【摘要】 针对M带小波包在分解过程中存在频率混淆现象,采取节点重排序、单子带重构改进算法等措施,减弱或消除频率混淆现象.将改进的M带小波包分析算法应用于高精度GPS数据序列,分析其误差特性,提取年周期、半年周期、月周期、半月周期等特征信息.通过小波包变换将GPS数据序列噪声和特征信息分离后,进一步提取了误差周期项和特征信息周期项.最后通过实验验证了该方法的有效性.更多还原

【Abstract】 Contraposing the frequency aliasing existing in the algorithm of M-band wavelet packet,measures such as reordering nodes,improved single sub-band reconstruction algorithm were taken to weaken or even eliminate the aliasing.The improved algorithm was used to analyze high-precision GPS data series to obtain the feature of GPS errors,and the feature information with annual cycle,semi-annual cycle,monthly cycle and semi-monthly cycle was extracted based on it.After departing the noise and feature information with wavelet packet transform,error cycle items and feature information cycle items were further extracted separately.The experiment verified the effectiveness of the method.更多还原