Practical adaptive security for resource-constrained IoT nodes

Abstract

Considered a key enabler on the Internet of Things (IoTs), Wireless Sensor Networks(WSNs) allow data to be collected from the surrounding environment. However, protecting the messages being transferred between these nodes is a complex issue. One of the reasons for security complexity in WSNs is resource limitations, such as energy and processing capability. Wireless nodes are usually powered by small batteries, which require reduced energy consumption so as to extend network lifetime. However, enabling security increases energy consumption due to the extra computation and transmission, required for encryption and authentication. Another reason is that the WSNs medium for transmission; and unattended operation make them vulnerable to various types of attack. The contradiction between the need for security and limited resources creates a trade-off between security and energy. There are several solutions to secure WSNs in the literature; however, few consider both efficiency and security. This research first evaluates the security overhead on IoT nodes. Then, it proposes a Practical Adaptive SEcurity architecture for Resource-constrained nodes (PASER). PASER uses the application preferences to select security level. Also, it contains a table for threat level, so that off the shelf threat detection systems can be easily used with PASER to feed that table and enhance the security decision. The per-packet evaluation shows that security overhead in terms of energy consumption fluctuates between 31.5% at a minimum level over non-secure packets and 60.4% at the top security level of IEEE802.15.4 specification. The evaluation is carried out using simulation and then validated using actual hardware. According to the evaluated scenario, the proposed architecture shows an energy saving of 11% compared to the static-security model. The results may vary with different scenarios according to the application requirements and threat level in the surrounding environment. Also, the proposed architecture shows an improvement in network throughput, lifetime by 25% and delay by almost 141%.

More information

Identifiers

ORCID for Nicholas Harris: orcid.org/0000-0003-4122-2219

Catalogue record

Export record