Marine science and Internet of Underwater Things applications rely significantly on collecting data from underwater sensors. Data collection using long-distance underwater acoustic communications consumes a lot of energy in underwater sensor nodes, which are powered by batteries. To achieve low-energy consumption, we can use the autonomous underwater vehicle (AUV) to move close to sensor nodes and exploit the short-range and high-rate communications. Most of the existing AUV-based data collection schemes consider the scenarios having the knowledge of node positions, where the cruising trajectory can be computed before the AUV’s departure. These schemes cannot apply to some scenarios such as turtle tracking for a certain sea area having no position information. To this end, we first propose a planning-while-detecting approach to dynamically detect the sensors on turtles and adjust the AUV cruising direction to collect data. To further improve data efficiency under the energy limit of the AUV, we group the sensors that can share the same trajectory using their detected directions. A grouping-based dynamic trajectory planning (GDTP) is then proposed to determine the next cruising direction that can visit the group of sensors having the largest amount of data and demanding the least cruising energy at the risk of detection errors. Simulation results show that GDTP achieves significantly higher data collection efficiency than the existing trajectory planning algorithms in dynamic scenarios, and as the communication range increases, it can even outperform the existing algorithms with node locations.

Underwater acoustic sensor networks are the efficient way to realize marine applications, which greatly rely on reliable data collection from underwater sensors. For this purpose, we exploit an autonomous underwater vehicle (AUV) as a mobile relay to reduce communication errors, and propose a frame error rate (FER)-based AUV relay algorithm. This algorithm will allow the AUV to avoid visiting the sensors that have sufficiently low FER to reduce cruising journey. For high-FER sensors, the AUV moves to the minimum-FER relay positions to relay their data. By analyzing the solution structure, we transform the FER minimization problem into an equation set to obtain the optimal relay position. The simulation verifies the validity of the equation set. Using the calculated optimal relay position, the FER-based relay achieves a lower average FER of data collection and lower energy consumption than the existing energy-consumption based relay algorithm.

Relaying is an effective method to achieve reliable and timely data collection, which is one of the most important parts of underwater acoustic sensor networks (UASNs). Considering that the relay position determines the reliability of relay communication, we study the problem of relay placement and place an autonomous underwater vehicle (AUV) to mobile relay data transmission and realize the reliable, low-latency, and low-energy data collection in UASNs. First, we formulate the relay positions that can meet a certain frame error rate (FER) requirement as the FER-restricted area (FRA), and approximate FRA with a three-dimensional geometry formula. The problem of reliable and timely data collection becomes planning a short AUV relaying trajectory under the FRA constraint. To this end, we propose a nearest-community (N-C) trajectory planning algorithm to design the AUV relay trajectory. A member grouping method and the necessity of position (NoP) concept are proposed to further reduce the relay positions and trajectory length of the AUV. Simulation results verify that the approximate FRA is more than 90% consistent with the real FRA and show that the N-C using NoP-based member grouping can successfully receive more packets per minute and consume fewer sensors’ energy than other algorithms.

The use of underwater unmanned vehicles has proven to be an efficient method for reliable marine data collection. For this purpose, we utilize an unmanned surface vehicle (USV) as a mobile sink and an autonomous underwater vehicle (AUV) as a mobile relay to reduce communication errors. As communication reliability depends on receiving and relay positions, our study focuses on the placement of mobile vehicles to ensure reliable collaborative communication. For shallow-water sensors, we formulate receiving positions that can meet a specific frame error rate (FER) requirement as the FER-restricted area (FRA), where USV can collect data directly. For deep-water sensors, we use the collaborative relay of USV and AUV to collect data. As mobile receiving and relaying offer more possibilities than fixed, we define all possible receiving and relay positions as the possible FER-restricted area (PFRA) for both USV and AUV, and approximate them with two geometry formulas. Within PFRA, reliable cooperative relaying can be achieved. Simulation results verify that the approximate FRA and PFRA are consistent with real values.

Signal detection is a crucial process for communication systems, and molecular communication (MC) is no exception. Current research on signal detection in MC focuses on either model-based detectors or data-driven detectors. In contrast, our work investigates the integration of both schemes. By incorporating prior knowledge from the model-based scheme, the data-driven neural network structure can be redesigned, significantly mitigating the impacts of inter-symbol interference and inter-link interference in molecular multiple-input multiple-output (MIMO) systems. The feasibility of such an integrated detector is validated through experimental data from a molecular MIMO prototype, where channel state information is no longer indispensable. Moreover, numerical results demonstrate its superiority over conventional schemes in various aspects.

As a privacy-preserving solution, Federated Learning (FL) demonstrates great potential in distributed model training, but limited bandwidth, particularly in NFC-based systems, emerges as a key bottleneck by restricting the number of participating clients. To address this challenge, over-the-air FL leverages the superposition property of wireless multiple-access channels, enabling faster model training and accommodating more clients, even in bandwidth-constrained scenarios like NFC. However, due to its analog-integrated nature, the FL performance is also affected by other factors, such as channel noise. These motivate us to consider how the selected client set and channel noise affect FL performance. To explore this concern, in this paper, we consider an over-the-air FL system with analog gradient aggregation and analyze the impact of the selected client set and channel noise on FL training performance. The theoretical analysis effectively shows the importance of the clients’ number and the power scaling factor to the FL training performance. Based on the theoretical analysis, we transform the global optimization problem into the client selection problem and propose a dynamic client selection scheme to optimize the training performance under the aggregation error constraint. Experimental results demonstrate that our proposed scheme can boost FL by speeding up the convergence of the global model (at least 35%) and saving energy consumption.