JoWUA

Abstract

The Internet of Things (IoT) has gained popularity in recent years by connecting physical objects to the Internet, enabling innovative applications. To facilitate communication in low-power and lossy networks (LLNs), the IPv6-based routing protocol for LLNs (RPL) is widely used. However, RPL’s lack of specified security models makes it vulnerable to security threats, particularly sinkhole attacks. Existing sinkhole attack detection techniques suffer from high detection delays and false positives. To overcome these limitations, in our research we propose a multidirectional trust-based detection approach for sinkhole attacks in the RPL routing protocol. Our model introduces a novel architecture that considers trust in parent, child, and neighbor directions, reducing detection delays. We enhance detection efficiency and reduce false positives by combining fuzzy logic systems (FLSs) and subjective logic (SL). Additionally, we introduce a new trust weight variable derived from Shannon's entropy method and multiattribute utility theory. We adaptively adjust the SL coefficient based on network conditions, replacing the constant coefficient value of SL theory. Our approach is compared to the most recent techniques, and we assess different indicators, such as false-positive rate, false-negative rate, packet delivery ratio, throughput, average delay, and energy consumption. Our results demonstrate superior performance in all these metrics, highlighting the effectiveness of our approach.