一种基于偏微分方程的车辆加速度信号自适应降噪方法

An Adaptive De-Noising Method for Vehicle's Acceleration Signal Based on PDE

提出一种针对MEMS加速度计信号的基于偏微分方程的自适应降噪方法,该方法不仅能有效克服由于传感器本身原因及车载环境振动噪声带来的影响,获得准确的加速度信号,而且实现容易、实时性好。通过对车辆加速度信号进行建模并叠加真实加速度噪声作为仿真信号,将该方法与选用db6小波基、heursure自适应阈值、4层分解的最佳小波进行降噪性能对比,证明在车辆正常行驶的加速度幅值下,该方法不仅能够取得和小波近似的降噪性能,而且很大程度上减少了运算时间。最后通过对实际车载加速度信号的降噪处理和倾角测量中的应用,再次证明该方法在滤除噪声的同时能够较好体现细节信息,很适合应用在对实时性和准确性要求高的实际工程中。

An adaptive de-noising method was put forward for MEMS accelerometer signal based on partial differential equation. This method can reduce the disturbance caused by vibration of vehicular environment and sensor itself, get acceleration accurately, and have advantages of easy realization and good real-time per- formance. A vehicle ＇s acceleration signal modeling was set up with real acceleration noises as simulation signals. Compared to the best wavelet in such signal with characters of db6 wavelet base,heursure adaptive threshold, and four decomposed layers, the adaptive de-noising method could perform quite same as wavelet in de-noising, but largely reduce the computation time. Experiments proved above result, while the car was in normal driving conditions. And experiments based on real vehicle acceleration signal and angle measurement proved again that this method could reflect the detailed information effectively while filtering noise, and it was very suitable for practical engineering which needs real-time applications and high accuracy.