基于阵列信号和小波降噪理论的路面平整度检测系统及其算法研究

【作者】 王淑；

【导师】 郭元术；

【作者基本信息】 长安大学 ， 交通信息工程及控制， 2003， 硕士

【摘要】 本论文将阵列信号处理理论和小波降噪理论运用到路面激光传感器检测系统，较准确的恢复出路面的平整度信息。文中详细介绍了多传感器阵列信号处理模型、小波降噪理论、路面平整度检测方法和评定指标、激光平整度检测系统及其检测原理、基于阵列和小波处理的路面平整度检测算法和应用。提出用阵列信号源的估计来判断激光检测系统硬件性能；用接收信号波达角的估计得出阵列信号最佳权向量；用权向量的加权求和获取期望信号的最大输出功率，同时基于David L.Donoho软阈值理论，进行多层小波降噪，重构原始路面信号。实验结论：基于阵列信号处理和小波降噪的算法应用于路面平整度检测可以得到较理想的断面曲线，换算国际平整度指标IRI与标定较吻合，符合处理要求。更多还原

【Abstract】 In this paper, a multi-functions laser detection system for pavement roughness is introduced. An accurate pavement roughness information is acquired with theory of array signals processing and de-noising by wavelet in this system. In detail, the model of multi-sensors array signals processing, the theory of de-noising by wavelet, the method of road roughness detection and evaluation indexes, multi-function laser-detection system and its principle are introduced in this paper. Further more, the really application to pavement roughness detection based on array signals processing and wavelet theory is analyzed too. In this paper, a method to evaluate hardware performance of laser detection system with the array signal estimation is proposed. And the optimally weight vector of array signal can be acquired by the DOA estimating of received signals. In order to acquire the maximum output power of expected signals, summation of weight vector is applied. Based on the based on David L.Donoho soft-thresholding theory, the roughness information is re-constructed with multi-levels decomposition and de-noising of wavelet. Experiment results prove that this algorithm is a practical and effective method, with which the ideal cross-section curve can be acquired. The result is coincidence to the standard IRI (the International Roughness Index) and meets the necessity.更多还原