Designing an efficient deployment method to guarantee optimal monitoring quality is one of the key topics in underwater sensor networks. At present, a realistic approach of deployment involves adjusting the depths of nodes in water. One of the typical algorithms used in such process is the self-deployment depth adjustment algorithm (SDDA). This algorithm mainly focuses on maximizing network coverage by constantly adjusting node depths to reduce coverage overlaps between two neighboring nodes, and thus, achieves good performance. However, the connectivity performance of SDDA is irresolute. In this paper, we propose a depth adjustment algorithm based on connected tree (CTDA). In CTDA, the sink node is used as the first root node to start building a connected tree. Finally, the network can be organized as a forest to maintain network connectivity. Coverage overlaps between the parent node and the child node are then reduced within each sub-tree to optimize coverage. The hierarchical strategy is used to adjust the distance between the parent node and the child node to reduce node movement. Furthermore, the silent mode is adopted to reduce communication cost. Simulations show that compared with SDDA, CTDA can achieve high connectivity with various communication ranges and different numbers of nodes. Moreover, it can realize coverage as high as that of SDDA with various sensing ranges and numbers of nodes but with less energy consumption. Simulations under sparse environments show that the connectivity and energy consumption performances of CTDA are considerably better than those of SDDA. Meanwhile, the connectivity and coverage performances of CTDA are close to those depth adjustment algorithms base on connected dominating set (CDA), which is an algorithm similar to CTDA. However, the energy consumption of CTDA is less than that of CDA, particularly in sparse underwater environments.

With the increasment of epilepsy patients, traditional epileptic seizure recognition is generally completed by encephalography (EEG) technicians, which is time-consuming and labor-intensive, so the automatic detection of seizure is imminent. This paper proposes a method which constructs a multi-layer network and extracts the same features in each network optimized by improved genetic algorithm (IGA). Among them, the multi-layer network refers to the three-layer network constructed by pearson correlation coefficient, mutual information and permutation disalignment index respectively. There is a lack of research on the fusion and comparison of different networks in previous studies. Therefore, this paper analyzes the effectiveness of different networks by studying the fusion relationship of different networks, and further uses IGA for iterative optimization with constraints to weight the network and features, and finally uses the random forest classifier to automatically detect epileptic seizures. On CHB-MIT database, accuracy (ACC), specificity (SPE), sensitivity (SEN) and F1 score (F1) of the method proposed in this paper reach 97.26%, 97.55%, 96.89% and 97.11%, respectively. On Siena scalp database, ACC, SPE, SEN and F1 reach 98.88%, 99.13%, 98.36% and 98.75%, respectively. The results show that the joint detection effect of the multi-layer network is better than the combined effect of other networks, and IGA can improve the effect of seizure detection.

Considering that deployment strategies for underwater sensor networks should contribute to fully connecting the networks, a Guaranteed Full Connectivity Node Deployment (GFCND) algorithm is proposed in this study. The GFCND algorithm attempts to deploy the coverage nodes according to the greedy iterative strategy, after which the connectivity nodes are used to improve network connectivity and fully connect the whole network. Furthermore, a Location Dispatch Based on Command Nodes (LDBCN) algorithm is proposed, which accomplishes the location adjustment of the common nodes with the help of the SINK node and the command nodes. The command nodes then dispatch the common nodes. Simulation results show that the GFCND algorithm achieves a comparatively large coverage percentage and a fully connected network; furthermore, the LDBCN algorithm helps the common nodes preserve more total energy when they reach their destination locations.