Multi-Point Co-Ordination Scheme for Traffic Control Systems

Abstract

Traffic management is one of the severe setbacks in metropolitan life style and is acutely realized in all major cities due to different factors such as overpopulation, exploitation, ecological contamination and so on. This is mainly due to the fact that the rate at which the change in infrastructure observed is drastically slower than that of the increase in the number of vehicles owing to constraints related to the space and cost. Certain research works conducted on management of traffic using Traffic Classification Approach (TCA) improved the classification performance by including the correlated information to the classification process. But using correlated information though classification performance was improved, the flow correlation analysis was not performed in an effective manner resulting in increase of the time taken to classify. With the increasing traffic patterns, high degree of flexibility and dynamic traffic Failure Independent Path Protecting (FIPP p-cycles) provided provisioning for dynamic protection but compromising the QoS. With the introduction of additional traffic management instruments, congestions arise whenever different measures of management of traffic are applied in the similar area. In this paper a multi-proxy co-ordination mechanism is designed for effective traffic management using fuzzy correlation analysis (MPFCA). A multi-proxy based traffic junction points is proposed to control the overall framework for intellectual transportation systems. Multi-proxy Co-ordination is formed by analyzing the nearest and adjacent traffic junction points to generate a coordinated traffic signal controlling across the region. A fuzzy correlation analysis is introduced in MPFCA in order to obtain the property of traffic intelligence to the system to minimize the time taken to classify. By representing different measures as intellectual proxies, the measures of the individual mechanisms are synchronized. By synchronizing the neighboring traffic junction points, the process performs superior as a whole to improve the QoS. Experimental evaluation is made with multiple image data scenes obtained from the traffic signals of specific region and also with the benchmark training vehicle image data extracted from the machine learning repositories. The performance of the proposed multi-proxy co-ordination using fuzzy correlation analysis (MPFCA) is measured in terms of execution time.