|
| Literature | Recovery method | Contribution | Limitation |
|
| [10] | Active fault detection, multi-AUV recovery | The proposed algorithm MECHR can adapt to a wide range of harsh scenes and multiconstrained 3D underwater environment. | Not applicable for large-scale damaged networks. |
| [11] | Active fault detection, node movement recovery based on clustering and sleep scheduling | The failure node, coverage vulnerability, coverage matrix, key position, and supplementary node are all considered in the proposed algorithm. | It requires the deployment of a large number of nodes in advance, leading to resource wastage. |
| [12] | Active fault detection, node movement recovery based on cat group optimization | The proposed algorithm CSO can predict a possible articulation point in the network, and it can search both locally and globally. | This method is only suitable for data collection and does not ensure permanent network connectivity. |
| [13] | Active fault detection reduces the destruction of sensors modifying the routing path | The proposed algorithm can prevent the unwanted loss of the data and decrease the damage of sensors. | This method does not detect vulnerabilities and connectivity restoration but only reduces the destruction of sensors by changing the route. |
| [14] | Active fault detection, multi-AUV recovery based on clustering and sleep scheduling | The proposed ReVOHPR algorithm can minimize the energy consumption and void hole avoidance. | This algorithm does not consider the security of data transmission, especially how to ensure low energy consumption and high quality of service under various attacks. |
| [15] | Active fault detection, node movement recovery based on clustering using hidden Poisson Markov model | The proposed HPM model provides an accurate region-based fault detection in the network, and the ANNP model for optimal node selection reduces the recovery time of the network. | This strategy does not address the error control mechanism during packet transmission. |
| [16] | Active fault detection, node movement recovery based on clustering using the Markov chain Monte Carlo process | The proposed 3D UWSN mechanism enhances coverage, connectivity, reliability, and network lifetime through static and mobile sensor deployment. | Not applicable for large-scale damaged networks. |
| [17] | Using node sinking for recovery based on the three-dimensional sphere packaging mode | The algorithm demonstrates superior coverage, network connectivity, and reduced time complexity compared to exhaustive search and peer-to-peer algorithms. | It requires the deployment of many nodes in advance, leading to resource wastage. |
| [18, 19] | Active fault detection, node movement recovery based on clustering using cluster head and candidate cluster head methods | Their algorithms all reduce the energy consumption and extend the network life. | The two algorithms are recovery responses to the failure of a single node. |
| [53] | Active fault detection, node movement recovery based on group nodes using multiobjective emperor penguin optimization | The DHD-MEPO algorithm utilizes group nodes for information management and uses a multiobjective optimization method for selecting repair node. | This method is not considered in real scenarios, and the sensing range of the sensor node is also affected by environmental changes and multiple obstacle obstructions, as well as the fact that its sensing range is variable over time in complex scenarios. |
| [58] | AUV deployment using an improved nondominated sorting genetic algorithm-II | This paper utilizes an NSGA-II metaheuristic approach with multipoint crossover and adaptive mutation for AUV deployment to achieve k-coverage and m-connectivity in UWSNs. | This method is not suitable for heterogeneous underwater cognitive sensor network where AUVs do not also have various sensing and communication capabilities. |
|
