Enhanced Adaptive Security Algorithm (EASA) for optimized performance in smart city networks

Abstract

The Enhanced Adaptive Security Algorithm (EASA) is crafted to bolster the robustness of smart city network security, specifically targeting the dynamic and complex nature of these networks. Its primary objective revolves around enhancing the adaptability of network security, focusing particularly on traffic management applications within smart cities. EASA emerges from the foundation of the Adaptive Multi-Layer Security Framework (AMLSF), integrating advanced deep learning techniques and leveraging optimized encryption for a more adaptive and efficient solution. In addressing the limitations of traditional security solutions, EASA exhibits superior performance in real-time responsiveness and efficient encryption compared to existing models, such as AMLSF. A comprehensive evaluation of EASA’s performance metrics reveals an adaptability rate of approximately 90%, underscoring its efficacy in adapting to varying network conditions and threats. The integration of machine learning algorithms in AMLSF, a pivotal aspect of EASA, facilitates dynamic security adaptation, crucial for real-time responsiveness and robust encryption in smart city networks. EASA’s advanced use of deep learning techniques and efficient data processing capabilities effectively complement and enhance the overall network security, addressing scalability issues and adding layers of security, especially in IoT environments within smart cities. Performance metrics such as threat detection accuracy (TDA), encryption efficiency (EE), key generation efficiency (KGE), adaptive response time (ART), system overhead score (SOS), and overall security efficiency (OSE) are employed to evaluate EASA. These metrics collectively reflect the algorithm’s ability to detect true threats, efficiently encrypt data, generate keys swiftly, respond adaptively to changes, manage system resources effectively, and provide an overall efficient security solution. EASA demonstrates impressive performance metrics, with an accuracy of 92%, precision of 91%, recall of 90%, and an F1 score of 90.5%, indicating its superior capability in smart city network security compared to AMLSF and CNN. This robust performance, coupled with its adaptability and efficiency, positions EASA as a promising solution for next-generation smart city security frameworks, advocating for user privacy and ethical data handling while encouraging collaborative efforts for continuous refinement.