Iron Tunnel Recognition Using Statistical Characteristics of Received Signals in Automotive Radar Systems

In this paper, we propose an iron tunnel recognition method using statistical characteristics of received signals in automotive radar systems. Iron tunnels have periodic steel-framed structures, which generate unwanted reflected signals called radar clutter. This clutter degrades the detection performance of the automotive radar and leads to misdetection of targets. To mitigate the adverse effect of the clutter, an efficient method to recognize iron tunnels using an automotive radar must be established in advance. Thus, we suggest an effective iron tunnel recognition method based on the concept that the statistical characteristics of signals received in iron tunnels differ from those on normal roads. Then, on the basis of these statistics, we use the support vector machine to distinguish iron tunnels from normal roads.