小波变换在行驶工况构建用车速数据处理中的应用研究

在以车载测试技术为主要手段的车辆实际道路行驶工况构建过程中，车速这一行驶数据具有数据量大、采样频率高等特点，需要借助计算机来完成相应的处理工作。原始车速数据中包含着大量的噪声和由于采样率高于所需精度产生的冗余，对信号进行一维离散小波变换，得到信号的高频和低频系数，并利用小波分层阈值降噪和小波分解域量化压缩实现信号的降噪与压缩。文中以采样周期7天的一段ECU车速数据为例阐述了一维离散小波变换在车速数据处理过程中的应用，结果表明采用该方法对车速数据进行降噪压缩具有较好的效果，可以增强短片段中速度、加速度等特征参数的差异性，改善聚类效果。证明一维离散小波变换在识别去除车速信号中噪声和冗余的有效性和正确性。

Study on Application of Vehicle Speed Data Processing in Driving Cycle Construction with Wavelet Transform

In the on-board measurement technology as the main means of the process of constructing a driving cycle, vehicle speed as a driving data need to be solved by computer because of its large amount of data and high sampling frequency. The original speed data contains a lot of noise and sampling rate is higher than the required accuracy of the resulting redundancy, one-dimensional discrete wavelet transform is used to obtain the high frequency and low frequency coefficients, the noise reduction and compression of the signal are realized by using the wavelet gradient threshold denoising and the wavelet domain quantization compression. In this paper, the application of one-dimension discrete wavelet transform in the vehicle speed data processing is illustrated with the example of ECU speed data of sampling period 7 days, the results show that the method has a good effect on noise reduction and compression of vehicle speed data, and can improve the difference of velocity, acceleration and other parameters of micro-trips, and improve the clustering effect. The validity and correctness of the noise and redundancy in the vehicle speed signal identification with one-dimensional discrete wavelet transform are verified.