Orthogonal time frequency space (OTFS) is a novel two-dimensional (2D) modulation technique that provides reliable communications over time- and frequency-selective channels. In underwater acoustic (UWA) channel, the multi-path delay and Doppler shift are several magnitudes larger than wireless radio communication, which will cause severe time- and frequency-selective fading. The receiver has to recover the distorted OTFS signal with inter-symbol interference (ISI) and inter-carrier interference (ICI). The conventional UWA OTFS receivers perform channel estimation explicitly and equalization to detect transmitted symbols, which requires prior knowledge of the system. This paper proposes a deep learning-based signal detection method for UWA OTFS communication, in which the deep neural network can recover the received symbols after sufficient training. In particular, it cascades a convolutional neural network (CNN) with skip connections (SC) and a bidirectional long short-term memory (BiLSTM) network to perform signal recovery. The proposed method extracts feature information from received OTFS signal sequences and trains the neural network for signal detection. The numerical results demonstrate that the SC-CNN-BiLSTM-based OTFS detection method performs with a lower bit error rate (BER) than the 2D-CNN, FC-DNN, and conventional signal detection methods.

Underwater acoustic (UWA) adaptive modulation (AM) requires feedback about channel state information (CSI) but the long propagation delays and time-varying features of UWA channels can cause the CSI feedback to be outdated. When the AM mode is selected by outdated CSI, the mismatch between the outdated CSI and the actual CSI during transmission degrades the performance and can even lead to communication failure. Reinforcement learning has the ability to learn the relationships between adaptive systems and the environment. This paper proposes a deep Q-network (DQN)-based AM method for UWA communication that uses a series of outdated CSI as the system input. Our study showed that it could extract channel information and select appropriate modulation modes in the expected channels more effectively than single Q-learning (QL) without needing a deep neural network structure. Furthermore, to mitigate any decision bias that was caused by partial observations of UWA channels, we improved the DQN-based AM by integrating a long short-term memory (LSTM) neural network, named LSTM-DQN-AM. The proposed scheme could enhance the DQN’s ability to remember and process historical input channel information, thus strengthening its relationship mapping ability for state-action pairs and rewards. The pool and sea experimental results demonstrated that the proposed LSTM-DQN-AM outperformed DQN-, QL- and threshold-based AM methods.

In this paper, we consider a practical underwater data collection network, where one destination needs to collect data from multiple underwater nodes. With the conventional automatic-repeat-request (ARQ) protocol, the destination requests retransmission from each node individually without any node cooperation. We propose two protocols, selective relay cooperation and dynamic network coded cooperation, utilizing the fact that underwater nodes can overhear the transmission of the others. In the selective relay cooperation, one node can be selected as a relay to transmit the data from another undecoded node in the retransmission phase. In the network coded cooperation, the selected relay nodes transmit network coded packets to the destination. The relay nodes participating the cooperation are selected by the destination based on the channel quality, as measured by the effective signal-to-noise ratio. In addition to simulation results, we have carried out several lake tests based on a full protocol implementation. Simulation and field testing results demonstrate that the proposed schemes can gain significant performance improvement compared with the conventional ARQ scheme.

Energy-efficient computation is an inevitable trend for unmanned aerial vehicles (UAVs)-enabled wireless powered mobile edge computing (MEC), while it has not been investigated when the hybrid passive and active transmissions (ATs) are considered for Internet of Things (IoT) nodes’ task offloading. In this article, we study the computation energy efficiency (EE) fairness among IoT nodes in a UAV-enabled wireless powered MEC network with hybrid passive and ATs, where the UAV serves as a dynamic energy source to support IoT nodes for backscatter communication (BackCom) and AT. Specifically, we formulate an optimization problem to maximize the computation EE of the worst IoT node by jointly optimizing the UAV’s transmit power and trajectory, the IoT nodes’ BackCom time and reflection coefficients, the IoT nodes’ AT power and time, as well as the IoT nodes’ local computing time and frequencies. The formulated problem is highly nonconvex and difficult to be solved optimally. To address it, we first obtain the closed-form expressions for the UAV’s transmit power and the IoT nodes’ local computing time by means of the proof by contradiction to simplify the problem, and then propose a Dinkelbach-based iterative algorithm to obtain the solution of other optimization variables. Specifically, based on the Dinkelbach’s method, the original fractional problem is transformed into the problem with the subtractive objective function. Then we further decouple the transformed problem into two subproblems based on the block-coordinated-decent (BCD) method and solve the transformed problem by the proposed BCD-based iterative algorithm, where the above two subproblems are solved by means of the existing convex optimization tools and the proposed successive convex approximation (SCA)-based iterative algorithm alternatively. Simulation results show that the proposed algorithms have a fast rate of convergence and that the proposed scheme outperforms other baseline schemes in terms of the computation EE fairness.

In the complex and dynamically varying underwater acoustic (UWA) channel, cooperative communication can improve throughput for UWA sensor networks. In this paper, we design a reasonable relay selection strategy for efficient cooperation with reinforcement learning (RL), considering the characteristics of UWA channel variation and long transmission delay. The proposed scheme establishes effective state and reward expression to better reveal the relationship between RL and UWA environment. Meanwhile, simulated annealing (SA) algorithm is integrated with RL to improve the performance of relay selection, where exploration rate of RL is dynamically adapted by SA optimization through the temperature decline rate. Furthermore, the fast reinforcement learning (FRL) strategy with pre-training process is proposed for practical UWA network implementation. The whole proposed SA-FRL scheme has been evaluated by both simulation and experimental data. The simulation and experimental results show that the proposed relay selection scheme can converge more quickly than classical RL and random selection with the increase of the number of iterations. The reward, access delay and data rate of SA-FRL can converge at the highest value and are close to the ideal optimum value. All in all, the proposed SA-FRL relay selection scheme can improve the communication efficiency through the selection of the relay nodes with high link quality and low access delay.