User agnostic error compensation for magnetic sensor in handheld device

This paper presents a novel method of user agnostic error compensation in MEMS based magnetic sensor for hand-held devices which lags in accuracy due to manufacturing abnormalities and interferences from external magnetic field in its vicinity. Proposed method performs an outlier correction and orientation calculated in a distributed matrix that will adaptively fit the magnetic sensor samples to find the compensation parameters. This method also takes care of automatically triggering error correction mechanism and re-trigger based on the notification when sensor accuracy goes below minimum threshold due to any magnetic anomaly creating scenarios. This method is generic and works for all MEMS based magnetic sensor chips. The dynamic error compensation mechanism presented here has a deviation of 1.2 μT RMSE (Root Mean Square Error) compared to 38.1μT RMSE in existing algorithms, which improves the performance in various applications which uses magnetic sensors.