they are widely available in smartphones, tablets, handheld devices and wearable products. However, they are not robust enough for wearable applications. The magnetometer is easily suffered from perturbation because of magnetic disturbance caused by surrounding hard-iron and soft-iron sources from indoor environments. The gyroscope is limited by various error sources that affect long-term and short-term performance, such as bias instability, scale factor and the angle random walk respectively. In this paper, we analyze the influence of surrounding magnetic disturbance on magnetometer output, and the gyroscope output drifting with time. Then, a method is proposed to rapidly calibrate the magnetometer and eliminate error of gyroscope. The Kalman filter algorithm is used during data fusing. In the final, the experiments are implemented using our proposed method and the results show that the accuracy has improved greatly