交通流状态辨识小波算法研究

针对智能交通系统的开发和交通流特性，应用小波多分辨分析理论的Mallat分解算法建立交通流状态辨识方法，利用多种小波系数与交通流参数之问的相应变化规律进行交通突变状态的辨识．交通流状态的突变多与交通事件直接相关，故采用事件和非事件条件下的模拟数据对算法参数进行了标定及离线测试．将算法与几种传统算法分别进行了性能比较，结果表明Mallat分解算法在交通流突变状态实时辨识方面具有很好的性能．

Study on Traffic Flow Condition Identification Using Wavelet Method

Traffic flow condition identification is one of important issues for ITS research, especially for Advanced Traffic Management System and Advanced Traveler Information System research. To develop Intelligent Transportation System, a traffic flow condition identification method based on fast Mallat algorithm of wavelet multiresolution analysis is presented. Utilizing the association between the wavelet coefficients and traffic flow, the condition of traffic flow can be extracted directly from the ap- proximate coefficients and detail coefficients of wavelet decomposition of traffic flow parameters. Traffic flow breakdown usually results from traffic incidents. Using data obtained froin the simulation under the condition of incident and non-incident, parameters of the algorithm are calibrated and an off-line test is made. The results show that the algorithm performs better than other algorithms in traffic flow breakdown identification in real time.