Abstract:

With the era of digital transformation, intelligent cities make use of networked systems to maximize urban efficiency, one of which is smart traffic management. As more systems are based on real-time information and wireless communications, they are more vulnerable to cyber attacks. Existing intrusion detection systems are typically inadequate in thoroughly analyzing sequential data or identifying complex attacks in real-time, hence leaving vulnerabilities in intelligent transportation infrastructures. Standard models, especially those that are not temporally aware, do not identify intricate, time-based anomalies that attackers leverage. In order to overcome these challenges, this paper introduces the STAB framework – Smart Traffic Management Systems using Bi-LSTM-Based Anomaly Detection Framework. The presented approach utilizes a Bidirectional Long Short-Term Memory (Bi-LSTM) neural network to inspect bi-directional interdependencies in traffic communication data to enable the identification of subtle and time-variant intrusion patterns. Experimental results on benchmark smart traffic datasets show that STAB possesses higher detection accuracy and lower false positive rates than traditional intrusion detection systems. The framework proves effective in identifying complex attacks, ensuring safer and more robust traffic management in the context of evolving smart city environments.