Search

The security landscape of Internet of Things (IoT) networks continues to draw significant attention from recent networks; the Version Number Attack (VNA) stands out as a prominent threat to IoT network resources. Routing Protocol for Low-Power and Lossy Network structures low-power and lossy networks through one or more Destination Oriented Directed Acyclic Graphs (DODAGs). Each DODAG is assigned a version number, serving the purpose of ensuring the existence of loop-free paths to the root node. In the event of any inconsistency detected by the root node, the version number is incremented, triggering a global repair process and the reconstruction of the Directed Acyclic Graph. A potential security concern arises when a malicious node advertises a false version number in its control message, compelling the initiation of a global repair process. This study conducts a comprehensive analysis of the VNA’s impact, considering various parameters that have not been explored in previous research articles. These parameters include an examination of the additional forwarding tasks introduced by the attack and an exploration of factors contributing to node hyperactivity (such as loops, local and global repairs, etc.). The article delves into the specific control messages that result in the detrimental effects of the attack, offering a detailed analysis of control overhead, energy consumption, Packet Delivery Ratio, and latency. The obtained results show that the DAO (Destination Advertisement Object) messages dominate the total control overhead and are the main contributor to the total consumer energy, where the closest the malicious node to the sink the more damaging is the attack.

Node localization is critical for accessing diverse nodes that provide services in remote places. Single-anchor localization techniques suffer co-linearity, performing poorly. The reliable multiple anchor node selection method is computationally intensive and requires a lot of processing power and time to identify suitable anchor nodes. Node localization in wireless sensor networks (WSNs) is challenging due to the number and placement of anchors, as well as their communication capabilities. These senor nodes possess limited energy resources, which is a big concern in localization. In addition to convention optimization in WSNs, researchers have employed nature-inspired algorithms to localize unknown nodes in WSN. However, these methods take longer, require lots of processing power, and have higher localization error, with a greater number of beacon nodes and sensitivity to parameter selection affecting localization. This research employed a nature-inspired crow search algorithm (an improvement over other nature-inspired algorithms) for selecting the suitable number of anchor nodes from the population, reducing errors in localizing unknown nodes. Additionally, the weighted centroid method was proposed for identifying the exact location of an unknown node. This made the crow search weighted centroid localization (CS-WCL) algorithm a more trustworthy and efficient method for node localization in WSNs, with reduced average localization error (ALE) and energy consumption. CS-WCL outperformed WCL and distance vector (DV)-Hop, with a reduced ALE of 15% (from 32%) and varying communication radii from 20 m to 45 m. Also, the ALE against scalability was validated for CS-WCL against WCL and DV-Hop for a varying number of beacon nodes (from 3 to 2), reducing ALE to 2.59% (from 28.75%). Lastly, CS-WCL resulted in reduced energy consumption (from 120 mJ to 45 mJ) for varying network nodes from 30 to 300 against WCL and DV-Hop. Thus, CS-WCL outperformed other nature-inspired algorithms in node localization. These have been validated using MATLAB 2022b. [ABSTRACT FROM AUTHOR]