Classification of ground vehicles based on acoustic signals can be employed effectively in battlefield surveillance, traffic control, and many other applications. The classification performance depends on the selection of signal features that determine the separation of different signal classes. In this paper, we investigate two feature extraction methods for acoustic signals from moving ground vehicles. The first one is based on spectrum distribution and the second one on wavelet packet transform. These two methods are evaluated using metrics such as separability ratio and the correct classification rate. The correct classification rate not only depends on the feature extraction method but also on the type of the classifier. This drives us to evaluate the performance of different classifiers, such K-nearest neighbor algorithm (KNN), and support vector machine (SVM). It is found that, for vehicle sound data, a discrete spectrum based feature extraction method outperforms wavelet packet transform method. Experimental results verify that support vector machine is an efficient classifier for vehicles using acoustic signals.
Different factors affect the process of choosing the appropriate traffic signal controller to solve the traffic conflict on an intersection. Important factors are; number of phases and vehicles arrival rates. Sequence of phases, timings of traffic signals and length of cycle are the most important parameters that all traffic signal controllers aim to optimize one or more of them. One of the major performance measures of traffic signal controller is the average waiting time of vehicles. To compare different kinds of traffic signal controllers, a discrete event simulation model of traffic signal controller on a single intersection is developed using Matlab/Simulink/Simevents. In this paper, three algorithms are proposed to reduce the average waiting time at intersections. The proposed algorithms are compared to the base-line fixed-time controller through extensive simulation experiments. All the proposed algorithms outperforms the base-line algorithm when there is a high variance on the traffic flow. One of the proposed algorithms that adapts both green intervals and cycle length, AW VariableC, outperforms other algorithms, including base-line, under all conditions, but this is on the expense of more computational overhead and more input parameters.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.