Modeling Cascading Failures and Invulnerability Analysis of Underwater Acoustic Sensor Networks Based on Complex Network

Due to the frequent deployment of underwater acoustic sensor networks (UASNs) in harsh and unattended environments, they are particularly vulnerable to sensor node failures, energy exhaustion, and malicious attacks. These threats can lead to cascading failures, resulting in the collapse of the entire sensor network. Therefore, appropriate cascading failure models are urgently needed to analyze the network failure process and help design invulnerability networks. In this article, we first introduce the nodes' dynamic distribution and underwater routing protocols to establish a complex network model of UASNs, to represent the characteristics and behaviors of network sensor nodes, dynamic topologies, and routing protocols. Furthermore, we propose the cascading failure model designed specifically for based-depth-routing (BDR) and based-energy-routing (BER) protocol UASNs. In this cascading failure model, we analyze the traffic characteristics of the two types of routing protocol UASNs and establish the corresponding load redistribution scheme. Through extensive simulations and comprehensive multiangle analysis, we have obtained some meaningful results: 1) in some special deployment situations, such as the node deployment distance is 0.9 times the node communication radius, the UASN nodes' dynamics can actually increase the invulnerability of the network; 2) the routing protocols of UASNs can resist small-scale attacks by self-organizing the distribution of network load; and 3) the network node deployed distance is a key factor affecting network invulnerability. These conclusions offer valuable assistance in the design of UASNs that possess enhanced invulnerability.