Secure Multi-Hop Assisted IoT Communications in Smart Cities
Muhammad Awais Javed, Mohammed Alkhathami, Abdulaziz AlMohimeed, Abeer Almujalli · IEEE Access · 2024
Security and reliability are the major requirements for the successful implementation of Internet of Things (IoT) applications in smart cities such as healthcare. IoT networks are subject to various types of network attacks that can put critical IoT data at risk. To mitigate these security issues, robust security algorithms based on Elliptic Curve Cryptography (ECC) may be utilized. However, using such cryptographic techniques requires additional delay for signing/encrypting and verification/decryption of the messages. Hence, there is a tradeoff between the security strength of the algorithm and packet delay. In this paper, we present a novel technique to adaptively select the key size of ECC-based algorithms for multi-hop IoT networks. The proposed technique uses k-means clustering to classify the secrecy rates of the potential relay nodes into k clusters. We allocate lower key sizes to the relay nodes that have the highest secrecy rates and thus have better physical layer security. The key sizes and security strengths are selected as per the recommendation of the National Institute of Standards and Technology (NIST). The second part of the work utilizes the Hungarian algorithm to select the best relay nodes for each source-destination transmission to minimize the total delay. The techniques are implemented in MATLAB and results show that the total delay of the proposed technique is improved by 57% as compared to other techniques in the literature.