CNN-CLFFA: Support Mobile Edge Computing in Transportation Cyber Physical System

Abstract

In the present scenario, the transportation Cyber Physical System (CPS) improves the reliability and efficiency of the transportation systems by enhancing the interactions between the physical and cyber systems. With the provision of better storage ability and enhanced computing, cloud computing extends transportation CPS in Mobile Edge Computing (MEC). By inspecting the existing literatures, the cloud computing cannot fulfill the requirements in transportation CPS like lower context-awareness and latency. For enhancing the context-awareness and reducing the latency in a realistic MEC environment, an efficient portable deep learning model: Convolutional Neural Network (CNN) with Chaotic Lévy Flight based Firefly Algorithm (CLFFA) is implemented in this article. In the CNN model, the CLFFA selects the appropriate hyper-parameters or reduces the redundant parameters that results in minimal model size and inference latency than the traditional CNN models. Additionally, the CNN-CLFFA model significantly outperformed the existing models by means of recall, accuracy, F1-score, and precision on the benchmark datasets like German Traffic Sign Recognition Benchmark (GTSRB), MIOvision Traffic Camera Dataset (MIO-TCD) classification, and VCifar-100 datasets. The numerical analysis demonstrates that the CNN-CLFFA model obtained maximum accuracy of 99.02%, 99.11%, and 99.03% on the VCifar-100, MIO-TCD, and GTSRB-T datasets, which are superior to the traditional models.

Keywords

• Convolutional neural network
• cyber physical system
• firefly algorithm
• intelligent transportation system
• mobile edge computing