Explainable Deep Learning-Enabled Malware Attack Detection for IoT-Enabled Intelligent Transportation Systems

The Internet of Things (IoT) has the potential to improve the complementary of communication, control, and information processing within the public transportation system. The IoT-enabled Intelligent Transportation System (ITS) ensures that automated transportation is networked and operated collaboratively. The IoT-enabled ITS has revolutionized the transportation industry by enabling the seamless integration of a wide range of devices and systems. It makes the strategic use of networked devices, sensors, and data analytics to improve transportation network efficiency, safety, and environmental friendliness. The usage of the IoT in the ITS has grown in popularity due to its capacity to improve traffic control, reduce congestion, facilitate live monitoring, and optimize transportation operations. The IoT-enabled ITS systems and devices must be protected from cyber-attacks for various reasons, including preserving sensitive data, guaranteeing privacy, preventing unauthorized access, and protecting against the risk of interruptions or manipulations. Malware attacks affect the working and performance of the deployed smart IoT devices. We propose a secure deep learning-enabled malware attack detection for IoT-enabled ITS (in short, SDLMA-IITS). The approach of explainable artificial intelligence (XAI) has been utilized for the effective detection of malware. A deep security analysis of the proposed SDLMA-IITS is presented to prove its security against various potential attacks. The comparative performance analysis of SDLMA-IITS is given with the other similar existing schemes. Finally, a practical implementation of SDLMA-IITS is provided to measure its impact on the security of the IoT-enabled ITS systems and devices.